All authors are or were Amgen employees and own stock or stock options in Amgen. Mr. Graham and Ms Warmington are listed as inventors on an Amgen, Inc. and UCB Pharma patent for sclerostin antibodies.
Sclerostin plays a major role in regulating skeletal bone mass, but its effects in articular cartilage are not known. The purpose of this study was to determine whether genetic loss or pharmacologic inhibition of sclerostin has an impact on knee joint articular cartilage.
Expression of sclerostin was determined in articular cartilage and bone tissue obtained from mice, rats, and human subjects, including patients with knee osteoarthritis (OA). Mice with genetic knockout (KO) of sclerostin and pharmacologic inhibition of sclerostin with a sclerostin-neutralizing monoclonal antibody (Scl-Ab) in aged male rats and ovariectomized (OVX) female rats were used to study the effects of sclerostin on pathologic processes in the knee joint. The rat medial meniscus tear (MMT) model of OA was used to investigate the pharmacologic efficacy of systemic Scl-Ab or intraarticular (IA) delivery of a sclerostin antibody–Fab (Scl-Fab) fragment.
Sclerostin expression was detected in rodent and human articular chondrocytes. No difference was observed in the magnitude or distribution of sclerostin expression between normal and OA cartilage or bone. Sclerostin-KO mice showed no difference in histopathologic features of the knee joint compared to age-matched wild-type mice. Pharmacologic treatment of intact aged male rats or OVX female rats with Scl-Ab had no effect on morphologic characteristics of the articular cartilage. In the rat MMT model, pharmacologic treatment of animals with either systemic Scl-Ab or IA injection of Scl-Fab had no effect on lesion development or severity.
Genetic absence of sclerostin does not alter the normal development of age-dependent OA in mice, and pharmacologic inhibition of sclerostin with Scl-Ab has no impact on articular cartilage remodeling in rats with posttraumatic OA.
Osteoarthritis (OA) is a common degenerative disease of the joints and a major healthcare burden in today's aging population (1–3). All structures within the joint may be affected during progression of the disease, but the underlying causes are not well understood (4). Many factors contribute to the disease, including age, alterations in joint mechanical stability, body mass index, inflammation, and genetic heritability estimates that vary depending on the affected site (5–7).
Loss of cartilage and subsequent narrowing of the joint space are features of worsening OA (8, 9). Degradation of the articular cartilage extracellular matrix is associated with changes in the anabolic and catabolic functions of embedded chondrocytes following exposure to multiple signals (10, 11). The potential role of the Wnt signaling pathway in the pathogenesis of OA has been reviewed (12–14), and recent studies suggest that changes in the activity of this pathway can occur in OA (15–17).
A critical role for canonical Wnt signaling is also well established in the process of osteogenesis and in the control of bone mass (18, 19). Several soluble inhibitors of Wnt signaling show profound effects on bone mass, including the effects of sclerostin (20). Mutations in the human SOST gene lead to sclerosteosis (21, 22), and deletion of sclerostin in mice results in a high bone mass phenotype (23). Preclinical studies (24, 25) and recent human clinical trials with sclerostin-neutralizing monoclonal antibody (Scl-Ab) therapy have shown beneficial effects on bone mineral density (BMD) and bone formation and resorption markers (26).
Recently, it was demonstrated that sclerostin is expressed in articular cartilage, and in cell-based studies, it was proposed to play a chondroprotective role in response to catabolic signals (27). In the current study, we also demonstrate expression of sclerostin in articular chondrocytes. However, neither genetic deletion of sclerostin in mice nor pharmacologic inhibition of sclerostin with Scl-Ab therapy in rats showed any major effects on articular cartilage.
MATERIALS AND METHODS
Histologic analysis of articular cartilage.
Knee joints were obtained from male sclerostin-knockout (KO) mice and wild-type (WT) littermates and from both male and female Sprague-Dawley rats. For cartilage and bone assessment at the femoral diaphysis and tibial plateau, individual femurs and tibias were dissected and fixed in 10% neutral buffered formalin (NBF). Two undecalcified frontal sections were evaluated per animal for cartilage histology and for determination of bone formation parameters. To ensure that similar regions were compared for analysis, the shape of the growth plate was used as a landmark for orientation of sections. Quantitative analysis of the dimensions of the mouse and rat cartilage was performed by one unblinded observer (QTN) using Osteomeasure bone-analysis software (Osteometrics). The cartilage area included the entire region between the 2 cartilage surfaces from the medial to the lateral aspect. The thickness of the cartilage reflected the average distance measured between the 2 cartilage surfaces. The percentage length of the cartilage surface from the medial to lateral aspect was calculated by dividing the length of the cartilage surface by the length of the bone surface.
For Mankin assessment of the severity of cartilage lesions (28, 29), both knee joints with intact femur and tibia from WT and sclerostin-KO mice were positioned at a standardized 90° angle and fixed in 10% NBF. Tissues were decalcified and frontal serial sections (n = 2 per animal) were stained with hematoxylin and eosin and Safranin O. Mankin scoring included analysis of the structure, cells, and matrix staining of the articular cartilage, as described previously (28, 29). The femur and tibia were scored independently, with the medial and lateral aspects combined for each tissue, and a final additive score for both tissues was generated for comparison between WT and sclerostin-KO mice.
Immunohistochemical and in situ hybridization analyses were performed on tissue specimens from male WT and sclerostin-KO mice (age 12 weeks, n = 3 per group) and male Sprague-Dawley rats (age 12 weeks, n = 3). Mankin scoring for the severity of histopathologic features of the mouse knee joints was performed in a blinded manner by an independent pathologist (MR). Immunohistochemical staining of human knee joint sections was also scored in a blinded manner. In the rat medial meniscus tear (MMT) studies (see below), a second independent pathologist performed the histologic scoring, also in a blinded manner.
Procurement of human tissue.
Human tissue samples were obtained from the tissue bank at Articular Engineering. The samples comprised knee joint tissue obtained from patients undergoing total knee replacement for OA (n = 5 men and n = 7 women, mean ± SD age 65 ± 3 years [range 60–69 years]) and from clinically normal cadaveric donors (n = 3 men and n = 3 women, mean ± SD age 58 ± 11 years [range 39–70 years]).
Slabs of human femoral condyle and tibial plateau tissue, measuring ∼30 mm in length, 20 mm in width, and 10 mm in depth, with overlying articular cartilage were fixed in 10% formalin for 48 hours. A circular saw was used to cut two 4-mm–thick slices, one centered on normal-appearing cartilage and the second on the worst-appearing OA cartilage lesion. Slices were decalcified and sections were stained with both hematoxylin and eosin and Safranin O. The severity of the histopathologic features of the cartilage was scored based on the Mankin protocol.
A semiquantitative assessment was performed to determine the number of sclerostin-positive cells in the human knee joint specimens showing cartilage and bone. The entire articular cartilage area contained within the section was used as the region of interest, and the percentage of sclerostin-positive chondrocytes was scored in increments of 10, on a scale from 0 to 100%, in the superficial, intermediate, deep, and calcified zones of the cartilage.
A separate pool of normal (n = 16) and human OA (n = 16) cartilage biopsy tissue samples was obtained for preparation of total RNA. Real-time polymerase chain reaction (PCR) assays were performed for messenger RNA (mRNA) expression of sclerostin (normal group n = 11 [7 men and 4 women, mean ± SD age 54 ± 14 years]; OA group n = 12 [6 men and 6 women, mean ± SD age 60 ± 6 years]) and DKK1 (normal group n = 11 [6 men and 5 women, mean ± SD age 56 ± 14 years]; OA group n = 12 [5 men and 7 women, mean ± SD age 60 ± 6 years]).
Immunohistochemical analyses to assess the expression of sclerostin were performed using standard protocols. In these experiments, 5-μm sections of knee joint tissue were formalin fixed and immunostained with a goat anti-mouse sclerostin antibody (AF1589; R&D Systems) at concentrations of 0.5–4 μg/ml.
In situ hybridization protocol.
Isotopic in situ hybridization analyses of sclerostin expression in the knee joint tissue were performed using standard protocols. For this procedure, specific complementary DNA templates for mouse (NM_024449, nucleotides 43–678), rat (AF32674.1, nucleotides 21–252), and human (NM_025237, nucleotides 11–771) sclerostin were used.
Real-time PCR was performed on purified human cartilage RNA using reagents for human sclerostin (forward primer GAA-TGA-TGC-CAC-GGA-AAT-CAT, reverse primer CGG-TTC-ATG-GTC-TTG-TTG-TTC-TC, probe 6-FAM–AC-CCC-GAG-CCT-CCA-CCG-GAG–TAM), while for DKK1, RT2 Profiler PCR arrays specific for the Wnt signaling pathway (PAHS-043A; SA Biosciences) were used.
Sclerostin-KO mouse studies and knee joint assessment.
All animal protocols and procedures were approved by the Institutional Animal Care and Use Committee of Amgen. Male sclerostin-KO mice and WT littermates (n = 12 per group) at 12 months of age were used for dynamic histomorphometric analysis of the distal femoral epiphysis and for histologic analysis of the articular cartilage. A separate cohort of older (age 16 months) male sclerostin-KO mice and WT littermates (n = 16 per group) was used for histologic analysis of the whole knee joints and for lesion assessment based on the Mankin scoring protocol.
Pharmacologic studies and analysis of articular cartilage after treatment.
In one pharmacologic study, 16-month-old male Sprague-Dawley rats were treated with Scl-Ab (25 mg/kg subcutaneously [SC], twice per week) or vehicle control (n = 10 per group) for 5 weeks. In a second study, 14-month-old ovariectomized (OVX) female Sprague-Dawley rats (7 months post-OVX) were treated with Scl-Ab (15 mg/kg SC, twice per week) or vehicle control (n = 9–10 per group) for 12 weeks. In both studies, dynamic histomorphometry and cartilage analyses were performed on the epiphyseal region of the proximal tibia.
Histomorphometric analysis of bone.
The knee joints of mice and rats were labeled with calcein, as previously described (23, 24). Subchondral bone and articular cartilage were evaluated at the epiphyseal region of either the distal femur or proximal tibia. Undecalcified frontal sections (4 μm in thickness) were used to measure bone parameters, including the bone volume (calculated as bone volume/total volume, in %), mineralizing surface (calculated as mineralized surface/bone surface, in %), and bone formation (calculated as bone formation rate/bone surface, in μm3/μm2/day), across the entire epiphyseal region.
Preparation of a sclerostin antibody–Fab (Scl-Fab) fragment.
Sclerostin antibody fragment Scl-Fab (MW 48 kd) and, as control, keyhole limpet hemocyanin (KLH)–derived Fab (KLH-Fab; MW 48 kd) were produced using standard cloning techniques. Fab fragments were tested for endotoxin contamination, and test results confirmed that the fragments contained <0.06 endotoxin units/mg. Based on the final formulation, the maximum amount of Fab that could be injected intraarticularly (IA) in a volume of 50 μl was 385 μg.
MC3T3-E1-STF assay for Scl-Ab activity.
Stable MC3T3-E1-SuperTopFlash (STF) osteoblast cells were grown in α-minimum essential medium (α-MEM; Gibco BRL) containing 10% fetal bovine serum and 1 μg/ml puromycin for selection. Cells were switched to differentiation medium containing α-MEM, 50 μg/ml ascorbic acid (Sigma), and 10 mM β-glycerophosphate (Sigma), with daily changes of medium for 4 days prior to evaluation of cell activity in luciferase assays. Rat sclerostin protein (0.2 μg/ml) was preincubated with Scl-Ab or Scl-Fab (at concentrations of 0.5, 1, and 2 μg/ml) at 37°C for 1 hour prior to the addition of complexes to the cells for 24 hours. Cells were treated with lysis buffer (Promega) for measurement of luciferase activity.
Ex vivo dual x-ray absorptiometry (DXA).
BMD of the whole femurs with intact epiphyses was assessed in 2 treatment groups of rats (vehicle-treated n = 15, Scl-Ab–treated n = 20) in the rat MMT study. BMD was determined using DXA, with the Piximus II system (GE/Lunar Medical Systems).
Pharmacologic studies in the rat MMT model of OA.
Studies using the rat MMT model of OA were performed at Bolder BioPATH in Colorado. These studies were approved by the Bolder BioPATH Institutional Animal Care and Use Committee. MMT surgery was performed as described previously (30, 31).
In one study, male Lewis rats weighing 260–286 grams (mean 276 grams) on day −1 prior to MMT surgery were injected with either Scl-Ab (25 mg/kg SC, twice per week) or vehicle control (n = 20 per group) beginning on the day of surgery and continuing until termination at 3 weeks postsurgery. At necropsy, the left (unoperated) femurs (n = 15 per group) were transferred to 70% ethanol for ex vivo DXA analysis. In a second study, male Lewis rats weighing 281–330 grams (mean 310 grams) on day −3 prior to MMT surgery were injected with either Scl-Fab (385 μg/50 μl IA, twice per week), KLH-Fab (385 μg/50 μl IA, twice per week), or vehicle control (n = 20 per group) beginning 3 days prior to surgery and continuing until termination at 3 weeks postsurgery. It was thought that the smaller-sized Scl-Fab fragment would facilitate its movement into the articular cartilage.
Histologic lesion assessment in the rat MMT model of OA.
In rats with MMT-induced OA, 3 frontal sections were cut from each operated knee at ∼200-μm steps, followed by staining of the sections with toluidine blue. All 3 sections of each knee were analyzed microscopically. A number of criteria were used to quantitatively assess the lesion severity, including 1) the cartilage degeneration width, 2) the cartilage degeneration score, 3) the lesion depth ratio, and 4) the osteophyte score (30, 31).
Weight-bearing assessment in the rat MMT model of OA.
The extent of weight-bearing of the rat knee joints following MMT surgery (left [unoperated] knee versus right [operated] knee) was recorded in each treatment group on day 20 or day 17 postsurgery, depending on the study, using an incapacitance meter. The difference in force (left knee minus right knee force) and the right-paw force (as a percentage of the total force exerted by both paws) were determined and compared between treatment groups, as was the percentage of the total body weight that was carried on the hind legs (32).
For human OA studies, two-way analysis of variance (ANOVA) was used for comparisons of the percentage of sclerostin-positive cells between the normal and OA cartilage groups and within different cellular compartments. For mouse KO and pharmacologic studies, Student's unpaired 2-tailed t-test was used for comparisons between 2 groups. One-way ANOVA followed by Tukey's post hoc test was used for statistical comparisons among 3 groups. For pharmacologic studies in the rat MMT model, data were analyzed using Student's t-test or Mann-Whitney U test (for nonparametric data). For comparisons of the ex vivo DXA findings, data were analyzed by Student's one-tailed t-test using a 2-sample equal variance assumption. Significance for all tests was set at P values less than 0.05.
Sclerostin expression in rodent articular chondrocytes.
In the long bones of 12-week-old WT mice, strong immunoreactivity for sclerostin was observed not only in cortical and cancellous bone osteocytes, as expected, but also in chondrocytes associated with the articular cartilage (Figure 1A). In sclerostin-KO mice, no expression of sclerostin was detected in the bone or cartilage by immunohistochemical staining (results not shown). In the WT mice, intense staining was apparent in the deep chondrocyte layer approaching the tidemark, whereas less staining was generally evident in the superficial zone chondrocytes. Strong expression of sclerostin in the deep layers of the WT mouse articular cartilage was confirmed by in situ hybridization (Figure 1B). Similar results were observed in the articular cartilage of 12-week-old rats (Figure 1B).
Sclerostin expression in normal and OA human articular cartilage.
Analysis of sclerostin immunoreactivity in normal human knee joint specimens clearly demonstrated the presence of sclerostin protein in the articular cartilage (Figure 1A). Staining was also evident in the underlying subchondral bone (results not shown). Within the normal human cartilage, sclerostin protein appeared to show variable expression, extending from chondrocytes located within the superficial zone to those throughout the deeper layers of the articular cartilage. Generally, it appeared that the least-intense staining for sclerostin was associated with cartilage regions of the middle zone. Sclerostin expression in normal human articular cartilage was confirmed by in situ hybridization (Figure 1B).
Sclerostin immunoreactivity was also evident in articular cartilage from patients with knee OA (Figure 1A). Similar to our observations in normal human cartilage specimens, there appeared to be variation in the intensity and location of the sclerostin expression across the different chondrocyte zones. Furthermore, sclerostin staining was also detected in the chondrocyte clusters that are often observed to be present in OA articular cartilage (Figure 1A). Sclerostin expression in OA cartilage specimens was confirmed by in situ hybridization (Figure 1B).
When the fraction of sclerostin-immunoreactive cells was counted in normal human knee specimens and compared to that in knee specimens from OA patients, there did not appear to be any significant differences in the percentages of sclerostin-positive cells across any of the chondrocyte layers, including the subchondral bone compartment (Figure 2A). Moreover, when the severity of the OA lesions was taken into account using Mankin scoring, there was still no dramatic difference in the percentage of sclerostin-positive cells across the different cartilage zones (Figure 2B).
Quantitative real-time PCR analysis of sclerostin mRNA expression in cartilage tissue showed no difference in expression between normal and OA human cartilage, despite more variation in the mRNA levels in OA cartilage specimens (Figure 2C). These results were consistent with the findings from regional immunohistochemical analysis of sclerostin protein expression in the OA cartilage (as shown in Figure 2A). In contrast, however, a significant increase in DKK1 mRNA expression was observed in cartilage from patients with OA compared to normal human cartilage, and the presence of OA was also associated with greater variation in DKK1 mRNA levels when compared to that in normal samples (Figure 2D). This increase in DKK1 mRNA expression in OA human cartilage is consistent with findings in previous studies (33, 34).
Histomorphometric analysis of the knee joints from 12-month-old male sclerostin-KO mice.
The results of bone histomorphometric analysis demonstrated a 65% increase in bone volume and 58% increase in bone formation rate in the distal femoral epiphysis of 12-month-old male sclerostin-KO mice compared to WT control mice (Figure 3A), consistent with previous findings (23). The bone marrow area was significantly decreased in sclerostin-KO mice compared to WT mice. However, detailed quantitative histologic analysis of the articular cartilage compartment indicated that both the cartilage area and the cartilage thickness were not significantly different between WT and sclerostin-KO mice (Figure 3A). In addition, the percentage of cartilage surface covering the joint surface was also unchanged.
Joint phenotype of 16-month-old male sclerostin-KO mice.
The femoral condyles of 16-month-old WT mice showed loss of articular cartilage, roughening of the articular surface, and substantial loss of proteoglycans as shown by Safranin O staining (Figure 3B). In addition, the knee joints showed evidence of osteophyte formation, consistent with indications of age-dependent onset of OA. Similar features were also observed in age-matched sclerostin-KO mice. However, a major difference in the sclerostin-KO mice was the increased amount of subchondral bone, which is a known feature associated with the high bone mass phenotype in these mice. However, despite the increased mass of subchondral bone in the sclerostin-KO mice, essentially all other joint features were similar to those in WT mice.
The severity of the joint phenotype was assessed quantitatively using the Mankin scoring system. Compared to WT mice, the sclerostin-KO mice did not show any significant difference in lesion severity (Figure 3C).
Pharmacologic treatment with Scl-Ab and effects on articular cartilage in aged male rats and OVX female rats.
In aged male rats and OVX female rats, treatment with Scl-Ab increased the bone volume and bone formation rate at the proximal tibial epiphysis (Figures 4A and B). However, despite these observations of acute (nongenetic) enhancement of bone mass in both male and female rats, no morphologic changes to the articular cartilage were noted after treatment. Both the area and the thickness of the cartilage were unchanged by treatment with Scl-Ab compared to that with vehicle control, suggesting that in aged male rats or osteopenic, OVX female rats, treatment with Scl-Ab will not alter the integrity of the articular surface.
Systemic effects of Scl-Ab in the rat MMT model of OA.
Results of the ex vivo DXA analysis of the intact (uninjured) left femurs of rats in the MMT model of OA confirmed that a significant increase in BMD occurred following treatment with Scl-Ab (Figure 5A), consistent with previous findings (24). As evident in the representative histologic images shown in Figure 5B, the knee joints from both treatment groups of rats showed development of typical joint damage in this MMT model of OA, with the greatest effects seen on the medial tibial plateau. However, no major differences in gross lesions were observed between the vehicle-treated and Scl-Ab–treated groups. In the Scl-Ab–treated animals, bone sclerosis was observed in both the medial and the lateral tibial subchondral bone compartments, as would be expected based on the known effects of Scl-Ab treatment in rodents.
The major criteria used to assess lesion severity in the rat knee joints are shown diagrammatically in Figure 5C. None of the key parameters of cartilage lesion severity was significantly different between the vehicle-treated and Scl-Ab–treated rats (Figure 5D). Moreover, no changes in lesion scores were noted, despite the development of bone sclerosis in rats following 3 weeks of Scl-Ab treatment.
No differences in the mean weight supported by the hind limbs (left [unoperated] versus right [operated] legs) were observed following treatment with either vehicle or Scl-Ab (Figure 5E). In addition, no difference in the total percentage of hind-limb force supported by the operated (injured) leg was observed following treatment with either vehicle or Scl-Ab (Figure 5E).
Intraarticular effects of Scl-Fab in the rat MMT model of OA.
The sclerostin-neutralizing activity of Scl-Fab was confirmed to be as effective as that of the parent Scl-Ab IgG in the rat MMT model of OA (Figure 6A). In contrast, preincubation with KLH-Fab had no effect on the Wnt inhibitory activity of sclerostin. Similar to our observations following treatment with Scl-Ab, there were no overt gross histologic differences between the control-treated groups and Scl-Fab–treated group (Figure 6B). No subchondral bone sclerosis was observed following local delivery of Scl-Fab. Furthermore, quantitative assessment of lesion severity revealed no significant effect of Scl-Fab on cartilage destruction or osteophyte formation when compared to that in the control groups (Figure 6C).
The dose of Scl-Ab used, when administered systemically, results in a maximum concentration (Cmax) of ∼200 μg/ml and >10,000-fold coverage of the target in the circulation (results not shown). The estimated Cmax for Scl-Fab, when administered via the IA route, is ∼7.7 mg/ml, suggesting that the Scl-Fab is present in very large excess to satisfy local target coverage. However, we do not have any data on the local concentration of sclerostin protein and the potential of the Scl-Fab to penetrate damaged cartilage as occurs in the rat MMT model.
Similar to the results reported above for Scl-Ab, the effects of Scl-Fab treatment on weight-bearing showed no difference when compared to either the vehicle-treated controls or the KLH-Fab–treated controls (Figure 6D). Moreover, there were no differences in gait between the treatment groups (results not shown).
The present study provides evidence of the expression of sclerostin mRNA and protein in rodent and human articular chondrocytes, including human OA cartilage. However, in contrast to the in vitro chondroprotective effects of sclerostin that have been reported previously (27), the current in vivo studies showed that neither genetic loss of sclerostin in mice nor pharmacologic inhibition of sclerostin in intact male or OVX female rats had an impact, implying that sclerostin may not have a critical role in articular cartilage. Furthermore, pharmacologic treatment of rats with Scl-Ab or Scl-Fab following MMT surgery also did not appear to affect either the histopathologic features of the knee joints or the development of cartilage lesions.
Expression of sclerostin is typically associated with osteocytes, particularly the more mature osteocytes that are surrounded by a mineralized matrix (35, 36). Nevertheless, some studies have demonstrated sclerostin expression in other cell types, such as hypertrophic chondrocytes in the growth plate and cementocytes (37–39). A recent study provided strong evidence for sclerostin expression in cartilage, particularly in the deeper chondrocyte cell layers of murine articular cartilage (27), and results of the present study support this finding. Species differences in the local regulation of sclerostin expression could explain the apparent regional differences observed in sclerostin expression between rodent and human cartilage. Since mechanical loading is known to influence sclerostin expression in bone, and loading is known to affect signal transduction in articular cartilage, it is possible that unique loading signals associated with different zonal regions in the articular cartilage may contribute to species-specific regulation of sclerostin expression.
In addition to the observations in normal human chondrocytes, strong sclerostin staining was also evident in the chondrocyte clusters that are often observed in damaged OA articular cartilage. This finding also supports the recent report describing the expression of sclerostin in some, but not all, human OA cartilage samples examined (27). Wnt signaling may play a role in this pathogenic adaptive response, as was demonstrated by the increased β-catenin activity observed in the Hartley guinea pig model of OA (40) and also in human OA cartilage (17, 41, 42). In addition, constitutively active β-catenin expression has been shown to accelerate cartilage destruction and the onset of OA in mice (17). However, in contrast to the activation of β-catenin, previous studies also showed that inhibition of β-catenin can lead to destruction of articular cartilage and increased chondrocyte apoptosis (43, 44).
When sclerostin expression was correlated with OA lesion severity in human OA cartilage specimens, there was no obvious relationship suggesting a role for sclerostin in the pathogenesis of OA. Previously, it was reported that sclerostin expression was modestly reduced in bone from lumbar zygapophyseal joints (45) and in femoral neck biopsy samples from human OA specimens (46), but a similar response was not observed in the human knee joint specimens in the current study. In agreement with our present findings, one study found that sclerostin expression appeared to be unchanged in the bone tissue of patients with OA, despite changes in other Wnt pathway genes (47). Our findings, based on a combination of real-time PCR, in situ hybridization, and immunohistochemistry, are in contrast to the recent findings from a microarray analysis of human OA cartilage, which showed a 14-fold increase in sclerostin mRNA expression in OA cartilage biopsy specimens (48). It is not clear why those findings with regard to sclerostin mRNA expression differed from ours, but the differences could be related to patient heterogeneity, cartilage biopsy sampling differences, or differences in the assay methods used.
The results of the present study indicate that sclerostin mRNA expression was unchanged in human OA cartilage, but DKK1 mRNA levels showed an increase, which is similar to the findings from other studies showing that the Wnt antagonist DKK1 was up-regulated in OA cartilage (33, 34, 49). However, although studies involving both inhibition of DKK1 in vitro and systemic administration of DKK1 antisense in vivo suggested a catabolic role for DKK1 (33, 49), the local overexpression of DKK1 specifically in the cartilage of mice with posttraumatic OA suggests that DKK1 has a chondroprotective role in OA (34). Interestingly, local overexpression of the related Wnt antagonist DKK2 had no effect in the same model, suggesting that Wnt antagonists have potentially unique local roles in the pathogenesis of OA.
Despite the observed increase in subchondral bone in sclerostin-KO mice, all of the remaining joint features associated with age progression were similar to those observed in WT control mice. Quantitatively, there appeared to be no differences in the knee-joint lesion scores, suggesting that life-long genetic loss of sclerostin does not compromise the integrity of the cartilage within the knee joint.
In addition to genetic loss of sclerostin, short-term pharmacologic treatment of intact, 16-month-old male rats or osteopenic, OVX female rats with Scl-Ab did not alter the morphologic features or integrity of the articular cartilage. These results imply that if the sclerostin produced by chondrocytes has an inhibitory effect on local Wnt signaling, then genetic or acute loss of the inhibitory activity of sclerostin, and any subsequent potential enhancement of Wnt signaling, is not sufficient to alter the normal morphology and integrity of the articular cartilage. Furthermore, it was previously reported that degenerative osteoarthropathy was not present in patients with sclerosteosis (50).
Our findings in sclerostin-KO mice showing that sclerostin does not play a central role in the loss of joint cartilage during normal aging was supported by our in vivo pharmacologic findings, showing that short-term treatment with Scl-Ab or Scl-Fab, administered either systemically or IA, respectively, had no effect on lesion development in the rat MMT model of OA. Lesion development, characterized by substantial loss of medial cartilage, occurs rapidly in the MMT model, but this was not altered following treatment with Scl-Ab or Scl-Fab. We do not know whether the Scl-Fab has the potential to penetrate intact or damaged cartilage as occurs in the rat MMT model, and therefore this is a limitation of the present study. Further work will be required to investigate the potential of Scl-Fab to impact chondrocytes following IA injection, and whether this may lead to any alteration in Wnt signaling and lesion development in the MMT model. Moreover, the role of sclerostin in articular cartilage could be further investigated in sclerostin-KO mice using induced models of OA.
The efficacy of systemic Scl-Ab in terms of its effects on subchondral bone formation was clearly evident, but despite the increase in bone mass, MMT lesion scores and osteophyte formation were unchanged in the Scl-Ab–treated rats compared to vehicle-treated control rats. This suggests that the accelerated subchondral bone formation that occurred during the 3-week timeframe had no impact on the normal cartilaginous lesions that develop in this model. These observations indicate that enhanced subchondral bone formation resulting from short-term pharmacologic inhibition or genetic loss of sclerostin is not associated with cartilage destruction in the knee joint. These in vivo results therefore do not support the hypothesis that accelerated wear of articular cartilage in OA is initiated by local subchondral bone sclerosis.
In summary, although we found evidence of sclerostin expression in rodent and human articular cartilage, genetic loss of sclerostin in mice had no effect on cartilage integrity in aging knee joints. Furthermore, pharmacologic treatment of rats with Scl-Ab or Scl-Fab, even following posttraumatic knee injury, had no effect on cartilage lesion development. These results suggest that if sclerostin plays a role in the biology of the articular cartilage, then its impact may be less powerful and/or less unique compared to its clearly central regulatory role in the control of bone mass. One possibility is that in the genetic absence of sclerostin or as a result of pharmacologic inhibition, a compensatory molecule (e.g., another Wnt signaling inhibitor) is up-regulated in the cartilage and masks the effects of sclerostin inhibition. Should such a compensatory molecule exist, uncovering its identity could lead to important further insights into the role of Wnt signaling in cartilage biology.
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. Babij 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. Li, Graham, Ke, Black, Hulme, Babij.
Analysis and interpretation of data. Roudier, Li, Niu, Pretorius, Graham, Yoon, Gong, Warmington, Ke, Black, Hulme, Babij.
ROLE OF THE STUDY SPONSORS
This study was supported by Amgen and UCB Pharma. All authors are current or former employees of Amgen. The sponsors had a role in the study design or in the collection, analysis, or interpretation of the data, the writing of the manuscript, or the decision to submit the manuscript for publication. Publication of this article was contingent upon approval by Amgen or UCB Pharma.
We thank Julie Hahn, Brenda Heron, and Noi Nuanamee for technical assistance, Alison Bendele for performing the studies at Bolder BioPATH, and Chris Paszty for reviewing the manuscript.