Our previous study1 demonstrated that chondrocytes undergo apoptotis
in response to mechanical injury to full-thickness cartilage explants.
To validate this response, several models of injury across species
and in vivo were examined.
Full-thickness cartilage explants were harvested from weight-bearing
portions of adult bovine femoral condyles, and 5-mm-diameter disks
were punched out with a dermal punch. Explants were allowed to stabilize
in Dulbecco modified Eagle medium supplemented with 10% fetal
bovine serum for forty-eight hours. Explants were then divided into
two groups: load and control. The load group underwent a single
500-msec injury load of 30% strain in radially unconfined
compression. The control group was not loaded. A 30% strain
was found to generate a more consistent injury than the previous loading
protocol1. At ninety-six hours
after injury, explants underwent histologic examination and the
number of apoptotic cells was counted with use of TUNEL (terminal
deoxynucleotidyl transferase-mediated dUTP nick-end labeling). Apoptosis
was confirmed in selected samples by electron microscopy and immunostaining
of a neo-epitope of cytokeratin. The experiment was repeated with
use of normal human articular cartilage harvested from the femoral
condyles of postmortem donors.
Osteochondral model: To determine whether the
presence of subchondral bone affected the cartilage response, whole
rabbit patellae and whole human patellae from donor cadavers were
harvested. Relatively flat areas were subjected to a load of 20
MPa. The areas of loading were documented with use of pressure-sensitive
film. Explants were harvested from the injured sites and from adjacent,
uninjured cartilage and were maintained in Dulbecco modified Eagle
medium for ninety-six hours, at the end of which apoptosis levels
were measured.
Cartilage defect model: Full-thickness 3-mm
drill-holes were made at regular intervals in the weight-bearing
portions of adult bovine femoral condyles. Cartilage disks of 5
mm in diameter were harvested such that the drill-hole was centrally
located. Disks were cultured for ninety-six hours, at the end of
which apoptosis levels were measured.
In vivo rabbit impact model: After the approval
of our Institutional Animal Review Committee was obtained, eight
rabbits were anesthetized and one patellofemoral joint was subjected
to an impact of 3 kg from a height of 50 cm in a drop-tower apparatus.
Ninety-six hours after injury, the rabbits were killed and the patellar
cartilage was harvested. Apoptotic levels in the injured patellar
cartilage were compared with those in the adjacent, uninjured cartilage
and on the contralateral, uninjured side.
All of the loaded samples in each injury model demonstrated significantly
higher apoptosis rates when compared with the matched controls.
Full-thickness cartilage explants: Bovine cartilage
explants demonstrated a higher mean apoptosis rate (37%±11%)
compared with control samples (12%±5%)
(Fig. 1).
Human cartilage explants also demonstrated a higher mean apoptosis
rate (32%±9%) compared with control samples
(9.1%±3%) (Fig. 1).
Osteochondral models: Both rabbit and human
whole patellae (Figs. 2 and 3) demonstrated a similar apoptotic
response to injury (15%±4% and
11%±3%, respectively). Thus, when
the subchondral bone had been left intact, apoptosis still occurred,
although at a lower rate than when full-thickness cartilage explants
were used.
Cartilage defect model: All of the defect explants
demonstrated apoptosis rates that were higher than those at the
adjacent, control sites. The edges of the lesion contained matrix
debris and fragmented cells. The overall apoptotic rates were 24%±9% for the
defect explants and 9%±4% for
control explants (Figs. 4-A, 4-B, and 5). The apoptotic rate was significantly
higher (55%±30%) in the 0.5-mm
region immediately surrounding the defect. However, some apoptosis
was also seen extending throughout the 5-mm-diameter explant.
In vivo rabbit impact model: The apoptotic rate
in the patellar cartilage from the impacted knees was 11%±3%,
whereas it was <1%±2% in
the contralateral patellar cartilage (Figs. 6 and 7).