Skip to main content
NCBI home page
Journal of Dental Research logoLink to Journal of Dental Research
. 2011 Nov;90(11):1331–1338. doi: 10.1177/0022034511421930

Role of Subchondral Bone during Early-stage Experimental TMJ Osteoarthritis

M Embree 1,2, M Ono 1, T Kilts 1, D Walker 1, J Langguth 1, J Mao 2, Y Bi 1, JL Barth 3, M Young 1,*
PMCID: PMC3188464  PMID: 21917603

Abstract

Temporomandibular joint osteoarthritis (TMJ OA) is a degenerative disease that affects both cartilage and subchondral bone. We used microarray to identify changes in gene expression levels in the TMJ during early stages of the disease, using an established TMJ OA genetic mouse model deficient in 2 extracellular matrix proteins, biglycan and fibromodulin (bgn-/0fmod-/-). Differential gene expression analysis was performed with RNA extracted from 3-week-old WT and bgn-/0fmod-/- TMJs with an intact cartilage/subchondral bone interface. In total, 22 genes were differentially expressed in bgn-/0fmod-/- TMJs, including 5 genes involved in osteoclast activity/differentiation. The number of TRAP-positive cells were three-fold higher in bgn-/0fmod-/- TMJs than in WT. Quantitative RT-PCR showed up-regulation of RANKL and OPG, with a 128% increase in RANKL/OPG ratio in bgn-/0fmod-/- TMJs. Histology and immunohistochemistry revealed tissue disorganization and reduced type I collagen in bgn-/0fmod-/- TMJ subchondral bone. Early changes in gene expression and tissue defects in young bgn-/0fmod-/- TMJ subchondral bone are likely attributed to increased osteoclast activity. Analysis of these data shows that biglycan and fibromodulin are critical for TMJ subchondral bone integrity and reveal a potential role for TMJ subchondral bone turnover during the initial early stages of TMJ OA disease in this model.

Keywords: temporomandibular disorders, osteoclasts, matrix biology, bone biology osteoarthritis, cartilage

Introduction

The temporomandibular joint consists of a fibrocartilaginous condyle, disc, synovial membrane, and subchondral bone. Temporomandibular joint osteoarthritis (TMJ OA) is a degenerative disease involving all TMJ tissues and leads to anatomical changes and severe pain (Scrivani et al., 2008). While the contribution of the cartilage to the disease pathology is well-studied, little is known about the role of subchondral bone. Emerging studies have implicated a potential role of subchondral bone during TMJ OA pathology, where sclerosis and increased bone metabolism are involved during cartilage degeneration (Jiao et al., 2011; Lories and Luyten, 2011). Furthermore, the development of new OA therapeutics that target both cartilage and bone highlights the necessity for a better understanding of the role of subchondral bone turnover during OA pathology (Karsdal et al., 2008).

The small leucine-rich repeat proteoglycans (SLRPs) are a family of extracellular matrix proteins highly expressed in both cartilage and bone (Embree et al., 2010). A mouse line doubly deficient in 2 members of the SLRP family, biglycan and fibromodulin (bgn-/0fmod-/-), is a well-established genetic model for TMJ OA (Wadhwa et al., 2005a,b; Embree et al., 2010). We have previously reported late-stage osteoarthritic changes in both bgn-/0fmod-/- TMJ cartilage and subchondral bone, including age-dependent cartilage degeneration, osteophyte formation, and sclerosis (Wadhwa et al., 2005a,b). One advantage of using an experimental, genetic TMJ OA model is the ability to identify early cellular mechanisms that may predispose the animal to late-onset TMJ OA disease phenotype. Specifically, we have shown that Bgn and Fmod modulate cartilage ECM breakdown at an early age, before overt histological cartilage damaged occurred (Embree et al., 2010). The goal of this study was to identify new genes in cartilage and bone involved during early-stage TMJ OA in this genetic model. Based on these data, we aimed to evaluate cellular and tissue defects in the bgn-/0fmod-/- TMJ cartilage and subchondral bone.

Materials & Methods

Animals

Male bgn-/0fmod-/- mice and their strain-matched WT counterparts (C57BL/6-129) were used with approval from the ACUC, National Institutes of Health (NIDCR-DIR-#07-414), and generated as previously reported (Embree et al., 2010).

RNA Isolation and Microarray Analysis

TMJ condyles with an intact cartilage/subchondral bone interface were dissected from 3-week-old WT and bgn-/0fmod-/- mice (Fig. 1A, black arrow). Tissue from 3 animals was pooled to create single samples, and 3 independent WT and bgn-/0fmod-/- pooled samples were prepared for analysis. RNA was isolated by Trizol (Invitrogen, Carlsbad, CA, USA), purified with RNeasy Mini Kit (Qiagen, Valencia, CA, USA), and treated with RNAse-free™ DNase (Ambion, Foster City, CA, USA). Synthesis of cDNA, in vitro transcription of biotin-labeled cRNA targets, and fragmentation of target cRNAs were performed following established Affymetrix® procedures (Allen et al., 2009) at the MUSC Proteomics Facility (Charleston, SC, USA). Fragmented cRNA samples were hybridized to Affymetrix® Mouse Genome 430 2.0 microarrays in accordance with manufacturer protocols. Data were normalized by the Bioconductor gcRMA algorithm (Sean and Meltzer, 2007) implemented with the ArrayQuest Web tool (Argraves et al., 2005) and imported into dChip software (Li and Wong, 2001) for comparative analyses. Genes differentially expressed between WT and bgn-/0 fmod-/- samples were defined by the criteria: (1) ‘present’ detection scores in ≥ 2 samples of either group; (2) fold change > 2; and (3) p < 0.05 (unpaired Student’s t test). False-discovery rate, estimated by permutation of sample groupings, approximated 19.4%.

Figure 1.

Figure 1.

Open in a new tab

Microarray analysis. (A) TMJ condylar cartilage (arrow) was dissected (black dashed lines) with intact subchondral bone interface from WT and bgn−/0fmod−/− mice. RNA was isolated for microarray and quantitative RT-PCR analysis. (B) Hierarchical clustering is shown for expression patterns of genes differentially expressed between WT and bgn-/0fmod-/- samples. Colorimetric scaling of standardized expression values (Z-standardization) is indicated at the bottom. (C) Quantitative RT-PCR for Cartpt, Ptprv, and Scl4a1 with RNA extracted from 3-week-old WT (open bars) and bgn-/0fmod-/- (black bars) TMJs with an intact cartilage/subchondral bone interface. Gene expression levels were normalized to housekeeping gene s29. Graph shown is representative of 3 biological replicates. *p < 0.01, bgn−/0fmod−/− vs. WT.

Histology and Immunohistochemistry

Paraffin-embedded sections from 3-week-old WT and bgn-/0fmod-/- TMJ condyles were treated as previously described (Embree et al., 2010). For tissue orientation, every 5th slide was stained with hematoxylin and eosin (H&E) or tartrate-resistant acid phosphatase (TRAP; Sigma 387A; Sigma, St. Louis, MO, USA). To quantify the percentage of TRAP-positive cells, we counted TRAP-positive cells and total cell number at the cartilage/subchondral bone interface from comparable tissue sections from 3 WT and 3 bgn-/0fmod-/- mice. Three serial sections within fixed areas were counted per mouse.

Immunohistochemistry was performed as previously described (Embree et al., 2010). Tissue sections were enzymatically treated with ABCase and incubated with primary antibodies and isotype-matched negative controls, including polyclonal rabbit anti-mouse Bgn (1:500 dilution, rabbit total serum, LF-106; Dr. Larry Fisher, NIH), polyclonal rabbit anti-mouse Fmod (1:500 dilution, rabbit total serum, LF-149; Dr. Larry Fisher, NIH), and polyclonal rabbit anti-human type I collagen (1:2000 dilution, rabbit antiserum, LF-68; Dr. Larry Fisher, NIH).

Quantitative RT-PCR

The cDNA was obtained by reverse-transcription of total RNA with TaqManTM reverse-transcription reagents (Applied Biosystems, Foster City, CA, USA). Primers used for the RT-PCR were designed with Beacon Designer Software (BioRad, Hercules, CA, USA) (Appendix Table 1). Real-time PCR was performed as previously described (Embree et al., 2010). Gene expression levels were normalized to the housekeeping gene s29.

Statistical Analysis

Data were collected from a minimum of 3 independent experiments. Statistical analyses were conducted with Student’s t test. Any p values < 0.05 were considered statistically significant.

Results

Differential Gene Expression in the bgn-/0 fmod-/- TMJ Cartilage/Subchondral Bone Interface

To identify potential new genes involved during early stages of TMJ OA in the bgn-/0fmod-/-genetic mouse model (Wadhwa et al., 2005a,b; Embree et al., 2010), we used a microarray analysis approach. Microarray analysis detected differential expression of 22 genes in bgn-/0fmod-/- vs. WT samples (Fig. 1B). Eight genes were down-regulated in bgn-/0 fmod-/- samples (Table) and included genes coding for extracellular matrix proteins involved in cartilage degeneration, such as procollagen, type IX, alpha 3 (Jackson et al., 2010), hyaluronan and proteoglycan link protein 1 (Urano et al., 2010), procollagen, type II, alpha 1 (Rintala et al., 1997), procollagen, type IX, alpha 1 (Hu et al., 2006), and matrilin 3 (van der Weyden et al., 2006). Mitochondrial ribosomal protein L30, a gene related to mitochondrial energy metabolism, was also down-regulated (Pagliarini et al., 2008). Interestingly, CART prepropeptide, a peptide of CART that functions to inhibit osteoclasts and bone resorption through the modulation of RANKL expression (Elefteriou et al., 2005), was also decreased in bgn-/0fmod-/- samples, suggesting that bone metabolism may be changed in the bgn-/0 fmod-/- mice. The 14 genes up-regulated in bgn-/0fmod-/- samples included 3 additional genes associated with osteoclast differentiation and/or function [secreted frizzled-related sequence protein 1 (Hausler et al., 2004), arylsulfatase K (Baron et al., 1990), solute carrier family 4, member 1 (Wu et al., 2008)], and 1 gene associated with bone metabolism [protein tyrosine phosphatase, receptor type V (Lee et al., 2007)] (Table, asterisks). Other genes up-regulated included those related to: cytokinesis of chondrocytes [profilin 1 (Bottcher et al., 2009)], beta secretase cleavage [beta-site APP cleaving enzyme 1 (Vassar et al., 1999)], stimulation of hematopoietic stem cells [angiopoietin-like 7 (Zhang et al., 2006)], inflammation and fat metabolism [ribosomal protein S19 (Filip et al., 2009)], AE binding protein 1 (Majdalawieh et al., 2006), erythropoiesis and blood cancer [erythroid-associated factor (Hardee et al., 2006)], and cell adhesion/motility [tetraspanin 33 (Lazo, 2007)]. Our microarray analysis demonstrated a relatively high false-discovery rate (FDR = 19.4%), indicating the probability that some false-positives were detected.

Table.

Differential Gene Expression by bgn-/0 fmod-/- vs. WT in TMJ Condyle with Intact Cartilage/Subchondral Bone Interface

Gene Gene Symbol Wild-type (mean) bgn-/0 fmod-/- (mean) Fold Change p-value Relative Function
Fibromodulin Fmod 5793 50 −117.0 0.0166 Negative Control
Fibromodulin Fmod 2353 68 −34.5 0.0037 Negative Control
Fibromodulin Fmod 2210 91 −24.4 0.0041 Negative Control
Fibromodulin Fmod 2241 95 −23.8 0.0052 Negative Control
Biglycan Bgn 3421 150 −22.6 0.0000 Negative Control
Biglycan Bgn 4270 187 −22.6 0.0000 Negative Control
Fibromodulin Fmod 2195 103 −21.4 0.0038 Negative Control
Biglycan Bgn 6165 385 −16.0 0.0002 Negative Control
Fibromodulin Fmod 1361 124 −10.9 0.0118 Negative Control
Procollagen, type IX, alpha 3 Col9a3 1105 300 −3.7 0.0430 Mutations result in multiple epiphyseal dysplasia and OA (Jackson et al., 2010)
Hyaluronan and proteoglycan link protein 1 Hapln1 1783 501 −3.5 0.0381 Polymorphism is associated with spinal disc degeneration (Urano et al., 2010)
Procollagen, type II, alpha 1 Col2a1 4390 1296 −3.4 0.0389 Col2a1-/- mice develop TMJ OA (Rintala et al., 1997)
Procollagen, type IX, alpha 1 Col9a1 2778 861 −3.2 0.0167 Col9a1-/- mice develop knee and TMJ OA (Hu et al., 2006)
Matrilin 3 Matn3 1783 592 −3.0 0.0419 Matn3-/- mice develop knee OA (van der Weyden et al., 2006)
RIKEN cDNA 3110079O15 gene LOC 344564 534 189 −2.8 0.0431
Mitochondrial ribosomal protein L30 Mrpl30 70 28 −2.5 0.0476 Energy metabolism (Pagliarini et al., 2008)
*CART prepropeptide Cartpt 498 229 −2.2 0.0172 Inhibits bone resorption by modulating Rankl expression (Elefteriou et al., 2005)
*Secreted frizzled-related sequence protein 1 Sfrp1 372 765 2.0 0.0402 Expressed in osteoblasts and inhibits osteoclast formation (Hausler et al., 2004)
Profilin 1 Pfn1 326 671 2.1 0.0073 Required for abscission during late cytokinesis of chondrocytes (Bottcher et al., 2009)
Beta-site APP cleaving enzyme 1 Bace1 370 798 2.1 0.0006 Proteolytic cleavage of the amyloid precursor protein (Vassar et al., 1999)
Ribosomal protein S19 Rps19 161 347 2.1 0.0102 Anti-inflammatory (Filip et al., 2009)
Angiopoietin-like 7 Angptl7 165 367 2.2 0.0225 Support expansion hematopoietic stem cells (Zhang et al., 2006)
*Arylsulfatase K Arsk 185 434 2.4 0.0049 Required for rapid inhibition of bone resorption by calcitonin (Baron et al., 1990)
RIKEN cDNA 1110020P15 gene 77 189 2.5 0.0288
Transcribed locus 108 265 2.5 0.0038
*Protein tyrosine phosphatase, receptor type V Ptprv 263 798 3.0 0.0059 Bone metabolism (Lee et al., 2007)
Tetraspanin 33 Tspan33 241 809 3.4 0.0390 Cell adhesion and motility (Lazo, 2007)
AE binding protein 1 Aebp1 109 419 3.8 0.0066 Macrophage cholesterol homeostasis and inflammation (Majdalawieh et al., 2006)
*Solute carrier family 4 (anion exchanger), member 1 Slc4a1 168 719 4.3 0.0423 Mediates osteoclast differentiation/function (Wu et al., 2008)
Erythroid-associated factor Ahsp 168 867 5.2 0.0455 Erythropoiesis and cancer (Hardee et al., 2006)
RIKEN cDNA 4833416E15 gene 4833416E15Rik 132 923 7.1 0.0005
Open in a new tab

Differential gene expression ± 2.0-fold and p < 0.05 (unpaired t test) for bgn-/0 fmod-/- vs. WT. False-discovery rate for this analysis approximates 19.4%. Mean indicates average normalized fluorescence measurement for 3 replicates. Asterisks denote genes related to osteoclast activity/function and bone metabolism.

Therefore, to confirm the findings from our microarray analysis, we tested 3 responding genes by quantitative RT-PCR (Fig. 1C). RT-PCR confirmed that CART prepropeptide (Cartpt) was down-regulated, whereas protein tyrosine phosphatase, receptor type V (Ptprv), and solute carrier family 4 (anion exchanger), member 1 (Slc4a1), were up-regulated in bgn-/0fmod-/- samples.

Increased Osteoclast Activity in the bgn-/0fmod-/- TMJ Cartilage/Subchondral Bone Interface

Given that the microarray analysis revealed changes in 4 genes related to osteoclast activity/function and 1 gene associated with bone metabolism in bgn-/0 fmod-/- samples (Table, asterisks), we speculated that osteoclast activity was altered in bgn-/0fmod-/- TMJ subchondral bone. We examined osteoclast activity in WT and bgn-/0fmod-/- by TRAP staining and quantitative RT-PCR for RANKL and OPG, 2 genes important for osteoclast activity. In comparison with WT TMJ (Fig. 2C, arrows), there was an increase in TRAP-positive regions along the cartilage/subchondral bone interface (Fig. 2, orange line) in bgn-/0fmod-/- TMJ (Fig. 2D, arrows). Quantification of TRAP-positive cells confirmed that there was a three-fold increase in bgn-/0fmod-/- TRAP-positive cells (Fig. 2E). Quantitative RT-PCR, with RNA isolated from WT and bgn-/0 fmod-/- TMJs, demonstrated a significant increase in RANKL (Fig. 2F) and OPG (Fig. 2G) gene expression levels in bgn-/0fmod-/-, with relatively greater up-regulation of RANKL (~4.5-fold increase) compared with OPG (~3.5-fold increase). Analysis of these data resulted in a 128% increase in RANKL/OPG ratio in bgn-/0fmod-/- and suggested that osteoclast activity was increased in bgn-/0fmod-/- TMJ subchondral bone. However, differential expression of RANKL and OPG was not detected in microarray studies (Appendix Table 2). This was likely due to the statistical limitations of the microarray analysis and the increased specificity of quantitative RT-PCR.

Figure 2.

Figure 2.

Open in a new tab

Increased osteoclast activity in the bgn-/0fmod-/-TMJ subchondral bone interface. (A,B) Low magnification and (C,D) high magnification of TRAP staining in the WT and bgn-/0fmod-/- TMJ cartilage/subchondral bone interface. Orange line indicates the TMJ cartilage/subchondral bone interface and divides TMJ into condylar cartilage (CC) and subchondral bone (SB). Green dashed lines show areas of high magnification. Arrows indicate TRAP-positive cells localized along the TMJ cartilage/subchondral bone interface. Bar = 50 µm. (E) The number of TRAP-positive cells at the TMJ cartilage/subchondral bone interface/fixed area tissue. Graph shown is representative of a total of 3 mice, with comparable tissue sections with a total of 3 serial sections per mouse. *P = 0.001, bgn−/0fmod−/− vs. WT. (F,G) Quantitative RT-PCR for RANKL and OPG with RNA from WT (open bars) and bgn-/0fmod-/- TMJs (black bars) in an intact TMJ cartilage/subchondral bone interface. Gene expression levels were normalized to housekeeping gene S29. Graph shown is representative of at least 3 independent experiments. *p < 0.01, bgn−/0fmod−/− vs. WT.

Identification of TMJ Subchondral Bone Defects in the bgn-/0fmod-/- TMJ

We proposed that increased osteoclast activity would cause phenotypic defects in bgn-/0fmod-/- TMJ subchondral bone. We examined the subchondral region from 3-week-old bgn-/0 fmod-/- TMJ compared with WT controls by histology (Figs. 3A-D) and immunohistochemistry (Figs. 3E-F). H&E staining of WT TMJ (Fig. 3A) showed distinct bone marrow cavities formed, which were surrounded by trabecular bone tissue (Fig. 3C; orange dashed lines indicate higher magnification). However, H&E stains of bgn-/0fmod TMJ (Fig. 3B) revealed significant tissue disorganization and sparse bone tissue present in comparable regions (Fig. 3D; orange dashed lines indicate higher magnification). To confirm the presence of subchondral bone defects in the bgn-/0fmod-/-, we examined type I collagen expression immunohistochemistry. Type I collagen was localized within WT TMJ subchondral bone (Fig. 3E, red), and a clear decrease was apparent in the bgn-/0fmod-/- TMJ (Fig. 3F, red), confirming that less type I collagen-rich bone was present. Interestingly, while type I collagen protein expression was decreased in bgn-/0fmod-/- TMJs, type I collagen mRNA was not detected, as differentially expressed by microarray analysis (Appendix Table 2) or by quantitative RT-PCR (Appendix Fig.), suggesting that type I collagen production is unchanged. Taken together, our histological findings suggested that the absence of Bgn and Fmod led to less bone in the TMJ subchondral bone region in mice. Immunohistochemistry showed that both Bgn (Fig. 3G, red, arrows) and Fmod (Fig. 3H, red, arrows) were both highly expressed and co-localized in WT TMJ at 3 wks in the bone regions shown to be affected by their absence.

Figure 3.

Figure 3.

Open in a new tab

Defects in TMJ subchondral bone in 3-week-old bgn-/0fmod-/- mice. (A,B) 20x magnification of H&E staining in 3-week-old WT and bgn-/0fmod-/- TMJs showing a condylar cartilage (CC) region and subchondral bone (SB) regions. Orange dashed box indicates a subchondral bone (SB) region that is shown in images below in higher magnification. Bar = 100 µm. (C,D) 40x magnification of H&E staining in 3-week-old WT and bgn-/0fmod-/- TMJ subchondral bone (SB) regions, indicated in orange dashed box. Bar = 50 µm. (E,F) Type I collagen immunostaining in 3-week-old WT and bgn-/0fmod-/- TMJ subchondral bone. Bar = 50 µm. (G,H) Biglycan (Bgn) and fibromodulin (Fmod) immunostainings in TMJ subchondral bone in 3-week-old WT mice. Bar = 50 µm.

Discussion

The bgn-/0fmod-/- mouse is an established TMJ OA genetic model demonstrating a chronological sequence of cellular/histological evidence that signifies TMJ OA pathology (Wadhwa et al., 2005a,b; Embree et al., 2010). We took advantage of this genetic model to identify early changes in gene expression levels that may predispose the animal to the late-onset TMJ OA phenotype. Using microarray analysis, we discovered 22 genes differentially expressed in young bgn-/0fmod-/- TMJs that could potentially be involved in disease initiation. Genes that correspond to other ECM proteins involved in cartilage degeneration in other genetic mouse models were also down-regulated in the bgn-/0 fmod-/- samples, including procollagen, type IX, alpha 3 (Jackson et al., 2010), procollagen, type II, alpha 1 (Rintala et al., 1997), procollagen, type IX, alpha 1 (Hu et al., 2006), and matrilin 3 (van der Weyden et al., 2006). Analysis of these data based on genetic mouse models highlights the possibility that, in humans, the loss/mutation of a single ECM-related gene may initiate pathology or increase susceptibility to TMJ disease. However, the factors that account for the majority of TMJ OA cases in humans, which occur with a high predilection in women during childbearing ages, remain unresolved (Scrivani et al., 2008). TMJ OA genetic mouse models with ECM-related mutations show pathology in aged males (Rintala et al., 1997; Wadhwa et al., 2005a,b; Hu et al., 2006), thus raising doubt as to whether existing genetic mouse models exhibit TMJ disease analogous to the human condition. It is possible that changes in sex-related hormones may act upstream to contribute to the abnormal expression patterns of ECM-related proteins (Hellio Le Graverand et al., 2000; Talwar et al., 2006; Salgado et al., 2009). Therefore, for a more comparable model of TMJ OA to be developed, it would be important to determine the effects of estrogen levels on TMJ disease in females, using bgn-/0fmod-/- mice or other genetic mouse models of TMJ OA.

The TMJ has unique properties compared with other articular joints like the knee, because it is made of fibrocartilage and acts both as an articular joint cartilage and as a site for endochondral ossification (Shen and Darendeliler, 2005). Consequently, the cellular events within the TMJ cartilage/subchondral bone interface are more likely to play a greater role in tissue homeostasis and disease in the TMJ than in the knee joint. TMJ OA is a degenerative disease of the entire joint, including the subchondral bone (Lories and Luyten, 2011). Phenotypic changes in the subchondral bone in osteoarthritic tissue from both animals and humans have been well-documented, including altered trabecular bone structure, sclerosis, and increased bone resorption/formation (Lories and Luyten, 2011). Our microarray analysis revealed that at least 5 genes are changed in bgn-/0fmod-/- samples that are related to osteoclast function/differentiation and bone turnover, including CART prepropeptide (Elefteriou et al., 2005), secreted frizzled-related sequence protein 1 (Hausler et al., 2004), arylsulfatase K (Baron et al., 1990), solute carrier family 4, member 1 (Wu et al., 2008), and protein tyrosine phosphatase, receptor type V (Lee et al., 2007). More specifically, CART prepropeptide inhibits bone resorption by modulating RANKL expression (Elefteriou et al., 2005). Similar to bgn-/0fmod-/- mice, mice deficient in CART also have low trabecular bone mass (Elefteriou et al., 2005). Furthermore, secreted frizzled-related sequence protein 1 is expressed by osteoblasts and inhibits osteoclast formation, suggesting that bone turnover may be higher in the bgn-/0fmod-/- TMJ subchondral bone (Hausler et al., 2004). In addition, solute carrier family 4, member 1 (Slc4a1), a bicarbonate/chloride exchanger closely related to Slc4a2, is also up-regulated in bgn-/0fmod-/- samples. This latter protein is a critical mediator of both osteoclast differentiation and function, where Slc4a2-/- osteoclasts are unable to resorb mineralized tissue and fail to express TRAP activity (Wu et al., 2008). Taken together, these microarray findings suggested that osteoclast activity and/or bone turnover may be increased in the bgn-/0fmod-/-TMJ subchondral bone.

To explore osteoclast activity in the bgn-/0fmod-/- mice, we examined TRAP stainings in the TMJ subchondral bone and also OPG and RANKL by quantitative RT-PCR. In agreement with our microarray findings, we discovered increased TRAP staining specifically localized along the bgn-/0fmod-/- TMJ cartilage/subchondral bone interface, suggesting active bone turnover in this region. Similarly, we discovered a 128% increase in OPG/RANKL gene expression ratio in bgn-/0fmod-/- samples, which could cause increased TMJ subchondral bone turnover in bgn-/0 fmod-/-mice. In support of our findings, previous studies have demonstrated that non-TMJ OA joints also display increased subchondral bone turnover (Goker et al., 2000). Moreover, increased osteoclast activity in bgn-/0fmod-/-mice was also coupled with increased chondrogenesis in 3-week-old mice (Embree et al., 2010). This concept raises the possibility that crosstalk between the cartilage and subchondral bone interface during younger ages is critical for maintaining tissue homeostasis. The disruption of bone and cartilage metabolism in younger bgn-/0fmod-/-mice could disrupt the overall TMJ tissue homeostasis and predispose the mice to the late-onset TMJ OA phenotype previously reported in the older bgn-/0fmod-/-mice, including osteophyte formation, TMJ subchondral bone sclerosis, and cartilage degeneration (Wadhwa et al., 2005a,b).

We further show that a possible consequence of high bone turnover is defective trabecular bone structure formed in bgn-/0 fmod-/- TMJ subchondral region, as evidenced by histology and type I collagen immunostaining. Given that the functional integrity of the cartilage can depend on the mechanical properties of the underlying bone (Bailey et al., 2004), it is possible that the disorganized and defective TMJ subchondral bone in bgn-/0 fmod-/- mice contributes significantly to TMJ OA pathology. Interestingly, while type I collagen protein expression was decreased in bgn-/0fmod-/- TMJs, type I collagen mRNA was unchanged. We speculate that the production of type I collagen mRNA is not changed, and that other things such as translational control of type I collagen mRNA or ECM resorption cause less type I collagen protein expression in bgn-/0fmod-/- TMJs. Given that TRAP-positive cells and RANKL/OPG gene expression levels are significantly higher in bgn-/0fmod-/- TMJs, it is likely that there is increased resorption and, therefore, less type I collagen protein expression. Taken together, we show for the first time that Bgn and Fmod are critical for the integrity of TMJ subchondral bone and reveal a potential role for subchondral bone turnover during early stages of TMJ OA in this mouse model.

Acknowledgments

We thank Victor Fresco at MUSC for technical assistance with microarray experimentation and Ms. Qiongfen Guo at Columbia University for laboratory assistance.

Footnotes

This investigation was supported by NIDCR grant 5F30 DE018072-04, the Division of Intramural Research, NIDCR, the Intramural Research Program, NIH, and NIH/NIDCR grant RC2DE020767. Microarray and bioinformatic resources of the MUSC Proteogenomics Facility were supported by NCRR grants RR16434 and RR16461.

The authors declare no potential conflicts of interest with respect to the authorship and/or publication of this article.

A supplemental appendix to this article is published electronically only at http://jdr.sagepub.com/supplemental.

References

  1. Allen DL, Bandstra ER, Harrison BC, Thorng S, Stodieck LS, Kostenuik PJ, et al. (2009). Effects of spaceflight on murine skeletal muscle gene expression. J Appl Physiol 106:582-595; erratum in J Appl Physiol 110:298, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Argraves GL, Jani S, Barth JL, Argraves WS. (2005). ArrayQuest: a Web resource for the analysis of DNA microarray data. BMC Bioinformatics 6:287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bailey AJ, Mansell JP, Sims TJ, Banse X. (2004). Biochemical and mechanical properties of subchondral bone in osteoarthritis. Biorheology 41:349-358 [PubMed] [Google Scholar]
  4. Baron R, Neff L, Brown W, Louvard D, Courtoy PJ. (1990). Selective internalization of the apical plasma membrane and rapid redistribution of lysosomal enzymes and mannose 6-phosphate receptors during osteoclast inactivation by calcitonin. J Cell Sci 97(Pt 3):439-447 [DOI] [PubMed] [Google Scholar]
  5. Bottcher RT, Wiesner S, Braun A, Wimmer R, Berna A, Elad N, et al. (2009). Profilin 1 is required for abscission during late cytokinesis of chondrocytes. EMBO J 28:1157-1169 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Elefteriou F, Ahn JD, Takeda S, Starbuck M, Yang X, Liu X, et al. (2005). Leptin regulation of bone resorption by the sympathetic nervous system and CART. Nature 434:514-520 [DOI] [PubMed] [Google Scholar]
  7. Embree MC, Kilts TM, Ono M, Inkson CA, Syed-Picard F, Karsdal MA, et al. (2010). Biglycan and fibromodulin have essential roles in regulating chondrogenesis and extracellular matrix turnover in temporomandibular joint osteoarthritis. Am J Pathol 176:812-826 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Filip AM, Klug J, Cayli S, Frohlich S, Henke T, Lacher P, et al. (2009). Ribosomal protein S19 interacts with macrophage migration inhibitory factor and attenuates its pro-inflammatory function. J Biol Chem 284:7977-7985 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Goker B, Doughan AM, Schnitzer TJ, Block JA. (2000). Quantification of progressive joint space narrowing in osteoarthritis of the hip: longitudinal analysis of the contralateral hip after total hip arthroplasty. Arthritis Rheum 43:988-994 [DOI] [PubMed] [Google Scholar]
  10. Hardee ME, Arcasoy MO, Blackwell KL, Kirkpatrick JP, Dewhirst MW. (2006). Erythropoietin biology in cancer. Clin Cancer Res 12:332-339 [DOI] [PubMed] [Google Scholar]
  11. Hausler KD, Horwood NJ, Chuman Y, Fisher JL, Ellis J, Martin TJ, et al. (2004). Secreted frizzled-related protein-1 inhibits RANKL-dependent osteoclast formation. J Bone Miner Res 19:1873-1881 [DOI] [PubMed] [Google Scholar]
  12. Hellio Le Graverand MP, Reno C, Hart DA. (2000). Heterogenous response of knee cartilage to pregnancy in the rabbit: assessment of specific mRNA levels. Osteoarthritis Cartilage 8:53-62 [DOI] [PubMed] [Google Scholar]
  13. Hu K, Xu L, Cao L, Flahiff CM, Brussiau J, Ho K, et al. (2006). Pathogenesis of osteoarthritis-like changes in the joints of mice deficient in type IX collagen. Arthritis Rheum 54:2891-2900 [DOI] [PubMed] [Google Scholar]
  14. Jackson GC, Marcus-Soekarman D, Stolte-Dijkstra I, Verrips A, Taylor JA, Briggs MD. (2010). Type IX collagen gene mutations can result in multiple epiphyseal dysplasia that is associated with osteochondritis dissecans and a mild myopathy. Am J Med Genet A 152:863-869 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Jiao K, Niu LN, Wang MQ, Dai J, Yu SB, Liu XD, et al. (2011). Subchondral bone loss following orthodontically induced cartilage degradation in the mandibular condyles of rats. Bone 48:362-371 [DOI] [PubMed] [Google Scholar]
  16. Karsdal MA, Leeming DJ, Dam EB, Henriksen K, Alexandersen P, Pastoureau P, et al. (2008). Should subchondral bone turnover be targeted when treating osteoarthritis? Osteoarthritis Cartilage 16:638-646 [DOI] [PubMed] [Google Scholar]
  17. Lazo PA. (2007). Functional implications of tetraspanin proteins in cancer biology. Cancer Sci 98:1666-1677 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Lee NK, Sowa H, Hinoi E, Ferron M, Ahn JD, Confavreux C, et al. (2007). Endocrine regulation of energy metabolism by the skeleton. Cell 130:456-469 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Li C, Wong WH. (2001). Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. Proc Natl Acad Sci USA 98:31-36 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Lories RJ, Luyten FP. (2011). The bone-cartilage unit in osteoarthritis. Nat Rev Rheumatol 7:43-49 [DOI] [PubMed] [Google Scholar]
  21. Majdalawieh A, Zhang L, Fuki IV, Rader DJ, Ro HS. (2006). Adipocyte enhancer-binding protein 1 is a potential novel atherogenic factor involved in macrophage cholesterol homeostasis and inflammation. Proc Natl Acad Sci USA 103:2346-2351 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Pagliarini DJ, Calvo SE, Chang B, Sheth SA, Vafai SB, Ong SE, et al. (2008). A mitochondrial protein compendium elucidates complex I disease biology. Cell 134:112-123 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Rintala M, Metsaranta M, Saamanen AM, Vuorio E, Ronning O. (1997). Abnormal craniofacial growth and early mandibular osteoarthritis in mice harbouring a mutant type II collagen transgene. J Anat 190(Pt 2):201-208 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Salgado RM, Favaro RR, Martin SS, Zorn TM. (2009). The estrous cycle modulates small leucine-rich proteoglycans expression in mouse uterine tissues. Anat Rec (Hoboken) 292:138-153 [DOI] [PubMed] [Google Scholar]
  25. Sean D, Meltzer PS. (2007). GEOquery: a bridge between the Gene Expression Omnibus (GEO) and BioConductor. Bioinformatics 23:1846-1847 [DOI] [PubMed] [Google Scholar]
  26. Shen G, Darendeliler MA. (2005). The adaptive remodeling of condylar cartilage–a transition from chondrogenesis to osteogenesis. J Dent Res 8:691-9 [DOI] [PubMed] [Google Scholar]
  27. Talwar RM, Wong BS, Svoboda K, Harper RP. (2006). Effects of estrogen on chondrocyte proliferation and collagen synthesis in skeletally mature articular cartilage. J Oral Maxillofac Surg 64:600-609 [DOI] [PubMed] [Google Scholar]
  28. Urano T, Narusawa K, Shiraki M, Sasaki N, Hosoi T, Ouchi Y, et al. (2010). Single-nucleotide polymorphism in the hyaluronan and proteoglycan link protein 1 (HAPLN1) gene is associated with spinal osteophyte formation and disc degeneration in Japanese women. Eur Spine J 20:572-577 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. van der Weyden L, Wei L, Luo J, Yang X, Birk DE, Adams DJ, et al. (2006). Functional knockout of the matrilin-3 gene causes premature chondrocyte maturation to hypertrophy and increases bone mineral density and osteoarthritis. Am J Pathol 169:515-527 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Vassar R, Bennett BD, Babu-Khan S, Kahn S, Mendiaz EA, Denis P, et al. (1999). Beta-secretase cleavage of Alzheimer’s amyloid precursor protein by the transmembrane aspartic protease BACE. Science 286:735-741 [DOI] [PubMed] [Google Scholar]
  31. Wadhwa S, Embree M, Ameye L, Young MF. (2005a). Mice deficient in biglycan and fibromodulin as a model for temporomandibular joint osteoarthritis. Cells Tissues Organs 181:136-143 [DOI] [PubMed] [Google Scholar]
  32. Wadhwa S, Embree MC, Kilts T, Young MF, Ameye LG. (2005b). Accelerated osteoarthritis in the temporomandibular joint of biglycan/fibromodulin double-deficient mice. Osteoarthritis Cartilage 13:817-827 [DOI] [PubMed] [Google Scholar]
  33. Wu J, Glimcher LH, Aliprantis AO. (2008). HCO3-/Cl- anion exchanger SLC4A2 is required for proper osteoclast differentiation and function. Proc Natl Acad Sci USA 105:16934-16939 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Zhang CC, Kaba M, Ge G, Xie K, Tong W, Hug C, et al. (2006). Angiopoietin-like proteins stimulate ex vivo expansion of hematopoietic stem cells. Nat Med 12:240-245 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Journal of Dental Research are provided here courtesy of International and American Associations for Dental Research

RESOURCES