Patients
Two biopsy specimens each were obtained from four patients who had
undergone autologous chondrocyte transplantation, according to the technique
of Brittberg et
al.26, with
Carticel Service (Genzyme, Cambridge, Massachusetts). One patient (Case 1) was
a thirty-three-year-old woman who had been treated for a primary isolated
femoral chondral defect measuring 2.5 cm2. The second patient (Case
2) was a thirtynine-year-old man who had been treated for a traumatic
osteochondral lesion with a defect area of 4 cm2. The third patient
(Case 3) was a twenty-eight-year-old man who had been treated for an
osteochondritis dissecans with a defect area of 2.5 cm2, and the
fourth patient (Case 4) was a thirty-seven-year-old woman who had been treated
for a traumatic femoral defect of 4 cm2. With informed consent,
each patient underwent second-look arthroscopy twenty-four months after
autologous chondrocyte transplantation. Two biopsy specimens of the grafted
areas were obtained from each patient. One of the two samples was used for
molecular biology analysis by real-time reverse transcriptase-polymerase chain
reaction, and the other was used for histological and immunohistochemical
examination. As controls, two biopsy samples of healthy cartilage were
obtained from two male multiorgan donors who had been twenty and thirty-seven
years old at the time of death and had not had a known history of arthritis or
other joint disorder.
Informed consent was obtained from all patients who entered the study, and
the work was approved by the Ethical Committee of the Istituti Ortopedici
Rizzoli.
Analysis of Cartilage-Specific Molecule Expression by Real-Time
Reverse Transcriptase-Polymerase Chain Reaction
RNA Extraction and Reverse Transcriptase
Chondrocytes were isolated from the biopsy samples by sequential digestion,
as previously
described27. A
total of 5 × 105 isolated cells were pelleted and lysed in
0.5 mL of RNAzol B reagent (Biotecx Laboratories, Houston, Texas); total RNA
was subsequently isolated according to the manufacturer's instructions.
Complementary DNA was synthesized from 1 µg of total RNA per sample with
forty-five minutes of incubation at 42°C, with use of Moloney murine
leukemia virus reverse transcriptase (PerkinElmer, Norwalk, Connecticut) and
oligo-(dT) priming.
Primer Design
Polymerase chain reaction primers for aggrecan, cathepsin B, and the
housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH), used as an
internal control, were obtained from published
references28-30.
Polymerase chain reaction primers for types I, II, and X collagen were
designed with use of the Primer3 software (Steve Rozen and Helen J. Skaletsky,
1998 Primer3;
)
and for Egr-1 and Sox-9 with use of the LightCycler Probe Design Software
(Roche Molecular Biochemicals, Mannheim, Germany). All primers were chosen to
span exon junctions. Specific primer pairs, polymerase chain reaction product
lengths, annealing temperatures, and references are reported in the
Appendix.
LightCycler Real-Time Polymerase Chain Reaction
Real-time polymerase chain reaction was run in a LightCycler Instrument
(Roche Biomechanicals) with use of the Quanti-Tect SYBR Green RT-PCR Kit
(Qiagen, Hilden, Germany) with the following protocol: initial activation of
HotStarTaq DNA polymerase at 94°C for fifteen minutes and then forty-five
cycles at 94°C for fifteen seconds, at 56°C to 60°C for twenty
seconds, and at 72° for ten seconds. The increase in polymerase chain
reaction product was monitored for each amplification cycle by measuring the
increase in fluorescence caused by the binding of SYBR Green-I dye to dsDNA.
The threshold cycle (CT) values (i.e., the cycle number at which
the detected fluorescence exceeds background levels) were determined for each
sample, and the specificity of the amplicons was confirmed by melting curve
analysis and agarose gel electrophoresis. Amplification efficiencies, as
calculated from the slopes of log input amounts plotted versus CT
values, were confirmed to be high (at least 88%), except for Sox-9 (see
Appendix).
Relative Quantification of Gene Expression Levels
To quantify the relative changes in gene expression between each patient
and the multiorgan donors, we used the comparative CT method, as
previously described by Livak and
Schmittgen31. This
method detects the fold changes in gene expression by the formula
2-??CT. The ?CT value
represents the difference between the CT of each target gene
(aggrecan; cathepsin B; type I, II, and X collagens; Egr-1; and Sox-9) and the
internal control (GAPDH). The ??CT value for each
specific gene represents the difference between ?CT values
for each patient and the mean value obtained in the multiorgan donors
(CtM). In the comparative CT method, the
?CT mean value obtained in the multiorgan donors is 0 and the
fold difference is 1.
Histochemistry
Samples for histological analysis were fixed in 10% buffered formalin,
washed, and decalcified with Formical-2000 (Decal Chemical, Congers, New York)
for two hours at room temperature. The samples were then dehydrated through a
graded series of alcohol and were embedded in paraffin. Sections, 4 µm
thick, were obtained from the cartilage specimens, and the slides were stored
at room temperature.
Safranin-O Staining
Slides were stained with 0.001% fast green (Sigma Chemical, St. Louis,
Missouri) in distilled water for three minutes at room temperature. The slides
were then quickly dipped in 1% acetic acid and stained with 0.1% safranin O
for five minutes at room temperature.
Immunohistochemistry
For immunohistochemical analysis, the following primary antibodies were
used: mouse monoclonal anti-human type-I collagen (MAB3391; Chemicon
International, Temecula, California); anti-human collagen-type-II mouse
monoclonal antibody (MAB8887; Chemicon International); anti-human-Egr-1 rabbit
affinity purified polyclonal antibody (sc-110; Santa Cruz Biotechnology, Santa
Cruz, California); anti-human-Sox-9 rabbit affinity purified polyclonal
antibody (AB5809; Chemicon International), and anti-human cathepsin-B goat
affinity purified polyclonal antibody (sc-6491, Santa Cruz Biotechnology).
Paraffin sections were deparaffinized and rehydrated. For epitope
unmasking, the samples were treated with 0.1% hyaluronidase (Sigma Chemical)
in phosphate-buffered saline solution at 37°C for five minutes. After they
were washed, the slides for the detection of type-I and II collagens, Egr-1,
and Sox-9 were incubated at room temperature for thirty minutes in
phosphate-buffered saline solution containing 5% of normal goat serum (Dako,
Carpinteria, California), while normal rabbit serum (Dako) was used for
cathepsin B, to prevent nonspecific bindings. The slides were incubated with
the anti-human type-I and type-II collagens, cathepsin-B, Egr-1, and Sox-9
primary antibodies diluted 1:20, 1:20, 1:400, 1:20, and 1:50, respectively, in
0.04-M Trizma base saline solution, pH 7.6, containing 1% bovine serum albumin
and 0.1% Triton X-100 for one hour at room temperature. The slides were washed
three times with 0.04-M Trizma base saline solution, pH 7.6, and were
incubated with goat anti-mouse and anti-rabbit immunoglobulins labeled with
dextran molecules-alkaline phosphatase (4017, EnVision; Dako) at room
temperature for thirty minutes for the detection of type-I and II collagens.
Anti-rabbit-biotinylated goat antibody (E0432; Dako) was used to detect Egr-1
and Sox-9 molecules, while anti-goat-biotin conjugated rabbit immunoglobulins
(Pierce, Rockford, Illinois) diluted 1:200 in Trizma base saline solution,
0.04 M, pH 7.6, were used for cathepsin B. For Egr-1, Sox-9, and cathepsin-B
immunostaining, a further incubation of one hour at room temperature with a
streptavidin-alkaline phosphatase conjugate (Boehringer Mannheim, Mannheim,
Germany) was also performed. After three washes with 0.04-M Trizma base saline
solution, pH 7.6, the reactions were developed with use of a new fuchsin kit
(New Fuchsin Substrate System; Dako) in the presence of 5 mM levamisole (Sigma
Chemical) to block endogenous alkaline phosphatase. Negative controls were
performed by omitting the primary antibody. Slides were counterstained with
hematoxylin and mounted in glycerol gel. All of the samples were analyzed with
use of a Zeiss Axioscope microscope (Carl Zeiss, Oberkochen, Germany).
Analysis of the Expression of Cartilage-Specific Molecules by
Real-Time Reverse Transcriptase-Polymerase Chain Reaction
Real-time reverse transcriptase-polymerase chain reaction monitoring with
the LightCycler with use of fluorescent dye allowed rapid and sensitive
detection of extracellular matrix molecules and regulatory factor mRNAs from
the patient samples and from the controls. Changes in mRNA for all target
genes relative to the internal control GAPDH are reported as
?CT values for the patient samples and the controls
(Table I). Fold changes in gene
expression calculated on the basis of ?CT values are shown in
Figure 1. Type-I collagen mRNA
was expressed in all of the samples evaluated, with higher values for one
patient (Case 1) than for the other patients and the controls
(Fig. 1, a). Type-II
collagen mRNA levels were lower in all patient samples evaluated with respect
to the controls (Fig. 1,
b). Type-X collagen mRNA was undetectable in all of the
samples and in the controls (data not shown). The mRNA levels for aggrecan
varied among the patient samples but, in all cases, were lower than the
control values (Fig. 1,
c). The mRNA levels for cathepsin B were particularly
high in one patient (Case 2) and were also higher in the other samples
compared with the controls (Fig. 1,
d). The Egr-1 and Sox-9 mRNA levels were scarce in all
patient samples (Fig. 1, e and
f).
Histochemistry and Immunohistochemistry
The samples obtained from the biopsy specimens of patients who underwent
autologous chondrocyte transplantation showed, in general, a structure that
was not well organized, with variable fibrocartilaginous features and various
degrees of tissue-remodeling. The superficial layer showed a typical green
staining due to collagen fibers, but, in samples from all four patients, the
thickness of this zone was greater than that observed in normal cartilage and
the fibers were oriented in various directions. The deep and intermediate
layers showed a normal content of proteoglycans, but the cells were not well
distributed and the tidemark was not always delineated
(Fig. 2-A).
Healthy cartilage showed the typical morphology of hyaline articular
cartilage, with a zonal variation through its depth and the cells tangentially
oriented to the superficial layer, homogeneously distributed in the middle and
deep zones and arranged in columns in the deep zone. A normal content of
glycosaminoglycans was highlighted by safranin-O staining, which showed a thin
superficial layer where collagen fibers were oriented tangentially to the
surface, a typical red staining due to the proteoglycans in the intermediate
and deep zones, and a well-defined tidemark
(Fig. 2-B). Type-I collagen was
slightly positive in all of the samples
(Fig. 2-C) and in the control
(Fig. 2-D). Type-II collagen
was evident in all of the samples, and the positivity was intracellular and
mainly localized in the deep zone (Fig.
2-E). The control sample showed uniformly distributed positivity
inside the matrix (Fig. 2-F).
Cathepsin B was positive in all of the samples in the entire cartilage
thickness (Fig. 3-A), while the
control was negative (Fig.
3-B). Egr-1 protein was detected in all of the samples with an
inverse correlation with the presence of type-II collagen; in fact, the
positivity was observed in particular in the middle zone
(Fig. 3-C), where type-II
collagen was negative. In the control, the cells were positive also in the
superficial layer (Fig. 3-D).
Sox-9 positive cells were found in all of the samples, with the positivity
evident in the superficial and middle layers
(Fig. 3-E), whereas in the
control the positivity was mainly superficial
(Fig. 3-F).
The use of autologous chondrocyte transplantation to repair cartilage
defects has begun a new era in the treatment of damage to articular
surfaces15,32.
Good clinical results, together with very few or no side effects, have
encouraged surgeons to continue this therapeutic
strategy20 and have
stimulated researchers to study the processes involved in cartilage
remodeling20-22.
A large number of studies aimed at evaluating the new cartilaginous tissues
after transplantation have been carried out both in
animals33-36
and in
humans20-22.
Those studies have shown that the treatment of cartilage defects by means of
autologous chondrocyte transplantation often results in the formation of a
repair tissue with different degrees of maturation, and the quality of the
regeneration improves with increasing time after the
grafting18,35,37.
However, there are some problems in performing more in-depth studies because
of the limited availability of human biopsy samples after transplantation.
We had the opportunity to evaluate cartilage biopsy specimens that were
obtained during a second-look operation performed twenty-four months after
autologous chondrocyte transplantation. It was inappropriate to perform a
biopsy of healthy cartilage from the same patients, so for normal cartilage
controls we used samples obtained from multiorgan donors. In this way, we
could use real-time reverse transcriptase-polymerase chain reaction to
determine the expression of the main extracellular matrix molecules in this
phase of tissue reconstruction and cathepsin B, which is a marker of cellular
dedifferentiation. Furthermore, we investigated two important transcription
factors, Egr-1 and Sox-9, that are known to be involved in the modulation of
gene expression. Egr-1 (also called KROX24 and zif-268) belongs to the class
of immediate early
genes38 and is
crucial in the regulation of growth factors, hormones, cytokines, and adhesion
molecules39.
Recently, a role for Egr-1 has been demonstrated in chondrocyte proliferation,
differentiation, and apoptosis, providing insight into the underlying
biochemistry of normal cartilage turnover and possibly the pathogenesis of
osteoarthritis40.
However, its target genes in chondrocyte differentiation have not been well
identified, and it has been proposed that Egr-1 can serve as a transcriptional
suppressor of a constitutively expressed collagen gene by preventing
interactions between Sp1 and the general transcriptional
machinery41. Sox-9
is a member of the family of Sox (Sry-type high-mobility-group box
transcription factor) genes that were first identified on the basis of a
region with high homology to that of Sry (sex-determining region Y). This
region encodes a 79-amino acid motif that is known as a high-mobility-group
box and is responsible for sequence-specific binding to DNA. It has been
demonstrated that Sox-9 has an active role in chondrogenesis by enhancing
type-II collagen gene
expression42-45
and aggrecan46,
even though the mechanism of this transcriptional regulation is not understood
in detail.
In our study, the biopsy specimens showed that all of the neocartilage
tissues expressed type-II collagen mRNA at very low levels, suggesting that
the transcriptional process, which regulates its expression, is not completely
active two years after cell transplantation. This finding was confirmed by the
low levels of Sox-9, which was scarcely expressed in the patient samples,
further suggesting also that this transcription factor is downregulated. Thus,
the pathway leading to type-II collagen expression is still "in
progress," as suggested also by the morphological appearance highlighted
by safranin-O staining, which revealed a zonal heterogeneity throughout the
thickness of the cartilage samples with fibrous and fibrocartilaginous
features, without evidence of hyaline tissue formation. These data are in
agreement with those reported by Richardson et
al.20. This
primitive organization is confirmed also by a low expression of
cartilage-specific molecules such as aggrecan and by lower levels of Egr-1 in
all of the samples, suggesting that the cells are not completely
differentiated40.
The higher presence of cathepsin B (in particular in Case 2) and type-I
collagen mRNAs in the samples further supports this hypothesis. In fact, it
has been demonstrated that both of these molecules increase in
dedifferentiated
chondrocytes23.
Type-X collagen mRNA was not detected in any of the samples, which confirms
that the presence of this molecule is restricted in particular to hypertrophic
chondrocytes47.
Our research was performed on a restricted number of small human biopsy
specimens because of the obvious difficulties in obtaining such material;
therefore, the observed results might be due to a sampling error that
underrepresents a possibly more robust remodeling repair response. However,
taking into account our initial data, we can point out that the biochemical
events occurring after autologous chondrocyte transplantation are regulated
both positively and negatively by sophisticated gene expression control
machinery. This results in a complicated, heterogeneous, and long process of
new cartilage formation that involves not only repair but also regeneration
and remodeling phases. Long-term follow-up investigations are probably needed
to verify whether, once these processes are completed, the newly formed tissue
will acquire the more typical features of hyaline cartilage.
A table presenting a description of the real-time reverse
transcriptase-polymerase chain reaction primers used in this study is
available with the electronic versions of this article, on our web site at
(go to
the article citation and click on "Supplementary Material") and on
our quarterly CD-ROM (call our subscription department, at 781-449-9780, to
order the CD-ROM). ?