EMBRYONIC STEM CELLS/INDUCED PLURIPOTENT STEM CELLS
Osteoblasts Derived from Induced Pluripotent Stem Cells Form
Calcified Structures in Scaffolds Both in Vitro and in Vivo
GANNA BILOUSOVA,a,b DU HYUN JUN,a,c KAREN B. KING,d STIJN DE LANGHE,e WALLACE S. CHICK,a
ENRIQUE C. TORCHIA,a,b KELSEY S. CHOW,a,c DWIGHT J. KLEMM,a,c DENNIS R. ROOP,a,b SUSAN M. MAJKAa,c
a
Charles C. Gates Regenerative Medicine and Stem Cell Biology Program, University of Colorado Denver,
Aurora, Colorado, USA; bDepartment of Dermatology, University of Colorado Denver, Aurora, Colorado, USA;
c
Department of Medicine, Cardiovascular Pulmonary Research Laboratory, University of Colorado Denver,
Aurora, Colorado, USA; dDepartment of Orthopaedics, Division of Bioengineering, University of Colorado
Denver, Aurora, Colorado, USA; eDepartment of Pediatrics, Division of Cell Biology, National Jewish Health,
Denver, Colorado, USA
Key Words. iPS cells • Differentiation • Osteoblasts • Mesenchymal lineages
ABSTRACT
Reprogramming somatic cells into an ESC-like state, or
induced pluripotent stem (iPS) cells, has emerged as a
promising new venue for customized cell therapies. In this
study, we performed directed differentiation to assess the
ability of murine iPS cells to differentiate into bone,
cartilage, and fat in vitro and to maintain an osteoblast
phenotype on a scaffold in vitro and in vivo. Embryoid
bodies derived from murine iPS cells were cultured in
differentiation medium for 8–12 weeks. Differentiation
was assessed by lineage-specific morphology, gene expression, histological stain, and immunostaining to detect matrix deposition. After 12 weeks of expansion, iPS-derived
osteoblasts were seeded in a gelfoam matrix followed by
subcutaneous implantation in syngenic imprinting control
region (ICR) mice. Implants were harvested at 12 weeks,
histological analyses of cell and mineral and matrix con-
tent were performed. Differentiation of iPS cells into mesenchymal lineages of bone, cartilage, and fat was
confirmed by morphology and expression of lineagespecific genes. Isolated implants of iPS cell-derived
osteoblasts expressed matrices characteristic of bone,
including osteocalcin and bone sialoprotein. Implants were
also stained with alizarin red and von Kossa, demonstrating mineralization and persistence of an osteoblast phenotype. Recruitment of vasculature and microvascularization
of the implant was also detected. Taken together, these
data demonstrate functional osteoblast differentiation
from iPS cells both in vitro and in vivo and reveal a
source of cells, which merit evaluation for their potential
uses in orthopedic medicine and understanding of molecular mechanisms of orthopedic disease. STEM CELLS 2011;
29:206–216
Disclosure of potential conflicts of interest is found at the end of this article.
INTRODUCTION
Mouse and human fibroblasts can be reprogrammed into an
ESC-like state by transduction with a combination of transcription factors (Oct 3/4, Sox2, Klf4, and c-Myc or alternatively Oct3/4, Sox2, Nanog, and Lin28) [1–5]. The induced
pluripotent stem (iPS) cells resulting from this manipulation
function in a manner indistinguishable from mouse ESCs. For
example, iPS cells are capable of differentiating into cell
types characteristic of the three germ layers in vitro and
in vivo, they express many of the markers associated with pluripotent cells, and they have an epigenetic status similar to that
of ESCs [1–5]. Although the differentiation potential of iPS
cell lines compared with that of ESC lines varies between similar to less potent depending on the lines tested and the directed
differentiation analyses performed [6], the accessibility and
therapeutic potential of patient’s own iPS cells provides a
G.B.: conception and design, collection and/or assembly of data, provision of study material or patients data analysis and interpretation,
manuscript writing; D.H.J.: conception and design, collection and/or assembly of data, data analysis and interpretation; K.B.K.:
provision of study material or patients, data analysis and interpretation, final approval of manuscript; S.D.L.: collection and/or assembly
of data; W.S.C.: conception and design; E.C.T.: collection and/or assembly of data, data analysis and interpretation; K.S.C.: collection
and/or assembly of data, data analysis and interpretation; D.J.K.: collection and/or assembly of data, data analysis and interpretation;
D.R.R.: financial support, provision of study material or patients; S.M.M.: conception and design, financial support, collection and/or
assembly of data, data analysis and interpretation, manuscript writing, final approval of manuscript. G.B. and D.H.J. contributed equally
to this article.
Correspondence: Susan M. Majka, Ph.D., University of Colorado Denver Health Sciences Center, Gates Center for Regenerative
Medicine and Stem Cell Biology, 12,800 E 1nineth Ave, P.O. Box 6511, Mail stop 8320, Aurora, Colorado 80045, USA. Telephone:
303-724-3957; Fax: 303-724-3051; e-mail: [email protected] Received August 3, 2010; accepted for publication November
C AlphaMed Press 1066-5099/2009/$30.00/0 doi: 10.1002/
9, 2010; first published online in STEM CELLS EXPRESS November 23, 2010. V
stem.566
STEM CELLS 2011;29:206–216 www.StemCells.com
Bilousova, Jun, King et al.
207
powerful tool for cell-based regenerative medicine, including
bone reconstructive surgery and other orthopedic procedures.
Although bone exhibits a remarkable regenerative
capacity, aging, disease, or injury often results in significant
bone loss, preventing natural replacement of this tissue in the
organism. Although bone autograft provides the best clinical
outcome for bone replacement therapy [7], it is associated
with severe pain at the site of removal and high morbidity
[8]. In turn, allogeneic transplants carry not only the risk of
immunological rejection but also the transmission of disease
from donor to recipient [9]. Autologous mesenchymal stem
cells (MSCs) derived from bone marrow offer a promising
source of cells for musculoskeletal regeneration because of
their potential to differentiate into bone, cartilage, and fat, as
well as their potent paracrine anti-inflammatory properties
[10, 11]. However, the use of MSCs in orthopedic reconstructive therapy may be restricted by their proliferative potential,
which significantly decreases with age [12]. Because of their
self-renewal capacity, patient-specific iPS cells would address
this limitation by providing an unlimited source of MSCs.
Recently, Marion et al. [13] demonstrated that iPS cells generated from both young and aged individuals have elongated
telomeres, and their telomers acquire the characteristics of
ESCs. Agarwal et al. [14] showed the importance of telomerase in the maintenance of iPS self-renewal. The MSC potential of iPS was described by Lian et al. [11]. The iPS cellderived mesenchymal cells were able to attenuate the injury
associated with hindlimb ischemia in a rodent model and to
contribute to tissue regeneration to a greater degree than bone
marrow-derived MSCs (BM-MSCs); however, anti-inflammatory properties of these iPS cell-derived MSCs remained
undefined in this study. Additional studies have also shown
that iPS cells can be differentiated into skeletal muscle, adipocytes, and vascular lineages in vitro [15–19]. Thus, similar to
BM-MSCs, iPS cell-derived MSCs may be used as an autologous graft in situations of fracture nonunion, osteoarthritis,
and intraoral defects to repair bone and cartilage and potentially reduce inflammation. Significant progress has recently
been made with regard to iPS cells in musculoskeletal regenerative medicine; however, the in vivo regenerative potential
of these iPS cell-derived lineages has not been addressed.
Several obstacles still have to be overcome before iPS
cells can be studied as a potential therapy for orthopedic medicine. Importantly, the persistence of differentiated phenotypes
in vivo must be demonstrated. Given the potential of pluripotent stem cells to be expanded and differentiate into multiple
lineages similar to ESCs, we hypothesized that clonal iPS
cells capable of generating mesenchymal tissues could differentiate to a functional osteoblast lineage, which when cultured
on a scaffold would give rise to ectopic mineralized tissue
nodules in vivo. Toward this hypothesis, we performed
directed differentiation of iPS cells to mesenchymal cells and
subsequently to the osteoblast lineage in vitro. These osteoblasts were seeded on gelatin scaffolds and studied in vitro
and in vivo using syngenic mice. The osteogenic scaffolds
were evaluated for stability of the osteoblast phenotype, proliferation, and osteogenic matrix production.
MATERIALS
AND
METHODS
Generation of iPS Cells
Mouse iPS cells were generated by transducing freshly isolated
ICR mouse dermal fibroblasts (Harlan Laboratories, Indianapolis,
IN) with four retroviruses encoding murine Oct3/4, Klf4, Sox-2,
and c-Myc as originally reported by Yamanaka [2]. Mouse Oct3/
www.StemCells.com
4, Klf4, and Sox2 pMxs retroviral vectors were obtained from
Addgene (Cambridge, MA), whereas the c-Myc construct was
generated by cloning mouse c-Myc cDNA into a MSCV-ires-GFP
vector (pMIG, Addgene). Viruses were prepared by transient
transfection of Phoenix-E cells together with the pCL-Eco packaging vector. Two days post-transduction, the transduced fibroblasts were plated onto a mouse embryonic fibroblast feeder cells
treated with mitomycin C (Sigma-Aldrich, St. Louis, MO) and
cultured in ESC medium containing Knockout Dulbecco’s modified Eagles medium (KO-DMEM) supplemented with 2 mM GlutaMAX-I, 0.1 mM minimum essential medium non essential
amino acids solution, 0.1 mM b-mercaptoethanol (Invitrogen,
Carlsbad, CA), 15% ESC tested fetal bovine serum (FBS; Tissue
Culture Biologicals, Seal Beach, CA), and 1,000 U/ml LIF
(ESGRO, Millipore, Billerica, MA) and in the presence of 0.5
mM valproic acid [20] (Chemicals purchased from Sigma-Aldrich
unless otherwise stated) until the ESC-like looking colonies
appeared in about 2 weeks.
Teratoma Formation
Mouse iPS cells (5 105 cells) were injected subcutaneously
into the flank of 6-week-old Foxn1nu nude mice (The Jackson
Laboratory, Bar Harbor, ME). The animals were monitored for 1
month for tumor formation.
Differentiation of iPS Cells into
Mesenchymal Tissues
Based on a modified protocol for embryoid body (EB) formation
[21–23], mouse iPS cells were detached from a feeder layer and
formed on a Petri dish in ESC medium without LIF. After 2
days, EBs were cultured in ESC medium in the presence of 107
M all-trans retinoic acid (ATRA) for 2 days. After 2 days, the
EBs were collected, cells dissociated via brief trypsinization, and
analyzed by flow cytometry or replated for 1 day on gelatincoated plates to promote commitment toward the mesenchymal
osteoblast, chondrocyte, and adipogenic lineages (3 days total)
[21–24]. This medium was replaced with lineage differentiation
medium for 3–4 weeks. In the case of adipocyte and osteoblast
differentiation, EB outgrowths were used in for differentiation
studies, whereas the remaining three-dimensional structures were
washed from the culture when the differentiation medium was
applied. To promote adipocyte differentiation, cells were cultured
in DMEM containing 10% fetal bovine serum supplemented with
500 lM isobutylmethylxanthine, 1 lM dexamethasone, and 100
U/ml Humulin (human recombinant insulin). Thereafter, the cells
were refed every 48 hours with this media supplemented with 1
lM troglitazone for 4 weeks when distinct lipid droplets were
observed by microscopy. Chondrocyte differentiation was performed by trypsinizing EB to a single cell suspension, diluting
cells to a final concentration of 2.5 105 cells/ml and forming
micromass pellets by centrifugation. Micromass was cultured as
nonadherent spheres in 15-ml conical tubes for 4 weeks. Media
was changed every 2 days. Differentiation media consisted of
high glucose (4.5 g/l), DMEM as base medium, and supplemented
with 10% FBS, 1 mM dexamethasone, 17 mM ascorbate-2-phosphate, 35 mM L-proline,1 mM sodium pyruvate, 1 insulin-transferrin-selenium, and 10 ng/ml TGF-b3 (R&D Systems, Minneapolis, MN) [25]. iPS cell-derived EBs were differentiated to the
osteoblast lineage by culturing cells to high density (90% confluence) followed by incubating with differentiation medium consisting of DMEM low glucose (Invitrogen), 10% FBS, 5% Pen/Strep,
1 mM dexamethasone, 0.1 M ascorbic acid, 1 M glycerol 2 phosphate. Media was changed every 2 days and 14–17 days later and
the differentiation documented by Alizarian red and von Kossa
staining. Differentiation was analyzed after 1, 4, and 8 weeks.
Lineage Phenotyping
Histochemical stains were performed as follows: Hematoxylin
stain (Vector Labs, Burlingame, CA) was applied directly on sections for 30 seconds followed by a 10-second rinse in water.
208
Sections were then counterstained with eosin Y (Thermo Scientific) as per standard techniques. In Alizarian red staining, cells
were first fixed in 10% formalin for 5 minutes, washed twice
with water, incubated with alizarian red solution for 10 minutes,
then rinsed several times in water, and allowed to dry. The von
Kossa staining cells were rinsed in water and stained with silver
nitrate for 30 minutes under a UV light. Cells were then rinsed in
water and allowed to dry. In Alcian blue staining, chondrocyte
micromass was fixed in 4% paraformaldehyde overnight. Micromass was then sectioned and slides were incubated with 1%
Alcian blue solution for 30 minutes. Slides were then washed
with 0.1 N HCl and allowed to dry.
Immunostaining of iPS cells was performed, where cells were
fixed in cold methanol (20 C) for 5 minutes and saturated with
phosphate-buffered saline containing 10% bovine serum albumin
(Sigma-Aldrich). The cells were then incubated with anti-Nanog
(Chemicon, Temecula, CA) antibody overnight at 4 C and secondary Alexa-conjugated fluorochromes 594 anti-rabbit antibody
(Invitrogen) for 1 hour at room temperature.
Tissues and micromass chondrocyte cultures were fixed in
4% paraformaldehyde, embedded in paraffin, and sectioned. For
immunofluorescence analysis, antigens were retrieved by boiling
sections in 10 mM sodium citrate for 10 minutes. The prepared
sections were then incubated with primary antibody overnight at
4 C and secondary antibody for 1 hour at room temperature and
mounted with hard set mounting medium (Vector Laboratories,
Burlingame, CA). The primary antibodies used were against keratin 14, Krt14 [26], cytokeratin Endo-A (TROMA-1), adult skeletal muscle myosin heavy chain (A4.1025; Developmental Studies
Hybridoma Bank, Iowa City, IA), and aggrecan (Santa Cruz Biotechnology, Santa Cruz, CA).
Implants were isolated as described, embedded in optimal
cutting temperature compound (OCT) and snap frozen in liquid
nitrogen. Frozen sections were cut at 25 lM followed by fixation
in 4% paraformaldehyde. Sections were incubated with primary
antibodies (bone sialoprotein [BSP] #WVID1(9C5) and osteocalcin clone M-15; Santa Cruz Biotechnology, Santa Cruz, CA) for
12 hours, followed by a 3-hour incubation with fluorescently labeled secondary antibody. The secondary antibody conjugates
used were Alexa-conjugated fluorochromes 594 or 488 antiguinea pig, anti-rat, and anti-mouse (Invitrogen). Coverslips were
mounted using Vectashied and nuclei identified with diamidino-2phenylindole (DAPI) (Vector Labs, Burlingame, CA).
Flow cytometry was performed to detect mesenchymal
markers CD 90, 73, 105, 106, and 133 and lack of the hematopoietic markers CD45 and c-kit as previously described [27] using a
Beckman Coulter Cyan ADP. Laser lines and emission filters
included 488 to detect fluorescein isothiocyanate using the 530/40
filter, phycoerythrin using 575/25 filter and 635 nm laser line to
detect allophycocyanin using the 665/20 filter. Gating strategy
included forward scatter/side scatter (FSC/SSC), doublet discrimination, live dead using DAPI, and single color analysis to detect
each marker. Negative gates were set to an unstained control and
compensation performed using beads (BD Pharmingen, Franklin
Lakes, NJ).
Gene Expression Analysis
Total RNA was extracted from fibroblasts, mouse CJ7 ESCs, iPS
cells, EB, and differentiated cells using RNeasy kit (Qiagen) per
manufacturers protocol. Reverse transcription of RNA to cDNA
followed the manufactures protocol (Invitrogen). Semiquantitative
polymerase chain reaction (PCR) reactions were conducted using
Taq Gold (Applied Biosystems, Foster City, CA) with 1 ll of
cDNA for 30 cycles. Primers used to detect the expression of endogenous nanog, oct3/4, and sox2 were previously published [2].
Glyceraldehyde-3-phosphate dehydrogenase (Gapdh) was detected
using the following oligos: 50 -CGTCCCGTAGACAAAATGGT-30
and 50 -TCTCCATGGTGGTGAAGACA. For quantitative (q)
PCR detection of Col1a1 (mm00801666_g1), spp1 (mm01204014_
m1), runx2 (mm00501584), sox9 (mn00448840_m1), acan
Derivation of Functional Osteoblasts from iPS Cells
(mn00545807_m1), flt-1 (mm01210866_ml), and pecam/CD31
(mm01242584_m1; Applied Biosystems), 50 ng of cDNA was
analyzed in triplicate under using the Light Cycler 480 System
(Roche Diagnostics, Basel, Switzerland). Levels were normalized
to Gapdh abundance (Applied Biosystems).
Three-Dimensional Culture and In Vivo
Analyses of Osteoblast Phenotype
Seeding of the scaffold was performed by cutting Gelfoam surgical sponges (Pfizer Pharmaceuticals) into one centimeter squares
using sterile technique. The sponges were impregnated with bone
differentiation medium. Differentiated osteoblasts at 8 weeks
were trypsinized to obtain a single cell suspension. A total of 8
106 cells were suspended in differentiation medium and
sponges added. Cells were allowed to adhere for 12 hours under
routine culture conditions. Sponges were then placed in a conical
tube containing fresh bone differentiation medium. Medium was
replaced every other day until the time of harvest.
For subcutaneous implantation of the scaffold, 12-week-old
ICR mice were purchased from Harlan Laboratories and housed
in the University of Colorado Denver central vivarium under
pathogen-free conditions. All procedures were performed according to the Animal Care and Use Committee guidelines at the
University of Colorado Denver. Mice were anesthetized with
inhaled isoflurane and hair shaved off the back of the recipient
mouse to minimize infection. Using aseptic technique a longitudinal 0.5-cm incision was made in the back of the mouse, the
skin separated from the underlying muscle with forceps, and the
Gelfoam per cell implant placed in this subcutaneous pouch.
The skin was closed with 3-0 nylon suture and tissue glue
applied over the suture to seal. One such pocket was made in
each mouse (using 15 mice). Animals were singly housed for 7
days following implant then housed in groups of 2–3 for the
remaining 12 weeks of the experiment. Undifferentiated iPS
cells produce teratomas, therefore, we did not include a control
group of undifferentiated cells. The groups were performed with
gelfoam controls (minus cells) or gelfoam seeded with osteoblasts 24 hours prior.
RESULTS
Generation of iPS Cells
We generated iPS cells by transducing primary mouse fibroblasts with retroviral vectors encoding four reprogramming
factors (Oct3/4, Sox2, Klf4, and c-Myc) [2]. Similar to the
observation originally made by Yamanaka [2], our iPS cells
exhibited an ESC-like morphology (Fig. 1A) and reactivated
expression of endogenous Oct3/4, Sox2, and Nanog, genes
normally expressed in mouse ESCs and silenced in somatic
cells, as determined by reverse transcription polymerase chain
reaction (RT-PCR; Fig. 1B). The reactivation of Nanog
expression in our iPS cell clones was further confirmed by
immunofluorescence analysis (Fig. 1C). The generated iPS
cells formed teratomas following subcutaneous injection into
nude mice. Tissues from all three germ layers were present in
these tumors as detected by immunofluorescence analysis
(Fig. 1D), thus confirming the pluripotency of our iPS cell
lines. We used Krt14 as a marker for ectoderm, heavy chain
myosin from skeletal muscles for mesoderm and cytokeratin
EndoA for endoderm.
Differentiation of iPS Cells into
Mesenchymal Cell Phenotypes
To determine the potential of a clonal iPS cell line to differentiate into the mesenchymal lineages of bone, cartilage,
Bilousova, Jun, King et al.
209
Figure 1. Generation of mouse iPS cells. (A): Skin fibroblasts from newborn ICR mice were transduced with retroviruses expressing reprogramming factors and cultured under ESC conditions. After 2 weeks in culture, colonies exhibiting an ESC morphology under phase-contrast microscopy emerged. (B): Reverse transcription polymerase chain reaction analysis with primers specific to endogenous mouse Nanog, Oct3/4, and
Sox2, as well as GAPDH as a control, was performed with total RNA extracted from lane 1 (primary ICR fibroblasts), lane 2 (ICR fibroblasts
transduced with four retroviral vectors and cultured for 5 days), lane 3 (mouse ESCs), and lane 4 (mouse iPS cells). (C): iPS cell colonies were
positive for Nanog, as determined by immunofluorescence analysis (red, left panel). The feeder cells used to maintain iPS cells served as a negative control for Nanog immunofluorescence (DAPl, right panel). (D): Teratomas were formed when iPS cells were injected subcutaneously into
nude mice. H&E staining are shown in top panels and consecutive sections labeled with various antibodies representing three germ layers, that
is, ectoderm (Krt14), mesoderm (MyHC), endoderm (cytokeratin Endo-A) are bottom panels. All images were taken with 20 objectives. Abbreviations: DAPI, diamidino-2-phenylindole; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; ICR, imprinting control region mouse; iPS,
induced pluripotent stem cell; MyHC, myosin heavy chain from skeletal muscles.
and fat, we employed differentiation protocols previously
formulated for ESCs. EB differentiated from iPS cells were
treated with retinoic acid in suspension culture to induce
cell commitment toward mesoderm, plated on gelatin followed by culture in lineage-specific differentiation medium
according to Kawaguchi [21, 22] with slight modifications.
www.StemCells.com
Fat differentiation was evident after 4 weeks by visualization
of lipid droplet accumulation (Fig. 2A). Cartilage differentiation was performed by culturing EBs as nonadherent cell
spheres or micromass in chondrogenic medium. Analysis of
mRNA extracted from iPS cells after 5 weeks of differentiation in chondrogenic medium showed expression of the
210
Derivation of Functional Osteoblasts from iPS Cells
obtained from 200 dissociated EB following the 3-day
ATRA treatment to analyze cell surface expression of
markers indicative of a mesenchymal cell. We found that
the cells express varying levels of the mesenchymal markers
CD 90, 73, 105, 106, and 133 and lack the hematopoietic
markers CD45 and ckit (Fig. 2G). Lung MSCs were used as
a positive control (not shown).
The differentiation of iPS cells into osteoblasts was performed by culturing mesodermal-induced iPS in osteogenic
medium for up to 8 weeks. Differentiation of iPS cells to
an osteoblast phenotype from day 0 (following retinoic acid
treatment) to 8 weeks was shown by histochemical staining
using alizarin red to indicate sites undergoing calcium deposition and mineralization (Fig. 3A–3D). Additionally, von
Kossa silver stain and alkaline phosphatase stain were performed at 8 weeks on the differentiated cells to demonstrate phosphate deposition and alkaline phosphatase activity
(Fig. 3E). Both phosphate and alkaline phosphatase activity
were present in the culture. 3T3E1 preosteocyte cells were
cultured in basal media or osteogenic differentiation media
and used to demonstrate the specificity of histochemical
staining (supporting information Fig. 1). Analysis of mRNA
extracted from iPS cells after 0, 1, 4, and 8 weeks of differentiation in osteogenic medium showed induction of the
osteoblast markers, the transcription factor runx2 (cbfa1),
extracellular matrix and structural proteins col1a1 (Collagen
type I pro alpha I), spp1 (bone sialoprotein or osteopontin
[OPN]), and flt-1 (vascular endothelial growth factor receptor 1 [28]; Fig. 3F). Controls for qRT-PCR primers
included a bone marrow as a negative and 3T3 E1 cells
cultured using osteogenic differentiation medium (supporting
information Fig. 2).
These data suggest the possibility of using approaches
developed for ESC differentiation to create and expand mesenchymal lineages such as osteoblasts and chondrocytes from
our iPS cell lines in vitro.
iPS Cell-Derived Osteoblasts Maintain Their
Phenotype in a Scaffold In Vitro and In Vivo
Figure 2. Differentiation of the mesenchymal adipocyte and chondrocyte lineages from iPS cells. Murine embryoid body derived from
iPS cells were treated with all-trans retinoic acid for 3 days to induce
mesoderm followed by the treatment with adiopgenic or chondrogenic
differentiation medium for an additional 4 weeks. Lipid droplets were
visualized in adipocytes using phase-contrast microscopy 10 objective (A). Chondrogenesis was performed in micromass cultures and
documented by quantitative reverse transcription polymerase chain
reaction detection of Sox9 and Aggrecan (B) and by Aggrecan immunostaining 4 objective (D) with DAPI nuclear stain (C). The phenotype of chondrocytes was documented in paraffin sections by H&E
(E) and Alcian blue histochemical stains and photographs taken with
the 20 objective (F). (G). Analysis of mesenchymal marker expression by dissociated iPS-derived embryoid bodies following 3-day
treatment with ATRA. Abbreviations: FSC, forward scatter; iPS,
induced pluripotent stem cell; SSC, side scatter.
chondrogenic differentiation factor, Sox-9, and the chondrocyte matrix protein Acan (Aggrecan; Fig. 2B). Further investigation demonstrated immunohistological staining for aggrecan (Fig. 2C, 2D). Differentiation into a chondrocyte
phenotype was demonstrated by staining with H&E and
Alcian blue, a cationic stain that highlights an extracellular
matrix rich in polyanionic glycosaminoglycans (Fig. 2E, 2F).
We also performed flow cytometric analysis on cells
The next step in the evaluation of iPS cell differentiation to a
bone-like phenotype was the ability to maintain a differentiated osteoblast phenotype when passaged and seeded into a
three-dimensional scaffold both in vitro and in vivo. After 4
weeks, the osteoblast cultures had reached confluence and
expressed high levels of runx2, col1a1, and spp1 (Fig. 3). At
that time, gelfoam carriers, a biodegradable gelatin matrix,
were seeded with 8 106 iPS cell-derived osteoblasts after 8
weeks of differentiation culture and maintained in differentiation media. Implanted sponges were analyzed in vitro after 48
hours and 2 weeks of culture to confirm the presence of cells
using hematoxylin staining (Fig. 4 A, 4B). After 2 weeks,
sponges stained with hematoxylin and alizarin red with a hematoxylin counterstain demonstrated an increase in cell number and secreted matrix (Fig. 4C–4E). Sponges seeded with
iPS cell-derived osteoblasts maintained the expression of
osteoblasts markers runx2, col1a1, and spp1 after 48 hours
with higher levels of col1a1 and spp1 2 weeks of differentiation in osteogenic medium (Fig. 4F).
Following passage osteoblast cells continue to express matrix and react to alizarin red, however, take another week to
increase the gene expression of more differentiated osteoblasts
as demonstrated by analysis of seeding on sponges presented
in Figure 4. Therefore, we waited until 8 weeks when the
osteoblasts had expanded in numbers before seeding gelfoam
for in vivo studies.
Bilousova, Jun, King et al.
211
Figure 3. Functional osteoblasts can be induced from iPS cells in vitro by the presence of dexamethasone and ascorbic acid. (A–D): Murine
embryoid body derived from iPS cells was treated with all-trans retinoic acid for 3 days followed by osteoblast differentiation medium. Commitment to the osteoblast lineage was documented by 4 and 8 weeks, as demonstrated by alizarin red stain, identified the increase in calcific deposit
by osteoblasts (red), and nuclei were visualized with a hematoxylin counterstain (blue). (E): Von Kossa stain (black) and alkaline phosphatase
colorimetric detection (red) were performed to characterize a differentiated osteoblast lineage. All images were taken with a 4 objective. (F): A
temporal increase in expression of runx2, col1a1, and spp1 genes was confirmed by quantitative reverse transcription polymerase chain reaction
analysis. Abbreviations: BM, bone marrow; D5, day 5; DIF, differentiation medium; iPS, induced pluripotent stem cells.
For in vivo analysis, 24 hours after seeding 8 106 iPS
cell-derived osteoblasts on gelfoam cubes, the sponges were
subcutaneously implanted on the dorsal surface of immunocompetent ICR mice. After 12 weeks, no tumor formation
was detected and the implants were extracted and analyzed
www.StemCells.com
for phenotypic markers of osteogenic cells and mineralization.
Implants containing osteoblasts were isolated from 8 of 15
mice. Grossly, the implants were identifiable between the cutaneous and peritoneal muscular layers and appeared to recruit
vasculature (Fig. 5A). Paraffin-sectioned implants (6 lm
212
Derivation of Functional Osteoblasts from iPS Cells
Figure 4. Induced pluripotent stem (iPS) cell-derived osteoblasts maintain their phenotype and deposit bone extracellular matrix on a scaffold in
vitro. One-centimeter cubes of gelfoam sponge were seeded with iPS cell-derived osteoblasts following 8 weeks of differentiation culture. (A, B):
After 48 hours, the gelfoam scaffolds were stained with hematoxylin to confirm the presence of cells. Images were taken with 10 and 20 objective. (C, D): After 2 weeks, the scaffolds were stained with alizarin red and hematoxylin counterstain, which indicated an increase in the cells
present within the scaffold, as well as matrix deposition by the osteoblasts. Images were taken with a 10 and 20 objective. (E): Nodules positive for alizarin red were detectable on gelfoam cubes in vitro after seeding with iPS cell-derived osteoblasts cultured in differentiation medium
for 2 weeks. A representative phase-contrast micrograph is presented. Scale bar ¼ 5 mm. (F): Maintenance or increased expression of runx2, col1a1,
and spp1 genes was documented using quantitative reverse transcription polymerase chain reaction analysis. Abbreviation: BM, bone marrow.
thickness) stained with H&E demonstrate vascularization of
the implant discernable by the presence of red blood cells in
the lumens of microvessels (Fig. 5B, 5C). Additional implants
were flash frozen in OCT and frozen sectioned to 25 lm
thickness to maintain tissue integrity and antigenicity for immunofluorescence. To determine the presumptive origin of the
microcirculation, we performed qRT-PCR analysis of to
detect gene expression indicative of vascular endothelial
differentiation. Using the marker pecam1/CD31, which is
expressed by both progenitors and differentiated endothelium,
we found no evidence of vasculature in the osteoblast cultures
or in vitro sponges (Fig. 5D).
Alizarin red and von Kossa stains localized osteoblasts
secreted matrix and calcium in the implants (Fig. 6A–6C).
Fluorescent immunostaining demonstrated the presence of
osteoblast markers osteocalcin and BSP in serial sections
(Fig. 6D–6G, supporting information Fig. 3). Thus, we
demonstrate that osteoblasts differentiated from iPS cells
Bilousova, Jun, King et al.
213
Figure 5. Subcutaneous implantation of osteoblast seeded scaffolds recruited vasculature in vivo. After 12 weeks, implants were identifiable
(A) as surrounded by vasculature. Paraffin sections stained with H&E demonstrated the subcutaneous localization of the implant (B) and the vascular supply within the implant, as evidenced by the presence of red blood cells in the lumens of microvessels ([C], arrows) using phase-contrast
microscopy. (D) Expression of pecam-CD31 was analyzed to determine whether vascular cells were present in the differentiating iPS osteogenic
cultures. Scale bar ¼ 50 lM. Abbreviations: BM, bone marrow; D5, day 5; iPS, induced pluripotent stem cells.
maintained their lineage-specific phenotype after passage onto
scaffold both in vitro and in vivo. The phenotype of these
osteoblast-derived iPS cells appears stable at both 4 and 8
weeks following directed differentiation. In vivo formation of
mineralized nodules using a mouse model suggests the persistence of a differentiated phenotype and highlights the
potential for these cells to be used in graft development and
cell-based therapy for musculoskeletal repair.
DISCUSSION
Cell-based therapy for repair of bone and cartilage offers a
promising treatment for both degenerative and genetic musculoskeletal diseases. However, engineered grafts are far from
being a standard of treatment for bone and cartilage repair.
BM-MSC, ESCs, and more recently, iPS cells have been demonstrated as having the potential to differentiate to bone and
cartilage, as well as other mesenchymal lineages. Each cell
type offers its own challenges. Although BM-MSC may function as an autologous graft, also suited for gene therapy, and
demonstrate anti-inflammatory properties, when obtained from
aged donors they may demonstrate a limited ability to prolifwww.StemCells.com
erate [11, 29, 30]. Although ESCs are pluripotent and exhibit
high differentiation and proliferation potential, their origin is
the topic of ethical debate as they must be derived from fertilized embryos. In addition, the repair of tissues with embryonic stem (ES)-derived cells may result in immune rejection,
as these cells are allogeneic. Data on immunological properties
of human and murine ESCs and their differentiated derivatives
are still controversial, ranging from those claiming unique
immune-privileged properties for ESCs [31, 32] to those which
refute these conclusions [33, 34]. iPS cells are the most recent
addition to this cast and they offer benefits of the aforementioned cell types yet they are the least well characterized.
Nevertheless, it has been shown that iPS cells function in a
manner similar to ESCs, that is, they are capable of forming
multiple cell types in vitro and in vivo, they express most (but
not all) of the markers associated with pluripotent cells, they
have an epigenetic and telomere status similar to that of ESCs,
and they can be used to make fertile mice [13, 35]. The immunological properties of iPS are not well characterized.
Here, we have assessed the ability of murine iPS cells
to be directionally differentiated to multipotent mesenchymal
cells and subsequently the osteoblast lineage in vitro. The
multipotent mesenchymal cell stage was demonstrated by
differentiation of iPS cell-derived mesenchymal cells into
214
Derivation of Functional Osteoblasts from iPS Cells
Figure 6. Osteoblast-seeded scaffolds maintain their phenotype and
deposit bone extracellular matrix in
vivo. After 12 weeks, gelfoam scaffolds were embedded in OCT, frozen, and 25-mM sections were
analyzed. Histochemical alizarin red
(A, B) and von Kossa (C) staining
confirmed the presence of matrix
and calcium within the nodules and
hematoxylin the presence of osteoblasts. (D–G): Consecutive sections
were immunoreactive to osteocalcin
and BSP. DAPI was used to label
nuclei. Scale bar ¼ 2.5 mm. Abbreviations: BSP, bone sialoprotein;
DAPI, diamidino-2-phenylindole; OCT,
optimal cutting temperature compound.
three characteristic mesenchymal lineages including osteoblast, adipocyte, and chondrocytes. Lian et al. recently demonstrated the ability of human iPS cells to give rise to mesenchymal cells with a cell surface phenotype and
differentiation potential similar to those of BM-MSC [11]
using clinically compliant culture techniques optimized for
ESCs [36] adapted from ES differentiation to MSC using
OP9 feeder cells [37]. While we also employed a differentiation strategy optimized for ESCs, however, the two strategies differ significantly. The differentiation strategy we
chose to evaluate was based on the premise that ATRA
treatment would increase osteoblast and chrondrocyte lineage
commitment of iPS-derived EB outgrowths through runx2
induction [21–24]. ATRA is known to increase neural crest
markers by ESCs that provides an origin for mesenchymal
elements [22]. Runx2 is expressed in neural crest-derived
mesenchyme by precursors to bones, teeth and chondrocytes
[38]. Runx2 deficiency or downregulation results in a shift
toward adipogenesis [39]. We made EB from iPS cells over
2 days using a hanging drop method followed by plating on
gelatin-coated dishes in ES medium with ATRA for 3 days.
After 3 days, the medium was changed to lineage-specific
differentiation medium and outgrowth evaluated for expression of lineage appropriate genes. Conversely, Lian et al.
cultured dissociated iPS cells, without an EB and ATRA
treatment stage, in medium to support MSC outgrowth containing defined factors such as basic fibroblast growth factor,
platelet-derived growth factor AB, and epidermal growth
factor [36, 37]. These culture conditions required 1 week to
detect a MSC phenotype confirmed by flow cytometric analysis to detect cell surface molecules including CD73,
CD105, and CD133 and the lack of hematopoietic markers
CD 45 and CD34 [11]. Lineage-specific differentiation for
both studies was 3–4 weeks. Following a 3-day treatment
with ATRA, we confirmed expression of the mesenchymal
markers CD 90 (50%), lower levels of CD 73, 105, 106,
and 133 and lack of hematopoietic markers CD45 and ckit.
The differences between iPS cell-derived mesenchymal
cells and BM-MSCs provide a compelling argument for the
use of iPS cell engineered cells. Lian et al. [11] demonstrated that iPS cell-derived mesenchymal cells could proliferate to 120 population doublings while maintaining a normal karyotype, had a 10-fold greater level of telomerase
activity and had more of a protective effect in a rodent
model of hindlimb ischemia than BM-MSCs. The difference
between cell types was attributed to the increased ability of
iPS-derived mesenchymal cells to survive, engraft, and promote de novo vasculogenesis and muscle differentiation. Independent groups have also demonstrated successful
directed differentiation of iPS cells into adipocyte, vascular,
Bilousova, Jun, King et al.
and skeletal muscle with efficiencies similar to those of
ESCs [16–19].
We demonstrate for the first time that iPS cells could be
terminally differentiated into functional osteoblasts that
would maintain their phenotype on a three-dimensional gelatin scaffold in vitro and in vivo. As iPS cells are somatic
cells manipulated to become pluripotent stem cells, it is vital
that their potential to terminally differentiate and maintain a
lineage-specific differentiated phenotype, including bone, is
documented in vivo otherwise tumorigenesis may become an
issue. The osteoblasts lineage was defined by increased
expression of the characteristic genes runx2, collagen typeIaI, and BSP/OPN. As anticipated runx2 expression
increased with ATRA treatment, whereas the expression of
flt-1, col1a1, and spp1 genes increased with differentiation
and was maintained over time in vitro [17, 24]. Flt-1 is a
marker of both osteoblasts and osteoclasts [28]. When the
osteoblasts were passaged and seeded on the gelatin scaffold, the early gene runx2 initially increased and was followed after 2 weeks by increased expression of matrix genes
collagen I and BSP in vitro. The osteoblasts production of
collagen may enhance further differentiation and secretion of
more mature osteogenic matrix [40]. Researchers have
shown that to engineer cell-based bone grafts, in vitro commitment or differentiation of the cells used is absolutely
necessary [41, 42]. Duan et al. [43] showed that enamel matrix derivatives increased runx2 expression and may be useful
in periodontal tissue regeneration. Osteogenic matrix production was confirmed by alizarin red stain and von Kossa to
detect mineralization. Isolated implant structures demonstrated
the presence of a microcirculation, which is necessary to support complete osteoblastic cell differentiation and functional
bone tissue generation. The cells present in the implant
formed a matrix, which consisted in part of BSP and osteocalcin. Additionally, remodeling occurred in the implant. Following implantation the osteoblasts were initially spread apart
on the sponge-like scaffold. Histological analysis of the tissue
revealed that the iPS-derived osteoblasts were densely packed
with detectable microvasculature and circulation. As the iPSderived osteoblast cultures lacked the endothelial marker
pecam/CD31, the circulation was likely host derived. The
iPS cell-derived osteoblasts did not form tumors in syngenic
or nonimmunocompromised mice, which indicates a stable
phenotype and no reversion to an embryonic stage of differentiation. These results are in direct contrast to the implantation of undifferentiated iPS or mixtures of undifferentiated
iPS and differentiated iPS-derived cells which under all circumstances when subcutaneously implated formed teratomas.
This study presents evidence that iPS cells can serve as
a potential source of osteoblasts. These iPS cell-derived
osteoblasts increase their expression of bone-specific genes
and osteogenic matrix when seeded on a gelatin scaffold in
vitro and in vivo, demonstrating their potential use in cell-
REFERENCES
1
2
3
Takahashi K, Tanabe K, Ohnuki M, et al. Induction of pluripotent
stem cells from adult human fibroblasts by defined factors. Cell 2007;
131:861–872.
Takahashi K, Yamanaka S. Induction of pluripotent stem cells from
mouse embryonic and adult fibroblast cultures by defined factors. Cell
2006;126:663–676.
Wernig M, Meissner A, Foreman R, et al. In vitro reprogramming of
fibroblasts into a pluripotent ES-cell-like state. Nature 2007;448:
318–324.
www.StemCells.com
215
based therapy. However, future studies are necessary to
extend and confirm these conclusions using additional criteria. First, in addition to in vitro commitment or differentiation of the cells used to engineer a bone graft, the environmental conditions in vivo must promote and support
osteogenesis [41, 42]. Additional factors may be impregnated
into the scaffold, which would enhance bona fide bone formation. Second, the differentiation of ESCs to osteoblasts is
affected or enhanced in culture with various scaffolds, including nanofiber [44, 45]. Ideal scaffold properties that support
terminal differentiation in vivo must be defined. Finally, the
issue of reproducibility of engineered tissues using human
cells still exists. To date, most studies have been performed
using human ESCs and BM-MSC. With multiple cell types
available, the most ideal selection markers for cells used to
seed grafts capable of inducing stable bone must be identified
[46]. Additionally, human iPS cells must be evaluated for the
same potential. The use of integrating virus in human cells
poses problems and risks, however, Somers et al. [47]
recently described the generation of human iPS from various
disease populations using a polycistronic construct which can
be removed form the host genome.
In summary, we performed directed differentiation of iPS
cells to mesenchymal cells and subsequently the osteoblast
lineage in vitro. These osteoblasts were seeded on gelatin
scaffolds where they demonstrated stability of the osteoblast
phenotype, proliferation, and osteogenic matrix production
both in vitro and in vivo. The maintenance of a stable osteoblast phenotype by iPS cell-derived osteoblasts in vivo spotlights these cells as a viable source for further study, in combination with ES and MSCs, for clinical cell-based therapy to
treat musculoskeletal diseases. Additionally, the expansion and
differentiation potential of patient-specific iPS cells also provides the scientific community with a valuable tool to study
the cell-based mechanisms of bone and cartilage diseases.
Taken together, a better understanding of these processes
promise to offer new therapeutic avenues in the long term.
ACKNOWLEDGMENTS
This work was supported by R03HL096382-01 and 1R01
HL091105-01 to S.M.M.; NIH grants AR052263 and AR50252
to D.R.R.; and a research grant from the Dystrophic Epidermolysis Bullosa Research Association (DebRA) International from
DebRA Austria to G.B. and D.R.R. Additional support was provided by DK NIDDK R01-078966.
DISCLOSURE OF POTENTIAL
CONFLICTS OF INTEREST
The authors indicate no potential conflicts of interest.
4
5
6
7
8
Yamanaka S. Strategies and new developments in the generation
of patient-specific pluripotent stem cells. Cell Stem Cell 2007;1:
39–49.
Yu J, Vodyanik MA, Smuga-Otto K, et al. Induced pluripotent stem cell
lines derived from human somatic cells. Science 2007;318:1917–1920.
Vogel G. Reprogrammed cells come up short, for now. Science 2010;
327:1191.
Damien CJ, Parsons JR. Bone graft and bone graft substitutes: A
review of current technology and applications. J Appl Biomater 1991;
2:187–208.
Arrington ED, Smith WJ, Chambers HG et al. Complications of iliac
crest bone graft harvesting. Clin Orthop Relat Res 1996;329:
300–309.
Derivation of Functional Osteoblasts from iPS Cells
216
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Kainer MA, Linden JV, Whaley DN, et al. Clostridium infections
associated with musculoskeletal-tissue allografts. N Engl J Med 2004;
350:2564–2571.
Granero-Molto F, Weis JA, Longobardi L et al. Role of mesenchymal
stem cells in regenerative medicine: Application to bone and cartilage
repair. Exp Opin Biol Ther 2008;8:255–268.
Lian Q, Zhang Y, Zhang J, et al. Functional mesenchymal stem cells
derived from human induced pluripotent stem cells attenuate limb ischemia in mice. Circulation 2010;121:1113–1123.
Stenderup K, Justesen J, Clausen C et al. Aging is associated with
decreased maximal life span and accelerated senescence of bone marrow stromal cells. Bone 2003;33:919–926.
Marion RM, Strati K, Li H, et al. Telomeres acquire embryonic stem
cell characteristics in induced pluripotent stem cells. 2009;4:141–154.
Agarwal S, Loh Y-H, McLoughlin EM, et al. Telomere elongation in
induced pluripotent stem cells from dyskeratosis congenita patients.
Nature 2010;464:292–296.
Li F, Bronson S, Niyibizi C. Derivation of murine induced pluripotent
stem cells (iPS) and assessment of their differentiation toward osteogenic lineage. J Cell Biochem 2010;109:643–652.
Mizuno Y, Chang H, Umeda K, et al. Generation of skeletal muscle
stem/progenitor cells from murine induced pluripotent stem cells.
FASEB J 2010;24:2245–2253.
Tashiro K, Inamura M, Kawabata K, et al. Efficient adipocyte and
osteoblast differentiation from mouse induced pluripotent stem cells
by adenoviral transduction. Stem Cells 2009;27:1802–1811.
Taura D, Noguchi M, Sone M, et al. Adipogenic differentiation of
human induced pluripotent stem cells: Comparison with that of human
embryonic stem cells. FEBS Lett 2009;583:1029–1033.
Taura D, Sone M, Homma K, et al. Induction and isolation of vascular cells from human induced pluripotent stem cells—Brief report.
Arterioscler Thromb Vasc Biol 2009;29:1100–1103.
Huangfu D, Osafune K, Maehr R, et al. Induction of pluripotent stem
cells from primary human fibroblasts with only Oct4 and Sox2. Nat
Biotechnol 2008;26:1269–1275.
Kawaguchi J. Generation of osteoblasts and chondrocytes from embryonic stem cells. Methods Mol Biol 2006;330:135–148.
Kawaguchi J, Mee PJ, Smith AG. Osteogenic and chondrogenic differentiation of embryonic stem cells in response to specific growth factors. Stem Cells 2005;36:758–769.
Dani C, Smith AG, Dessolin S et al. Differentiation of embryonic
stem cells into adipocytes in vitro. J Cell Sci 1997;110:1279–1285.
Wiper-Bergeron N, St-Louis C, Lee JM. CCAAT/enhancer binding
protein {beta} abrogates retinoic acid-induced osteoblast differentiation via repression of Runx2 transcription. Mol Endocrinol 2007;21:
2124–2135.
van Haaften T, Byrne R, Bonnet S, et al. Airway delivery of mesenchymal stem cells prevents arrested alveolar growth in neonatal lung
injury in rats. Am J Respir Crit Care Med 2009;180:1131–1142.
Yuspa SH, Kilkenny AE, Steinert PM, Roop DR. Expression of murine epidermal differentiation markers is tightly regulated by restricted
extracellular calcium concentrations in vitro. J Cell Biol 1989;109:
1207–1217.
Martin J, Helm K, Ruegg P, et al. Adult lung side population cells
have mesenchymal stem cell potential. Cytotherapy 2008;10:140–151.
28 Maes C, Carmeliet G. VEGF in development: Vascular and nonvascular roles of VEGF in bone development (Madame Curie Bioscience
Database). 2008, Bookshelf ID: NBK6134.
29 Bin NH, Jin AH, Lee W-J et al. The molecular signature of in vitro
senescence in human mesenchymal stem cells. Genes Genomics 2010;
32:87–93.
30 Roobrouck VD, Ulloa-Montoya F, Verfaillie CM. Self-renewal and
differentiation capacity of young and aged stem cells. Exp Cell Res
2008;314:1937–1944.
31 Zavazava N. Embryonic stem cells and potency to induce transplantation tolerance. Expert Opin Biol Ther 2003;3:5–13.
32 Koch CA, Geraldes P, Platt JL. Immunosuppression by embryonic
stem cells. Stem Cells 2008;26:89–98.
33 Grinnemo KH, Kumagai-Braesch M, Månsson-Broberg A, et al.
Human embryonic stem cells are immunogenic in allogeneic and xenogeneic settings. Reprod Biomed Online 2006;13:712–724.
34 Swijnenburg RJ, Tanaka M, Vogel H, et al. Embryonic stem cell immunogenicity increases upon differentiation after transplantation into
ischemic myocardium. Circulation 2005;112(9 suppl):I166–I172.
35 Okita K, Ichisaka T, Yamanaka S. Generation of germline-competent
induced pluripotent stem cells. Nature 2007;448:313–317.
36 Lian Q, Lye E, Suan Yeo K, et al. Derivation of clinically compliant
MSCs from CD105þ, CD24- differentiated human ESCs. Stem Cells
2007;25:425–436.
37 Barberi T, Willis LM, Socci ND et al. Derivation of multipotent mesenchymal precursors from human embryonic stem cells. Plos Med
2005;2:e161.
38 James MJ, Järvinen E, Wang XP et al. Different roles of Runx2 during early neural crest–derived bone and tooth development. J Bone
Miner Res 2006;21:1034–1044.
39 Enomoto H, Furuichi T, Zanma A, et al. Runx2 deficiency in chondrocytes causes adipogenic changes in vitro. J Cell Sci 2004;117:417–425.
40 Xiao G, Wang D, Benson MD et al. Role of the Œ62-Integrin in
osteoblast-specific gene expression and activation of the Osf2 transcription factor. J Biol Chem 1998;273:32988–32994.
41 Müller A, Mehrkens A, Schäfer DJ et al. Towards an intraoperative engineering of osteogenic and vasculogenic grafts from the stromal vascular fraction of human adipose tissue. Eur Cell Mater 2010;3:127–135.
42 Tremoleda J, Forsyth NR, Khan NS et al. Bone tissue formation from human
embryonic stem cells in vivo. Cloning Stem Cells 2008;10:119–132.
43 Duan X, Tu Q, Zhang J, et al. Application of induced pluripotent
stem (iPS) cells in periodontal tissue regeneration. J Cell Physiol
2010;226:150–157.
44 Naujoks C, Langenbach F, Berr K et al. Biocompatibility of osteogenic predifferentiated human cord blood stem cells with biomaterials
and the influence of the biomaterial on the process of differentiation.
J Biomater Appl 2010.
45 Smith LA, Liu X, Hu J et al. The enhancement of human embryonic
stem cell osteogenic differentiation with nano-fibrous scaffolding. Biomaterials 2010;31:5526–5535.
46 Pelttari K, Wixmerten A, Martin I. Do we really need cartilage tissue
engineering? Swiss Med Wkly 2009;139:602–609.
47 Somers A, Jean JC, Sommer CA, et al. Generation of transgene-free
lung disease-specific human iPS cells using a single excisable lentiviral stem cell cassette. Stem Cells 2010;28:1728–1740.
See www.StemCells.com for supporting information available online.