VOL. 83-B, NO. 2, MARCH 2001 289
G-I. Im, MD
D-Y. Kim, MD
J-H. Shin, MD
C-W. Hyun, MD
W-H. Cho, MD
Department of Orthopaedics, Hallym University Hospital, 153 Kyo-Dong,
Chunchon 200-060, Korea.
Correspondence should be sent to Dr G-I. Im.
©2001 British Editorial Society of Bone and Joint Surgery
0301-620X/01/210495 $2.00
Repair of cartilage defect in the rabbit with
cultured mesenchymal stem cells from bone
marrow
Gun-Il Im, Do-Young Kim, Joo-Ho Shin, Cheol-Won Hyun,
Won-Ho Cho
From Hallym University, Chunchon, Korea
I
n 16 mature New Zealand white rabbits
mesenchymal stem cells were aspirated from the
bone marrow, cultured in monolayer and implanted
on to a full-thickness osteochondral defect artificially
made on the patellar groove of the same rabbit. A
further 13 rabbits served as a control group. The
rabbits were killed after 14 weeks. Healing of the
defect was investigated histologically using
haematoxylin and eosin and Safranin-O staining and
with immunohistochemical staining for type-II
collagen. We also used a reverse
transcription-polymerase chain reaction (RT-PCR) to
detect mRNA of type-I and type-II collagen.
The semiquantitative histological scores were
significantly higher in the experimental group than in
the control group (p < 0.05). In the experimental
group immunohistochemical staining on newly formed
cartilage was more intense for type-II collagen in the
matrix and RT-PCR from regenerated cartilage
detected mRNA for type-II collagen in mature
chondrocytes. These findings suggest that repair of
cartilage defects can be enhanced by the implantation
of cultured mesenchymal stem cells.
J Bone Joint Surg [Br] 2001;83-B:289-94.
Received 20 August 1999; Accepted after revision 18 February 2000
Cartilage is a highly differentiated tissue and therefore has
a limited capacity for self-repair. Recently, several stud-
ies
1-5
have been reported of the successful repair of osteo-
chondral defects by the transplantation of cultured
chondrocytes, but the method requires that a sufficient
number of cells are obtained from the donor site in the
articular cartilage. Osteochondral progenitor cells or
mesenchymal stem cells are undifferentiated cells found in
small numbers in the periosteum or in the bone marrow
which are capable of differentiation to chondroblasts or
osteoblasts. These cells are able to re-establish chondro-
genesis during the developmental process when implanted
in osteochondral defects.
6-9
We have investigated by histo-
logical, immunohistological, and molecular biological
methods the feasibility of repairing cartilage defects with
mesenchymal stem cells from bone marrow.
Materials and Methods
We used 29 skeletally mature New Zealand white rabbits
between six months and one year old and with a mean
weight of 3.52 ± 0.43 kg (SD). There were 16 rabbits in the
experimental group and 13 in the control group. They were
given tap water and food ad libitum. They were kept in
separate cages and allowed to move freely. No specific
immobilisation was applied after operation. The experi-
ments had been approved by the University Committee for
Animal Experimentation.
Isolation and culture of bone-marrow-derived mesenchy-
mal cells. We used a method similar to that of Wakitani et
al.
9
Under anaesthesia induced by intramuscular administra-
tion of 50 mg/kg of ketamine and 10 mg/kg of xylazine, the
right leg was shaved, disinfected. A 1 cm incision was made
on the medial aspect of the right proximal tibia, centred
b tween the anterior and medial border, beginning from the
level of the kne and extending downwards. A drill hole of
2m in diam ter was created with a hand drill and 2 ml of
bone marrow were aspirated with an 18-gauge needle on a
5ml syr nge cont ining 0.1 ml of heparin (3000 U/ml). After
the addition of 0.7 ml of Ficoll-Hypaque (Sigma, St Louis,
Missouri), the aspirate was centrifuged at 200 times gravity
for 15 minutes. The buffy coats were isolated and red blood
cells (RBC) destroyed by RBC lysis buffer (Sigma) contain-
ing 0.83% ammonium chloride. After suspension with Ham
F-12 medium (Gibco, Green Island, New York) the cells
were washed by centrifuging three times for five minutes at
200 times gravity. The nucleated cells were cultured in Ham
F-12 medium containing 10% fetal calf serum (Gibco) and
antibiotics (penicillin G, 100 U/ml; streptomycin, 0.1 mg/ml;
amphotericin B, 0.25
g/ml) at 37°C in 5% CO
2
in a 25 cm
2
culture flask.

The media were first changed at the third day of culture,
and then every other day thereafter. Non-adherent cells
were removed subsequently. The cells achieved confluence
after 10 to 20 days when they were separated from the
culture flask by treating with 0.25% trypsin and 0.001M
EDTA, washed three times with Ham F-12 medium and
counted by a haemocytometer. The mean number of cells
was 3.4 ± 1.5  10
6
before and 2.3 ± 0.6  10
6
after
culture.
Autologous implantation of cultured cells on the osteo-
chondral defect. The rabbits were anaesthetised and the
left leg was prepared as described above. A 3 cm medial
parapatellar incision was made over the left knee and the
patella was everted. We made a full-thickness defect of
diameter 3 mm and depth of 2 mm on the articular cartilage
of the patellar groove of the left distal femur with a hand
drill. A flap of about 6  6mm was removed from the
fascia overlying the quadriceps muscle and sutured to the
peripheral rim of the artificial defect with 6-0 catgut.
In the experimental group we then injected approx-
imately 1  10
6
cells suspended in 0.025 ml of Ham F-12
medium into the defect through the sutured fascia while the
control group received 0.025 ml of Ham F-12 medium
without cells.
Histological evaluation. Fourteen weeks after the implan-
tation of the cultured stem cells onto the osteochondral
defect the rabbits were killed by an intravenous injection of
a lethal dose of barbiturate. The distal femora were resected
en bloc and parts of the regenerated cartilage were removed
and deep frozen at -70°C. After fixation with 10% buffered
formalin, the distal femora were decalcified with 0.5M
EDTA solution. Sections were prepared sagittally and
stained with haematoxylin and eosin and Safranin-O/Fast
Green. The specimens were graded semiquantitatively
based on the predominant nature of the repair tissue, matrix
staining, regularity of the surface, structural integrity, thick-
ness of the repair, apposition between the repaired cartilage
and surrounding normal cartilage, freedom from degener-
ative signs in repair tissue, and freedom from degenerative
changes of the surrounding normal cartilage as described
by O’Driscoll, Keeley and Salter (Table I). Differences in
histological scoring between groups were analysed using
the Wilcoxon rank-sum test.
Immunohistochemistry. Sections were deparaffinised with
xylene, rehydrated with decreasing solutions of ethanol and
rinsed three times for five minutes each with 0.01M phos-
phate-buffered saline (PBS, pH 7.2). They were processed
with 0.25% trypsin for 15 minutes and 3% H
2
O
2
for 30
minutes to suppress intrinsic activity of peroxidase. A
blocking solution containing normal rabbit serum was
added to the sections for 30 minutes and they were then
incubated with polyclonal mouse antibody for type-II colla-
gen (Monosan, Uden, The Netherlands) at optimal dilution
in 0.01M PBS with 1% w/v bovine serum albumin for two
hours. The sections were rinsed with 0.01M PBS three
times for five minutes each and were then incubated with
biotinylated anti-mouse goat immunoglobulin (Biorad, Her-
cules, California) for two hours. They were rinsed three
imes with PBS, treated for 45 minutes with streptavidine-
h rseradish peroxidase complex, rinsed three times with
PBS and treated for ten minutes with 0.02% diaminobenzi-
dine in 0.01 M Tris buffer (pH 7.6) containing 0.005%
H
2
O
2
.
Rev r e tr nscription-polymerase chain reacti n (RT-
PCR) for mRNA of type-I and type-II collagen. The
regenerated cartilage which had been removed was homo-
genised and the total RN was isolated by an RNA-binding
matrix method using a total RNA Isolation Kit (5prime-
>3prime; Boulder, Colorado). The main templates of the
RT-PCR were total RNA isolated from regenerated cartil-
age from the experimen al and the control groups. The
oligonucleotide primers were ordered from Bioneer
(Chungbuk, Korea) based on the sequence available from
the G n Bank (National Centre for Biotechnology Informa-
290 GUN-IL IM, DO-YOUNG KIM, JOO-HO SHIN, CHEOL-WON HYUN, WON-HO CHO
THE JOURNAL OF BONE AND JOINT SURGERY
Table I. Histological and histochemical grading scale
Score
Nature of the predominant tissue
Cellular morphology
Hyaline articular cartilage 4
Incompletely differentiated mesenchyme 2
Fibrous tissue or bone 0
Safranin-O staining of the matrix
Normal or nearly normal 3
Moderate 2
Slight 1
None 0
Structural characteristics
Surface regularity
Smooth and intact 3
Superficial horizontal lamination 2
Fissure - 25% to 100% of the thickness 1
Severe disruption, including fibrillation 0
Structural integrity
Normal 2
Slight disruption, including cysts 1
Severe disi tegration 0
Thickness
100% of normal adjacent cartilage 2
50% to 100% of normal cartilage 1
0% to 50% of normal cartilage 0
Bonding to the adjacent cartilage
Bonded at both ends of graft 2
B nded at one end, or partially at both ends 1
Not bonded 0
Freedom from cellular changes of degeneration
Hypocellularity
Normal 3
Slight 2
Moderat 1
Severe 0
Chondrocyte clustering
No clusters 2
<25% of the cells 1
25% to 100% of the cells 0
Freedom from degenerative changes in adjacent cartilage
Normal cellularity, no clusters, normal staining 3
Normal cellularity, mild clusters, moderate staining 2
Mild or moderate hypocellul rity, slight staining 1
Severe hypocellularity, poor or no staining 0