Objective—To elucidate tissue inhibitor of metalloproteinase
(TIMP)-mediated effects on chondrocytes.
Sample Population—Articular cartilage from humeral
heads of 6 dogs.
Procedure—Chondrocytes from harvested specimens
were cultured in 3-dimensional (3-D) agarose at
106 cells/mL. We prepared 3-D constructs exposed to
only tumor necrosis factor (TNF)-α (50 ng/mL).
Recombinant human TIMP-1 (255nM), -2 (285nM), or
-3 (250nM) was added to liquid media bathing 3-D
constructs cultured with TNF-α. Chondrocytes cultured
without TIMP or TNF-α served as control samples.
Samples of liquid media were collected on days
6, 9, 15, and 21 of culture for evaluation of glycosaminoglycan
(GAG) and nitric oxide concentrations.
The 3-D constructs were collected on days 9,
15, and 21 for evaluation of GAG, hydroxyproline (HP),
and DNA contents.
Results—GAG content in control samples increased
significantly during the study, whereas GAG content
in 3-D constructs cultured with TNF-α or TNF-α plus
TIMP did not increase. On day 9, GAG release from
3-D constructs cultured with TNF-α was significantly
higher than that in other constructs. The HP content
in control samples increased during the study and
was significantly higher than that in all other constructs
on day 21. Concentrations of nitric oxide were
significantly lower in control samples on day 6, compared
with concentrations for all other constructs.
Conclusions and Clinical Relevance—Addition of
TIMPs did not counteract suppression of GAG and HP
accumulation in 3-D constructs exposed to TNF-α.
Apparently, adverse effects on chondrocytes exposed
to TNF-α cannot be prevented by addition of TIMP
alone. (Am J Vet Res 2004;65:1611–1615)
Objective—To characterize chondrocytes from naturally
occurring osteochondrosis (OC) lesions of the
humeral head of dogs.
Sample Population—15 cartilage specimens from
13 client-owned dogs with humeral head OC and 10
specimens from the humeral head of healthy dogs
Procedure—Chondrocytes were isolated and cultured
in a 3-dimensional system. On days 7, 10, 15,
20, and 25, glycosaminoglycan and hydroxyproline
content and cytologic characteristics were evaluated.
Expression of collagen types I, II, and X was assessed
by use of immunohistochemistry.
Results—Chondrocytes from OC lesions were less
viable, compared with control chondrocytes.
Glycosaminoglycan content in the OC group was significantly
less than in the control group on all days
except day 20. Hydroxyproline content was also significantly
less in the OC group on days 10, 20, and 25.
Expression of collagen type II was significantly less in
the OC group, compared with the control group on all
days, whereas expression of collagen type I was significantly
greater in the OC group on days 20 and 25.
Expression of collagen type X was significantly less in
the OC group on all days except day 25.
Conclusions and Clinical Relevance—Chondrocytes
from naturally occurring OC lesions of the
humeral head of dogs cultured in a 3-dimensional system
were less viable and less capable of producing
appropriate extracellular matrix molecules than chondrocytes
from unaffected dogs. Alterations in the synthetic
capabilities of chondrocytes from OC-affected
cartilage may be a cause or an effect of the disease
process. (Am J Vet Res 2002;63:186–193)
Objective—To determine the effects of interleukin
(IL)-1β on matrix synthesis and degradation by chondrocytes
cultured in a 3-dimensional (3-D) gel medium.
Sample Population—Chondrocytes from 7 dogs.
Procedure—Articular chondrocytes were harvested
and cultured in 3-D gel medium alone or with 10 or 20
ng IL-1βml that was added beginning on day 0, 3, 6,
or 9. On days 3, 6, 12, and 20 of 3-D culture, samples
of the liquid medium were evaluated for glycosaminoglycan
(GAG), prostaglandin E2 (PGE2), and
matrix metalloprotease (MMP)-3 content. The 3-D
plug in each well was evaluated for histologic characteristics
of viability, cell morphology, and proteoglycan
staining, immunohistochemically stained for collagen
type II, and spectrophotometrically analyzed for GAG
Results—Significant differences for all variables were
detected between controls and each IL-1β group,
among groups with different IL-1β concentrations, and
among groups with IL-1β added at various time points.
Chondrocytes exposed to IL-1β had loss of GAG,
increased PGE2 and MMP-3 concentrations, and lack
of collagen type-II synthesis. These IL-1β effects
appeared to be time and concentration dependent.
Conclusions—Addition of IL-1β to chondrocytes in 3-
D gel medium results in time- and concentrationdependent
effects on matrix synthesis and degradation
and provides an appropriate in vitro model for
many of the pathophysiologic events associated with
osteoarthritis. (Am J Vet Res 2000;61:766–770)
Objective—To evaluate biocompatibility and effects
of implantation of 3-dimensional chondrocyte-agarose
autografts in tibial defects in rabbits and to compare
in vitro and in vivo chondrocyte-agarose constructs
with respect to cell viability, differentiation, and matrix
Animals—24 adult New Zealand White rabbits.
Procedure—Three-dimensional constructs with
(grafted group) or without (control group) autogenous
chondrocytes were implanted into tibial defects of
rabbits and cultured in vitro. During an 8-week period,
defects were evaluated radiographically, grossly, histologically,
biochemically, and immunohistochemically.
In vitro constructs were evaluated histologically,
biochemically, and immunohistochemically.
Results—Tibial defects had significantly higher radiographic
densitometry values at 4 and 6 weeks after
implantation in grafted group rabbits, compared with
control group rabbits. Number of observed centers of
endochondral ossification was significantly greater in
defects of grafted group rabbits, compared with control
group rabbits. On day 14, glycosaminoglycan concentration
was significantly higher in tibial defects of
grafted group rabbits, compared to defects of control
group rabbits or in vitro constructs. At weeks 2, 4, and
8, glycosaminoglycan concentrations were significantly
lower in the in vitro control constructs, compared
with other groups. Collagen type I was present
in bone and bony callous in defects of grafted and
control group rabbits. Collagen type II was identified
in cartilaginous tissues of grafted and control group
rabbits. Collagen type X was associated with hypertrophic
chondrocytes. Only type II collagen was found
in the in vitro chondrocyte constructs.
Conclusion and Clinical Relevance—Chondrocyte-agarose
grafts are biocompatible in large tibial defects
and appear to provide a cell source for augmenting
endochondral ossification. (Am J Vet Res 2003;64:12–20)
Objective—To determine glycosaminoglycan (GAG)
concentration and immunohistochemical staining
characteristics of type-I, -II, and -X collagen from cartilage
affected by osteochondritis dissecans (OCD) in
Animals—31 dogs with OCD and 11 clinically normal
Procedure—Cartilage samples were evaluated
microscopically, and GAG content was determined.
Immunohistochemical staining was performed for
type-I, -II, and -X collagen. Sections were subjectively
evaluated for location and intensity of staining.
Results—Cartilage affected by OCD had a variety of
pathologic changes and significantly lower GAG concentrations
than did normal cartilage. Normal cartilage
had no detectable type-I collagen. For dogs < 9
months of age, cartilage affected by OCD had significantly
more type-I collagen but significantly less type-
X collagen than did control cartilage. For dogs > 12
months of age, cartilage affected by OCD contained
significantly more type-I collagen than did control cartilage.
There was a significant negative correlation
between immunoreactivity of type-I collagen and that
of type-II and -X collagen. A significant positive correlation
was found between immunoreactivity of type-II
and -X collagen.
Conclusions and Clinical Relevance—Cartilage
affected by OCD contains less GAG, more type-I collagen,
and less type-X collagen, compared with normal
cartilage. A direct correlation between these
changes and the etiopathogenesis of OCD was not
established. (Am J Vet Res 2001;62:876–881)
Objective—To determine effects of carprofen and
dexamethasone on chondrocytes in a culture model
of osteoarthritis (OA).
Sample Population—Chondrocytes isolated from
articular cartilage of the humeral head of 5 adult dogs.
Procedure—Chondrocytes were harvested, cultured
and subcultured in monolayer, and then cultured in a
3-dimensional (3-D) medium. Cells from each dog
were distributed into 6 groups with differing content
of liquid medium for each 3-D construct (agarose
[AG], AG plus interleukin [IL]-1β, AG plus carprofen [4
μg/mL], AG plus dexamethasone [1 mg/mL], AG plus
IL-1β [20 ng/mL] plus carprofen [4 μg/mL], and AG plus
IL-1β (20 ng/mL) plus dexamethasone
(1 mg/mL). On days 3, 6, 12, and 20 of culture, samples
from all groups were collected. Liquid media
were assayed for glycosaminoglycan, prostaglandin
(PG)E2, matrix metalloprotease (MMP)-3, and MMP-
13 concentrations. All 3-D constructs were evaluated
for viability, cell morphology, proteoglycan staining,
and collagen type-II concentration. Total glycosaminoglycan
content in each 3-D construct was quantitated
by spectrophotometric assay.
Results—Addition of IL-1β caused a significant loss of
cell viability and matrix production. Addition of carprofen
or dexamethasone caused significant decreases
in PGE2 in the liquid media, and each was minimally
effective in protecting chondrocytes against negative
effects of IL-1β.
Conclusions and Clinical Relevance—Human
recombinant IL-1β resulted in loss of cell viability,
alterations in extracellular matrix components, and
production of PG and MMP. Carprofen and dexamethasone
had little effect on cell and matrix variables
but did decrease PGE2 concentrations and primarily
affected the inflammatory pathway of
osteoarthritis. (Am J Vet Res 2002;63:1363–1369)
Objective—To determine immunoreactivity of matrix
metalloproteinase (MMP)-1, -3, and -13 in cartilaginous
tumors of dogs, correlate expression of MMP
with histologic grade of tumors and clinical outcome
of dogs, and compare MMP immunoreactivity
between chondrosarcomas and chondromas.
Sample Population—Formalin-fixed, paraffin-embedded
tissues obtained from samples of naturally occurring
chondrosarcomas (n = 31) and chondromas (8) of
dogs that were submitted to our veterinary medical
Procedure—Histologic sections from each sample
were stained with H&E and monoclonal antibody to
MMP-1, -3, and -13 by use of an avidin-peroxidase
immunohistochemical technique. For each section, histologic
grade (I, II, or III) and immunohistochemical
expression (0, 1, 2, or 3) were evaluated. Clinical outcome
was obtained from medical records or interviews
with referring veterinarians and scored as a good outcome,
moderate outcome, or poor outcome.
Correlations among variables and differences between
chondrosarcomas and chondromas were analyzed.
Results—Samples from chondrosarcomas had significantly
higher immunoreactivity of MMP-1 and -13,
compared with immunoreactivity in samples from
chondromas. In chondrosarcomas, a significant positive
correlation (r, 0.386) was found between MMP-1
and -13 immunoreactivities, and a significant negative
correlation (r, –0.390) was detected between MMP-3
and -13 immunoreactivities.
Conclusion and Clinical Relevance—A significant
increase in expression of collagenases (MMP-1 and -
13) in chondrosarcomas, compared with expression in
chondromas, suggests that collagenases may play an
important role in tumor progression, and possibly
metastasis, in chondrosarcomas of dogs. (Am J Vet
Objective—To evaluate the clinical and pathologic
characteristics of mammary duct ectasia in dogs.
Animals—51 dogs with mammary duct ectasia.
Procedure—Information regarding body condition,
history, number and location of affected mammary
glands, appearance of lesions, surgical treatment,
nonsurgical treatment, and evidence of recurrence or
development of mammary neoplasia was obtained
from surveys sent to referring veterinarians. Results
of information from examination of histologic sections
and referring veterinarians were evaluated for all
mammary duct ectasia biopsies performed between
1992 and 1999.
Results—Duct ectasia was the primary diagnosis in
51 of 1,825 (2.8%) mammary biopsy specimens and
comprised 48% of nonneoplastic mammary diseases.
Affected dogs were evenly distributed over a range of
1 to 13 years of age, with a mean age at the time of
diagnosis of 6.1 ± 3.1 years. All dogs were female (31
sexually intact, 20 spayed); 10 of 26 had whelped.
Duct ectasia was described as nodular (26 dogs), cystic
(13), and multiglandular (11) and located in caudal
(31) more often than cranial (14) or middle glands (10).
Ectasia recurred in 3 dogs. One dog had a history of
previously excised mammary adenocarcinoma; another
subsequently developed mammary carcinoma.
Conclusions and Clinical Relevance—Duct ectasia
affected mature, sexually intact and spayed female
dogs over a wide age range. Certain breeds were
affected more commonly than expected. Increased
risk for mammary neoplasia was not evident. Duct
ectasia should be considered as a cause for mammary
enlargement, especially in young dogs or when its
cystic nature is evident. Mastectomy is usually curative,
and neoplasia should be ruled out in dogs with
ectasia. (J Am Vet Med Assoc 2001;218:1303–1307)