Rheumatology 2003;42:162–165
doi:10.1093/rheumatology/keg024, available online at www.rheumatology.oupjournals.org
Embryonic stem cells injected into the mouse
knee joint form teratomas and subsequently
destroy the joint
S. Wakitani, K. Takaoka, T. Hattori1, N. Miyazawa2,
T. Iwanaga2, S. Takeda2, T. K. Watanabe2 and A. Tanigami3
Objective. To determine whether the joint space is a suitable environment for
embryonic stem (ES) cells to grow and form cartilage.
Method. We transplanted ES cells into the knee joint and a subcutaneous space of
mice with severe combined immunodeficiency.
Results. Teratomas formed in both areas. Those in the joints grew and destroyed
the joints. The incidence of cartilage formation was the same in the knee joint and
subcutaneous space, but the ratio of cartilage to teratoma was higher in the knee
joint than in the subcutaneous space. The teratomas were proved to have been
derived from the transplanted ES cells by detection of the neomycin-resistance
gene that had been transfected into the ES cells.
Conclusions. It is currently not possible to use ES cells to repair joint tissues.
Further optimization of donor ES cells to differentiate as well as inhibit tumour
growth may help to meet these challenges.
KEY WORDS: ES cells, Joint space, Teratoma, Chondrocyte, Joint destruction.
The capacity of articular cartilage for repair is limited.
Many attempts have been made to repair articular
cartilage defects, including transplantation of various
tissues or cells from joint tissues, bone marrow and periosteum, but none of these has been widely accepted for
clinical use w1x. Recently, autologous cultured chondrocyte transplantation has been shown to improve symptoms and to result in some degree of repair w2x. However,
this method involves the collection of autologous
cartilage, which causes cartilage defects in the peripheral
area. Thus, the search for new cell sources is continuing.
Embryonic stem (ES) cells are thought to be a
possible source of tissue regeneration because they are
self-renewing pluripotent cells that can differentiate into
any tissue or cell type w3, 4x. Many attempts have been
made to induce in vitro differentiation into many cell
types, including haematopoietic precursors w5x, heart
and skeletal muscle w6x, endothelium w7x and neural cells
w8, 9x. In the orthopaedic field, ES cells are considered
useful for regenerating articular cartilage, but thus far it
has remained impossible to induce the formation of
chondrocytes from ES cells in vitro w10x. It was reported,
however, that ES cells could differentiate into neural
cells in vivo when transplanted into the spinal cord w11x.
This led us to conclude that the environment is one of
the important factors influencing cell differentiation. We
therefore transplanted ES cells into the knee joint and a
subcutaneous space of mice with severe combined immunodeficiency (SCID mice) and observed their growth
and chondrogenesis under different circumstances.
Materials and methods
ES cell preparation
AB2.2 prime ES cell kits were purchased from Lexicon
Genetics (Houston, TX, USA). The ES cells had been
obtained from 129uSvuEv mice. They were cultured
according to the instructions for the kit. Briefly, cells
were cultured on the ESQ feeder cells that were supplied
with the kit. Just before transplantation, the ES cells
were embedded in collagen solution (type I collagen
Department of Orthopaedic Surgery, Shinshu University School of Medicine, Matsumoto, 1Department of Orthopaedic Surgery, Osaka-Minami
National Hospital, Kawachinagano and 2Otsuka GEN Research Institute, Otsuka Pharmaceutical Co. Ltd, Tokushima and 3Fujii Memorial
Research Institute, Otsuka Pharmaceutical Co. Ltd, Otsu, Japan.
Submitted 23 May 2002; revised version accepted 27 May 2002.
Correspondence to: S. Wakitani, Department of Orthopaedic Surgery, Shinshu University School of Medicine, Asahi 3-1-1, Matsumoto 3908621, Japan. E-mail: [email protected]
162
ß 2003 British Society for Rheumatology
ES cells form teratoma in mouse knee joint
obtained from porcine Achilles tendon; Nitta Gelatin,
Osaka, Japan) at 48C and at a cell density of 107 cellsuml.
Surgery
Twenty-five female 5-week-old SCID mice (Fox Chase
SCID C.B-17uIcr-scid Jcl; Nihon Clea, Osaka, Japan)
were anaesthetized by intramuscular injection of ketamine (150 mgukg; Sankyo, Tokyo, Japan) and xylazine
(15 mgukg; Bayer, Tokyo, Japan). Ten microlitres of the
collagen solution containing the ES cells was injected
into each right knee and into a subcutaneous space on
the back of each mouse.
163
Evaluations
Five mice were killed at each of the time-points of 1, 2,
4, 6 and 8 weeks after surgery. The knee joints were
collected, fixed with 10% buffered formalin, decalcified,
embedded in paraffin and sectioned. The area of the
back surrounding the injected subcutaneous space was
examined, and the mass in the space itself was collected,
fixed with 10% buffered formalin, embedded in paraffin
and sectioned. Specimens were stained with haematoxylin–eosin and toluidine blue and analysed histologically. The widths of the tumours and of the cartilaginous
area in the tumours were measured using a Scion image
(Scion, Frederick, MD, USA).
Using Dexpad (Takara, Kyoto, Japan), DNA was
extracted from two paraffin sections (5 mm thick) from
each of the 15 selected preparations, i.e. all eight knees
with tumours and seven subcutaneous tumours. The
DNA was then amplified with the polymerase chain
reaction to detect the neomycin (Neo)-resistance gene
that had been transfected into the AB2.2 prime ES cells.
Detection of the X11 gene served as a positive control as
it was present in both the ES cells and the SCID mice.
The primers for the Neo-resistance gene were set
between neo p4 (59-AGGATCTCGTCGTGACCCATG39) and neo int2 (59-TCAGAAGAACTCGTCAAGAA
GGC-39), and the size of the product was 250 base pairs.
For the X11 gene, the primers were set between X11
KO-Hd-1
(59-TGGGAGGGTGAACGCTATAC-39)
and X11 KO-Hd-2 (59-CTCACTGCGCGCTCATTTTG-39), and the size of the product was 260 base
pairs.
For statistical analysis of data on the incidence of
mass formation and the percentage of cartilaginous
mass we used the Mann–Whitney U-test. Probability
values less than 5% were considered significant.
StatView (SAS Institute, Cary, NC, USA) was used
for statistical analyses.
The procedure was approved by the Institutional
Review Board.
FIG. 1. Tumours formed in the knee joints and subcutaneous
spaces. (Top) A tumour in a knee joint 4 weeks after
transplantation. Toluidine blue staining; bar=1 mm. A
tumour (arrow) containing three cartilaginous masses (arrowheads) was observed. The tumour measured 2.33 mm3 and the
cartilaginous areas 0.18, 0.13 and 0.02 mm3. The proportion of
cartilaginous areas to tumour was 14%. (Middle) A large
tumour had destroyed the knee structure 8 weeks after
transplantation. Haematoxylin–eosin staining; bar=1 mm.
The tumour (arrows) had grown inside the knee joint and
destroyed the knee structure. (Bottom) Tissues thought to have
originated from ectoderm (skin), mesoderm (cartilage) or
endoderm (mucous gland), 4 weeks after transplantation in a
subcutaneous space. Toluidine blue staining; bar=100 mm.
Dotted arrow indicates skin (zonal detachment of cells that
have lost their nuclei); solid arrow indicates cartilage (round
cells surrounded by matrix with metachromatic staining);
arrowhead indicates a mucous gland (cells containing round,
homogeneous materials arranged side by side facing the open
space).
164
S. Wakitani et al.
Results
In the case of subcutaneous injection, small white
fibrous tissues were found 1 week after transplantation,
and with time these tissues became larger and the colour
changed to black or red.
At short follow-up, it was difficult to detect tumours
in the knees after injection with ES cells. When tumours
had grown, they became identifiable. We found eight
knees containing a tumour, and three of these tumours
contained cartilage (Fig. 1, top). In two knee joints at
8 weeks, ES cells formed a large tumour that extruded
from the knee joint and destroyed the knee structures
(Fig. 1, middle).
Some tumours were thought to be teratomas because
they contained tissues that showed characteristics of
tissues originating from the ectoderm, mesoderm and
endoderm (Fig. 1, bottom).
The incidence of tumour formation in the knee (total
of 8 tumours for 25 injections: 1u5, 1u5, 3u5, 1u5 and 2u5 at
1, 2, 4, 6 and 8 weeks after injection respectively) was
much lower than that after subcutaneous transplantation (total of 22 for 25 injections: 3u5, 4u5, 5u5, 5u5 and
5u5 at 1, 2, 4, 6 and 8 weeks after injection respectively).
The volume of the tumour became larger with time in
both the knee joint and the subcutaneous space, but it
was much larger in the subcutaneous space. The mean
width of tumours in the knee joints was 0.24, 0.20, 1.31,
2.06 and 12.22 mm2 at 1, 2, 4, 6 and 8 weeks after
injection respectively. Those in subcutaneous spaces
were 3.65, 10.12, 18.23, 53.07 and 101.90 mm2 at 1, 2, 4,
6 and 8 weeks after injection respectively.
There was no significant difference between the
incidence of cartilage formation in tumours in the knee
joint (3u8) and in tumours in the subcutaneous space
(6u22). The mean size of a single cartilage mass in a tumour
in the knee joint (0.09 mm2) was almost the same as that in
the subcutaneous space (0.08 mm2). Several cartilage
masses were usually observed in a single tumour. Because
the tumours in the subcutaneous transplantation sites
were larger, the ratio of the width of the cartilaginous to
the total width of the tumour in the knee joints (14.16,
10.50 and 2.67% in the three samples from joints) was
significantly larger than that in the subcutaneous spaces
(0.30, 1.06, 1.46, 0.49, 0.20 and 0.40% in the six samples
from subcutaneous spaces) (P=0.0201).
The DNA analysis revealed that all samples contained
the Neo-resistance gene, which had been transfected into
the ES cells but was not present in the SCID mice,
indicating that some cells in the histological sections had
been derived from the transplanted cells (Fig. 2).
Discussion
The ES cells formed tumours both in the knee joints and
in the subcutaneous spaces in SCID mice. This is the
first report of ES cell transplantation into the knee joint.
The incidence of tumour formation in the knee joint was
low. The growth of tumour in the knee joint was slower
FIG. 2. Gel electrophoresis of DNA amplified by the polymerase chain reaction. Lanes 1–7, DNA extracted from
sections of subcutaneous tumours; lanes 8–15, DNA extracted
from sections of knee joints with tumours; lane 16, DNA
extracted from AB2.2 cells; lane 17, genomic DNA of an SCID
mouse; lane 18, negative control.
than in the subcutaneous space. It thus appears that the
joint environment is not optimal for the ES cells to grow
and form a tumour. It is also conceivable that the
confined space in the knee joint may have interfered
with growth. However, some of these tumours subsequently grew larger and destroyed the knee joint.
We concluded that these tumours were teratomas
because we identified tissues that showed characteristics
of tissues originating from the ectoderm, mesoderm and
endoderm. There was no significant difference between
the incidence of cartilage formation in the tumours in
the knee joint and that in the subcutaneous space. The
ratio of cartilage to tumour was significantly greater for
the joint space than the subcutaneous space. However,
we consider that the joint environment can hardly be
described as chondrogenic for ES cells. The volume of
cartilage in the tumour was almost the same in the joint
space as in the subcutaneous space. Even when the
tumour became larger, the size of the cartilage mass
within it remained constant. Thus, differences in the
proportion of cartilage were due mainly to tumour size,
not cartilage size.
We embedded the ES cells in type I collagen, which is
not a constituent of articular cartilage. It is possible that
type II collagen, which is a cartilage-derived collagen,
may have a chondrogenic effect on ES cells. We did not
use type II collagen, however, for two reasons. First, we
wanted to make a clear distinction between the two
environments. In the joint, but not in the subcutaneous
space, the injected cells could have come into contact
with type II collagen. If we had embedded the ES cells in
type II collagen, the difference between the two
environments would not have been clearly observed.
Secondly, because type II collagen does not produce
a hard gel, the embedded cells may not be arranged in
three-dimensional position in a gel.
The Neo-resistance gene, which had been transfected
into the ES cells, was not present in the cells of the SCID
mice we used. However, it was detected in the
histological sections, indicating that some cells in the
histological sections had been derived from the ES cells.
We examined all areas of all histological sections and
ES cells form teratoma in mouse knee joint
found no abnormal tissues except for the tumours. This
strongly suggests that the tumours were generated by the
ES cells.
It is currently not possible to use ES cells to repair
articular cartilage defects because tumours eventually
grow and extrude from the joints and because cartilage
formation is not satisfactory. Further optimization of
donor ES cells to differentiate into cartilage w10x and
inhibit tumour growth w12x may help to meet these
challenges in the development of an ES-derived therapy
for chondral defects.
References
1. Buckwalter JA, Mankin HJ. Articular cartilage repair and
transplantation. Arthritis Rheum 1998;41:1331–42.
2. Peterson L, Minas T, Brittberg M, Nilsson A, SjogrenJansson E, Lindahl A. Two- to 9-year outcome after
autologous chondrocyte transplantation of the knee. Clin
Orthop 2000;374:212–34.
3. Evans MJ, Kaufman MH. Establishment in culture of
pluripotential cells from mouse embryos. Nature 1981;
292:154–6.
4. Thomson JA, Itskovitz-Eldor J, Shapiro SS et al.
Embryonic stem cell lines derived from human blastocysts.
Science 1998;282:1145–7.
165
5. Wiles MV, Keller G. Multiple hematopoietic lineages
develop from embryonic stem (ES) cells in culture.
Development 1991;111:259–67.
6. Klug MG, Soonpaa MH, Koh GY, Field L. Genetically
selected cardiomyocytes from differentiating embryonic
stem cells form stable intracardiac grafts. J Clin Invest
1996;98:216–24.
7. Lumelsky N, Blondel O, Laeng P, Velasco I, Ravin R,
McKay R. Differentiation of embryonic stem cells to
insulin-secreting structures similar to pancreatic islets.
Science 2001;292:1389–94.
8. Brustle O, Jones KN, Learishm RD et al. Embryonic stem
cell-derived glial precursors: a source of myelinating
transplants. Science 1999;285:754–6.
9. Schuldiner M, Eiges R, Eden A et al. Induced neuronal
differentiation of human embryonic stem cells. Brain Res
2001;913:201–5.
10. Kramer J, Hegert C, Guan K, Wobus AM, Muller PK,
Rohwedel J. Embryonic stem cell-derived chondrogenic
differentiation in vitro: activation by BMP-2 and BMP-4.
Mech Dev 2000;92:193–205.
11. McDonald JW, Liu XZ, Qu Y et al. Transplanted
embryonic stem cells survive, differentiate and promote
recovery in injured rat spinal cord. Nat Med 1999;5:
1410–2.
12. Zhang X, Morham SG, Langenbach R, Baggs RB, Young
DA. Lack of cyclooxygenase-2 inhibits growth of
teratocarcinomas in mice. Exp Cell Res 2000;254:
232–40.