| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
EMBRYONIC STEM CELLS: CHARACTERIZATION SERIES |
aRoy J. and Lucille A. Carver College of Medicine, University of Iowa and Department of Veterans Affairs Medical Center;
bImmunology Graduate Program, Iowa City, Iowa, USA
Key Words. Embryonic stem cells • Engraftment • Apoptosis • Mixed chimerism • Retinoic acid early inducible-1
Correspondence: Nicholas Zavazava, M.D., Ph.D., University of Iowa Hospitals and Clinics, Department of Internal Medicine, 200 Hawkins Drive, C51-F, Iowa City, Iowa 52242, USA. Telephone: 319-384-6577; Fax: 319-356-8280; e-mail: Nicholas-zavazava{at}uiowa.edu
Received January 11, 2006;
accepted for publication June 16, 2006.
First published online in STEM CELLS EXPRESS June 22, 2006.
 |
ABSTRACT |
|---|
 Top Abstract IntroductionExperimental ProceduresResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionExperimental ProceduresResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
Typical phenotypic characteristics of ESCs include stage-specific-embryonic antigen (SSEA) expression, Oct-3/4, and Nanog-1, which are transcription factors required for ESC self-renewal. SSEA-1 is defined in mice by a monoclonal antibody [11] that reacts with carbohydrate chains bearing a common terminal Lex structure, Galb1–4(Fuca1–3)GlcNAcb1-R [12, 13]. This structure is thought to play a role in cell-cell recognition and adhesion in developing mouse embryos [14, 15]. Expression of SSEA-1 is known to change characteristically during mouse embryogenesis [11]. Mouse- and rat-derived ESCs express SSEA-1, whereas human ESCs express SSEA-3 and SSEA-4, and nonhuman primate ESCs express SSEA-4 [16]. Octamer-binding transcription factor 4 (Oct-4) is characteristic of ESCs in addition to high telomerase activity. Oct-4, a homeobox transcription factor encoded by Pou5f1, is required to maintain the totipotency of embryonic stem cells and is expressed in embryonic primordial germ cells and in spermatogonia in juvenile mice [17].
Data on preimplantation stage stem cells [18] and on human embryonic stem cells [19] have suggested that ESCs may be immune-privileged. Although the published data are of potential interest in the use of ESCs in cellular therapies, very little progress has been made over the past few years to substantiate these data in other experimental models. Recently, data on ESC-derived cells were published showing that ESCs are less likely to provoke an immunological reaction due to the low expression of major histocompatibility complex (MHC) antigens [20]. In addition, the use of ESCs to treat myocardial infarction revealed that that ESCs elicit an immune response, which is unexpected from the in vitro lack of MHC expression detected on ESCs [21, 22]. The mechanism by which these cells become immunogenic in vivo is unclear. It is possible that under inflammation, such as during a myocardial infarction as seen in the two studies cited above, inflammatory cytokines may stimulate ESCs to upregulate MHC antigens. Here, we further studied the susceptibility of ESCs to alloreactive cytotoxic T cell (CTL) and natural killer (NK) lysis and show that these cells show limited susceptibility to killing by either cell type.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TopAbstractIntroductionExperimental ProceduresResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
Phenotyping of ESCs by Flow Cytometry and by Immunohistochemistry
An important characteristic of mouse ESCs is their low expression of MHC class I and lack of expression of MHC class II antigens. ESCs possess the property of quickly differentiating in culture. Thus, to ensure that nondifferentiated cells were used in our studies, ESC expression of SSEA-1 and MHC expression were measured before each experiment. The SSEA-1 antibody was purchased from the University of Iowa Hybridoma Facility, and the MHC antibodies were purchased from BD Pharmingen (San Diego, http://www.bdbiosciences.com/pharmingen). Only cultures with >80% SSEA-1 and <5% MHC class I expression were used in our experiments. Briefly, 2 x 105 ESCs were incubated with 1 µg of each antibody, respectively. After incubating for 30 minutes at 4°C, the cells are washed three times in phosphate-buffered saline (PBS) and further incubated with a fluorescein-conjugated goat anti-mouse serum for another 30 minutes. Subsequently, the cells are washed again and cell fluorescence measured on a FACScan (BD Pharmingen).
To determine alkaline phosphatase expression, the AEC kit (DakoCytomation, Carpinteria, CA, http://www.dakocytomation.com) was used according to the manufacturer's instructions. Upon oxidation, AEC forms a red end product. Oct-4 expression was determined by the use of the anti-Oct-4 antibody, purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, http://www.scbt.com), using the BCIP/NBT stain (DakoCytomation) as instructed by the manufacturer.
Transplantation of ESCs
We previously infused rat ESCs in rats via the portal vein [18]. This is technically challenging and accompanied with a high mortality and bleeding complication rate. Thus, to establish an easier and more reliable technique for the infusion of ESCs, we established the i.v. infusion of mouse ESCs via the supraorbital vein. Cell numbers of 106, 2 x 106, and 5 x 106, of all three cell lines, were infused in 6–8-week-old MRL mice (n = 10), respectively. To enhance ESC engraftment, recipient mice were sublethally irradiated (550 cGy) 24 hours prior to ESC infusion. Multilineage mixed chimerism was studied by flow cytometry. Venous blood was drawn from the supraorbital vein and stained with fluorescein-conjugated antibodies against CD3, CD19, and CD16/32 (granulocytes and monocytes). All antibodies were purchased from BD Pharmingen. The cells were washed and further incubated by a lysis buffer containing ammonium chloride to lyse the erythrocytes. After 10 minutes, the cells were washed in PBS, and cell fluorescence was measured using a flow cytometer (Becton, Dickinson and Company, Mountain View, CA, http://www.bd.com). The animals were either left untreated, irradiated with 500 cGy prior ESC infusion, or injected 15 mg/kg cyclosporine A once at the time of ESC infusion (n = 10 in each group).
ESC Susceptibility to NK Killing
To determine mouse ESC susceptibility to NK cells, ESCs were used as target cells in a 4-hour 51Cr-release assay. The cells were labeled with 51Na2CrO4 (GE Healthcare, Little Chalfont, Buckinghamshire, U.K., http://www.gehealthcare.com/) and incubated with splenocytes separated from MRL mice. ESCs were incubated with NK cells at a serial dilution of target:effector cell ratios. 51Cr release was measured in a gamma counter. To activate NK cells in vivo before use, MRL mice were infused 200 µl of tilorone dihydrochloride, 95% (Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com) 24 hours before the NK assay. This reagent strongly induces interferon-
(IFN-) production, thus enhancing NK activity.
ESC Susceptibility to CTL Killing
To determine whether mouse ESCs are susceptible to CTL killing, cytotoxic T cells against 129/SvJ were generated according to the standard protocol, described in this laboratory previously [29, 30]. Briefly, splenocytes from the donor strain mice were irradiated with 20 Gy and used as stimulator cells in bulk cultures with splenocytes derived from the MRL strain as responder cells. The responder:target cell ratio was maintained at 1:1 and the culture conditions kept generally as established in this laboratory [29, 30]. After 2–3 restimulations, the CTL were used as effector cells against ESCs. Positive target cell controls were Con A-derived blast cells from the two donor strains C57BL/6 and 129/SvJ. Third-party controls were BALB/c (H-2d)-derived Con A blast cells. We hypothesized that lack of killing of the ESCs by CTL could be due to low MHC expression by ESCs. Therefore, ESCs were treated with 100 µg of recombinant IFN- 24 hours prior to the cytotoxicity assay to boost MHC expression and then used as target cells as well. Syngeneic and BALB/c-derived third party Con A blast cells were used as controls.
Mixed Lymphocyte Cultures
Mixed lymphocyte cultures were performed as we have previously described in great detail [30].
Detection of Donor-Derived Cells and Apoptosis in Peripheral Organs
Chimeric animals were sacrificed on day 7 post-ESC infusion, and the thymus, spleen, and liver were harvested (n = 6). The organs were cut in half; one-half was frozen for immunohistochemistry and the other half was minced and a free suspension of lymphocytes prepared by Ficoll gradient separation. The cells were stained with the anti-H-2b antibody and measured by flow cytometry to determine the percentage of donor-derived cells. Frozen sections of the thymus, liver, and spleen were prepared and stained with the anti-H-2b antibody or with an anti-CD3 antibody to determine donor cells or T cells, respectively. To determine apoptotic cells in the spleen, thymus, and liver, frozen sections of chimeric animals were stained using the terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay (Trevigen, Inc., Gaithersburg, MD, http://www.trevigen.com) according to the manufacturer's instructions. Control tissue was also stained by the TUNEL assay to determine background apoptosis. To determine the involvement of Fas/FasL in apoptosis induction, Con A blast cells were co-incubated with ESCs. Apoptosis was measured using the TUNEL assay. Furthermore, to determine the involvement of the Fas/FasL system, apoptosis was neutralized by an anti-FasL antibody (BD Biosciences, San Diego, http://www.bdbiosciences.com) in a concentration-dependent manner.
Immunomodulation of alloresponses in chimeric animals was determined using the interleukin 2 (IL-2) enzyme-linked immunospot assay (BD Biosciences) as instructed by the manufacturer. Statistical analysis was performed using the Graphpad-Prism software package, version 3.0 (Graphpad Software, San Diego, http://www.graphpad.com/ecommerce/buynow.htm).
Karyotyping
Karyotyping is now well-established and was performed according to standard procedure. The cells were preplated on slides, lysed, fixed, and stained with the Giemsa stain. The number of chromosomes was counted under a light microscope.
|
RESULTS |
|---|
|
TopAbstractIntroductionExperimental ProceduresResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
|
stimulation upregulated MHC class I expression in ESCs, but not that of class II antigen. In addition, we investigated whether ESCs express NKG2D ligand retinoic acid early inducible-1 (RAE-1ß), as a possible target for NK cells. In mice, the RAE-1 family of proteins has been identified as high-affinity ligands for NKG2D [33]. For example, it was previously shown that RAE-1ß expression by class I-deficient cells enhanced NK cell killing [34]. The Rae1 family comprises 
, ß, , and possibly 
isoforms [35] and is induced in embryonal carcinoma cells upon exposure to retinoic acid, but the mRNA for these molecules is not detected in normal adult organs by Northern blot analysis [36]. These RAE-1 molecules are predominantly expressed in embryonic tissues and certain tumors, but hardly in adult healthy tissues. We therefore wondered whether ESCs express RAE-1 proteins, making them susceptible to NK cells. Indeed, flow cytometric analysis of ESCs showed high expression of RAE-1 but a lack of expression on splenocytes (Fig. 2A).
|
inducer tilorone, intraperitoneally, 24 hours prior to the NK assay. This reagent, augments NK killing in rodents. ESCs were now more susceptible to NK killing (Fig. 2B), although to a much lesser extent than the Yac-1 target cells. The poor susceptibility of ESCs to NK killing despite high expression of Rae-1 allows speculation on whether ESCs induced apoptosis to NK cells based on their high expression of FasL.
Finally, to determine whether mouse ESCs are susceptible to CTL killing, anti-H-2b CTL were generated by coculturing irradiated 129SvJ splenocytes (H-2b), with MRL splenocytes (H-2k) as responders. Con A blast cells of the 129SvJ or third-party BALB/c-derived splenocytes and the 129SvJ RW-4 ESCs were used as target cells in a 4-hour 51Cr release assay. The 129SvJ Con A blast cells were lysed as expected (Fig. 2C), but not the BALB/c-derived Con A blast cells. Interestingly, the 129SvJRW-4 ESCs were not lysed by the CTL, presumably due to a lack of MHC expression. Thus, ESCs were stimulated by IFN- to upregulate class I expression. Stimulated ESCs were more susceptible to CTL killing than nonstimulated ESCs, suggesting that under stimulation or inflammatory conditions, as might be the case after allograft transplantation, ESCs could be rejected.
The immunogenicity of ESCs was further investigated in vitro using mixed lymphocyte reaction assays. Irradiated 129SvJ RW-4 ESCs or 129SvJ-derived splenocytes were used as stimulator cells and MRL-derived splenocytes as responder cells in a 72-hour mixed lymphocyte reaction. Splenocyte-stimulated responder cells responded strongly; however, ESCs failed to induce T-cell proliferation (Fig. 2D). To determine whether ESCs abrogate an ongoing allostimulation, ESCs were added to a mixture of MRL and irradiated 129SvJ splenocytes. The alloresponse was completely abrogated, suggesting a strong downregulatory capacity of alloresponses by ESCs (Fig. 2D). This effect was shown to require cell-cell contact since transwell cultures, where ESCs were cultivated in the upper chambers above mixed lymphocyte cultures, had no effect on T-cell proliferation (Fig. 2E). Although transforming growth factor-ß has been shown to be secreted by ESCs and could potentially downregulate T-cell proliferation, these experiments appear to suggest that soluble molecules may not be involved, but that there is a requirement for cell-cell contact, which agrees with recent data in a xenogeneic model of human embryonic stem cells [37].
ESCs Induce Mixed Chimerism in Allogeneic Recipients
It has been established that the homing of hematopoietic stem cells to the liver and the degree of established engraftment is ideal after intraportal infusion of hematopoietic stem cells rather than after i.v. infusion [38]. Here, i.v. mouse ESC infusion was used instead of the portal vein infusion, which, although more efficient at promoting stem cell engraftment, is technically less feasible for large studies. The goal here was to determine the robustness of ESC tolerogenicity when ESCs were infused in a peripheral vein instead of the liver. Some studies have suggested that the liver is immune-privileged [39, 40], potentially blurring the tolerogenicity of ESCs. However, Bumgardner et al. have shown that allogenic hepatocytes are rejected in recipient mice [41, 42]. Therefore, the supraorbital vein of the MRL mice was used for the infusion of 2 x 106 ESCs. This cell number was determined to be ideal for the mouse in our own preliminary studies (data not shown). Only donor-derived B cells, T cells, and NK cells were detected at levels on the order of 2.0%–5.0% altogether (Fig. 3A). No granulocytes or macrophages were detectable as measured by the anti-CD16/32 antibodies. Only approximately 30%–40% of the animals showed this modest degree of mixed chimerism. Engraftment was not improved by the use of either cyclosporine A or rapamycin. We hypothesized that there might be a requirement to create space in the lymphoid tissues. Therefore, recipient mice were sublethally irradiated before ESC infusion (Fig. 3B). Clearly, higher numbers of both lymphoid and myeloid cells were measured in peripheral blood after ESC infusion, suggesting that myeloid niches required to be emptied through irradiation to allow ESC engraftment and subsequent engraftment. Using this protocol, >75% of the recipient animals showed mixed chimerism, which, however, was no longer detectable after 4 weeks, suggesting that eventual sensitization and rejection or lack of transcription factors are required hematopoietic renewal. A strategy to induce permanent engraftment will likely involve ablation and immunosuppressive treatment of the host or the use of transcription factors, such as HoxB4, that have been shown to induce hematopoietis [43–45]. A concern has been raised about the formation of teratomas after ESC transplantation. In our series, less than 2% (n = 200) of the treated animals developed teratomas. This was not strain-specific, because when we used Balb/c alternatively as recipient mice, the percentage of animals developing tumors remained the same. This low frequency in teratoma formation is partially due to the route of ESC infusion to the very stringent monitoring of ESC cultures and the generally low cell numbers transplanted in this series. We used 1–2 x 106 cells, far fewer cells than the 10 x 106 or more used by others [46].
|
ESCs Populate the Thymus, Spleen, and Liver After Intravenous Infusion in Allogeneic Recipients
To track mouse ESC-derived cells in lymphoid organs, recipient animals were sublethally irradiated to create "space," and ESCs were infused. Chimeric animals were sacrificed 14 days after ESC infusion to track ESC progenitors. For histological studies, frozen sections of the thymus, liver, and spleen were stained for donor MHC class I antigens (H-2b), whereas minced samples were analyzed by flow cytometry after Ficoll gradient separation. A high proportion of donor-derived cells were identified in the splenic marginal zones (Fig. 4A). These cells were confirmed by the use of an antibody that stains marginal zone macrophages (MOMA; BD Biosciences; data not shown). Flow cytometric analysis of splenocytes indicated 11.7% donor-derived cells (Fig. 4B). Interestingly, ESC-derived cells in the thymus were more uniformly distributed within the organ (Fig. 4B). Only 3.0% of thymic cells were donor-derived (Fig. 4D). However, in the liver, most donor-derived cells were detected around the large blood vessels (Fig. 4E) and had a high proportion of approximately 25% of donor cells as detected by flow cytometry (Fig. 4F). Tissues from nonchimeric animals were used at all times as controls and remained negative. The immunohistochemical stains in Figure 4 were all based on donor class I staining. We therefore wondered whether more definitive proof could be attained by a double stain strategy. Here, enhanced yellow-fluorescent protein (E-YFP)-transduced ESCs were infused in recipient mice, and the animals were sacrificed after 14 days for use in staining for both class I and class II antigens. Our data indicate that we obtained both class I and class II staining cells which were E-YFP-positive (Fig. 4G). Thus, clearly these were donor-derived cells that had differentiated in thymic tissue into class I- and class II-expressing cells.
|
ESCs Induce Apoptosis in T Cells of Lymphoid Organs, Regulating Alloresponses
The high expression of FasL by ESCs is an interesting phenotypic characteristic. We determined whether ESCs induce apoptosis to T cells in lymphoid organs, protecting themselves from T-cell-mediated rejection. Frozen sections obtained 14 days after ESC infusion in irradiated allogeneic MRL mice were stained for apoptotic cells using the TUNEL assay. In the spleens, apoptosis was prominent in the T-cell region of the marginal zones (Fig. 5A), following the pattern of distribution of ESC-derived cells detected in Figure 4 above. This finding suggested that ESC-induced apoptosis to T cells, as confirmed by costaining with an anti-CD3 antibody (Fig. 5C). Similarly, the thymus was stained for apoptotic cells (Fig. 5B) and then costained for T cells (Fig. 5D, 5E). The inset of Figure 5E, at higher magnification, clearly indicates that the apoptotic cells are doubly stained for CD3. These results suggested that ESCs may delete activated alloreactive T cells. Splenic histologic sections from control irradiated, but nonchimeric animals showed low levels of apoptosis, as expected (Fig. 5F, 5G). These control experiments are summarized in Figure 5H, which indicates that the number of apoptotic cells in thymuses of chimeric animals 14 days after ESC infusion was clearly much higher than in control animals. To further determine whether the apoptotic cells were a result of the FasL expression by ESCs, we coincubated Con A blast cells with ESCs and measured the number of apoptotic cells using the TUNEL assay. At a ratio as low as 1:100 (ESCs:blast cells), a high proportion of the cells was apoptotic (Fig. 5I). Apoptosis was blocked by an anti-FasL antibody, clearly confirming that the FasL expressed by ESCs was a powerful apoptosis inducer.
|
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionExperimental ProceduresResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
Consistent with previously published data by us and others, ESCs express low MHC class I and no class II MHC antigens [18, 19]. This property could be exploited for their use in the induction of mixed chimerism. However, this lack of MHC expression could be influenced by cytokines. To determine the ability of ESCs to upregulate MHC antigen expression, we stimulated ESCs with IFN-, a potent inducer of MHC expression. MHC class I, but not class II, were upregulated by IFN- stimulation. We therefore tested whether ESCs are susceptible to NK and alloreactive CTL killing in vitro. Clearly, ESCs were not lysed by nonactivated NK cells, nor were they lysed by alloreactive T cells. However, after NK activation, there was modest killing. Similarly, after IFN- stimulation of ESCs that upregulated MHC class I expression, alloreactive CTL lysed ESCs modestly compared with Con A blast target cells. Studies on human embryonic stem cells appear to confirm, too, that ESCs express low MHC class I levels [19]. In contrast to these studies that suggested that ESCs lack NK receptor ligands, and are therefore not susceptible to NK killing, here we show that ESCs indeed are susceptible to both NK and CTL killing, although sensitivity to these cytotoxic cells was modest. Interestingly, ESCs abundantly express RAE-1 molecules. This group of molecules has been reported to make cells susceptible to NK cell killing, as they are ligands for NKG2D [33, 34, 48]. Furthermore, ESCs blocked alloresponses in vitro by a process that required cell-cell contact as detected in transwell experiments.
The results so far therefore suggested that ESCs could be protected from rejection in vivo. Subsequently, we tested whether ESCs engraft in naïve allogeneic mice, without any preconditioning. Indeed, the ESCs engrafted differentiating into lymphoid, but not myeloid cells at a low percentage of donor-derived cells, less than 5%. In contrast, after sublethal irradiation of recipient mice, ESC-derived cells were abundantly detected in the major lymphoid organs, spleen, thymus, and liver. In the spleen, ESC-derived cells were detected in the splenic marginal zones, in contrast with the pattern of donor cells in recipient mice after bone marrow transplantation, which were uniformly distributed within the spleen (not shown). The presence of ESC-derived cells in the marginal zones was further confirmed by intravital imaging, which indicated ESCs homing in the splenic subcapsular region, in particular around the follicles. The multiphoton microscopy allows tracking cells in vivo in real time. In both the thymus and liver, ESC-derived cells were more homogenously distributed in the tissues. The presence of ESC-derived cells in the thymus is significant, since the thymus is the site of T-cell education and maturation. Furthermore, the identification of ESC-induced apoptotic T cells in the thymus, spleen, and liver suggests possible clonal deletion of alloreactive T cells. This is confirmed by our observations that engraftment of ESCs in thymectomized mice was not successful (data not shown).
Our recently published data indicate that ESCs fail to engraft in Fas-deficient mice, suggesting that the constitutively expressed FasL by ESCs contributes to their ability to engraft in allogeneic recipients [32]. This might be the explanation for the requirement for cell-cell contact in the abrogation of alloresponses by ESCs. Indeed, ESCs induced apoptosis in vivo to T cells as indicated by immunofluorescence and TUNEL staining of the spleen, thymus, and liver. These results show that ESCs effectively use the constitutively expressed FasL to protect themselves from T-cell rejection. In further support of these data is the observation that ESCs engraft in nonpreconditioned recipient mice. Although the observed degree of mixed chimerism was low, generally approximately 2%–5%, and only of the lymphoid population, ESC-induced mixed chimerism was elevated by sublethal irradiation of the recipient mice. This treatment allowed differentiation of ESCs into both lymphoid and myeloid cells. T cells of chimeric animals were shown to be nonresponsive to donor MHC, as reflected by low responses to alloantigen in the ELISPOT assay. These findings suggest that the use of ESCs to induce mixed chimerism in allogeneic recipients with minimal treatment of recipient animals is a viable mechanism for generating hematopoietic cells in vivo. However, hematopoietic mixed chimerism was not long-term, as the animals lost mixed chimerism within 4 weeks, suggesting lack of renewal capability. Thus, HoxB4, a transcription factor that has been shown to promote hematopoiesis, may be critical for improved long-term engraftment of ESCs [43–45]. Transfection of ESCs with HoxB4 combined with sublethal irradiation that appears to free up hematopoietic niches might significantly improve ESC engraftment that is stable to protect allografts from rejection. Despite these low numbers of ESC-derived hematopoietic cells in peripheral blood of transplanted mice, a recent report suggests that murine ESCs pretreated with fibroblast growth factor were able to allow differentiation and engraftment of ESC-derived parenchymal cells in the liver across MHC barriers [49]. Sufficient numbers of cells engrafted to correct factor IX deficiency in mice. These findings show the enormous potential for the application of ESCs in cellular transplants, although there may be tissue-specific differences, such as those observed in the induction of hematopoietic mixed chimerism.
|
DISCLOSURES |
|---|
|
TopAbstractIntroductionExperimental ProceduresResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TopAbstractIntroductionExperimental ProceduresResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionExperimental ProceduresResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. C. Wu, A. S. Boyd, and K. J. Wood Embryonic Stem Cells and Their Differentiated Derivatives Have a Fragile Immune Privilege but Still Represent Novel Targets of Immune Attack Stem Cells, August 1, 2008; 26(8): 1939 - 1950. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.-M. Chan, S. Bonde, H. Klump, and N. Zavazava Hematopoiesis and immunity of HOXB4-transduced embryonic stem cell-derived hematopoietic progenitor cells Blood, March 15, 2008; 111(6): 2953 - 2961. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
