Engraftment of Embryonic Stem Cells and Differentiated Progeny by Host Conditioning with Total Lymphoid Irradiation and Regulatory T Cells

Report

Engraftment of Embryonic Stem Cells and Differentiated Progeny by Host Conditioning with Total Lymphoid Irradiation and Regulatory T Cells Graphical Abstract

Authors Yuqiong Pan, Dennis B. Leveson-Gower, ..., Joseph C. Wu, Robert S. Negrin

Correspondence [email protected]

In Brief ESCs hold tremendous promise to treat serious medical conditions; however, their utility is limited by immune rejection after allogeneic transplantation. Pan et al. find that host conditioning with either TLI and ATS or TLI plus Treg can promote the engraftment of ESCs. Their findings provide clinically translatable strategies for inducing tolerance to adoptively transferred ESCs for cell replacement therapy.

Highlights d

ESCs and differentiated progeny are readily rejected in immunocompetent animals

d

TLI, ATS, and Treg are used to induce engraftment of allogeneic ESC transplantation

d

TLI and ATS promote the engraftment of ESC and differentiated progeny allografts

d

TLI plus Treg promote the engraftment of ESC and differentiated progeny allografts

Pan et al., 2015, Cell Reports 10, 1793–1802 March 24, 2015 ª2015 The Authors http://dx.doi.org/10.1016/j.celrep.2015.02.050

Cell Reports

Report Engraftment of Embryonic Stem Cells and Differentiated Progeny by Host Conditioning with Total Lymphoid Irradiation and Regulatory T Cells Yuqiong Pan,1 Dennis B. Leveson-Gower,1 Patricia E. de Almeida,2 Antonio Pierini,1 Jeanette Baker,1 Mareike Florek,1 Hidekazu Nishikii,1 Byung-Su Kim,1 Rong Ke,4 Joseph C. Wu,2,3 and Robert S. Negrin1,* 1Division

of Blood and Marrow Transplantation, Department of Medicine, Stanford University, Stanford, CA 94305, USA of Medicine and Radiology, Stanford University, Stanford, CA 94305, USA 3Stanford Cardiovascular Institute & Institute for Stem Cell Biology and Regenerative Medicine, Stanford University, Stanford, CA 94305, USA 4Peninsula Pathologists Medical Group, Inc., South San Francisco, CA 94080, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.celrep.2015.02.050 This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 2Departments

SUMMARY

Embryonic stem cells (ESCs) hold promise for the treatment of many medical conditions; however, their utility is limited by immune rejection. The objective of our study is to establish tolerance or promote engraftment of transplanted ESCs as well as mature cell populations derived from ESCs. Luciferase (luc+)-expressing ESCs were utilized to monitor the survival of the ESCs and differentiated progeny in living recipients. Allogeneic recipients conditioned with fractioned total lymphoid irradiation (TLI) and anti-thymocyte serum (ATS) or TLI plus regulatory T cells (Treg) promoted engraftment of ESC allografts after transplantation. Following these treatments, the engraftment of transplanted terminally differentiated endothelial cells derived from ESCs was also significantly enhanced. Our findings provide clinically translatable strategies of inducing tolerance to adoptively transferred ESCs for cell replacement therapy of medical disorders. INTRODUCTION Embryonic stem cells (ESCs) hold tremendous promise for the treatment of a broad range of serious medical conditions, including organ failure and neurodegenerative disorders (Murry and Keller, 2008). However, their utility is limited by immunological rejection upon transplantation of the allogeneic ESCs or differentiated progeny. Rejection has been observed in animal modeling (Pearl et al., 2012; Wu et al., 2008). One alternative is to use autologous induced pluripotent stem cells (iPSCs) (Wu and Hochedlinger, 2011). However, when autologous iPSCs are not available, transplantation of allogeneic iPSCs can also provoke an immune response. Further concerns remain about the ability to derive differentiated functional cells and tissues from iPSCs. Therefore, overcoming the immunogenicity or inducing tolerance to ESCs and differentiated progeny remains a

critical issue before the promise of ESC-based treatments can be realized. Regulation of immune function has significant potential for the treatment of a variety of disorders, including following hematopoietic cell transplantation (HCT) for controlling graftversus-host disease (GVHD), as well as for inducing tolerance to transplanted organs and tissues. One strategy, developed in murine models and extended to the clinic, involves host conditioning with fractioned total lymphoid irradiation (TLI) and anti-thymocyte globulin (ATG). This approach has been successfully employed to reduce the risk of GVHD after allogeneic HCT and to induce tolerance to transplanted solid organ allografts (Kohrt et al., 2009; Lowsky et al., 2005; Scandling et al., 2008). The mechanism underlying the protective effects of TLI/ATG conditioning is through the altered ratio of natural killer–T (NK-T) cells that are relatively radioresistant as compared with conventional CD4+ and CD8+ T cells (Tcon), resulting in a much greater NK-T/Tcon cell ratio in TLI/ ATG conditioned hosts. NK-T cells that express a common T cell receptor recognize targets through engagement of CD-1 and appear to require regulatory T cells for optimal effects (Hongo et al., 2012; Nador et al., 2010; Schneidawind et al., 2014). Another strategy to suppress alloimmune responses and induce tolerance is through the activation or adoptive transfer of regulatory T cells (Treg). Distinct populations of regulatory T cells are capable of controlling immune reactions such as CD4+CD25+Foxp3+ natural Treg (Fontenot et al., 2003; Sakaguchi et al., 2008). Adoptive transfer of Treg suppresses GVHD in allogeneic animal models of HCT and in humans (Brunstein et al., 2011; Colonna et al., 2011; Di Ianni et al., 2011; Edinger et al., 2003; Nguyen et al., 2008). The goal of our study was to evaluate these clinically utilized strategies to promote the engraftment of undifferentiated and differentiated ESCs. We utilized luciferase (luc+) expressing ESCs to non-invasively track the fate of ESCs in living animals by bioluminescent imaging (BLI) and to explore both the engraftment of ESC populations in allogeneic as compared with syngeneic and severe combined immunodeficiency (SCID) hosts and assess the impact of host

Cell Reports 10, 1793–1802, March 24, 2015 ª2015 The Authors 1793

Figure 1. TLI plus ATS and TLI and ATS Alone Promoted the Engraftment of Undifferentiated ESC Allografts (A) The schema of ESC regimen of allogeneic host conditioning with TLI/ATS. 5 3 104 undifferentiated luc+ C57BL/6 ESCs were IM injected into the allogeneic Balb/c recipients. Ten doses of 240 cGy each were administered to the recipient animals starting on the day following ESC transplantation using 5 doses per week. Rabbit ATS was administered by intraperitoneal injection to the recipient animals with 0.05 ml of ATS in 0.5 ml saline on days 0, 1, 2, 3, and 4 after ESC transplantation. Both TLI and ATS were performed to TLI plus ATS group.

(legend continued on next page)

1794 Cell Reports 10, 1793–1802, March 24, 2015 ª2015 The Authors

conditioning with TLI/ATS and regulatory T cells on their survival. RESULTS ESCs Were Rejected in Allogeneic Recipients after Transplantation We utilized BLI to noninvasively track the engraftment of ESCs in living animals after transplantation. The C57BL/6 mESC line (H-2Kb) was transfected with the luc gene driven by a constitutive CMV promoter. There was a clear correlation between cell number and reporter gene luc activity (Figure S1A). The transduced luc+ ESCs had similar characteristics of selfrenewal and pluripotency to the wild-type C57BL/6 ESCs (data not shown). To assess the survival of the luc+ mESCs in vivo, 1 3 104, 5 3 104, 1 3 105, and 5 3 105 ESCs were injected into the left gastrocnemius muscle of allogeneic Balb/c (H-2Kd) recipients. As controls, the same number of ESCs were injected into C57BL/6 syngeneic, Balb/c SCID, and Balb/c mice previously transplanted with C57BL/6 TCD-BM (BMT). The engraftment of ESCs was non-invasively assessed by BLI every 3–5 days. In agreement with previous studies (Pearl et al., 2011; Swijnenburg et al., 2008a; Wu et al., 2008), the ESCs were rejected in the allogeneic recipients, with BLI signal gradually increasing until 10–14 days after transplantation and then decreasing, until reaching background levels at around 21 days. However, in the syngeneic, SCID and BMT groups, the transplanted ESCs proliferated well and formed teratomas with BLI signal continuously increasing (p < 0.0001) (Figure S1B). We found that 5 3 104 ESCs demonstrated the most consistent rejection and engraftment patterns after transplantation (other cell doses data not shown), and this dose of luc+ C57BL/6 ESCs was utilized in all subsequent experiments. Non-transduced C57BL/6 mESCs were studied in parallel studies, and engraftment was assessed by the size of the teratomas by calipers where a similar rejection pattern was observed, demonstrating that the rejection was not due to immunoreactivity against luc (data not shown). TLI plus ATS and TLI and ATS Alone Promoted the Engraftment of Undifferentiated ESC Allografts Clinically validated strategies utilized to promote the engraftment of ESCs following transplantation by preparing Balb/c hosts with TLI and ATS (Figure 1A) resulted in significant engraftment of undifferentiated ESC allografts (p < 0.05) (Figure 1B). TLI and ATS alone resulted in similar effects as the combination of TLI and ATS in promoting engraftment of ESCs when compared with the allogeneic group (p < 0.05 and p < 0.0001, respectively) (Figure 1C). Representative images of individual animals are shown in Figure 1D.

TLI plus ATS and TLI and ATS Alone Promoted the Engraftment of Differentiated ESC-EB and ESC-TC Allografts As we are concerned that undifferentiated ESCs form teratomas, which is clearly not the ultimate goal for ESC therapy, we explored the engraftment of differentiated progeny derived from ESCs. We used in vitro differentiated ESC-derived EB cells (ESC-EBs) and in vivo differentiated ESC-derived teratoma cells (ESC-TCs) to examine the impact of TLI/ATS conditioning on engraftment of these differentiated cells (Figure 2A). After differentiation, there were still an average of 65% undifferentiated SSEA-1+ ESCs in the day 15 EBs and 10% undifferentiated ESCs in the ESC-TCs (Figures 2A and S2A). These undifferentiated SSEA-1 positive cells were depleted for subsequent studies (Figure 2A). Compared with the low-level expression of MHC class I (H-2Kb) in the undifferentiated ESCs, the ESC-EBs and ESC-TCs have slightly increased expression of MHC-I, with very little MHC-II expression detected (Figure S2A). Compared with the allogeneic group, TLI and ATS conditioning also promoted engraftment of ESC-EBs (p < 0.0001) (Figure 2B). TLI alone and ATS alone both showed similar effects (both p < 0.05) (Figure 2C). A similar impact was also observed following TLI plus ATS conditioning on improving the engraftment of differentiated ESC-TCs (p < 0.05) (Figure 2D). However, compared with the allogeneic group, the effect of TLI and ATS alone had limited impact on the engraftment of the more mature ESCTCs, yet the combination of TLI and ATS did result in engraftment of this cell population (Figure 2E). Forty days after transplantation, we harvested the differentiated ESC grafts from the TLI and ATS treated groups and performed histological analysis. The engrafted tissue still consisted of derivatives from all three lineages, indicating that the strategy of conditioning with TLI and ATS retains the capacity to promote engraftment of differentiated cell types derived from ESCs (Figure S2B). Regulatory T Cells Promoted the Engraftment of Undifferentiated ESCs and Differentiated ESC-EB and ESC-TC Allografts Conditioning with TLI and ATG has been utilized in the clinic and is generally well tolerated (Kohrt et al., 2009; Lowsky et al., 2005; Scandling et al., 2008). However, ATG has significant toxicities, which could have deleterious effects in the clinic that although tolerated for cancer patients may be less desirable in other settings. Therefore, we explored an alternative strategy utilizing Treg. After TLI conditioning, 5 3 105 fluorescence-activated cell sorting (FACS)-purified natural CD4+CD25+Foxp3+ Treg were infused into Balb/c hosts (day 12 after the ESC transplantation; Figure 3A). Compared with the allogeneic group, TLI plus donor type Treg promoted the engraftment of undifferentiated ESC allografts, which is similar to the effect in the TLI alone group

(B) The quantitative BLI signal showed that the undifferentiated ESCs were routinely rejected in the untreated allogeneic group, while in the untreated Balb/c SCID and C57BL/6 syngeneic control groups, the ESCs were well engrafted. However, following host conditioning with TLI plus ATS, significant engraftment of undifferentiated ESC allografts was demonstrated. (C) TLI and ATS alone resulted in similar effects in promoting engraftment of ESCs. (D) The representative in vivo BLI images from one experiment are shown. n = 5 per group, *p < 0.05, ****p < 0.0001, compared with the allogeneic group, respectively. All values are expressed as mean ± SEM. The analysis is by ANOVA test. Each experiment was repeated twice, and one representative experiment is shown. See also Figure S1.

Cell Reports 10, 1793–1802, March 24, 2015 ª2015 The Authors 1795

Figure 2. TLI plus ATS and TLI and ATS Alone Promoted the Engraftment of Differentiated ESC-EB and ESC-TC Allografts (A) The schema of the regimen of allogeneic host conditioning with TLI/ATS. 5 3 104 differentiated luc+ C57BL/6 ESC-EBs and ESC-TCs were IM injected into the allogeneic Balb/c recipients. To deplete the undifferentiated ESCs from the differentiated progenies, only FACS-sorted live SSEA-1 negative cells were injected. (B) Compared with the untreated allogeneic group, TLI and ATS conditioning promoted the engraftment of ESC-EBs. (C) Both TLI alone and ATS alone showed similar effects as TLI plus ATS. (D) A similar impact was also observed following TLI plus ATS conditioning on improving the engraftment of ESC-TCs.

(legend continued on next page)

1796 Cell Reports 10, 1793–1802, March 24, 2015 ª2015 The Authors

(p < 0.0001, compared with the allogeneic group) (Figure 3B). However, with the differentiated ESC-TCs and ESC-EBs, the infusion of donor type Treg after TLI resulted in improved engraftment of these differentiated cells compared with the TLI alone group (p < 0.01) (Figure 3C). TLI/ATG conditioning could enhance iNK-T cells, which appeared to interact with host Treg for optimal effects on tolerance induction to the solid organ allografts (Hongo et al., 2012; Nador et al., 2010). Since donor Treg are generally not available, we evaluated the impact of host Treg on the engraftment of ESC-derived cells. Similar results were observed with the differentiated ESC-EBs, where the impact of TLI alone treatment was very limited and was significantly improved with the addition of host type Treg (p < 0.01) (Figure 3D). TLI and ATS and TLI plus Regulatory T Cells Promoted the Engraftment of Fully Differentiated ESC-Derived Endothelial Cell Allografts Differentiated ESC-EBs and ESC-TCs are composed of precursor cells with the capacity to differentiate into all three lineages; however, they also have the potential for teratoma growth. Since the goal of ESC-based therapies will likely involve the use of differentiated cell populations, we explored the impact of these approaches on engraftment of fully differentiated cells. We chose fully differentiated endothelial cells (ECs) as the target cells since protocol is available for their derivation (Huang et al., 2010) and substantial numbers of cells could be obtained. C57BL/6 luc+ ESCs (H-2Kb) were first cultured to form EBs in suspension culture. EBs were re-plated after 3 days and further differentiated for 3 weeks in the presence of vascular endothelial growth factor (VEGF). At the end of the culture period, the differentiated cells had characteristic endothelial cell morphology (Figure 4A1) and expressed specific endothelial marker-VE-cadherin and CD31 on the SSEA-1 negative cells (Figure 4B). The differentiated ECs also possessed endothelial cell functions in vitro, such as forming endothelial tube-like networks when cultured on Matrigel or gelatin (Figure 4A2), and the uptake of low-density lipoprotein (LDL) (Figure 4A4). FACS analysis demonstrated that 15% of the cells were VE-cadherin+SSEA-1 cells from the differentiation cultures (Figures 4B and 4D). These cells showed much higher MHC class I (H-2Kb) expression than undifferentiated ESCs, without expression of MHC class II (Figures 4C and S2A). The VE-cadherin+SSEA-1 cells were purified to obtain terminally differentiated ESC-derived endothelial cells (ESC-ECs) (post-sort purity 99%), which also completely depleted the undifferentiated SSEA-1+ ESCs (Figure 4D). After FACS, 5 3 104 purified ESC-ECs were injected into the left gastrocneimus muscle of allogeneic Balb/c (H-2Kd) recipients. We compared allogeneic host conditioning with TLI alone, TLI and ATS, and TLI plus host Balb/c type CD4+CD25+Foxp3+ Treg infusion to the Balb/c recipients with unprepared Balb/c recipients. The positive control groups also consisted of Balb/c SCID and C57BL/6 syngeneic recipients without conditioning (data

not shown). The cell number, transplantation, and cell survival tracking experiments were performed as previous experiments (Figure 4D). After transplantation, the survival of ESC-ECs was tracked by in vivo BLI every 5 to 7 days. Unlike undifferentiated ESC, differentiated ESC-EBs, and differentiated ESC-TCs, these fully differentiated ESC-ECs first grow very slowly in all the recipients during the first 3 weeks, with weak BLI signals (Figure 4E). After approximately 3 weeks, the engraftment of ESC-ECs was observed in the TLI and ATS treated group, compared with the untreated allogeneic group (p < 0.0001). Although the ECs grew rapidly during this period, their overall BLI intensity was much lower than from the engrafted undifferentiated ESCs and ESCEBs (Figures 1B and 2B). When 5 3 105 natural host Balb/c type Treg purified by FACS were infused to the allogeneic recipients after 10 doses of TLI at day 12 of ESC-ECs post-transplantation (Figure 4D), the BLI results showed that TLI plus Treg treatment can also significantly improve the engraftment of ESC-ECs (p < 0.0005, compared with the allogeneic group) (Figure 4E). The Treg infusion after TLI significantly enhanced the engraftment of differentiated ESC-ECs when compared with animals treated with TLI alone, where the cells engrafted minimally by 7 weeks and then quickly rejected later at 9 weeks (p < 0.01; Figure 4E). All of the ESC-EC allografts survived for 2 months after TLI/ATS and TLI plus host Treg treatments. After 2 months, we harvested the allografts and performed histological analysis. A high density of capillary tube-like structures could be clearly observed among the muscle fibers. Simultaneously, immunofluorescence staining of luc and the endothelial marker CD31 demonstrated the presence of donor-derived luc and CD31 double-positive cells in the allografts, which indicated the engraftment of transplanted luc+ ESC-ECs (Figure 4F). DISCUSSION Embryonic stem cells have demonstrated ability to self-renew and differentiate into a broad array of functional cells making these cells attractive for clinical translation. However, rejection of ESCs after transplantation into allogeneic recipients is a major limitation. We confirmed that ESCs are readily rejected in agreement with other studies (Pearl et al., 2011; Swijnenburg et al., 2008a, 2008b; Wu et al., 2008). To overcome rejection, we utilized clinically relevant immune regulatory strategies to promote the engraftment of ESCs and differentiated cell populations after allogeneic transplantation without the need for additional donor bone marrow cells. TLI and ATS conditioning either alone or in combination had a significant impact on promoting the engraftment of ESC allografts but not differentiated progeny where TLI and ATS together were required for similar engraftment observed in syngeneic hosts. Treg infusion after TLI treatment also significantly promoted the engraftment of differentiated progeny from ESCs, especially the fully differentiated ESCs in the allogeneic recipients. This also demonstrated that the adoptive transfer of Treg can suppress rejection to allogeneic

(E) However, compared with the allogeneic group, TLI and ATS alone had a limited effect on promoting the engraftment of the more mature ESC-TCs. In the Balb/c SCID and C57BL/6 syngeneic control groups, the differentiated ESCs were well engrafted. n = 5 per group, *p < 0.05, ***p < 0.0005, ****p < 0.0001, compared with the allogeneic group, respectively. All values are expressed as mean ± SEM. The analysis is by ANOVA test. Each experiment was repeated twice, and one representative experiment is shown. See also Figure S2.

Cell Reports 10, 1793–1802, March 24, 2015 ª2015 The Authors 1797

Figure 3. Treg Promoted the Engraftment of Undifferentiated ESCs and Differentiated ESC-EB and ESC-TC Allografts (A) The schema of the Treg infusion regimen. 5 3 104 undifferentiated luc+ C57BL/6 ESCs and FACS-sorted differentiated SSEA-1 negative ESC-EBs and ESCTCs were IM injected into allogeneic Balb/c recipients, the same as TLI/ATS experiments. (B–D) After 10 doses of TLI conditioning, 5 3 105 FACS-purified C57BL/6 donor (B, C) or Balb/c host (D) type CD4+CD25+Foxp3+ Tregs were infused to Balb/c recipients (day 12 after ESC transplantation). (B) The quantitative BLI signal results showed that compared with the allogeneic group, TLI plus donor C57BL/6 type Tregs infusion promoted the engraftment of undifferentiated ESC allografts, which is similar to the impact from the TLI alone group. (C) With the differentiated ESC-TCs, the infusion of donor type Tregs after TLI resulted in improved engraftment of these differentiated cells compared with the TLI alone group. (D) Similar results were also observed with the differentiated ESC-EBs, where the impact of TLI alone treatment was limited and significantly improved with the addition of host Balb/c type Treg. n = 5 per group, **p < 0.01 comparison between TLI + Treg and TLI alone; *p < 0.05, ***p < 0.0005, ****p < 0.0001, compared with the allogeneic group. All values are expressed as mean ± SEM. The analysis is by ANOVA test. Each experiment was repeated twice, and one representative experiment is shown.

ESC transplanted across a major-mismatched histocompatibility complex barrier. We further explored the impact of these clinically translatable regimens not only to the undifferentiated ESCs, but also to the differentiated cells derived from ESCs. The TLI/ATG regimen has been successfully used in the clinic to induce tolerance to transplanted kidney allografts in combina-

tion with bone marrow transplantation from the same donor (Scandling et al., 2008) and to suppress GVHD after allogeneic HCT for the treatment of patients with hematological malignancies (Kohrt et al., 2009; Lowsky et al., 2005; Messina et al., 2012). In murine models, host TLI and ATS conditioning can induce tolerance to combined solid organ and bone marrow

1798 Cell Reports 10, 1793–1802, March 24, 2015 ª2015 The Authors

allografts (Hongo et al., 2012; Nador et al., 2010). Mechanistic studies have demonstrated that this regimen promotes tolerance by changing the balance of regulatory and conventional naive CD4 T cells in the host, with TLI/ATS favoring NK-T cells in the recipient that interact with CD4+CD25+ Treg in the graft (Nador et al., 2010). Graft acceptance and donor cell engraftment required host Treg production of IL-10 that was in turn enhanced by the host iNK-T cell production of IL-4. Depletion of Treg before conditioning resulted in rapid rejection of bone marrow and organ allografts (Hongo et al., 2012; Nador et al., 2010). The TLI and ATS regimens have never before been applied to ESC transplantation. Previous studies have shown that the rejection emerging in ESC allografts is due to CD4 and CD8 T cells that infiltrate the graft (Robertson et al., 2007; Swijnenburg et al., 2008a; Wu et al., 2008). In our experiments, TLI/ATS host conditioning had a significant impact on promoting the engraftment of both undifferentiated and differentiated ESC allografts. The TLI and ATS alone treatments had similar effects on improving the engraftment of undifferentiated ESCs as combined treatment with TLI/ATS. When more differentiated ESCTCs and ESC-ECs were utilized, the effects of TLI or ATS alone became more limited. In addition, when TLI/ATG conditioning is utilized in the solid organ tolerance induction experiments and clinical trials, it is administered in combination with bone marrow transplantation. In our ESC transplantation experiments, TLI plus ATS, without bone marrow transplantation, could promote the engraftment of ESC allografts, which is consistent with previous studies (Robertson et al., 2007; Wu et al., 2008) that ESCs have lower immunogenicity than conventional terminally differentiated cells. These studies imply that there may be a lower barrier to induce tolerance to ESCs than to solid organ transplantation. To evaluate the impact of these approaches on the transplantation of more differentiated cell populations, which is closer to the clinical situation, we first transplanted SSEA-1 negative differentiated progeny from the ESC-EBs and ESC-TCs and examined the impact of TLI/ATS and Treg on their subsequent engraftment. We further examined these immune regulatory strategies on the engraftment of terminally differentiated ESCderived endothelial cells. Importantly, conditioning with TLI/ ATS or TLI/Treg also promoted the engraftment to fully differentiated ESC-derived endothelial cells. Furthermore, the induced tolerance in both treated groups can be maintained for as long as 2 months. Transplanted luc+ ESC-EC derived capillary tubelike structures, which are luc+CD31+, could be observed among the muscle fibers in the allografts, while teratoma formation was not observed. Naturally occurring Treg have been widely used in suppressing GVHD and autoimmune diseases in animal models with supportive studies emerging in the clinic (Brunstein et al., 2011; Di Ianni et al., 2011; Edinger et al., 2003; Nguyen et al., 2008). In the setting of ESC transplantation, very few previous studies have indicated that Treg play an important role in acceptance of ESC-derived tissues transplanted across major mismatched histocompatibility barriers (Lui et al., 2010). Co-stimulation blockade enhanced human ESC engraftment into the testis and heart of immunocompetent mice, and regulatory T cells were postulated to be involved in tolerance induction (Grinnemo

et al., 2008). In our current experiments, infusion of Treg after host TLI conditioning resulted in enhanced engraftment of ESC allografts, both to undifferentiated and differentiated ESCs, which confirmed the previous study implicating an important role for Treg in ESC transplantation and also suggested that infusion of naturally arising Treg can be a promising strategy for future ESC-based treatments. The combination of TLI conditioning with Treg may also reduce the toxicity of the combined TLI/ATG regimen. Our findings highlighting the functional role of Treg combined with TLI, which promotes engraftment of allogeneic transplanted ESCs, has some differences from our previous study (Pearl et al., 2011). Because of the mechanistic difference of immunosuppression strategies utilized, blocking leukocyte costimulatory molecules promotes engraftment of differentiated progeny derived from ESCs where in that setting Treg did not play a major role (Pearl et al., 2011). We also did not observe that Treg alone promoted engraftment of any ESC-derived tissue. In our current studies, the adoptive transfer of Treg after host TLI conditioning made a significant contribution to promote engraftment of differentiated ESC progeny, which is in agreement with previous iNK-T/Treg mechanistic findings from TLI/ATS tolerance induction animal experiments (Hongo et al., 2012; Nador et al., 2010). These conditions are quite different and highlight the importance of the specific approach to promote engraftment of these differentiated cells and tissues when clinical translation is contemplated. In conclusion, our findings demonstrate that TLI plus ATS and TLI plus Treg infusion can promote the engraftment of ESC and differentiated progeny allografts, which not only provides insights into the mechanisms of the immune rejection of ESCs but also develops clinically translatable strategies of inducing tolerance to adoptively transferred ESCs for the future treatment of medical disorders. EXPERIMENTAL PROCEDURES Culturing, Transduction, Transplantation, and Host Conditioning with TLI, ATS, Treg Infusion, and In Vivo BLI Tracking of ESCs The C57BL/6 mESC cell line derived from wild-type C57BL/6 (H-2Kb) mice purchased from ATCC (Manassas, VA) was cultured and transduced with a reporter gene construct carrying the firefly luc driven by a constitutive CMV promoter using techniques as previously described (Pearl et al., 2011; Swijnenburg et al., 2008a, 2008b). 5 3 104 luc+ undifferentiated C57BL/6 mESCs, FACS-sorted SSEA-1 negative in vitro differentiated ESC-EBs, in vivo differentiated ESC-TCs, and in vitro differentiated ESCECs (Huang et al., 2010) were injected intramuscularly (IM) into the left gastrocnemius muscle of allogeneic Balb/c (H-2Kd) mice recipient and controls including C57BL/6 syngeneic, Balb/c SCID, and lethally irradiated Balb/c mice previously transplanted with T cell depleted bone marrow (TCD-BM) cells from C57BL/6 mice (BMT). Allogeneic Balb/c host conditioning with 10 doses of 240 cGy TLI and 5 doses of ATS was performed as previously described (Hongo et al., 2012; Nador et al., 2010). 5 3 105 donor C57BL/6 or host Balb/c type Treg were injected intravenously into Balb/c recipients after TLI and at 12 days after ESCs transplantation. Isolation and purification of Treg was performed as previously described (Edinger et al., 2003; Nguyen et al., 2008). Each group consisted of at least five mice and each experiment was repeated at least two times. Animal protocols were approved by the Institutional Animal Care and Use Committee of Stanford University. The engraftment of ESCs was non-invasively assessed by serial BLI every 3–5 days. BLI was performed as described previously (Swijnenburg et al., 2008a) with an IVIS 200 or IVIS spectrum charge-coupled device

Cell Reports 10, 1793–1802, March 24, 2015 ª2015 The Authors 1799

Figure 4. TLI and ATS and TLI plus Treg Promoted the Engraftment of Fully Differentiated ESC-Derived Endothelial Cell Allografts (A) In vitro characterization of cultured luc+ C57BL/6 ESC-derived endothelial cells (ESC-ECs) demonstrating endothelial morphology after the 3-week VEGF induced differentiation (A1), and typical EC functions in vitro, such as forming endothelial tube-like networks (A2), uptake of low-density lipoprotein (LDL), where the Rodamine labeled Dil-ac-LDL can be observed in the ECs under the fluorescence microscope (A4), when compared with the phase contrast image of all the cells on the left (A3). (B) ESC-ECs expressed the specific endothelial marker-VE-cadherin and CD31. (C) ESC-ECs expressed increased MHC I (H-2Kb) and did not express MHC II.

(legend continued on next page)

1800 Cell Reports 10, 1793–1802, March 24, 2015 ª2015 The Authors

imaging system (Xenogen). Images were analyzed with Living Image Software 4.1 (Caliper Life Sciences). Differences in BLI signals among different groups were analyzed with the ANOVA test using GraphPad Prism v6.0 software, and p < 0.05 was considered statistically significant. More detailed methodological information is available in the Supplemental Experimental Procedures. SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures and two figures and can be found with this article online at http://dx.doi.org/ 10.1016/j.celrep.2015.02.050. AUTHOR CONTRIBUTIONS Y.P. designed and performed research, analyzed data, and wrote the manuscript. D.B.L.-G., P.E.de A., A.P., J.B., M.F., H.N., B.K., and R.K. helped perform the research; J.C.W. provided experimental advice, helped write the manuscript, and provided funding support; R.S.N. designed research, provided overall guidance, helped write the manuscript, and provided funding support. ACKNOWLEDGMENTS This work was supported by the California Institute of Regenerative Medicine (CIRM) grants RM1-01725 (R.S.N.), DR2-05394, and TR3-05556 and NIH grant R01 AI085575 (J.C.W.). We wish to thank David Hongo in the laboratory of Dr. Samuel Strober for the technical help with TLI/ATS and Rosalie Nolley in the Department of Urology for the kind help with immunohistochemistry staining. We thank all the members of the R.S.N. laboratory, especially Maite Alvarez Rodriguez, Dominik Schneidawind, and Emanuela I. Sega, for valuable help and discussions throughout the course of this work. We also thank the Stanford Shared FACS Facility for providing support for FACS analysis and the Stanford Center for Innovation in In-Vivo Imaging (SCI3) for in vivo BLI imaging.

Di Ianni, M., Falzetti, F., Carotti, A., Terenzi, A., Castellino, F., Bonifacio, E., Del Papa, B., Zei, T., Ostini, R.I., Cecchini, D., et al. (2011). Tregs prevent GVHD and promote immune reconstitution in HLA-haploidentical transplantation. Blood 117, 3921–3928. Edinger, M., Hoffmann, P., Ermann, J., Drago, K., Fathman, C.G., Strober, S., and Negrin, R.S. (2003). CD4+CD25+ regulatory T cells preserve graft-versustumor activity while inhibiting graft-versus-host disease after bone marrow transplantation. Nat. Med. 9, 1144–1150. Fontenot, J.D., Gavin, M.A., and Rudensky, A.Y. (2003). Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat. Immunol. 4, 330–336. Grinnemo, K.H., Genead, R., Kumagai-Braesch, M., Andersson, A., Danielsson, C., Ma˚nsson-Broberg, A., Dellgren, G., Stro¨mberg, A.M., Ekberg, H., Hovatta, O., et al. (2008). Costimulation blockade induces tolerance to HESC transplanted to the testis and induces regulatory T-cells to HESC transplanted into the heart. Stem Cells 26, 1850–1857. Hongo, D., Tang, X., Dutt, S., Nador, R.G., and Strober, S. (2012). Interactions between NKT cells and Tregs are required for tolerance to combined bone marrow and organ transplants. Blood 119, 1581–1589. Huang, N.F., Niiyama, H., Peter, C., De, A., Natkunam, Y., Fleissner, F., Li, Z., Rollins, M.D., Wu, J.C., Gambhir, S.S., and Cooke, J.P. (2010). Embryonic stem cell-derived endothelial cells engraft into the ischemic hindlimb and restore perfusion. Arterioscler. Thromb. Vasc. Biol. 30, 984–991. Kohrt, H.E., Turnbull, B.B., Heydari, K., Shizuru, J.A., Laport, G.G., Miklos, D.B., Johnston, L.J., Arai, S., Weng, W.K., Hoppe, R.T., et al. (2009). TLI and ATG conditioning with low risk of graft-versus-host disease retains antitumor reactions after allogeneic hematopoietic cell transplantation from related and unrelated donors. Blood 114, 1099–1109. Lowsky, R., Takahashi, T., Liu, Y.P., Dejbakhsh-Jones, S., Grumet, F.C., Shizuru, J.A., Laport, G.G., Stockerl-Goldstein, K.E., Johnston, L.J., Hoppe, R.T., et al. (2005). Protective conditioning for acute graft-versus-host disease. N. Engl. J. Med. 353, 1321–1331. Lui, K.O., Boyd, A.S., Cobbold, S.P., Waldmann, H., and Fairchild, P.J. (2010). A role for regulatory T cells in acceptance of ESC-derived tissues transplanted across an major histocompatibility complex barrier. Stem Cells 28, 1905–1914.

Received: January 27, 2014 Revised: January 15, 2015 Accepted: February 22, 2015 Published: March 19, 2015 REFERENCES

Messina, G., Giaccone, L., Festuccia, M., Irrera, G., Scortechini, I., Sorasio, R., Gigli, F., Passera, R., Cavattoni, I., Filippi, A.R., et al. (2012). Multicenter experience using total lymphoid irradiation and antithymocyte globulin as conditioning for allografting in hematological malignancies. Biol. Blood Marrow Transplant. 18, 1600–1607.

Brunstein, C.G., Miller, J.S., Cao, Q., McKenna, D.H., Hippen, K.L., Curtsinger, J., Defor, T., Levine, B.L., June, C.H., Rubinstein, P., et al. (2011). Infusion of ex vivo expanded T regulatory cells in adults transplanted with umbilical cord blood: safety profile and detection kinetics. Blood 117, 1061–1070.

Murry, C.E., and Keller, G. (2008). Differentiation of embryonic stem cells to clinically relevant populations: lessons from embryonic development. Cell 132, 661–680.

Colonna, L., Sega, E.I., and Negrin, R.S. (2011). Natural and expanded CD4+CD25+ regulatory T cells in bone marrow transplantation. Biol. Blood Marrow Transplant. 17, S58–S62.

Nador, R.G., Hongo, D., Baker, J., Yao, Z., and Strober, S. (2010). The changed balance of regulatory and naive T cells promotes tolerance after TLI and anti-Tcell antibody conditioning. Biol. Blood Marrow Transplant. 10, 262–272.

(D) The regimen of allogeneic host conditioning with TLI/ATS and TLI plus Treg. 5 3 104 fully differentiated luc+ C57BL/6 ESC-ECs were injected (IM) into allogeneic Balb/c recipients. To deplete the undifferentiated ESCs from the differentiated progenies, only purified VE-cadherin+SSEA-1 cells were injected. The post-sort purity was greater than 99%. 5 3 105 purified host type natural CD4+CD25+Foxp3+ Treg were infused to the allogeneic recipients after 10 doses of TLI on day 12 of ESC-ECs transplantation. (E) BLI results demonstrating that fully differentiated ESC-ECs first grew slowly in all recipients over the first 3 weeks. Engraftment of ESC-ECs was observed in the TLI and ATS treated group, compared with the untreated allogeneic group (p < 0.0001). TLI plus host Balb/c type Treg infusion also significantly improved the engraftment of ESC-ECs (p < 0.0005, compared with the unprepared group) than TLI alone, where the cells engrafted minimally by 7 weeks, then were quickly rejected by 9 weeks (p < 0.01). (F) After 2 months, allografts from TLI/ATS and TLI plus Treg groups were harvested. Histological analysis and immunohistochemistry staining were performed. H&E stains demonstrated that the high density of capillary tube-like structures (indicated by the arrows) could be observed among the muscle fibers. Immunofluorescence staining of luciferase (Luc; green) and endothelial marker CD31 (red) demonstrated the presence of donor-derived Luc and CD31 double positive ESC-ECs in the allografts, indicated by the arrowheads in the higher magnification from the boxed area on the right. Nuclear staining is identified by DAPI (blue). The negative controls including staining with isotype antibodies or secondary antibodies alone showed negative results. Scale bars represent 100 mm (A); 50 mm (CD31, Luc, DAPI, and Merge in F); 25 mm (High Mag in F); and 200 mm (H&E in F), respectively. n = 5 per group, **p < 0.01, ***p < 0.0005, ****p < 0.0001. All values are expressed as mean ± SEM. The analysis is by ANOVA test. Each experiment was repeated twice, and one representative experiment is shown.

Cell Reports 10, 1793–1802, March 24, 2015 ª2015 The Authors 1801

Nguyen, V.H., Shashidhar, S., Chang, D.S., Ho, L., Kambham, N., Bachmann, M., Brown, J.M., and Negrin, R.S. (2008). The impact of regulatory T cells on T-cell immunity following hematopoietic cell transplantation. Blood 111, 945–953.

Schneidawind, D., Pierini, A., Alvarez, M., Pan, Y., Baker, J., Buechele, C., Luong, R.H., Meyer, E.H., and Negrin, R.S. (2014). CD4+ invariant natural killer T cells protect from murine GVHD lethality through expansion of donor CD4+CD25+FoxP3+ regulatory T cells. Blood 124, 3320–3328.

Pearl, J.I., Lee, A.S., Leveson-Gower, D.B., Sun, N., Ghosh, Z., Lan, F., Ransohoff, J., Negrin, R.S., Davis, M.M., and Wu, J.C. (2011). Short-term immunosuppression promotes engraftment of embryonic and induced pluripotent stem cells. Cell Stem Cell 8, 309–317.

Swijnenburg, R.J., Schrepfer, S., Cao, F., Pearl, J.I., Xie, X., Connolly, A.J., Robbins, R.C., and Wu, J.C. (2008a). In vivo imaging of embryonic stem cells reveals patterns of survival and immune rejection following transplantation. Stem Cells Dev. 17, 1023–1029.

Pearl, J.I., Kean, L.S., Davis, M.M., and Wu, J.C. (2012). Pluripotent stem cells: immune to the immune system? Sci. Transl. Med 4, 164ps25. Robertson, N.J., Brook, F.A., Gardner, R.L., Cobbold, S.P., Waldmann, H., and Fairchild, P.J. (2007). Embryonic stem cell-derived tissues are immunogenic but their inherent immune privilege promotes the induction of tolerance. Proc. Natl. Acad. Sci. USA 104, 20920–20925. Sakaguchi, S., Yamaguchi, T., Nomura, T., and Ono, M. (2008). Regulatory T cells and immune tolerance. Cell 133, 775–787. Scandling, J.D., Busque, S., Dejbakhsh-Jones, S., Benike, C., Millan, M.T., Shizuru, J.A., Hoppe, R.T., Lowsky, R., Engleman, E.G., and Strober, S. (2008). Tolerance and chimerism after renal and hematopoietic-cell transplantation. N. Engl. J. Med. 358, 362–368.

Swijnenburg, R.J., Schrepfer, S., Govaert, J.A., Cao, F., Ransohoff, K., Sheikh, A.Y., Haddad, M., Connolly, A.J., Davis, M.M., Robbins, R.C., and Wu, J.C. (2008b). Immunosuppressive therapy mitigates immunological rejection of human embryonic stem cell xenografts. Proc. Natl. Acad. Sci. USA 105, 12991– 12996. Wu, S.M., and Hochedlinger, K. (2011). Harnessing the potential of induced pluripotent stem cells for regenerative medicine. Nat. Cell Biol. 13, 497–505. Wu, D.C., Boyd, A.S., and Wood, K.J. (2008). Embryonic stem cells and their differentiated derivatives have a fragile immune privilege but still represent novel targets of immune attack. Stem Cells 26, 1939–1950.

1802 Cell Reports 10, 1793–1802, March 24, 2015 ª2015 The Authors