Therapeutic Efficacy of Human Embryonic Stem
Cell–Derived Endothelial Cells in Humanized Mouse
Models Harboring a Human Immune System
Heung-Mo Yang,* Sung-Hwan Moon,* Young-Sil Choi, Soon-Jung Park, Yong-Soo Lee,
Hyun-Joo Lee, Sung-Joo Kim,† Hyung-Min Chung†
Downloaded from http://atvb.ahajournals.org/ by guest on June 15, 2017
Objective—Allogeneic transplantation of human embryonic stem cell (hESC) derivatives has the potential to elicit the
patient’s immune response and lead to graft rejection. Although hESCs and their derivatives have been shown to have
advantageous immune properties in vitro, such observations could not be determined experimentally in vivo because
of ethical and technical constraints. However, the generation of humanized mice (hu-mice) harboring a human immune
system has provided a tool to perform in vivo immunologic studies of human cells and tissues. Using this model, we
sought to examine the therapeutic potential of hESC-derived endothelial cells, human embryonic fibroblasts, and cord
blood–derived endothelial progenitor cells in a human immune system environment.
Approach and Results—All cell types transplanted in hu-mice showed significantly reduced cell survival during the first
14 days post-transplantation compared with that observed in immunodeficient mice. During this period, no observable
therapeutic effects were detected in the hindlimb ischemic mouse models. After this point, the cells demonstrated
improved survival and contributed to a long-term improvement in blood perfusion. All cell types showed reduced
therapeutic efficacy in hu-mice compared with NOD scid IL2 receptor gamma chain knockout mice. Interestingly, the
eventual improvement in blood flow caused by the hESC-derived endothelial cells in hu-mice was not much lower than
that observed in NOD scid IL2 receptor gamma chain knockout mice.
Conclusions—These findings suggest that hESC derivatives may be considered a good source for cell therapy and that humice could be used as a preclinical in vivo animal model for the evaluation of therapeutic efficacy to predict the outcomes
of human clinical trials. (Arterioscler Thromb Vasc Biol. 2013;33:2839-2849.)
Key Words: endothelial cells ◼ human embryonic stem cells ◼ hindlimb ischemia ◼ humanized mouse
◼ therapeutic efficacy
H
uman embryonic stem cells (hESCs) are capable of differentiating into almost all types of cells in the body.1 For
this reason, hESC derivatives are widely recognized as potential therapeutic agents for the treatment of various degenerative
medical conditions. A growing number of translational studies
has shown that hESC derivatives promote functional recovery
by either secreting humoral factors or directly engrafting into
the injured site.2,3 Thus, long-term survival of grafted hESC
derivatives is key to promote stable, functional recovery at the
injured site.4 One of the factors affecting the in vivo survival
of hESC derivatives is graft rejection, which occurs as a result
of the activity of the recipient’s immune system.5,6 hESCs
express distinct major histocompatibility complex (MHC)
antigens, the expression of which is elevated during differentiation.7 For this reason, transplantation of MHC-mismatched
hESC derivatives could activate the recipient’s immune system through alloantigen recognition.8 However, the immunogenicity of hESCs and their derivatives remains debatable, as
they theoretically have beneficial immune properties because
of their low expression of MHC class I molecules.9 Several
studies have been performed to examine the immunologic
potential of hESCs, but because of ethical and technical constraints, such studies have only been performed in xenogeneic
immunocompetent mice.10,11 Therefore, more extensive studies are required to fully predict the immunologic potential of
hESCs in the human body. To do so, there is a genuine need
for better preclinical models, which will ultimately diminish
the risk for treated individuals.
For many years, mouse models have constituted useful in
vivo tools for preclinical studies of various human therapies.
Received on: July 13, 2012; final version accepted on: September 15, 2013.
From the Department of Surgery, Samsung Medical Center (H.-M.Y., S.-J.K.), and Samsung Biomedical Research Institute, Seoul, South Korea (Y.-S.C.,
Y.-S.L., H.-J.L.); School of Medicine, KonKuk University, Seoul, South Korea (S.-H.M., S.-J.P., H.-M.C.); and Stem Cell Research Laboratory, CHA Stem
Cell Institute, CHA University, Seoul, South Korea (S.-H.M., S.-J.P., H.-M.C.).
*These authors contributed equally to this article as first authors.
†These authors contributed equally to this article as corresponding authors.
The online-only Data Supplement is available with this article at http://atvb.ahajournals.org/lookup/suppl/doi:10.1161/ATVBAHA.113.302462/-/DC1.
Correspondence to Hyung-Min Chung, PhD, Stem Cell Research Laboratory, CHA Stem Cell Institute, CHA University, 606-16 Yoeksam 1-dong,
Gangnam-gu, Seoul 135-081, Korea (e-mail [email protected]); or Sung-Joo Kim, MD, PhD, Department of Surgery, Samsung Medical Center,
Sungkyunkwan University School of Medicine, Seoul 135-710, Korea (e-mail [email protected]).
© 2013 American Heart Association, Inc.
Arterioscler Thromb Vasc Biol is available at http://atvb.ahajournals.org
2839
DOI: 10.1161/ATVBAHA.113.302462
2840 Arterioscler Thromb Vasc Biol December 2013
Nonstandard Abbreviations and Acronyms
CB-EPC
hEF
hESC
hESC-EC
hu-mice
IFN-γ
MHC
cord blood–derived endothelial progenitor cell
human embryonic fibroblast
human embryonic stem cell
hESC-derived endothelial cell
humanized mice
interferon-γ
major histocompatibility complex
Downloaded from http://atvb.ahajournals.org/ by guest on June 15, 2017
However, because of obvious physiological differences
between mice and humans, such models do not necessarily
represent outcomes in humans.12 To provide more suitable in
vivo models, mice have been humanized to recreate complex
human physiological processes in small animals through the
transplantation of functional human tissues into immunodeficient mice.13 Different models of humanized mice (humice) have been generated for various types of biomedical
research, such as studies of cancer, hematology, and HIV/
AIDS.14–16 In particular, the technology for creating hu-mice
harboring human immune systems has advanced in recent
years and now permits the long-term engraftment of human
hematopoietic stem cells and the production of hematopoietic lineages, including B and T cells.13,17 This type of
hu-mouse model is thought to provide a useful preclinical
model to test the safety and efficacy of hESC-based regenerative medicine.18
The potential therapeutic application of hESC-derived
endothelial cells (hESC-ECs) and cord blood–derived endothelial progenitor cells (CB-EPCs) has been suggested because
of their ability to induce neovascularization under ischemic
conditions.2,19–21 These studies have been largely performed
in immunodeficient mice, which bypasses the immunologic
defense mechanisms that could reduce overall therapeutic
efficacy in the human body.
In this study, we evaluated the therapeutic efficacy of human
embryonic fibroblasts (hEFs), CB-EPCs, and hESC-ECs to
promote neovascularization in hindlimb ischemia using mice
with different immune competencies. Graft rejection was
observed when all cell types were transplanted into xenogeneic immunocompetent mice 6 days post-transplantation, and
this transfer therefore demonstrated no therapeutic effects. In
immunodeficient mice, we observed prolonged survival and
a significant improvement in blood perfusion in the ischemic
region as early as 6 days post-transplantation. The survival of
all cell types was limited in hu-mice during the first 14 days
post-transplantation, and there was no observable functional
improvement. However, after this point, the cells demonstrated improved survival and led to a stable increase in blood
perfusion for 2 weeks. Surprisingly, the eventual improvement in blood flow caused by hESC-ECs in hu-mice was not
much different from that observed in immunodeficient mice.
Our findings suggest that hu-mice could provide a valuable
in vivo preclinical tool to predict the likelihood of functional
improvement in humans transplanted with hESC derivatives.
Materials and Methods
Materials and Methods are available in the online-only Supplement.
Results
Confirmation of a Reconstituted Human
Immune System in Immunodeficient Mice
To create an in vivo human immune system in the mouse,
we generated hu-mice according to the schematic shown in
Figure 1A. The distribution of human immune cells in the
peripheral blood of NOD scid IL2 receptor gamma chain
knockout (NSG) mice was analyzed by measuring CD45+
cells every 4 weeks postinjection (4 weeks: n=72; 8 weeks:
n=67; 12 weeks: n=52; and 16 weeks: n=25; Figure 1B). After
a 12-week generation process, ≈72% (n=52/72) existed with
the exception of formerly used hu-mice (n=11), and ≈31.92%
cells were shown to be CD45+ by fluorescence-activated cell
sorting analysis of the peripheral blood (Figure 1B). Only a
small proportion of T cells (CD3) was detected during the first
8 weeks after generation, but the T-cell frequency gradually
increased to ≈19.7% and 32.6% by 12 and 16 weeks, respectively (Figure 1C). In addition, we analyzed the reconstitution of human T (CD3), B (CD19), NK (CD56), and myeloid
(CD33) cells within the CD45+ cell population in the bone
marrow, spleen, and liver of NSG mice at 16 weeks post-transplantation with hUCB-CD34+ cells (Figure 1D and 1E; n=14).
According to the fluorescence-activated cell sorting analysis
of the human immune cell populations within the 3 organs, we
confirmed that hu-mice possessing a human immune system
were generated successfully from NSG mice.
Survival of hEFs, CB-EPCs, and
hESC-ECs in Mouse Strains With
Different Immune Competencies
Before cell transplantation experiments, we have analyzed
the cellular characterization of hESC-EC (Figure I in the
online-only Data Supplement), CB-EPC (Figure III in the
online-only Data Supplement), and hEFs (Figure IV in the
online-only Data Supplement) and their proliferation rate
in vitro (Figure V in the online-only Data Supplement).
Especially, hESC-ECs were generated from a karyotypically
normal H9 cell line, according to the methods listed in our
previous studies.2,3,22 In addition, we identified for sorted hESECs at passage 3 before cell transplantation (Figure IC–IE in
the online-only Data Supplement). To analyze cell survival,
all cell types were transplanted into immunocompetent mice
(C57BL/6, n=7), immune-deficient mice (NSG, n=7), and
mice harboring a functional human immune system (hu-mice,
n=8). For the analysis of noninvasive cell survival, all cell
types were prelabeled with DiD-cy5.5, and their survival was
measured by their signal intensity using Xenogen (Figure II
in the online-only Data Supplement). As expected, the transplantation of all cell types in C57BL/6 mice resulted in limited
cell survival (Figure 2A); each cell type showed dramatic cell
death, as the signal intensities of these cells decreased significantly 3 days after transplantation (hEFs=33.37±3.35%,
CB-EPCs=21.28±2.53%, and hESC-ECs=35.31±3.84%;
Figure 2D). Complete graft rejection of all cell types was
observed 7 days after transplantation, as the signal intensities dropped to an undetectable level (hEFs=2.31±0.73%,
CB-EPCs=0.87±0.45%,
and
hESC-ECs=5.73±2.33%;
Figure 2D). In contrast, when the cells were transplanted into
Yang et al Evaluation of hESC-ECs in hu-Mice 2841
Downloaded from http://atvb.ahajournals.org/ by guest on June 15, 2017
Figure 1. Reconstitution of humanized mice (hu-mice). A, Schematic of the generation of hu-mice. As described in the Methods in the
online-only Data Supplement, hCD34+ cells were isolated from human umbilical cord blood (hUCB). Isolated hUCB-CD34+ cells (2×105
cells) were intravenously injected into adult NSG mice (n=72) 24 hours after preconditioning with busulfan. B, Kinetic analysis of the
reconstitution of human CD45+ cells in humanized NSG mice (hu-mice). Peripheral blood was collected from the hu-mice at different
time points post-transplantation (4 weeks: n=72; 8 weeks: n=67; 12 weeks: n=52; and 16 weeks: n=25). C, Reconstitution of CD45+
cell–derived human T cells (CD3+ cells; white bar) and B cells (CD19+ cells; mosaic bar) in the peripheral blood of hu-mice during the
first 16 weeks post-transplantation with hUCB-CD34+ cells. D and E, Reconstitution of human T (CD3), B (CD19), NK (CD56), and
myeloid (CD33) cells within the CD45+ cell population in the bone marrow (BM), spleen and liver of hu-mice (n=14). Values represent the
mean±SD. MACS indicates magnetic-activated cell sorting; and MNC, mononuclear cell.
NSG mice, we observed prolonged cell survival (Figure 2B);
compared with C57BL/6 mice, significantly more viable hEFs
and hESC-ECs were detected, as shown by the gradual loss
of signal intensity during the first 14 days post-transplantation (hEFs=80.28±3.91% and hESC-ECs=77.71±2.45%;
Figure 2E). The majority of viable cells remained alive, and
both hEFs and hESC-ECs maintained ≈57% signal intensity
when measured on day 28 post-transplantation (Figure 2B and
2E). The survival of the CB-EPCs was also prolonged compared with that observed in C57BL/6 mice (Figure 2B, blue
arrowheads). However, when compared with the hEFs and
hESC-ECs in NSG mice, the survival of the CB-EPCs was
lower at each measured interval (≈71% and ≈45% at 14 and 28
days post-transplantation, respectively; Figure 2E). Each cell
type transplanted into hu-mice demonstrated similar survival
patterns (Figure 2C), and cells experienced a dramatic level of
cell death during the first 14 days post-transplantation, with
CB-EPCs demonstrating the most dramatic loss of viability
(hEFs=35.33±5.11%, CB-EPCs=23.70±9.73%, and hESCECs=47.33±3.57%; Figure 2F). The cells continued to lose
viability, albeit at a reduced rate, as measured by recording the
signal intensity on day 28 post-transplantation (Figure 2F).
The percentage of remaining viable hESC-ECs on day 28 was
≈35%, which indicated superior survival when compared with
hEFs (≈15%) and CB-EPCs (≈5%; Figure 2F).
Different Immunogenicities of hEFs, CB-EPCs,
and hESC-ECs to the Immune Cells of hu-Mice
The innate production of interferon-γ (IFN-γ) is the first event
to occur after the immune detection of pathogens.23 To examine whether the impaired survival of the 3 cell types examined in hu-mice was attributable to an elevated production of
2842 Arterioscler Thromb Vasc Biol December 2013
Downloaded from http://atvb.ahajournals.org/ by guest on June 15, 2017
Figure 2. Comparison of the viability of human embryonic fibroblasts (hEFs), cord blood–derived endothelial progenitor cells (CB-EPCs),
and human embryonic stem cell–derived endothelial cells (hESC-ECs) in mice with different immune competencies. A, Representative
Xenogen images for the detection of DiD-cy5.5–labeled cells in immunocompetent C57BL/6 mice (n=7) at days 0, 3, and 7 post-transplantation (black arrowhead: hEFs; blue arrowhead: CB-EPCs; and red arrowhead: hESC-ECs). B, Representative Xenogen images for the
detection of DiD-cy5.5–labeled cells in immunodeficient NSG mice (B; n=7) and hu-mice (C; n=7) at days 0, 14, and 28 post-transplantation. D–F, Graphical representation of the bioluminescence intensity of the cell types in each mouse model (black line: hEFs; blue line:
CB-EPCs; and red line: hESC-ECs). Values represent the mean±SD. Student t test: *P<0.05 and **P<0.01.
IFN-γ, we first cocultured each cell type with hu-mice splenocytes and measured IFN-γ production on days 1, 3, and 5.
No significant level of IFN-γ was detected on day 1 in each
coculture group compared with the control splenocyte culture (Figure 3A). However, IFN-γ production by splenocytes
cocultured with hEFs significantly increased compared with
the control group on day 3 (2.0-fold increase; Figure 3A).
Splenocytes cocultured with CB-EPCs and hESC-ECs also
exhibited a significant elevation in IFN-γ production compared with the control group (1.3- and 1.2-fold increases,
respectively), but this elevation was less dramatic than that
observed in splenocytes cocultured with hEFs (Figure 3A).
On day 5, splenocytes cocultured with hEFs and CB-EPCs
demonstrated 2.9- and 1.6-fold increases in IFN-γ production,
Yang et al Evaluation of hESC-ECs in hu-Mice 2843
Downloaded from http://atvb.ahajournals.org/ by guest on June 15, 2017
Figure 3. Comparison of the in vitro immunogenicities of each
cell type in relation to interferon-γ (IFN-γ) production and the
expression of major histocompatibility complex (MHC) class
molecules. A, Analysis of IFN-γ production in hu-mice splenocytes after a 1-way mixed lymphocyte reaction by ELISA. Each
cell type (2×105 cells) was treated with mitomycin C (25 mg/mL)
and cocultured with 2×105 hu-mice splenocytes. IFN-γ production
was measured on days 1, 3, and 5. A gradual increase in IFN-γ
production was observed after 5 days for the human embryonic
fibroblast (hEF) and cord blood–derived endothelial progenitor cell
(CB-EPC) coculture groups compared with the control splenocyte
culture. About the human embryonic stem cell–derived endothelial
cell (hESC-EC) coculture group, a slight increase in IFN-γ production was observed on day 3, but there was no significant difference in production on day 5. B, Each cell type (2×105 cells) was
treated with mitomycin C (25 mg/mL) and cocultured with 2×105
CD3+ T-cell populations of hu-mice splenocytes. IFN-γ production was measured on days 5. IFN-γ production by CD3+ T cell
was measured by ELISA. C, Analysis of the expression levels of
the human MHC molecules (human leukocyte antigen [HLA]-ABC
and HLA-DR) on the cell surface of hEFs, CB-EPCs, and hESCECs using fluorescence-activated cell sorting. Data are representative as the mean fluorescence intensity±SD.
respectively, compared with the control group (Figure 3A).
Interestingly, IFN-γ production by splenocytes cocultured
with hESC-ECs was not significantly different from that of
the control culture (Figure 3A). After that, T cells (CD4+ and
CD8+ cells) were confirmed by intracellular IFN-γ staining
as a major source of IFN-γ within the splenocyte (Figure VIII
in the online-only Data Supplement) and IFN-γ production
by CD3 T cells cocultured with hEFs significantly increased
compared with cells cocultured with CB-EPCs and hESC-ECs
group on day 5 (Figure 3B). Next, we analyzed the expression
level of MHC class I (human leukocyte antigen [HLA]-ABC)
and class II (HLA-DR) molecules on each cell type by fluorescence-activated cell sorting and found that hEFs expressed
MHC class I molecules (mean fluorescence intensity [MFI],
414.18±15.59) at a level that was significantly higher than
that observed for CB-EPCs and hESC-ECs (MFI, 67.95±2.58
and 69.2±3.52, respectively; Figure 3C). As expected, all cell
types were devoid of MHC class II molecule expression (hEFs:
MFI, 4.19±0.21; CB-EPCs: MFI, 3.84±0.58; and hESCECs: MFI, 4.21±0.1), as these molecules are only expressed
on antigen-presenting cells, such as B cells, macrophages,
and dendritic cells.24,25 Moreover, to confirm the relationship between MHC class I molecule expression and immunogenicity, we have established lentiviral β2 microglobulin
(β2M)–specific shRNA expression system in hEF cells for the
MHC class I–suppressed situation. hEF cells stably expressing β2M-specific shRNA reduced β2M surface expression by
≈70% in culture compared with the expression rate in control
virus–transduced cells (Figure IXA in the online-only Data
Supplement). Also from the result of IFN-γ secretion assay,
hu-mice splenocytes were incubated with several target cells,
such as hEF cells, control lentivirus–transduced hEF cells
(hEF-shControl), and β2M-specific shRNA-expressed hEF
cells (hEF-shβ2M). In the presence of hEF cells presenting
normal levels of MHC class I expression, hu-mice splenocytes
produced higher IFN-γ levels compared with the level when
incubated with MHC class I–suppressed cells (Figure IXB in
the online-only Data Supplement). These results suggest that
there is relationship between MHC class I expression level
and immunogenicity in hu-mice. From these observations,
we hypothesized that the different survivabilities of each cell
type may be related to the induction of the cellular immune
response (IFN-γ secretion), and we concluded that the low
immunogenicity of the allogenic hESC derivatives was consistent with their low level of MHC class I expression.
Comparison of Therapeutic Efficacies
for Improving Blood Flow in Models of
Hindlimb Ischemia in NSG and hu-Mice
Translational studies of CB-EPCs and hESC-ECs have shown
their potential to be used in cellular therapies to treat various forms of ischemic diseases.2,21,26,27 However, these studies
could not predict accurate therapeutic outcomes in humans,
as these observations were based in immunodeficient mouse
models. In an attempt to provide a more reliable prediction,
we assessed the therapeutic efficacy of CB-EPCs and hESCECs after transplantation of these cells into hu-mice after
surgically creating hindlimb ischemia (Figure 4A). Before
injecting the cells into the ischemic region, blood perfusion
was measured at day 0 to verify that the surgery had been successful in generating hindlimb ischemia (Figure 4B–4F). NSG
mice injected with cells into the ischemic region were used
as controls, and we compared functional recovery by examining blood flow in the ischemic region using a Laser Doppler
2844 Arterioscler Thromb Vasc Biol December 2013
perfusion imaging system. hEFs were ineffective at improving
blood flow when transplanted into the ischemic region of humice compared with NSG mice (Figure 4B). In NSG mice, a
significant improvement in blood flow was observed 14 days
post-transplantation (≈23%), but no further improvements
were observed when blood flow was measured again on day
28 (≈29%; Figure 4C). In hu-mice, no significant improvement in blood flow was observed during the course of treatment (Figure 4C), which are similar to the result of phosphate
buffered saline-only treated group without cells (Figure VI in
the online-only Data Supplement).
CB-EPCs were shown to be more effective at recovering
restricted blood flow in the ischemic regions of NSG and
hu-mice. We observed a noticeable increase in blood flow
at 7 days post-transplantation (≈12%) when CB-EPCs were
transplanted into NSG mice (Figure 4D), and this blood flow
was shown to increase further at a steady rate for an additional 2 weeks (from ≈12% to ≈62% between 7 and 21 days
Downloaded from http://atvb.ahajournals.org/ by guest on June 15, 2017
Figure 4. Therapeutic efficacy of each cell type to improve blood flow during hindlimb ischemia in NSG and humanized mice (hu-mice) after
transplantation. A, Schematic depicting the transplantation of the 3 cell types into each mouse model after the surgical generation of hindlimb
ischemia. Representative images and graph showing the improvement in blood flow in the ischemic regions of each mouse group after
human embryonic fibroblast (hEF; B and C), cord blood–derived endothelial progenitor cell (CB-EPC; D and E), and human embryonic
stem cell–derived endothelial cell (hESC-EC; F and G) transplantation. The blood flow was analyzed every 7 days after transplantation for
28 days. Values represent the mean±SD. Student t test: *P<0.05 and **P<0.01.
Yang et al Evaluation of hESC-ECs in hu-Mice 2845
Downloaded from http://atvb.ahajournals.org/ by guest on June 15, 2017
post-transplantation; Figure 4E) before reaching a plateau at
day 28 post-transplantation (≈67%; Figure 4E). A reduction in
the therapeutic efficacy of CB-EPCs was observed when these
cells were transplanted into the ischemic region of hu-mice. In
these animals, the blood flow remained unchanged at 7 days
post-transplantation but increased up to ≈13% by 14 days
post-transplantation (Figure 4E). This improvement was comparable with that observed in NSG mice at 7 days post-transplantation with CB-EPCs, indicating a delayed therapeutic
effect in the hu-mice (NSG≈12% of 7 days and hu-mice≈13%
of 14 days; Figure 4E). In addition, the improved blood flow
observed at 14 days post-transplantation in hu-mice was significantly less than that detected in NSG mice at the same time
point (NSG≈33% and hu-mice≈13%; Figure 4E). CB-EPCs
continued to contribute to blood flow improvement in hu-mice
until 28 days post-transplantation, but this improvement was
significantly less than that observed in NSG mice (NSG≈67%
and hu-mice≈42%; Figure 4E).
Next, we compared the therapeutic efficacy of hESC-ECs
using the same mouse models (Figure 4F). Remarkably,
we observed a ≈16% increase in blood flow at 7 days posttransplantation of hESC-ECs into NSG mice, which was significantly greater than that observed after hEF and CB-EPC
transplantation (Figure 4G). From this time on, blood flow
increased sharply at a steady rate and achieved ≈73% limb
perfusion by 21 days post-transplantation before reaching a
plateau with ≈80% limb perfusion on day 28 post-transplantation (Figure 4G). In comparison, the therapeutic effects of
hESC-ECs transplanted in hu-mice were less dramatic; the
improvement in blood flow was gradual during the initial
14-day period after transplantation (≈16%; Figure 4G) but
was followed by a sharp increase in blood flow until day 28
post-transplantation (≈70%; Figure 4G). Interestingly, in line
with the analysis of the therapeutic effects of CB-EPCs in both
mouse models, the percentages of limb perfusion at each measured interval in the hu-mice were less than those observed in
NSG mice (Figure 4G). However, unlike CB-EPCs, the eventual improvement in blood perfusion mediated by hESC-ECs
on day 28 post-transplantation was relatively high in hu-mice
compared with NSG mice (Figure 4G).
hESC-EC Transplantation Improved
Tissue Regeneration in Ischemic Limbs
The histological examination of muscle degeneration and
fibrosis involved harvesting ischemic limbs at 4 weeks posttransplantation (Figure 5A), and hematoxylin and eosin staining of the hEF-transplanted limbs demonstrated massive
muscle degeneration and the presence of abnormal structures.
In contrast, muscles from the hESC-EC and CB-EPC transplantation groups had degenerated only slightly, especially
those in the hESC-EC group, which exhibited near-normal
structures (Figure 5A, first row). Furthermore, Masson’s trichrome staining of the hEF group showed serious muscle fibrosis, whereas the muscle tissue damage in the hESC-EC group
was markedly attenuated (Figure 5A; second row). Arteriole
and capillary enrichment was analyzed by performing mouse
endothelial cell antigen, smooth muscle actin, and platelet
endothelial cell adhesion molecule (PECAM) stains on the
harvested tissues and using normal limb muscles as a control
(Figure 5A, third, fourth, and fifth rows, respectively). These
immunohistochemical examinations and the quantification of
(mouse endothelial cell antigen)-stained vessels revealed that
the implantation of hESC-ECs inhibited vessel damage after
ischemia (35.85±9.73/mm2) when compared with the implantation of CB-EPCs (31.57±7.27/mm2; Figure 5B). In addition,
immunohistochemical staining for smooth muscle actin and
the quantification of arteriole density demonstrated significantly enhanced arteriole formation mediated by hESC-ECs
(23.57±13.53/mm2) compared with CB-EPCs (14.28±7.22/
mm2; Figure 5C). Staining for PECAM and the quantification of capillary density also revealed dramatic improvements
in mice treated with hESC-ECs (118.14±65.95/mm2) compared with those treated with CB-EPCs (86.71±49.32/mm2;
Figure 5D). These quantification results revealed that transplantation of hESC-ECs dramatically enhanced smooth muscle actin-positive arteriole formation and PECAM-positive
density in ischemic regions compared with the transplantation
of hEFs or CB-EPCs.
Next, to investigate the engraftment of transplanted Dillabeled hESC-ECs, we fluorescently stained ischemic limb
tissues. Staining with BS-1 lectin revealed that transplanted
hESC-ECs could be detected in capillaries near muscle tissues in ischemic regions and that these cells had been incorporated into the vessels between muscle tissues (Figure 5E;
white arrowhead). This result indicated that neovessels
induced by hESC-EC transplantation were functional blood
vessels that contributed to blood perfusion. PECAM immunostaining showed that Dil- and PECAM-positive cells were
present (Figure 5F), and the capillary networks comprised
PECAM-positive transplanted cells (Figure 5F, white arrowhead) and PECAM-positive nontransplanted cells (Figure 5F,
black arrowhead) in ischemic muscles. Moreover, mouse-specific smooth muscle actin and human-specific human nuclear
antigen staining confirmed the presence of hybrid blood vessels composed of transplanted hESC-ECs (Figure 5G and 5I,
white arrowhead) and mouse ECs (Figure 5G, black arrowhead). Collectively, hESC-ECs survived after transplantation,
engrafted into mouse tissue, and induced the formation of vascular networks in ischemic muscles (Figure 5G and 5I; Movie
I in the online-only Data Supplement). These results highlight
the therapeutic contribution of hESC-ECs to neovascularization in hu-mice.
Discussion
The therapeutic efficacy of allogeneic hESC derivatives in
humans remains largely elusive, partly because of the potential for graft rejection by the recipient’s immune system.6
These types of therapy carry significant importance, as Food
and Drug Administration–approved human clinical trials of
hESC derivatives are currently underway (www.advancedcell.
com). It has been accepted widely that the administration of
immunosuppressive drugs can inhibit the complications associated with the immunologic rejection of hESC derivatives
post-transplantation.28 However, this practice is not ideal, as
prolonged use of such drugs can leave patients vulnerable to
opportunistic infections.29 Moreover, it has been reported that
2846 Arterioscler Thromb Vasc Biol December 2013
Downloaded from http://atvb.ahajournals.org/ by guest on June 15, 2017
Yang et al Evaluation of hESC-ECs in hu-Mice 2847
Downloaded from http://atvb.ahajournals.org/ by guest on June 15, 2017
long-term use of these drugs may not be required because of
(1) the low expression levels of MHC class I molecules on
hESCs and their derivatives, which render them less susceptible to immune attack9 and (2) the fact that short-term use
of immunosuppressants at the onset of treatment is sufficient
to induce long-term survival and engraftment of transplanted
hESC derivatives.30 Previous studies have attempted to determine the immunologic potential of hESCs and their derivatives in xenogeneic settings, but because of differences in
physiological and biological processes between humans and
mice, outcomes of these studies have not necessarily provided
a true representation of human outcomes.10,11 Therefore, the
main aim of this study was to evaluate the influence of a functional human immune system on the therapeutic efficacy of
hESC-ECs using hu-mice as an in vivo tool.
All cell types demonstrated dramatically reduced viability
during the first 14 days post-transplantation in hu-mice compared with NSG mice (Figure 2E and 2F). Because the possession of a human immune system is the only difference between
these 2 mouse models, this distinction may account for the differences in the survival of the transplanted cell types. To understand the major cell population related to the immune response
in hu-mice, we performed IFN-γ secretion assay with hu-mice
splenocytes. In this experiment, IFN-γ was detected in the several kinds of immune cells and we found that CD8+ T cell was
the major cell population secreting IFN-γ within a coculture
(Figure VIII in the online-only Data Supplement). However,
there was no detection of IFN-γ–secreting CD56+ NK cell and
CD68+ MØ. Although the immune response mechanisms were
not specifically evaluated in the hu-mice, transplanted cells
seemed to be reduced by CD8+ T cell response. As previously
reported, short-term inhibition of the T-cell–mediated immune
response is important for engraftment of hESCs,30 and T-cell–
specific immunosuppression is known to significantly prolong
xenogenic hESC survival in immunocompetent mice.11 In
addition, NK cells are known to uniquely target cells with low
expression of MHC class I molecules31; however, only T-cell–
deficient animals fail to reject hESCs,32 and human NK cells
do not recognize effectively the hESCs in vitro.33 From these in
vitro results, it seems that human NK cells in hu-mice are not
major immune responsor to hESC transplantations, and that
the naturally low expression level of MHC class I molecules on
hESC derivatives may result in a minimal host cellular immune
response. In this study, we observed that the cell death of hEFs
and CB-EPCs in hu-mice 14 days post-transplantation was
significantly greater than that of hESC-ECs (Figure 2F). As
hEFs and CB-EPCs stimulated 2.9- and 1.6-fold increases in
IFN-γ production by splenocytes from hu-mice, respectively,
compared with control hu-mice splenocytes (Figure 3A), we
hypothesized that the cellular immune response (in terms of
IFN-γ production) may be directly linked with the different
survivabilities of transferred hEFs and CB-EPCs. Therefore, it
is plausible that the increased IFN-γ production observed after
the transplantation of hEFs and CB-EPCs into hu-mice may
have activated a cell-mediated immune response against the
transplanted cell types.
As expected, we observed no significant functional improvement during the period in which the CB-EPCs and hESC-ECs
underwent dramatic reductions in viability (Figure 4E and 4G).
However, when cell survival was more stable, we were able to
observe a stable increase in blood perfusion for the CB-EPCs
and hESC-ECs (Figure 4E and 4G). Strikingly, although the
therapeutic effects of hESC-ECs were observed at a delayed
time point, the eventual improvement in blood perfusion by
28 days post-transplantation was similar to that observed in
NSG mice (Figure 4G). This phenomenon was not observed
with CB-EPCs, as the eventual improvement in blood perfusion in hu-mice was significantly less than that observed in
NSG mice (Figure 4E). However, unlike NSG mice, improvement of blood flow at day 14 was not as notable in hu-mice
when transplanted with all 3 cell types. Figure 2F reveals that
the survivals of injected cells were affected by immune rejection of hu-mice compared with the NSG mice (Figure 2E) for
14 days, which gave rise to differences in therapeutic effects.
Basically, hESC-ECs showed higher survival than CB-EPCs
or hEFs in hu-mice (Figure 2E), which was detected in the
ischemic region at day 28 (Figure 5I). Unfortunately, we could
not evaluate the remaining cell population at the end point post
cell transplantation. However, to provide the therapeutic property of hESC-ECs for significant improvement of blood flow
after 14 days post cell transplantation, we performed ELISA
to measure the secretions of humoral angiogenic factors as a
paracrine ability of cells. As a result, we found that hESC-ECs
secreted higher amounts of vascular endothelial growth factor and angiopoietin-1 in comparison with hEFs or CB-EPCs
(Figure VII in the online-only Data Supplement). Such survival property and paracrine effects of hESC-ECs resultantly
contributed to the recovery of ischemia in hu-mice at end
point, which was similar to that of immunodeficient NSG
mice. In addition, histological analysis of hindlimb ischemic
regions demonstrated that hESC-ECs were more efficient at
attenuating muscle degeneration and fibrosis compared with
CB-EPCs (Figure 5A). Functional engraftment of hESC-ECs
was also observed in the ischemic region of hu-mice, indicating that hESC-ECs are capable of recovering blood perfusion
by directly incorporating into the injured area in the presence
of a human immune system. These findings suggest that the
therapeutic efficacy of CB-EPCs is significantly reduced by
the presence of a human immune system.
Based on the data presented in this study, it is plausible to
suggest that the human immune system reduces the efficacy
of hESC-ECs to restore functionality in diseased conditions
Figure 5 (Continued). Improvement in tissue regeneration in ischemic limbs after human embryonic stem cell–derived endothelial cell
(hESC-EC) implantation. A, Hematoxylin and eosin staining for the analysis of muscle degeneration (first row), Masson’s trichrome staining to detect fibrosis in the ischemic region (second row), and immunohistochemical staining with anti-MECA, anti-smooth muscle actin
(SMA) or anti-PECAM antibodies to quantify the microvessels present in the ischemic tissues (third to fifth rows). B–D, Quantification
of MECA-, SMA-, and PECAM-positive microvessels in the ischemic regions of the hESC-EC group compared with those of the human
embryonic fibroblast (hEF) and normal control groups. Values represent the mean±SEM. Student t test: *P<0.05. E and F, Expression pattern of the endothelial-specific markers BS1-lectin (E) and PECAM (F) in Dil-labeled hESC-ECs. F, Many DiI-positive cells also expressed
PECAM. G, Detection of mouse-specific SMA-expressing cells around DiI-positive cells within the vessel-like structures. I, Localization of
human-specific human nuclear antigen (HNA)-expressing cells in the PECAM-expressing vessels.
2848 Arterioscler Thromb Vasc Biol December 2013
Downloaded from http://atvb.ahajournals.org/ by guest on June 15, 2017
rather than rejecting the graft entirely, as previously thought.
We also think that improved survival, and therefore engraftment, could be achieved if HLA matching is available for
patients receiving hESC treatment.34 Based on the HLA type,
it has been estimated that ≈170 hESC lines would be required
to provide 1 hESC line carrying only a single HLA mismatch
to 80% of the patients in Japan.35 Therefore, the generation of
patient-specific stem cell lines has been proposed in the form
of induced pluripotent stem cells to avoid complications associated with immunogenicity.36 Although induced pluripotent
stem cells are reprogrammed from a patient’s own somatic
cells, 1 study demonstrated graft rejection of mouse induced
pluripotent stem cells, which has prompted more extensive
studies on the immunologic potential of induced pluripotent stem cells for human clinical application.37 Therefore,
the establishment of a hESC bank containing cell lines with
diverse MHC expression presents a potential solution to the
need for safe and efficient cell therapy treatments.38
In conclusion, the current study successfully mimicked the
human immune system in mice through the transplantation of
hematopoietic stem cells and also examined the immunogenic
potential of hESC-ECs. The therapeutic efficacy of the hESCECs was reduced, but compared with hEFs and CB-EPCs, these
cells were shown to be less susceptible to immune responses
and possess somewhat beneficial immune properties. These
results suggest that the application of hESC-ECs for treatment
of ischemic diseases could be achieved with minimal use of
immunosuppressive drugs, and that the integration of hu-mice
for the validation of hESC derivatives should promote the
application of these types of cellular therapy.
Acknowledgments
We thank Hye-Jin Lee (CHA Stem Cell Institute, CHA Hospital) and
Ok-Jung Kim (Medical Science, Boston University) for technical assistance with the histological analysis, Won-Woo Lee (CHA Stem Cell
Institute, CHA Hospital) for technical assistance with the Xenogen
imaging system, and Dr Daekyeong Bae (Chabio & Diostech) for
reviewing the article. In addition, we thank Professor Eui-Cheol Shin
(KAIST Institute for the BioCentury, KAIST) for technical assistance
with IFN-γ intracellular staining and FACS analysis.
Sources of Funding
This research was supported by the Bio & Medical Technology
Development Program of the National Research Foundation, which
is funded by the Republic of Korea (MEST; no. 2012-0006107).
Disclosures
None.
References
1. Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ,
Marshall VS, Jones JM. Embryonic stem cell lines derived from human
blastocysts. Science. 1998;282:1145–1147.
2. Cho SW, Moon SH, Lee SH, Kang SW, Kim J, Lim JM, Kim HS, Kim
BS, Chung HM. Improvement of postnatal neovascularization by human
embryonic stem cell derived endothelial-like cell transplantation in a
mouse model of hindlimb ischemia. Circulation. 2007;116:2409–2419.
3. Moon SH, Kim JS, Park SJ, Lee HJ, Do JT, Chung HM. A system for treating ischemic disease using human embryonic stem cell-derived endothelial cells without direct incorporation. Biomaterials. 2011;32:6445–6455.
4. Hagell P, Brundin P. Cell survival and clinical outcome following intrastriatal transplantation in Parkinson disease. J Neuropathol Exp Neurol.
2001;60:741–752.
5.Charron D, Suberbielle-Boissel C, Al-Daccak R. Immunogenicity and
allogenicity: a challenge of stem cell therapy. J Cardiovasc Transl Res.
2009;2:130–138.
6. Sarić T, Frenzel LP, Hescheler J. Immunological barriers to embryonic
stem cell-derived therapies. Cells Tissues Organs. 2008;188:78–90.
7. Drukker M, Katz G, Urbach A, Schuldiner M, Markel G, Itskovitz-Eldor
J, Reubinoff B, Mandelboim O, Benvenisty N. Characterization of the
expression of MHC proteins in human embryonic stem cells. Proc Natl
Acad Sci USA. 2002;99:9864–9869.
8.Grinnemo KH, Sylvén C, Hovatta O, Dellgren G, Corbascio M.
Immunogenicity of human embryonic stem cells. Cell Tissue Res.
2008;331:67–78.
9.Drukker M, Katchman H, Katz G, Even-Tov Friedman S, Shezen E,
Hornstein E, Mandelboim O, Reisner Y, Benvenisty N. Human embryonic stem cells and their differentiated derivatives are less susceptible to
immune rejection than adult cells. Stem Cells. 2006;24:221–229.
10. Li L, Baroja ML, Majumdar A, Chadwick K, Rouleau A, Gallacher L, Ferber
I, Lebkowski J, Martin T, Madrenas J, Bhatia M. Human embryonic stem
cells possess immune-privileged properties. Stem Cells. 2004;22:448–456.
11. Swijnenburg RJ, Schrepfer S, Govaert JA, Cao F, Ransohoff K, Sheikh AY,
Haddad M, Connolly AJ, Davis MM, Robbins RC, Wu JC. Immunosuppressive
therapy mitigates immunological rejection of human embryonic stem cell
xenografts. Proc Natl Acad Sci USA. 2008;105:12991–12996.
12. Pearson T, Greiner DL, Shultz LD. Humanized SCID mouse models for
biomedical research. Curr Top Microbiol Immunol. 2008;324:25–51.
13. Mosier DE, Gulizia RJ, Baird SM, Wilson DB. Transfer of a functional
human immune system to mice with severe combined immunodeficiency.
Nature. 1988;335:256–259.
14. Hesselton RM, Greiner DL, Mordes JP, Rajan TV, Sullivan JL, Shultz LD.
High levels of human peripheral blood mononuclear cell engraftment and
enhanced susceptibility to human immunodeficiency virus type 1 infection
in NOD/LtSz-scid/scid mice. J Infect Dis. 1995;172:974–982.
15. Hudson WA, Li Q, Le C, Kersey JH. Xenotransplantation of human lymphoid malignancies is optimized in mice with multiple immunologic
defects. Leukemia. 1998;12:2029–2033.
16.Yaccoby S, Barlogie B, Epstein J. Primary myeloma cells growing
in SCID-hu mice: a model for studying the biology and treatment of
myeloma and its manifestations. Blood. 1998;92:2908–2913.
17.Ito M, Hiramatsu H, Kobayashi K, Suzue K, Kawahata M, Hioki K,
Ueyama Y, Koyanagi Y, Sugamura K, Tsuji K, Heike T, Nakahata T. NOD/
SCID/gamma©(null) mouse: an excellent recipient mouse model for
engraftment of human cells. Blood. 2002;100:3175–3182.
18. Shultz LD, Ishikawa F, Greiner DL. Humanized mice in translational biomedical research. Nat Rev Immunol. 2007;7:118–130.
19. Kocher AA, Schuster MD, Szabolcs MJ, Takuma S, Burkhoff D, Wang J,
Homma S, Edwards NM, Itescu S. Neovascularization of ischemic myocardium by human bone-marrow-derived angioblasts prevents cardiomyocyte
apoptosis, reduces remodeling and improves cardiac function. Nat Med.
2001;7:430–436.
20. Kakinuma S, Tanaka Y, Chinzei R, Watanabe M, Shimizu-Saito K, Hara
Y, Teramoto K, Arii S, Sato C, Takase K, Yasumizu T, Teraoka H. Human
umbilical cord blood as a source of transplantable hepatic progenitor cells.
Stem Cells. 2003;21:217–227.
21. Li Z, Han Z, Wu JC. Transplantation of human embryonic stem cell-derived
endothelial cells for vascular diseases. J Cell Biochem. 2009;106:194–199.
22. Park SJ, Moon SH, Lee HJ, Lim JJ, Kim JM, Seo J, Yoo JW, Kim OJ, Kang
SW, Chung HM. A comparison of human cord blood- and embryonic
stem cell-derived endothelial progenitor cells in the treatment of chronic
wounds. Biomaterials. 2013;34:995–1003.
23.Murray HW. Interferon-gamma in infection and immunoparalysis.
Intensive Care Med. 1996;22(Suppl 4):S455.
24. Khandelwal S, Roche PA. Distinct MHC class II molecules are associated
on the dendritic cell surface in cholesterol-dependent membrane microdomains. J Biol Chem. 2010;285:35303–35310.
25. Harding CV, Geuze HJ. Class II MHC molecules are present in macrophage lysosomes and phagolysosomes that function in the phagocytic processing of Listeria monocytogenes for presentation to T cells. J Cell Biol.
1992;119:531–542.
26.Zhang L, Yang R, Han ZC. Transplantation of umbilical cord bloodderived endothelial progenitor cells: a promising method of therapeutic
revascularisation. Eur J Haematol. 2006;76:1–8.
27. Kim JY, Song SH, Kim KL, Ko JJ, Im JE, Yie SW, Ahn YK, Kim DK, Suh
W. Human cord blood-derived endothelial progenitor cells and their conditioned media exhibit therapeutic equivalence for diabetic wound healing. Cell Transplant. 2010;19:1635–1644.
Yang et al Evaluation of hESC-ECs in hu-Mice 2849
28. Toriumi H, Yoshikawa M, Matsuda R, Nishimura F, Yamada S, Hirabayashi
H, Nakase H, Nonaka J, Ouji Y, Ishizaka S, Sakaki T. Treatment of
Parkinson’s disease model mice with allogeneic embryonic stem cells: necessity of immunosuppressive treatment for sustained improvement. Neurol Res.
2009;31:220–227.
29. López MM, Valenzuela JE, Alvarez FC, López-Alvarez MR, Cecilia GS,
Paricio PP. Long-term problems related to immunosuppression. Transpl
Immunol. 2006;17:31–35.
30. Pearl JI, Lee AS, Leveson-Gower DB, Sun N, Ghosh Z, Lan F, Ransohoff
J, Negrin RS, Davis MM, Wu JC. Short-term immunosuppression promotes engraftment of embryonic and induced pluripotent stem cells. Cell
Stem Cell. 2011;8:309–317.
31. Ljunggren HG, Kärre K. In search of the ‘missing self’: MHC molecules
and NK cell recognition. Immunol Today. 1990;11:237–244.
32. Suhr ST, Chang EA, Rodriguez RM, Wang K, Ross PJ, Beyhan Z, Murthy
S, Cibelli JB. Telomere dynamics in human cells reprogrammed to pluripotency. PLoS One. 2009;4:e8124.
33. Tseng HC, Arasteh A, Paranjpe A, Teruel A, Yang W, Behel A, Alva JA,
Walter G, Head C, Ishikawa TO, Herschman HR, Cacalano N, Pyle AD,
Park NH, Jewett A. Increased lysis of stem cells but not their differentiated
cells by natural killer cells; de-differentiation or reprogramming activates
NK cells. PLoS One. 2010;5:e11590.
34. Lee JE, Kang MS, Park MH, Shim SH, Yoon TK, Chung HM, Lee DR.
Evaluation of 28 human embryonic stem cell lines for use as unrelated
donors in stem cell therapy: implications of HLA and ABO genotypes.
Cell Transplant. 2010;19:1383–1395.
35. Nakajima F, Tokunaga K, Nakatsuji N. Human leukocyte antigen matching estimations in a hypothetical bank of human embryonic stem cell lines
in the Japanese population for use in cell transplantation therapy. Stem
Cells. 2007;25:983–985.
36.Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda K,
Yamanaka S. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell. 2007;131:861–872.
37. Zhao T, Zhang ZN, Rong Z, Xu Y. Immunogenicity of induced pluripotent
stem cells. Nature. 2011;474:212–215.
38. Taylor CJ, Bolton EM, Pocock S, Sharples LD, Pedersen RA, Bradley JA.
Banking on human embryonic stem cells: estimating the number of donor
cell lines needed for HLA matching. Lancet. 2005;366:2019–2025.
Downloaded from http://atvb.ahajournals.org/ by guest on June 15, 2017
Significance
Allogeneic transplantation of various human stem cells for cellular therapy has the potential to elicit the patient’s immune response and lead
to graft rejection. Although human stem cells have shown to be advantageous for immune properties in vitro, such observations could not
be determined experimentally in vivo because of ethical and technical constraints. The generation of humanized mice harboring a human
immune system has provided a tool to perform in vivo immunologic studies. Using this model, we found that the survival and therapeutic
potential of human embryonic stem cell–derived endothelial cells are more impactful than cord blood–derived endothelial progenitor cells in
a human immune system environment. Therefore, humanized mice could be used as a preclinical in vivo animal model to search for a good
source for cell therapy and evaluate therapeutic efficacy to predict the outcomes of human clinical trials.
Downloaded from http://atvb.ahajournals.org/ by guest on June 15, 2017
Therapeutic Efficacy of Human Embryonic Stem Cell−Derived Endothelial Cells in
Humanized Mouse Models Harboring a Human Immune System
Heung-Mo Yang, Sung-Hwan Moon, Young-Sil Choi, Soon-Jung Park, Yong-Soo Lee,
Hyun-Joo Lee, Sung-Joo Kim and Hyung-Min Chung
Arterioscler Thromb Vasc Biol. 2013;33:2839-2849; originally published online October 3,
2013;
doi: 10.1161/ATVBAHA.113.302462
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272
Greenville Avenue, Dallas, TX 75231
Copyright © 2013 American Heart Association, Inc. All rights reserved.
Print ISSN: 1079-5642. Online ISSN: 1524-4636
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://atvb.ahajournals.org/content/33/12/2839
Data Supplement (unedited) at:
http://atvb.ahajournals.org/content/suppl/2013/10/03/ATVBAHA.113.302462.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Arteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the
Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for
which permission is being requested is located, click Request Permissions in the middle column of the Web
page under Services. Further information about this process is available in the Permissions and Rights
Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online
at:
http://atvb.ahajournals.org//subscriptions/
1
Materials and Methods
2
3
Mice
4
NOD.Cg-PrkdcscidIl2rgtmlWjl/Sz (NOD-SCID IL2rγnull; NSG) mice were purchased from the Jackson Laboratory
5
(Bar Harbor, ME, USA) and housed under specific pathogen-free conditions in accordance with the Principles
6
of Laboratory Animal Care and the Guide for the Use of Laboratory Animals of Samsung Biomedical Research
7
Institute and CHA University.
8
9
Generation of humanized-mice (hu-mice) with CD34+ cells
10
Umbilical cord blood (UCB) samples were obtained from normal full-term deliveries after receiving informed
11
parental consent, according to the institutional guidelines of CHA University (Seoul, Korea). Purification of
12
CD34+ cells from the UCB and the establishment of hu-mice were performed as previously described 1. Briefly,
13
mononuclear cells were isolated by Ficoll-Hypaque density gradient centrifugation and bound by anti-hCD34
14
immunomagnetic beads (Miltenyi Biotec, Bergisch Gladbach, Germany). Selection of CD34+ cell fractions was
15
accomplished using AutoMACSTM (Miltenyi Biotec) according to the manufacturer’s manual. The cells were
16
then stained with anti-hLin-FITC (Becton Dickinson, Franklin Lakes, NJ, USA) and anti-hCD34-PE (BD
17
PharmingenTM, CA, USA) antibodies. Purity assessments by flow cytometry showed that the cell isolations
18
contained ≥ 95% CD34+ cells in all experiments (Figure 1A). Twenty-four hours after the intraperitoneal
19
injection of liquid busulfan solution (30 mg/kg; Busulfex, Ben Benue Laboratories, Inc., Bedford, OH, USA)
20
into NSG mice, hCD34+ cells (2 x 105 cells in 100 μl PBS) were delivered by intravenous (IV) inoculation. To
21
prevent infection, the subjected mice received 100 mg/L ciprofloxacin (CJ Pharma, Seoul, Korea) in their
22
drinking water for 4 weeks. At the end of the antibiotic treatment, the mice were subsequently used for cell
23
transplantation and disease modeling as hu-mice, and these experiments lasted for 12-16 weeks.
24
25
Flow cytometry analysis
26
The following human-specific monoclonal antibodies (hmAbs), which were purchased from eBioscience (San
27
Diego, CA, USA), were used: anti-CD3-, anti-CD33- and anti-hHLA-ABC-FITC; anti-CD34-, anti-CD31-, anti-
28
CD19-, anti-CD56- and anti-HLA-DR-PE; and anti-CD45-APC. Human lymphocytes in hu-mice were
29
examined through multicolor cytometric analysis using a FACSCalibur flow cytometer (BD Biosciences, San
1
1
Jose, CA, USA). Cell reconstitution was monitored every 4 weeks through the collection of peripheral blood
2
from the tail venous plexus into heparinized tubes, and the blood cells were lysed using BD Pharm Lyse lysis
3
buffer (BD Biosciences). Harvested cells were labeled with fluorescently conjugated hmAbs. At the time of
4
sacrifice, single-cell suspensions were prepared from the spleen, liver, and bone marrow (BM) by mincing the
5
tissues through nylon mesh and flushing the tibiae and femurs with PBS containing 2% FBS (Gibco BRL,
6
Gaithersburg, MD) using a 27-gauge needle. After the final wash, the cells were subjected to flow cytometric
7
analysis. The proportion of each lineage was calculated using CELL Quest (BD Biosciences) software.
8
9
10
hESC culture and endothelial differentiation
2
Undifferentiated hESCs (H9 hESCs, 44+XX)
were grown on mitotically inactivated STO cells in
11
DMEM/F12 (50:50; Gibco) supplemented with 20% (v/v) serum replacement (Gibco) and basic ES medium
12
components, including 1 mM L-glutamine (Gibco), 1% nonessential amino acids (Gibco), 100 mM beta-
13
mercaptoethanol (Gibco), and 4 ng/mL bFGF (Invitrogen, Grand Island, NY). The medium was changed every
14
24 hours, and the hESCs were transferred to new feeder cells every 7 days using dissecting pipettes. To induce
15
the differentiation of hESCs into endothelial cells, the hESCs were allowed to form hEBs in suspension with 20
16
ng/mL BMP4-containing hESCs culture medium for 2 days. Then, the hESCs were split on Matrigel-coated
17
plates and cultured in DMEM supplemented with 10% FBS (Gibco) for 10 days 3. Isolation of ECs from other
18
differentiated cells was achieved through cell sorting using a FACS Vantage flow cytometer (BD Biosciences)
19
with anti-hCD31-PE conjugated antibodies (BD Biosciences)
20
for 3 passages in EGM-2MV (Clonetics, San Diego, CA), which have been identified for cellular characteristics
21
before cells transplantation (Supplementary Figure 1).
4, 5
. The sorted cells were subsequently cultured
22
23
hEFs and CB-EPCs culture
24
Human embryonic fibroblasts (hEFs; IMR90) (ATCC, Manassas, VA) served as a negative cell control, cultured
25
in DMEM containing 10% FBS for optimal proliferation condition 6. In addition, CB-EPCs served as a positive
26
cell control,
27
cultured in EGM-2MV medium (Clonetics)
28
characterization of hEFs and CB-EPCs, respectively (Supplementary Figure 3 and 4).
derived from cord blood samples donated by CHA General Hospital (Seoul, South Korea) and
7
. Before experiments, we have analyzed the cellular
29
2
1
In vitro IFN-γ production assay
2
Mitomycin C (MMC; 50 ug/ml)-treated hEFs, CB-EPCs and hESC-ECs (2 x 105/well) were plated in 12-well
3
plates. Non-adherent splenocytes (SPCs) and hCD3-MACS sorted T cells from hu-mice spleen (2 x 105/well)
4
were plated with and without PMA (10 ng/ml) and ionomycin (1 μg/ml; Sigma-Aldrich, St Louis, MO) in the
5
absence and presence of hEFs, CB-EPCs and hESC-ECs. After 1, 3 and 5 days, secreted IFN-γ was measured
6
using an ELISA kit (BD Biosciences) according to the manufacturer’s instructions.
7
8
Mouse limb ischemia and in vivo imaging
9
Male mice (C57BL/6, NSG and hu-mice; body weight 25-30 g) were anesthetized using rompun (20 mg/kg)
10
and ketamine (100 mg/kg) for ligation of the femoral artery and its branches through a skin incision with 6-0
11
silk (Ethicon, Somerville, NJ). The external iliac artery and all arteries above it were then ligated, and the
12
femoral component was excised from its proximal origin as a branch of the external iliac artery to the distal
13
point where it bifurcates into the saphenous and popliteal arteries 8. Prior to transplantation, hESC-ECs, hEFs,
14
and CB-EPCs were labeled with CM-DiI (DiI, Molecular probes) and DiIC18(5)-DS (DiD-cy5.5, Molecular
15
Probes). In vivo fluorescent images of transplanted cells were captured using a Xenogen IVIS imaging system
16
(Xenogen Corp., Alameda, CA) 9, and the optimal number of cells was determined by examining cell survival
17
using signal measurements of different quantities (Supplementary Figure 2). The optimal number of cells (3 x
18
106 cells/mouse) was suspended in 200 l of DMEM and then injected into the dorsal region of the hu-mice to
19
examine survivability. Subsequently, functionality was investigated by intramuscular injection into four sites of
20
the gracilis muscle in the limbs of the ischemic hu-mice.
21
22
23
Laser Doppler imaging analysis
Laser Doppler imaging analysis was performed as described previously
10
. A laser Doppler perfusion imager
24
(Moor Instruments, Devon, UK) was used to measure blood flow in the hindlimbs on days 0, 7, 14, 21, and 28
25
after treatment. The digital color-coded images were analyzed to quantify blood flow in the region from the knee
26
joint to the toe, and the mean perfusion values were calculated.
27
28
29
Histological and immunohistochemical analyses
For tissue staining, mice were sacrificed, and ischemic limb tissues were retrieved after 4 weeks. Specimens
3
1
were fixed in 10% (v/v) buffered formaldehyde, dehydrated with a graded ethanol series, and embedded in
2
paraffin. The samples were then sliced into 4-m sections and stained with hematoxylin and eosin (H&E).
3
Masson’s trichrome collagen staining was performed to assess the presence of fibrosis in the ischemic tissues.
4
Normal limb muscle that was not surgically modified was used as a positive control. To estimate microvessel
5
density, tissue sections were stained using anti-mouse endothelial cell antigen (MECA) (Millipore), anti-
6
PECAM (DAKO) and anti-SM -actin (SMA) (DAKO). An anti-human specific nucleus (HNA) was used to
7
detect for transplanted human cells (Millipore). The staining signal was visualized using avidin-biotin complex
8
immunoperoxidase (Vectastain ABC kit, Vector Laboratories, Burlingame, CA) and 3,3-diaminobenzidine
9
substrate solution kits (Vector Laboratories). Arterioles were recognized as vascular structures with lumens
10
containing one or more continuous layers of smooth muscle cells 11. Capillaries and arterioles in ischemic areas
11
were counted under a light microscope. Five fields from two muscle samples of each mouse were randomly
12
selected for counting
13
embedded in O.C.T. compound (TISSUE-TEK 4583, Sakura Finetek USA Inc., Torrance, CA), frozen, and cut
14
into 8-m-thick sections at -20°C. The tissue sections were immunofluorescently stained with FITC-labeled BS-
15
1-lectin (Sigma), anti-PECAM (DAKO) or mouse-specific anti-SMA (Millipore) antibodies. The staining
16
signals for PECAM and SMA were visualized with FITC-conjugated anti-rabbit IgG and anti-goat IgG
17
secondary antibodies (Molecular Probes), respectively. All fluorescence images were acquired using a LSM 510
18
META confocal microscope (Carl Zeiss, Inc).
12
. For immunofluorescent staining to trace human ECs, the remaining specimens were
19
20
21
22
Statistical analysis
Quantitative data are expressed as the mean values ± SD. Significant differences were assessed using an
unpaired t-test at a significance level of *p<0.05.
23
24
25
4
1
References
2
3
1.
Choi B, Chun E, Kim M, Kim ST, Yoon K, Lee KY, Kim SJ. Human b cell development and antibody
4
production in humanized nod/scid/il-2rgamma(null) (nsg) mice conditioned by busulfan. J Clin
5
Immunol. 2011;31:253-264
6
2.
7
8
Embryonic stem cell lines derived from human blastocysts. Science. 1998;282:1145-1147
3.
9
2011;32:6445-6455
4.
12
13
Moon SH, Kim JS, Park SJ, Lee HJ, Do JT, Chung HM. A system for treating ischemic disease using
human embryonic stem cell-derived endothelial cells without direct incorporation. Biomaterials.
10
11
Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS, Jones JM.
Levenberg S, Golub JS, Amit M, Itskovitz-Eldor J, Langer R. Endothelial cells derived from human
embryonic stem cells. Proc Natl Acad Sci U S A. 2002;99:4391-4396
5.
Sone M, Itoh H, Yamashita J, Yurugi-Kobayashi T, Suzuki Y, Kondo Y, Nonoguchi A, Sawada N,
14
Yamahara K, Miyashita K, Park K, Shibuya M, Nito S, Nishikawa S, Nakao K. Different differentiation
15
kinetics of vascular progenitor cells in primate and mouse embryonic stem cells. Circulation.
16
2003;107:2085-2088
17
6.
18
19
Suhr ST, Chang EA, Rodriguez RM, Wang K, Ross PJ, Beyhan Z, Murthy S, Cibelli JB. Telomere
dynamics in human cells reprogrammed to pluripotency. PLoS One. 2009;4:e8124
7.
Park SJ, Moon SH, Lee HJ, Lim JJ, Kim JM, Seo J, Yoo JW, Kim OJ, Kang SW, Chung HM. A
20
comparison of human cord blood- and embryonic stem cell-derived endothelial progenitor cells in the
21
treatment of chronic wounds. Biomaterials. 2013;34:995-1003
22
8.
23
24
Couffinhal T, Silver M, Zheng LP, Kearney M, Witzenbichler B, Isner JM. Mouse model of
angiogenesis. Am J Pathol. 1998;152:1667-1679
9.
Moon SH, Kim JS, Park SJ, Lim JJ, Lee HJ, Lee SM, Chung HM. Effect of chromosome instability on
25
the maintenance and differentiation of human embryonic stem cells in vitro and in vivo. Stem Cell Res.
26
2011;6:50-59
27
10.
Cho SW, Moon SH, Lee SH, Kang SW, Kim J, Lim JM, Kim HS, Kim BS, Chung HM. Improvement
28
of postnatal neovascularization by human embryonic stem cell derived endothelial-like cell
29
transplantation in a mouse model of hindlimb ischemia. Circulation. 2007;116:2409-2419
5
1
11.
Madeddu P, Emanueli C, Pelosi E, Salis MB, Cerio AM, Bonanno G, Patti M, Stassi G, Condorelli G,
2
Peschle C. Transplantation of low dose cd34+kdr+ cells promotes vascular and muscular regeneration
3
in ischemic limbs. FASEB J. 2004;18:1737-1739
4
12.
Yang F, Cho SW, Son SM, Bogatyrev SR, Singh D, Green JJ, Mei Y, Park S, Bhang SH, Kim BS,
5
Langer R, Anderson DG. Genetic engineering of human stem cells for enhanced angiogenesis using
6
biodegradable polymeric nanoparticles. Proc Natl Acad Sci U S A. 2010;107:3317-3322
7
8
6
Supplementary information
Manuscript No.
ATVB/2012/300088
Title: Therapeutic efficacy of human embryonic stem cell-derived endothelial cells in
humanized mouse models harboring a human immune system
Supplementary Materials and Methods
Gene silencing with shRNA lentivirus
Lentiviral vectors pLKO.1-puro, and pLKO.1-puro containing MISSION shRNA targeting
β2 microglobulin (shβ2M) or non-targeting shRNAs (shControl) were purchased from
Sigma-Aldrich. Five distinct sequences per gene were assessed for knockdown (clone #3 :
TRCN0000057253, #4 : TRCN0000057254, #5 : TRCN0000057255, #6 : TRCN0000057256,
#7 : TRCN0000057257), shControl sequence not known to target any human genes
(SHC002) served as negative control. The infectious viral supernatants were collected in viral
harvest medium at 24 hours after transfection in HEK293T cells. In order to establish a gene
knockdown model, the hEF cells were infected with shβ2M and shControl lentivirus using 8
μg/ml of polybrene (Sigma-Aldrich) to increase infection efficiency. Infected cells were
selected with 1.5~3.0 μg/ml puromycin (Sigma-Aldrich). Knockdown of β2M was measured
by FACS analysis.
Intracellular staining of IFN-γ and FACS analysis
Mitomycin C (MMC; 50 ug/ml)-treated hEFs, CB-EPCs and hESC-ECs (2 x 105/well) were
plated in 12-well plates. Non-adherent splenocytes (SPCs) from hu-mice spleen (2 x
105/well) were plated with and without PMA (10 ng/ml) and ionomycin (1 μg/ml; SigmaAldrich, St Louis, MO) in the absence and presence of hEFs, CB-EPCs and hESC-ECs. After
5 days, harvested splenocytes were stained with anti-hCD4, -hCD8, -hCD68 and -hCD56
antibody (BD Biosciences). For anti-hIFN-γ antibody intracellular staining was measured
using an ELISA kit (BD Biosciences) according to the manufacturer’s instructions.
Supplementary Figure I. (A) To generate endothelial cells derived from hESCs (hESC-ECs),
we treated the cells with BMP4 for 2 days in suspended conditions and subsequently attached
the cells to Matrigel-coated plates. Cell expressing the endothelial-specific surface marker
CD31 were specifically localized around clusters at day 10 post-plating. (B) To purify the
endothelial cells after hESC differentiation (day 12), we performed FACS on mechanically
isolated clusters to collect CD31-expressing cells. More than 11% of the endothelial cells
were purified and exhibited homogeneous cobblestone-like shapes, which was consistent
with the general morphology of endothelial cells. (C) The chromosomal stability at passage 3
was also determined by staining. (D) To analyze the characterization of hESC-ECs, we
performed immunocytochemistry. After passage 3, purified hESC-ECs were strongly
expressed with endothelial specific markers such as PECAM and vWF. (E) To investigate the
functional properties of hESC-ECs in vitro, we concurrently performed Matrigel and ac-LDL
uptake assays. Most of the cells formed tubule structures that took up ac-LDL in Matrigel
(Fig. 2D).
Supplementary Figure II. Before in vivo experiments, we determined the optimal number
of cells by signal measurements of different quantities using a Xenogen IVIS imaging system.
Supplementary Figure III. (A and B) Phenotypic analysis of CB-EPCs by FACS showed
higher expression of representative EPC-positive markers, such as CD34, CD31, and CD146,
complemented with
lower expression of EPC-negative markers, such as CD45, CD90, and
CD73. (C) Immunocytochemistry showing strong expression of endothelial specific markers
such as PECAM and vWF in CB-EPCs.
Supplementary Figure IV.
hEF analysis for endothelial cell characteristics bymatrigel
assay and FACS analysis. (A) Inability of
(B) The absence of
hEFs to form tubule structures on the matrigel.
endothelial or hematopoietic lineage specific markers in hEFs such as
CD34, CD31, CD146, CD45, and CD73, but strong expression of fibroblast specific marker.
Supplementary Figure V. Proliferative ability of hESC-ECs, hEFs, and CB-EPCs. hESCECs and EPCs were cultured in EGM2MV medium, and hEFs were cultured in DMEM
containing 10% FBS for optimal proliferation condition. Cell number and viability was
measured by a hemocytometer and trypan blue staining, respectively. hESC-ECs exhibited
higher growth rate compared to the CB-EPCs or hEFs when cultured in vitro. After 3 days,
the number of hESC-ECs was approximately 1.5-fold higher than that of CB-EPCs. In
addition, CB-EPCs proliferated less than did hEFs.
Supplementary Figure VI. Therapeutic efficacy of non-cell treated group following
injection of PBS into hu-mice after surgically creating hindlimb ischemia. Representative
images and graph were indicated the ineffective at improving blood flow when PBS injected
into the ischemic region of hu-mice.
Supplementary Figure VII. Examination
levels of angiogenic factors released from hEFs,
CB-EPCs and hESC-ECs, through secretion measurements
of FGF-2, VEGF, and Ang-1 in
the supernatant of cultured media using enzyme-linked immunosorbent assay (ELISA). For
ELISA analysis, the hESC-ECs, CB-EPCs, and hEFs were cultured for 3 days in growth
factor-free medium (EBM, Clonetics). The supernatants were harvested a cell-free solution,
which were analyzed the ELISA for FGF-2, VEGF, and Ang-1 using a Quantikine
Immunoassay Kit (R&D Systems Inc., Minneapolis, MN), according to the manufacturer’s
instructions. All measurements were performed in duplicate from 3 different experiments,
and EBM medium served as a control. As a result, hESC-ECs secreted
VEGF and Ang-1 compared to
high amounts of
CB-EPCs and hEFs. On the other hand, amount of FGF-2
secretion by hESC-ECs were lower in comparison to the other groups.
Supplementary Figure VIII. In vitro IFN-γ secretion assay. Splenocytes (2x105) of humice were co-culture with hEFs, CB-EPCs and hESC-EC cells (2x105) for 5 days.
PHA/Ionomycin treated or co-cultured splenocytes were harvested and stained with hCD4,
hCD8, hCD56, hCD68 and hIFN-γ antibody for FACS analysis.
Supplementary Figure IX. β2 microglobulin (β2M)-specific silencing of MHC class I in
hEF cells. (A) Flow cytometry analysis results obtained with hEF cells transduced with
empty and β2m specific shRNAs expressing lentiviral vectors are shown. Cells were stained
with an anti-β2m PE-labeled antibody was used to evaluate β2m expression (representative
for expression of MHC class I). (B) Prestimulated or non-stimulated hu-mice splenocytes
were incubated with different target cells: hEFs, hEF-empty and hEF-sh β2M (#4) for 5 days.
Splenocyte proliferation was detected with CCK-8.