ARTHRITIS & RHEUMATISM
Vol. 65, No. 1, January 2013, pp 159–166
© 2013, American College of Rheumatology
Sirtuin 1 Enzymatic Activity Is Required for
Cartilage Homeostasis In Vivo in a Mouse Model
Odile Gabay,1Christelle Sanchez,2Mona Dvir-Ginzberg,3Viktoria Gagarina,1
Kristien J. Zaal,1Yingjie Song,4Xiao Hong He,5and Michael W. McBurney5
Objective. We and others previously demon-
strated that sirtuin 1 (SIRT-1) regulates apoptosis and
cartilage-specific gene expression in human chondro-
cytes and mouse models. This study was undertaken to
determine if SIRT-1 enzymatic activity plays a protec-
tive role in cartilage homeostasis in vivo, by investigat-
ing mice with SIRT-1 mutations to characterize their
Methods. Articular cartilage was harvested from
the paws and knees of 5- and 6-month-old wild-type
(WT) mice and mice homozygous for SIRT-1tm2.1Mcby
(SIRT-1y/y), an allele carrying a point mutation that
encodes a SIRT-1 protein with no enzymatic activity
(y/y mice). Mice ages 2 days old and 6–7 days old were
also examined. Mouse joint cartilage was processed for
histologic examination or biochemical analyses of chon-
Results. We found that articular cartilage tissue
sections from y/y mice of up to 6 months of age
contained reduced levels of type II collagen, aggrecan,
and glycosaminoglycan compared to sections from WT
mice. In contrast, protein levels of matrix metallo-
proteinase 8 (MMP-8), MMP-9, and MMP-13 were
elevated in the cartilage of y/y mice. In addition, chon-
drocyte apoptosis was elevated in SIRT-1 mutant mice
as compared to their WT littermates. Consistent with
these observations, protein tyrosine phosphatase 1b was
elevated in the y/y mice.
Conclusion. Our in vivo findings in this animal
model demonstrate that mice with defective SIRT-1 also
have defective cartilage, with elevated rates of cartilage
degradation with age. Hence, normal cartilage homeo-
stasis requires enzymatically active SIRT-1 protein.
Osteoarthritis (OA) is a multifactorial and com-
plex degenerative disease of the cartilage. Different
mechanisms are involved in cartilage degradation, in-
cluding inflammation, apoptosis, and breakdown of
major extracellular matrix (ECM) components, such as
type II collagen and aggrecan (1). Aging is one of the
most important risk factors linked to OA susceptibility
(2,3). More than 50% of adults ages 65 or older reported
an arthritis diagnosis, making OA one of the most
common diseases in developed countries (4).
Sirtuin 1 (SIRT-1) has been shown to regulate
the lifespan and aging in simple eukaryotes (2,5,6). In
mammals, SIRT-1 has been reported to play an impor-
tant role in age-related diseases, such as osteoporosis,
diabetes, and cancer (6–8). We have previously shown in
vitro that SIRT-1 modulates gene expression in human
chondrocytes. SIRT-1 elevates the expression of genes
encoding the cartilage ECM in human chondrocytes (9)
and enhances the survival of human OA chondrocytes
by repressing apoptosis (10,11). Recently, SIRT-1 has
been reported to be involved in the pathogenesis of OA
by modulating chondrocyte gene expression and hyper-
trophy (12). Finally, SIRT-1 plays an antiinflammatory
role in different tissues by inhibiting the transcription of
proinflammatory genes (6).
So far, in vitro studies have suggested a protective
role of SIRT-1 in cartilage; however, very few studies
have demonstrated these roles in vivo. We previously
created SIRT-1–null mice (13,14) and found that they
were smaller than wild-type (WT) mice, had cranio-
Supported by the Intramural Research Program of the
National Institute of Arthritis and Musculoskeletal and Skin Diseases,
1Odile Gabay, PhD, Viktoria Gagarina, MD, PhD, Kristien J.
Zaal, PhD: National Institute of Arthritis and Musculoskeletal and
Skin Diseases, NIH, Bethesda, Maryland;2Christelle Sanchez, PhD:
University of Lie `ge, Lie `ge, Belgium;
Hebrew University, Hadassah Ein Kerem, Jerusalem, Israel;4Yingjie
Song, MS: Uniformed Services University of the Health Sciences,
PhD: Ottawa Hospital Research Institute, Ottawa, Ontario, Canada.
Address correspondence to Odile Gabay, PhD, NIH, Carti-
lage Biology and Orthopedic Branch, National Institute of Arthritis
and Musculoskeletal and Skin Diseases, NIH, 50 South Drive,
Bethesda, MD 20892. E-mail: firstname.lastname@example.org.
Submitted for publication March 2, 2012; accepted in revised
form October 9, 2012.
3Mona Dvir-Ginzberg, PhD:
5Xiao Hong He, MS, Michael W. McBurney,
facial abnormalities, and their long bones mineralized
slower than normal (15). These observations were con-
sistent with the notion that SIRT-1 may play a role in
cartilage development and homeostasis. Additional ana-
lysis of heterozygous SIRT-1 mice (SIRT-1?/?) showed
that they have elevated levels of apoptotic chondrocytes
and develop more severe OA with age compared to WT
In this study, we investigated the articular car-
tilage of mice carrying a SIRT-1 protein lacking enzy-
matic activity (SIRT-1tm2.1Mcbyor SIRT-1y/y) and found
that these mutant mice were predisposed to develop
OA, especially with age. Our results showed an altered
cartilage phenotype and an acceleration of the OA
process in these SIRT-1 mutant mice due to a higher
level of cartilage breakdown and apoptosis (17).
MATERIALS AND METHODS
Animals. Animals carrying a point mutation, SIRT-
1tm2.1Mcby(SIRT-1y/y), which ablates SIRT-1 enzymatic activ-
ity, and their littermates were used in the experiments. The
point mutation encodes a SIRT-1(H355Y) protein with unde-
tectable enzymatic activity, as described by Seifert et al (17).
Animals homozygous for this point mutation are referred to as
SIRT-1y/y(y/y) mice. All animals were females ages 2 days, 6–7
days, 5 months, or 6 months, and a minimum of 3 animals per
strain were used in each experiment. Animals were housed
under standard conditions of temperature and light and were
fed a standard laboratory diet, with water ad libitum. All
procedures were performed in accordance with the National
Institutes of Health (NIH) Committees for Animal Use and
Primary mouse cell culture and cell count. Methods
for the isolation of costal cartilage from 6–7-day-old y/y or
WT mice and chondrocyte plating were adapted from those
described previously (18). Briefly, pieces of costal cartilage
were incubated in 3 mg/ml collagenase for 1 hour and 30
minutes at 37°C, carefully isolated from soft tissue in phos-
phate buffered saline (PBS), and incubated in 0.5 mg/ml collage-
nase overnight. Chondrocytes were washed, centrifuged, and
plated for a 7–10-day period during which they were allowed
to propagate. Chondrocytes from passages 0 and 1 were used
in the experiments because chondrocytes from these passages
maintain their phenotypes. Monolayer cultures were main-
tained in culture conditions as previously described (18), and
growth medium was replaced every 3 days. Prior to analy-
sis, cells were rinsed with PBS, trypsinized with trypsin 0.25%,
resuspended in 10 ml of growth medium, and counted with a
hemocytometer using a trypan blue exclusion assay.
Protein extraction and immunoblotting. Whole-cell
protein extracts from chondrocytes from the hyaline cartilage
of 6–7-day-old mice were obtained using the method of
Dvir-Ginzberg et al (9). The protein extracts were resolved by
sodium dodecyl sulfate–polyacrylamide gel electrophoresis
(10 ?g protein/lane) and transferred onto PVDF membranes
for immunoblotting. The blots were processed as previously
described (9) and probed with antibodies. Blots were devel-
oped using an alkaline phosphatase–conjugated secondary
antibody and BCIP/nitroblue tetrazolium solution (Invitro-
gen). Antibodies against protein tyrosine phosphatase 1b
(PTP1b), matrix metalloproteinase 13 (MMP-13), MMP-8, and
?-actin were purchased from Abcam, SIRT-1 antibody was
from Upstate/Millipore, MMP-9 antibody was from R&D
Systems, and acetylated NF-?B p65 antibody was from Sigma-
Histologic and immunohistochemical analyses. For
immunohistochemical analysis, cartilage samples were fixed in
4% paraformaldehyde for 24–48 hours, dehydrated in a graded
series of ethanol baths, embedded in paraffin, and cut into
5-?m sections. Slides were rehydrated and incubated with
antibodies specific for mouse anti-aggrecan; mouse anti–type II
collagen (Millipore); mouse anti-PTP1b (Abcam); mouse anti–
MMP-1, MMP-3, MMP-8, MMP-9, MMP-13, ADAMTS-4, or
ADAMTS-5 (all from Abcam); mouse anti–acetylated H4K16
(Cell Signaling Technology); or rabbit anti–acetylated NF-?B
p65 (Sigma-Aldrich) and visualized using a broad-spectrum
immunohistochemistry kit (diaminobenzidine; Invitrogen).
Sections were stained with Alcian blue to test for the
presence of glycosaminoglycans (GAGs) and with Safranin
O–light green to examine the subchondral bone and detect
OA. Joints of the paws and knees were examined at 5 and 6
months. To identify cells undergoing apoptosis, TUNEL label-
ing was performed on paraffin-embedded sections, using a
TUNEL Apoptosis Detection Kit (for Paraffin-embedded
Tissue Sections, Biotin-labeled POD) according to the recom-
mendations of the manufacturer (GenScript). Cell prolifera-
tion was examined by bromodeoxyuridine (BrdU) incorpora-
tion, carried out with an Amersham cell proliferation labeling
reagent, according to the recommendations of the manufac-
turer (GE Healthcare).
Cartilage defects at 6 months were assessed in accor-
dance with the OA Research Society International (OARSI)
histopathology initiative recommendations for histologic as-
sessment of OA in mice (19) and recommendations for grading
and scoring of OA cartilage histopathology (20). Immuno-
histochemical analysis of cartilage breakdown was performed
using the Coll2-1 peptide biomarker for collagen breakdown
(Artialis), and the results were analyzed according to the
method of Ameye et al (21).
Total skeletal staining was performed with basic ethanol–
KOH:glycerol alizarin red–Alcian blue, as described by Depew
et al (22). Results were analyzed using the Color Atlas of Fetal
Skeleton of the Mouse, Rat, and Rabbit (23).
Statistical analysis. Results are presented as the
mean ? SD. Statistical analysis was performed using one-way
analysis of variance, assuming confidence levels of ?95% (P ?
0.05) to be statistically significant. For immunoblot analysis, 3
different experiments were each repeated twice. For immuno-
histochemical analysis, the percentage of positively stained
cells was determined; stained cells reflected a ?10-fold inten-
sity above the background, as determined by scanning densi-
tometry (ImageJ software [NIH; online at http://rsbweb.nih.
gov/ij/]). A scoring method adapted from Lee et al (24) was
used to evaluate the staining of the tissue matrix. An average
of 3 fields in 3 separate sections (1 field per section) from 3
different animals per genotype (y/y or WT) was assessed. Each
field was read by 2 different readers in a blinded manner.
Cartilage defects were scored in accordance with the recom-
160GABAY ET AL
mendations of the OARSI histopathology initiative for scoring
OA in mice (19).
Reduced cartilage SIRT-1 enzymatic activity and
presence of defects in the cartilage of the skull, spine,
rib cages, and joints in y/y mice. We first confirmed a
reduction in SIRT-1 enzymatic activity in the cartilage of
the y/y mouse strain by comparing the acetylation of
histones in the y/y and WT mice (Figure 1A). Consistent
with mouse genotyping, H4K16 staining showed a higher
level of acetylation in the cartilage of y/y mice, indicating
a lower level of deacetylation of the SIRT-1 histone
Skeletal staining of the y/y mice (Figure 1C)
confirmed that they were smaller than their normal
littermates and had craniofacial and eye defects, as
previously reported (13,15). The finding that SIRT-1
enzymatic activity was reduced in the y/y mice (Figure
1A) confirmed that lack of SIRT-1 enzymatic activity
plays a role in cartilage-dependent developmental de-
fects (Figure 1B).
We further examined the effect of SIRT-1 enzy-
matic activity on the skeleton and cartilage (results are
available from the corresponding author upon request).
The y/y mice had less cartilage in the skull, rib cage,
spine, and knee joints, which was attributed to their
smaller size (Figure 1C). They exhibited less bone
mineralization, suggesting that cartilage hypotrophy re-
sults in delayed osteogenesis. Interestingly, the sutures
of the calvariae in the y/y mice were not fixed, unlike
those in the WT mice. These findings were similar in all
of the y/y mice analyzed and in all of the WT mice
analyzed (n ? 3 mice per group). Interestingly, the y/y
mice had more severe matrix disorganization of the rib
cage, defects of the eyes, and were smaller compared to
WT mice of the same age.
Reduced levels of type II collagen, aggrecan, and
GAG in joint articular cartilage tissue from y/y mice
compared to WT mice. We investigated the skeletal
structure of adult y/y mice (17). Alcian blue staining was
carried out to visualize GAG content, which is a hall-
mark of healthy cartilage. There was less GAG in the
paws and knees of y/y mice than in the paws and knees
of WT mice at 5 months (Figures 2A and B). The
6-month-old adult y/y animals also had less GAG in the
paws and knees than their WT littermates (results not
shown). To examine levels of aggrecan and type II
collagen, immunohistochemical staining of cartilage was
carried out and quantified using densitometry and sta-
tistical analysis (Figures 2A and B). Overall, the joints
Figure 1. Cartilage phenotype of SIRT-1y/y(y/y) mice. A, Top, Immunohistochemical staining for acetylated H4K16 in knee sections from
5-month-old wild-type (WT) and y/y mice. Top panels show the cartilage surface; bottom panels show the cartilage midzone. Original magnification
? 20. Bottom, Number of acetylated H4K16–stained cells per field in WT and y/y mice. An average of 6 fields per section in 3 sections from 3
separate mice per genotype was assessed. The mean number of positively stained cells was increased 1.5 times in the y/y mouse chondrocytes
compared to the WT mouse chondrocytes, indicating that the enzymatically inactive SIRT-1 in y/y mice is unable to deacetylate its histone target.
Bars show the mean ? SD. ? ? P ? 0.007 versus y/y mice. B, Comparison of the size of a 2-day-old WT mouse and a 2-day-old y/y mouse before
skeletal staining. C, Alizarin red–Alcian blue skeletal staining of 2-day-old animals, showing cartilage disparities in the skull, spine, and large joints;
disorganization of the rib cages; and major defects of the eyes and all joint cartilage in the y/y mouse.
CARTILAGE PHENOTYPE IN MICE LACKING SIRT-1 ENZYMATIC ACTIVITY 161
of y/y mice showed less Alcian blue, aggrecan, and
type II collagen staining (65%, 50%, and 19% reduction,
respectively) than WT mice at 5 months of age.
Elevated MMP-8, MMP-9, and MMP-13 protein
levels in y/y mice compared to WT mice. To determine
whether the catalytic enzymes involved in cartilage
degeneration were elevated in the y/y mice, we per-
formed immunohistochemical staining for MMP-1,
MMP-3, MMP-8, MMP-9, MMP-13, ADAMTS-4, and
ADAMTS-5, all of which have been shown to be ele-
vated in OA (25). Comparisons of the WT and y/y mice
at 5 months revealed a statistically significant increase
in MMP-8, MMP-9, and MMP-13 protein expression in
the cartilage of y/y mice, as determined by immuno-
staining and Western blot analysis (Figures 3A–C). The
y/y mice and WT mice showed similarly low levels of
immunostaining for MMP-1, MMP-3, ADAMTS-4, and
ADAMTS-5 at 5 months of age (results not shown).
Bone defects and moderate local inflammation of
the joint in y/y mice. To examine mouse knee articular
cartilage, we used Safranin O–light green staining. As
shown in Figure 4A, y/y mice exhibited differences in
articular cartilage, growth plate, and subchondral bone
morphology compared to WT mice at 6 months. More
specifically, the cartilage appeared thinner and eroded
in the y/y mice, and the subchondral bone had less
trabecular bone volume and thicker trabeculae than in
the WT mice. A distinct difference in growth plate
thickness was also observed, with a thicker growth plate
in the y/y mice (Figure 4A). Immunohistochemical ana-
lysis using acetylated NF-?B p65 antibody showed ele-
vated nuclear staining in the y/y mice as compared to
their littermates, suggesting that NF-?B may be hyper-
activated in the y/y chondrocytes, illustrating a local
inflammation due, at least in part, to the absence of
Figure 3. Elevated levels of matrix metalloproteinase 8 (MMP-8),
MMP-9, and MMP-13 protein in SIRT-1y/y(y/y) mouse cartilage. A,
Immunohistochemical staining for MMP-8, MMP-9, and MMP-13 in
the paws of 5 month-old wild-type (WT) and y/y mice. Hematoxylin
counterstained; original magnification ? 20. B, Percent of cells stained
positively for MMP-8, MMP-9, and MMP-13 in the WT and y/y mice.
The percent of positively stained cells per field was determined for an
average of 6 fields in 3 sections from 3 separate mice per genotype.
Bars show the mean ? SD. ? ? P ? 5.0416 ? 10?6for MMP-8,
P ? 7.651 ? 10?6for MMP-9, and P ? 1.7189 ? 10?5for MMP-13,
versus y/y mice. C, Immunoblotting of protein extracts from WT and
y/y mice for MMP-8, MMP-9, and MMP-13 protein levels. Results are
representative of 3 different experiments and were consistent with the
results of the immunohistochemical analysis.
Figure 2. Lower glycosaminoglycan (GAG), aggrecan, and type II
collagen content in articular cartilage from SIRT-1y/y(y/y) mice than
their wild-type (WT) littermates. A, Alcian blue staining (for GAG
content), aggrecan staining, and type II collagen staining of paw
sections from 5-month-old WT and y/y mice. Alcian blue, aggrecan,
and type II collagen staining intensity were reduced by 65%, 50%, and
19%, respectively, in the paws of y/y mice compared to those of WT
mice. Arrows indicate the difference in staining intensity for each
marker. Original magnification ? 10. B, Alcian blue staining (for GAG
content), aggrecan staining, and type II collagen staining of knee
sections from 5-month-old WT and y/y mice. Staining intensity was
reduced in the knees of the y/y mice, similar to the findings in paws.
Arrows indicate the difference in staining intensity for each marker.
Original magnification ? 16.
162 GABAY ET AL
SIRT-1 histone deacetylase activity. Moderate local
inflammation was seen in 6-month-old WT mice, consis-
tent with an early stage of OA; however, the local
inflammation was significantly more severe in the y/y
mice (Figure 4B). No systemic inflammation was ob-
served at this age.
Significantly higher level of cartilage breakdown
in y/y mice than in WT mice. The findings described
above suggest that the cartilage degeneration in y/y mice
occurred through elevated levels of MMPs and reduced
levels of GAGs and type II collagen. We further exam-
ined cartilage degeneration using Coll2-1, which is a
degradation product of type II collagen and thus serves
as a biomarker for cartilage degeneration (20). The WT
mouse cartilage showed Coll2-1 cellular localization, but
the y/y mouse cartilage showed diffuse matrix staining
(Figure 5A). Intracellular staining by Coll2-1 is charac-
teristic of healthy cartilage, whereas diffuse staining of
the ECM surrounding the cells is characteristic of un-
healthy cartilage (21). Additional analyses were carried
out, and cartilage degradation was scored in accordance
with the OARSI initiative recommendations for scoring
of OA in adult mice (Figures 5B and C). The y/y mice
had elevated levels of cartilage degeneration at 6
months, with a mean score of 3.5, corresponding to
moderate OA, whereas their WT littermates had a mean
score of 1, corresponding to early-stage OA (Figure 5C).
Enhanced apoptosis and elevated PTP1b levels
in the cartilage of y/y mice compared to WT mice. A
TUNEL assay of WT and y/y mouse cartilage sections
showed significantly more chondrocytes positive for
apoptosis in y/y mouse cartilage than in WT mouse
cartilage (Figure 6A). Further quantification of chon-
drocyte numbers within a given cartilage section and in
equivalent cartilage regions revealed that the number of
chondrocytes per cartilage area was significantly lower in
the y/y mice than in the WT mice (Figure 6B). These
data support the notion that SIRT-1 contributes to
chondrocyte proliferation, and lack of SIRT-1 activity
leads to limited growth and repair of cartilage. To
confirm these findings, we monitored the rate of chon-
drocyte division by BrdU incorporation in histologic
sections and found a reduced rate of BrdU incorpora-
tion in the y/y mouse chondrocyte cell culture as com-
pared to the WT controls (results are available from the
corresponding author upon request). As previously re-
ported by Gagarina et al (10), SIRT-1 may prevent
chondrocyte apoptosis in a PTP1b–dependent manner.
To examine PTP1b expression in vivo, immunohisto-
chemical analysis was carried out. Immunohistochemical
analysis of articular cartilage from 5-month-old y/y mice
showed PTP1b levels ?10-fold higher than those in WT
mouse cartilage (Figure 6C).
A growing body of evidence suggests that the
protein deacetylase SIRT-1 plays an important role in
cartilage biology (9,10,15,26,27). The present study dem-
onstrates that mice lacking SIRT-1 enzymatic activity
possess an altered cartilage phenotype, which is consis-
tent with our previous findings in 9-month-old SIRT-
1?/?heterozygous mice (16). Further, we described two
different phenomena underlying this phenotype: an in-
crease in chondrocyte apoptosis, and an accelerated rate
of cartilage breakdown in these SIRT-1–deficient mice.
The SIRT-1 mutant mice, called y/y mice, were
smaller in size compared to WT mice of the same age,
which is consistent with the results of previous studies
(13,14,28). We hypothesize that the skeletal defects seen
in y/y mice at 6 months could be a result of altered
cartilage metabolism attributed to variations in skeletal
growth and leading to loss of cartilage homeostasis and
cellularity with age.
Figure 4. Bone defects and local inflammation in the SIRT-1y/y(y/y)
mice. A, Histologic staining of knee sections from 6-month-old wild-type
(WT) and y/y mice with Safranin O–light green, showing some erosion
and a thinner cartilage layer in the y/y mice. The subchondral bone
showed less trabecular bone volume, thicker trabeculae, and thicker
growth plate (arrows) in y/y mice compared to WT mice. Original
magnification ? 16. B, Left, Immunohistochemical staining of knee
sections from 6-month-old WT and y/y mice for acetylated NF-?B p65.
Arrows indicate the difference in staining intensity. Original magnifica-
tion ? 20. Right, Percent of acetylated p65–stained cells per field in WT
and y/y mice. An average of 3 fields in 3 sections from 3 separate mice per
genotype was assessed. The mean percent of acetylated p65–stained cells
per field was increased 1.7 times in the y/y mice compared to the WT
mice, indicating that the enzymatically inactive SIRT-1 in the y/y mice is
unable to deacetylate its histone target. Bars show the mean ? SD. ? ?
P ? 0.042 versus y/y mice.
CARTILAGE PHENOTYPE IN MICE LACKING SIRT-1 ENZYMATIC ACTIVITY163
In this study, we used y/y mice, which express a
SIRT-1 point mutation that ablates SIRT-1 enzymatic
activity (17). Overall, our findings in 6-month-old y/y
mice were similar to those in 9-month-old SIRT-1?/?
animals. Both had severe OA and elevated chondrocyte
cell death with age (16). Here we conclude that much of
the age-related cartilage phenotype can be attributed to
SIRT-1 enzymatic activity. In order to evaluate a possi-
ble cartilage phenotype in the y/y mice, we first analyzed
SIRT-1 activity in the cartilage of these mice by moni-
toring histone acetylation of the SIRT-1 target H4K16 in
histology slides of articular cartilage derived from y/y
mice and WT mice at 5 months. The results showed a
33% increase in acetylated H4K16 in y/y mice versus WT
control (P ? 0.0073), consistent with the notion that
SIRT-1 enzymatic activity is abrogated in y/y mouse car-
tilage (Figure 1A). It is now well known that SIRT-1
preferentially deacetylases H4K16, even if H4K16 is not
its only target. In this experiment, we confirmed the re-
duction of enzymatic activity in y/y mouse cartilage.
Since the cartilage phenotype is strong in this strain, we
conclude that SIRT-1 enzymatic activity is necessary for
cartilage protection, and that the lack of SIRT-1 plays a
major role in the acceleration of the OA process with age.
We previously showed that SIRT-1 enhances
the survival of human OA chondrocytes by repressing
PTP1b and activating the insulin-like growth factor
receptor pathway (10,15). Previous studies have con-
sistently shown that inhibitors of SIRT-1 and SIRT-2
activity induce cell death through hyperacetylated p53
(29). Di Renzo et al recently proposed a cascade linking
skeletal defects and apoptosis to histone hyperacetyla-
tion (30). Accordingly, and consistent with our results,
altering SIRT-1 activity and levels could lead to en-
hanced cell death via multiple mechanisms and diminish
the protective effect SIRT-1 has on chondrocyte viability
in vivo as well as in vitro.
To date, very few studies have demonstrated
the role of SIRT-1 in vivo with regard to cartilage
biology or age-induced arthritis. Our recent study of WT
and haploinsufficient SIRT-1?/?mice compared muscu-
loskeletal features, OA severity scores, and apoptosis in
the cartilage in 1-month-old and 9-month-old mice. The
results showed a significant decrease in SIRT-1 protein
levels in the heterozygous mice compared to their
littermates at 1 month. Interestingly, both strains ceased
to express full-length SIRT-1 at 9 months with the
appearance of the cleaved inactive 75-kd SIRT-1 variant
(16,26,27). Also, 9-month-old SIRT-1?/?mice pre-
sented enhanced OA severity and chondrocyte apoptosis
compared to age-matched WT mice.
An accelerated cartilage degeneration process
was observed in y/y mice. Accumulation of MMPs in the
cartilage and low levels of type II collagen and aggrecan
could predict very weak mechanical properties of carti-
lage in these mice and susceptibility to cartilage degen-
Figure 5. Elevated rates of cartilage breakdown in SIRT-1y/y(y/y) mice. A, Immunohistochemical staining of knee sections from 6-month-old y/y
and wild-type (WT) mice with antibody against the Coll2-1 breakdown peptide. An average of 6 fields in 3 sections from 3 separate mice per
genotype was assessed, and representative fields are shown. Erosion of cartilage surface (arrows) and diffuse brown staining around the defects was
observed in the y/y mice, whereas the staining was strongly intracellular in the WT mice. Original magnification ? 20. B, Anti-aggrecan antibody
staining of knee sections from 6-month-old y/y and WT mice, showing significantly higher cartilage degradation in the y/y mice than in the WT mice,
with cartilage defects (arrows), such as erosions of the cartilage surface, deeper clefts, and fibrillations. Thirty-six knee sections from each genotype
were examined; representative results are shown. Original magnification ? 16. C, Cartilage degradation grade, determined in Safranin O–light
green–stained sections, for y/y and WT mice. The scoring system was adapted from the Osteoarthritis Research Society International (OARSI)
semiquantitative scoring system for OA severity. A score of 0 indicates no degradation; a score of 5 indicates denuded cartilage surface with sclerotic
bone. The mean scores were 3.41 for the y/y mice and 1.2 for the WT mice. Twelve samples for each genotype (2 sections of 2 different knees from
3 different mice per genotype) were scored, showing more severe grades for the y/y knees. Bars show the mean ? SD. ? ? P ? 0.05 versus WT mice.
164GABAY ET AL
eration. Our study showed increased MMP levels, with
reduced ECM content in the y/y mice until 6 months of
age. However, it is not yet clear if the decreased intensity
of aggrecan, type II collagen, and GAG staining ob-
served in the present study was due to the decrease in
the cellularity within the cartilage tissue in y/y mice.
These data are consistent with the increased severity
of OA observed in these mice, especially with age.
However, mechanically induced OA, such as the com-
monly established destabilization of the medial meniscus
or anterior cruciate ligament transection models, could
possibly show more extreme variations in OA severity
in these SIRT-1 mutant mice versus WT controls, rather
than natural cartilage occurrences attributed to SIRT-1.
Whereas MMP levels were elevated in y/y mice,
ADAMTS-4 and ADAMTS-5 aggrecanase levels were
slightly increased in both y/y and WT mice and corre-
lated with cartilage breakdown in these mouse strains. It
appears that type II collagen breakdown and MMP-13
were most affected and correlated well with OA severity
in the SIRT-1–deficient mice.
Recently, the involvement of SIRT-1 in age-
related diseases has been described in humans, as well as
the role of physical activity in increasing SIRT-1 protein
levels (31). SIRT-1 activators have been extensively
studied and have demonstrated in vitro attributes (32).
However, the relatively high concentrations required
to obtain pharmacologic effects have led to a search
for more potent SIRT-1 activators. Three SIRT-1–
activating molecules have been identified and studied in
the treatment of diabetes in mice (33) and may show
promise in retarding cartilage destruction in OA.
SIRT-1 plays a role in inflammation (34) and
in immune tolerance (35). Therefore, we tested the role
of SIRT-1 in secondary inflammation related to OA
by measuring the serum levels of the proinflamma-
tory cytokines interleukin-1? (IL-1?), tumor necrosis
factor ?, IL-6, and granulocyte–macrophage colony-
stimulating factor. We performed these analyses during
the natural age-induced OA observed at 6 months but
did not observe any difference in the levels of these
cytokines between the y/y mice and WT mice, indicating
that there was no systemic inflammation at 6 months.
However, using NF-?B p65 pathway checking, we ob-
served significantly higher local inflammation in the
y/y mice than in their WT littermates. Further study
of inflammation in the OA process in this mouse strain
could provide more evidence of the involvement of
SIRT-1 in cartilage protection, cartilage degeneration,
and inflammation tolerance associated with age.
The findings of the present study highlight the
fact that SIRT-1 and its enzymatic activity are necessary
for the normal development and homeostasis of carti-
lage in vivo. In the absence of SIRT-1 enzymatic activity,
mice develop characteristics similar to those of OA,
one of many late-onset diseases common in aging. This
study establishes that SIRT-1 plays a crucial role in
normal healthy cartilage homeostasis as well as during
skeletal development. The data obtained so far support
conducting further studies to develop novel pharmaceu-
tical targets for SIRT-1 activation in cartilage suscepti-
ble to damage and degeneration.
The authors wish to acknowledge the following indi-
viduals for their valuable assistance: Dr. Yongqing Chen for
mouse pictures, Dr. Vittorio Sartorelli for reviewing the man-
Figure 6. Increased number of apoptotic chondrocytes in SIRT-1y/y
(y/y) mice compared to wild-type (WT) mice. A, Top, Sections of knee
cartilage from 6-month-old WT and y/y mice. Sections were subjected
to a TUNEL assay, and the number of chondrocytes per cartilage
region was determined. Original magnification ? 10. Bottom, Number
of chondrocytes per field in WT and y/y mice. The number of positively
stained cells was determined in 3 different sections in 2 different knees
of 3 separate mice per genotype. A significantly higher level of
apoptosis was observed in the y/y mice than in the WT mice. Bars show
the mean ? SD. ? ? P ? 0.00238 versus WT mice. B, Number of
chondrocytes per cartilage area in the superficial zone of the knee in
WT and y/y mice. Cells were counted in 50-?m2fields in 3 histologic
panels for 3 separate mice per genotype. Significantly fewer cells were
seen in the y/y mice. Bars show the mean ? SD. ? ? P ? 0.002 versus
WT mice. C, Top, Immunohistochemical staining of knee sections
from 5-month-old WT and y/y mice for protein tyrosine phosphatase
1b (PTP1b). Original magnification ? 10. Bottom, Percent of posi-
tively stained cells per field in WT and y/y mice. An average of 6 fields
from 3 sections of 3 separate mice per genotype was assessed.
Enhanced PTP1b intensity was observed in the y/y mice as compared
to the WT mice of the same age (10-fold increase). Bars show the
mean ? SD. ? ? P ? 0.0002426 versus y/y mice.
CARTILAGE PHENOTYPE IN MICE LACKING SIRT-1 ENZYMATIC ACTIVITY 165
uscript, Evelyne Ralston for providing imaging platforms, and Download full-text
Youngmi Ji for conducting the NF-?B experiment. The authors
also acknowledge Dr. David Engel for editing the manuscript and
Mr. Richard Booth for assistance in statistical analyses.
All authors were involved in drafting the article or revising it
critically for important intellectual content, and all authors approved
the final version to be published. Dr. Gabay had full access to all of the
data in the study and takes responsibility for the integrity of the data
and the accuracy of the data analysis.
Study conception and design. Gabay, Dvir-Ginzberg.
Acquisition of data. Gabay, Sanchez, Zaal, Song, He, McBurney.
Analysis and interpretation of data. Gabay, Dvir-Ginzberg, Gagarina.
1. Goldring MB. The role of cytokines as inflammatory mediators in
osteoarthritis: lessons from animal models. Connect Tissue Res
2. Zhang WG, Bai XJ, Chen XM. SIRT1 variants are associated with
aging in a healthy Han Chinese population. Clin Chim Acta
3. Loeser RF. Age-related changes in the musculoskeletal system and
the development of osteoarthritis. Clin Geriatr Med 2010;26:
4. Lawrence RC, Felson DT, Helmick CG, Arnold LM, Choi H,
Deyo RA, et al, for the National Arthritis Data Workgroup. Es-
timates of the prevalence of arthritis and other rheumatic condi-
tions in the United States: part II. Arthritis Rheum 2008;58:26–35.
5. Tissenbaum HA, Guarente L. Increased dosage of a sir-2 gene
extends lifespan in Caenorhabditis elegans. Nature 2001;410:
6. Michan S, Sinclair D. Sirtuins in mammals: insights into their
biological function. Biochem J 2007;404:1–13.
7. Backesjo CM, Li Y, Lindgren U, Haldosen LA. Activation of Sirt1
decreases adipocyte formation during osteoblast differentiation of
mesenchymal stem cells. J Bone Miner Res 2006;21:993–1002.
8. Zeng L, Chen R, Liang F, Tsuchiya H, Murai H, Nakahashi T,
et al. Silent information regulator, sirtuin 1, and age-related
diseases. Geriatr Gerontol Int 2009;9:7–15.
9. Dvir-Ginzberg M, Gagarina V, Lee EJ, Hall DJ. Regulation of
cartilage-specific gene expression in human chondrocytes by SirT1
and nicotinamide phosphoribosyltransferase. J Biol Chem 2008;
10. Gagarina V, Gabay O, Dvir-Ginzberg M, Lee EJ, Brady JK, Quon
MJ, et al. SirT1 enhances survival of human osteoarthritic chon-
drocytes by repressing protein tyrosine phosphatase 1B and acti-
vating the insulin-like growth factor receptor pathway. Arthritis
11. Takayama K, Ishida K, Matsushita T, Fujita N, Hayashi S, Sasaki
K, et al. SIRT1 regulation of apoptosis of human chondrocytes.
Arthritis Rheum 2009;60:2731–40.
12. Fujita N, Matsushita T, Ishida K, Kubo S, Matsumoto T, Ta-
kayama K, et al. Potential involvement of SIRT1 in the pathogen-
esis of osteoarthritis through the modulation of chondrocyte gene
expressions. J Orthop Res 2011;29:511–5.
13. McBurney MW, Yang X, Jardine K, Hixon M, Boekelheide K,
Webb JR, et al. The mammalian SIR2? protein has a role in
embryogenesis and gametogenesis. Mol Cell Biol 2003;23:38–54.
14. McBurney MW, Yang X, Jardine K, Bieman M, Th’ng J, Lemieux
M. The absence of SIR2? protein has no effect on global gene
silencing in mouse embryonic stem cells. Mol Cancer Res 2003;1:
15. Lemieux ME, Yang X, Jardine K, He X, Jacobsen KX, Staines
WA, et al. The Sirt1 deacetylase modulates the insulin-like growth
factor signaling pathway in mammals. Mech Ageing Dev 2005;126:
16. Gabay O, Oppenhiemer H, Meir H, Zaal K, Sanchez C, Dvir-
Ginzberg M. Increased apoptotic chondrocytes in articular carti-
lage from adult heterozygous SirT1 mice. Ann Rheum Dis 2012;
17. Seifert EL, Caron AZ, Morin K, Coulombe J, He XH, Jardine K,
et al. SirT1 catalytic activity is required for male fertility and
metabolic homeostasis in mice. FASEB J 2012;26:555–66.
18. Gabay O, Gosset M, Levy A, Salvat C, Sanchez C, Pigenet A, et al.
Stress-induced signaling pathways in hyalin chondrocytes: inhibi-
tion by avocado-soybean unsaponifiables (ASU). Osteoarthritis
19. Glasson SS, Chambers MG, Van Den Berg WB, Little CB. The
OARSI histopathology initiative—recommendations for histolog-
ical assessments of osteoarthritis in the mouse. Osteoarthritis
Cartilage 2010;18 Suppl 3:S17–23.
20. Pritzker KP, Gay S, Jimenez SA, Ostergaard K, Pelletier JP,
Revell PA, et al. Osteoarthritis cartilage histopathology: grading
and staging. Osteoarthritis Cartilage 2006;14:13–29.
21. Ameye LG, Deberg M, Oliveira M, Labasse A, Aeschlimann JM,
Henrotin Y. The chemical biomarkers C2C, Coll2-1, and Coll2-
1NO2provide complementary information on type II collagen
catabolism in healthy and osteoarthritic mice. Arthritis Rheum
22. Depew MJ. Analysis of skeletal ontogenesis through differential
staining of bone and cartilage. Methods Mol Biol 2008;461:37–45.
23. Yasuda M, Yuki T, Tanimura T. Color atlas of fetal skeleton of the
mouse, rat, and rabbit. Congenit Anom (Kyoto) 1996;36:263–5.
24. Lee JH, Fitzgerald JB, DiMicco MA, Cheng DM, Flannery CR,
Sandy JD, et al. Co-culture of mechanically injured cartilage with
joint capsule tissue alters chondrocyte expression patterns and
increases ADAMTS5 production. Arch Biochem Biophys 2009;
25. Verma P, Dalal K. ADAMTS-4 and ADAMTS-5: key enzymes in
osteoarthritis. J Cell Biochem 2011;112:3507–14.
26. Dvir-Ginzberg M, Gagarina V, Lee EJ, Booth R, Gabay O, Hall
DJ. Tumor necrosis factor ?–mediated cleavage and inactivation
of SirT1 in human osteoarthritic chondrocytes. Arthritis Rheum
27. Oppenheimer H, Gabay O, Meir H, Haze A, Kandel L, Liebergall
M, et al. 75-kd sirtuin 1 blocks tumor necrosis factor ?–mediated
apoptosis in human osteoarthritic chondrocytes. Arthritis Rheum
28. Cheng HL, Mostoslavsky R, Saito S, Manis JP, Gu Y, Patel P, et al.
Developmental defects and p53 hyperacetylation in Sir2 homolog
(SIRT1)-deficient mice. Proc Natl Acad Sci U S A 2003;100:
29. Peck B, Chen CY, Ho KK, Di Fruscia P, Myatt SS, Coombes RC,
et al. SIRT inhibitors induce cell death and p53 acetylation
through targeting both SIRT1 and SIRT2. Mol Cancer Ther 2010;
30. Di Renzo F, Broccia ML, Giavini E, Menegola E. VPA-related
axial skeletal defects and apoptosis: a proposed event cascade.
Reprod Toxicol 2010;29:106–12.
31. Corbi G, Conti V, Scapagnini G, Filippelli A, Ferrara N. Role of
sirtuins, calorie restriction and physical activity in aging. Front
Biosci (Elite Ed) 2012;4:768–78.
32. Howitz KT, Bitterman KJ, Cohen HY, Lamming DW, Lavu S,
Wood JG, et al. Small molecule activators of sirtuins extend
Saccharomyces cerevisiae lifespan. Nature 2003;425:191–6.
33. Milne JC, Lambert PD, Schenk S, Carney DP, Smith JJ, Gagne
DJ, et al. Small molecule activators of SIRT1 as therapeutics for
the treatment of type 2 diabetes. Nature 2007;450:712–6.
34. Sequeira J, Boily G, Bazinet S, Saliba S, He X, Jardine K, et al.
sirt1-null mice develop an autoimmune-like condition. Exp Cell
35. Zhang J, Lee SM, Shannon S, Gao B, Chen W, Chen A, et al. The
type III histone deacetylase Sirt1 is essential for maintenance of
T cell tolerance in mice. J Clin Invest 2009;119:3048–58.
166GABAY ET AL