SEARCH

SEARCH BY CITATION

Keywords:

  • primary cilium;
  • articular cartilage;
  • patella;
  • cilia length;
  • osteoarthritis;
  • bovine

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Osteoarthritis (OA) is a common joint disease characterized by articular cartilage degeneration. The etiology of OA is unknown. Because several previous studies have shown that primary cilia play critical roles in joint development, this study examined the incidence and morphology of primary cilia in chondrocytes during joint degeneration in a naturally occurring bovine model of OA. Primary cilia were detected using antibodies to acetylated α-tubulin in normal cartilage as well as in mild and severe OA tissue. In normal cartilage, cilia number and length were lowest in the superficial zone and increased with distance from the articular surface. In OA tissue, the incidence and length of cilia increased at the eroding articulating surface, resulting in an overall increased proportion of cilia. This is the first study to show that primary cilia are present on chondrocytes throughout OA progression and that the overall percentage of ciliated cells within the degenerating cartilage increases with OA severity. Developmental Dynamics 237:2013–2020, 2008. © 2008 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Osteoarthritis (OA) is a chronic degenerative disease that affects the articular cartilage of synovial joints, mainly in the elderly population. Pathological changes in the extracellular matrix (ECM) during cartilage degeneration that are associated with early OA include increased water content, loosening of the collagen network, and tissue swelling, as well as disruption of the articulating surface layer of cartilage. As the degeneration progresses, the cartilage matrix becomes fibrillated and the cartilage slowly erodes into the joint cavity. In the later stages of OA, there is extensive loss of cartilage matrix resulting in exposure and ebonation of the underlying subchondral bone (Aigner et al.,2007; Goldring and Goldring,2007).

Concurrent with these matrix changes, the chondrocytes plus their pericellular microenvironment (termed chondrons) undergo a three-phase remodeling process in an attempt to recapitulate cartilage growth, and the repair of the osteoarthritic lesion (Poole et al.,1996,1997a; Ross et al.,2006). During the first phase, the pericellular microenvironment of the chondron begins to swell as the fine type II collagen fibrils are catabolized and the concentrated pericellular proteoglycans begin to expand. In the second phase, the normally quiescent chondrocytes divide. Cells migrate apart and secrete new pericellular macromolecules. The final phase is characterized by unrestricted cell division and chondrocyte cluster formation within a remodeled pericellular matrix. These clusters are ultimately shed into the joint cavity along with fibrillated matrix fragments.

In normal articular cartilage, all chondrocytes monitor the pericellular, territorial, and interterritorial matrices and alter ECM turnover in response to mechanical load (Palmoski and Brandt,1984; Hunziker,1992; Kim et al.,1994). Although many studies have clearly shown that the mechanical environment is critical to normal chondrocyte function (Buschmann et al.,1995; Dumont et al.,1999), the mechanisms underlying these mechanosensory and biosynthetic responses are only partially understood. The mechanical environment within an osteoarthritic joint is also severely affected and osteoarthritic lesions are often localized to weight-bearing cartilage or to sites of trauma (Englund and Lohmander,2004; Roos,2005). Although the etiology of OA is unknown, repetitive mechanical joint loading, in conjunction with age-related altered responsiveness of chondrocytes to catabolic and anabolic stimuli are likely to play a significant role. The chondron is clearly altered both morphologically and metabolically during OA, but it is still unclear whether chondrocytes are the primary drivers of disease progression, or whether they react secondarily to breakdown of the ECM associated with continuous mechanical loading.

Our previous studies have strongly suggested that the chondrocyte primary cilium operates within a mechanotransductory sensory feedback loop, translating extracellular information to the cell to facilitate secretion of ECM (Poole et al.,1985). The primary cilium is a highly conserved, single cytoplasmic organelle found in virtually all eukaryotic cells. The chondrocyte primary cilium projects into the pericellular matrix of the chondron and interacts with matrix macromolecules such as collagens type II and VI by means of receptors that include integrins and NG2 (Wilsman and Fletcher,1978; Wilsman et al.,1980; Poole et al.,1997b,2001; Jensen et al.,2004; McGlashan et al.,2006a). Defects in perichondrial, chondroblast, and chondrocyte primary cilia have recently been implicated in both skeletal patterning and growth plate abnormalities related to alterations in ECM secretion and sequestration (Zhang et al.,2003; Gouttenoire et al.,2007; Haycraft et al.,2007; Koyama et al.,2007; McGlashan et al.,2007; Ruiz-Perez et al.,2007; Song et al.,2007). Because the mechanical properties of the pericellular matrix, as well as chondrocyte function, are altered during the development of OA, the current study aimed to investigate the fate of the chondrocyte primary cilium during the progression of OA.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Grading and Histology of OA Cartilage

The tissue was graded on the macroscopic appearance of the articular surface using an established system (Broom et al.,2001). Normal tissue (Grade 0) showed a clearly defined smooth, shiny surface with no evidence of erosion. Mild OA tissue (Grade 1) showed surface roughness and splitting with a thickened articular surface, whereas severe OA tissue (Grade 2) exhibited clear fibrillation extending into the deeper zones. No Grade 3 samples of complete cartilage loss and bone ebonation were used.

Histologically, normal grade 0 cartilage showed the characteristic four distinct zones. The superficial zone chondrocytes were ellipsoid and oriented parallel to the articular surface (Fig. 1A); middle zone chondrocytes often formed a pair of chondrons (Fig. 1A); deep zone chondrocytes were found in chondron columns oriented perpendicular to the articular surface and tidemark (Fig. 1D). A tidemark was present and contained chondrocytes surrounded by a calcified matrix (Fig 1D; asterisk). In mild grade 1 OA tissue, there was a loss of the ellipsoidal surface cells and the surface was roughened with small fibrillations. Chondrocytes were present in clusters of two to six cells along the eroding articulating surface (Fig. 1B; arrows) and cluster formation occasionally extended into the middle zone. Deep zone chondrocytes were present in chondrons oriented perpendicular to the tidemark, similar to normal tissue (Fig. 1E). In severe Grade 2 OA tissue, the entire superficial and mid zones were lost. The tissue remaining at the articular surface showed large fibrillations extending into the deep zone. At the eroding surface, chondrocytes were present as clusters comprising four to eight cells, although occasional clusters contained more than ten cells (Fig. 1C; arrows). Within the deep zone of severe OA tissue, there was a reduced density of chondron columns (Fig. 1F; arrowheads), as well as evidence of chondron swelling (Fig 1F; arrowheads) as compared to normal and mild OA deep zone cartilage.

thumbnail image

Figure 1. A–F: Histology of the articulating surface (A–C) and the deep zone (D–F) of normal articular cartilage and mild and severe osteoarthritic articular cartilage. Sections were stained with Safranin O/Fast Green. Arrowheads indicate normal chondrons (A,D–F); arrows indicate cell clusters (B,C); the asterisk indicates the faintly stained tidemarks (D–F). Scale bar = 50 μm.

Download figure to PowerPoint

Proportion of Ciliated Chondrocytes During OA Progression

Primary cilia labeled with an antibody to α-acetylated tubulin antibody were identified in all three grades of cartilage tissue (Figs. 2, 3). Cilia were also detected on the chondrocytes in the calcified cartilage zone, and on osteoblasts and osteocytes within the subchondral bone (data not shown). The percentage of ciliated chondrocytes assessed from full-thickness sections of normal cartilage (articulating surface to tidemark) showed an overall mean percentage of 46% (Table 1). The proportion of ciliated cells was significantly greater in the deep zone than in both the superficial (P < 0.001) and the middle zone (P < 0.001; Table 1). No differences in cilia incidence were found between superficial and middle zones in normal Grade 0 cartilage (P > 0.05).

thumbnail image

Figure 2. A,C,E: Differential interference contrast images of chondrocytes in normal articular cartilage in the superficial zone at the articular surface (A, arrowhead), the middle zone (C), and the deep zone (E). B,D,F: Corresponding images of primary cilia (arrows) labeled with acetylated α-tubulin antibody. Scale bars = 10 μm.

Download figure to PowerPoint

thumbnail image

Figure 3. Chondrocyte clusters at the eroding articulating surface in six different (not corresponding) examples of severe osteoarthritic articular cartilage. A–C: Differential interference contrast images of clusters containing two cells (A), three cells (B), and multiple cells (C). D–F: Confocal z-projections of primary cilia (arrows) labeled with actetylated α-tubulin antibody. Cilia (arrows) are oriented toward the center of the cluster within the expanding chondron. Scale bars = 5 μm.

Download figure to PowerPoint

Table 1. Percentage of Ciliated Cells and Cilia Length (μm) in Normal and Osteoarthritic Articular Cartilagea
  Tissue grade
  NormalMildSevere
  • a

    Values are expressed as the mean ± SE (n > 40 for each group).

% ciliated cellsFull depth46.3 ± 3.152.2 ± 1.868.5 ± 2.0
1st 100 μm of articulating surface37.8 ± 5.651.1 ± 3.367.8 ± 5.0
Middle zone39.0 ± 5.056.4 ± 4.3NA
Deep zone62.1 ± 2.849.1 ± 4.269.2 ± 3.8
Cilia length (μm)1st 100 μm of articulating surface1.1 ± 0.11.3 ± 0.11.4 ± 0.1
Middle zone1.3 ± 0.11.2 ± 0.1NA
Deep zone1.5 ± 0.11.4 ± 0.11.4 ± 0.1

In both mild and severe OA tissue, the proportion of ciliated cells from the articulating surface to the tidemark increased markedly (Table 1). The percentage of cilia detected on OA chondrocytes at the articulating surface increased with grade severity, with an almost twofold rise from normal to severe OA cartilage (Table 1; P < 0.001). There were also significant differences between the incidence of cilia in deep zone cells, both between normal and mild OA (P < 0.05) and mild and severe OA (P < 0.01). However, no differences were found in cilia incidence between normal and severe OA deep zone cells.

Primary Cilia Length and Orientation in OA Cartilage

In normal cartilage, the length of superficial zone chondrocyte cilia were statistically significantly shorter than deep zone cilia with a mean length of 1.1 μm compared with 1.5 μm in deep zone cells (P < 0.05). Cilia showed a clear orientation in relation to the zone of origin; superficial zone chondrocyte cilia were consistently oriented away from the articular surface, (i.e., on the inferior aspect of the cell with respect to the articular surface; Fig. 2A,B; arrow). Middle and deep zone cilia were located either between two cells within a chondron (Fig. 2D,F; arrows) or on the medial or lateral cell membranes (with respect to the longitudinal axis from the articular surface to the subchondral bone; Fig. 2F; arrow).

In OA tissue, cilia were present on over 50% of cells within clusters with a mean length of 1.3 μm for mild OA and 1.4 μm for severe OA (Table 1). There was no significant difference in cilia length between mild and severe OA cells at the eroding surface. Most of the primary cilia in cell clusters were oriented toward the center of the cluster within the pericellular matrix of the OA chondron (Fig. 3D–F; arrows). Similarly, there was no significant difference in cilia length of deep zone chondrocytes from mild and severe OA cartilage compared to normal (Table 1). Osteoarthritic deep zone chondrocyte cilia showed a similar orientation to the orientation of normal deep zone chondrocyte cilia.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Primary cilia are sensory organelles found on many different cell types throughout the body. Because defects in both cilia structure and function are known to be associated with several key developmental processes and pathologies (Christensen et al.,2007; Hildebrandt and Zhou,2007; Pan and Snell,2007; Satir and Christensen,2007; Weimbs,2007; Yoder,2007), this study investigated whether there are also defects associated with primary cilia of articular chondrocytes during the pathological development of OA.

Although primary cilia have previously been reported in osteoarthritic cartilage (Kouri et al.,1998; Capin-Gutierrez et al.,2004), this is the first study to fully characterize these organelles in a model of primary OA. The study used mature bovine patella articular cartilage (Broom et al.,2001; Broom and Flachsmann,2003; Thambyah and Broom,2007) that correlated well with the progressive stages of the disease reported in human degenerative OA (Buckwalter and Mankin,1998b). This progression included cartilage softening, fibrillation, fissures, and chondrocyte clustering.

It is well documented that different zones of articular cartilage show distinct zone-dependent properties related to biomechanics, cell morphology, gene expression, and ECM composition (Muir et al.,1970; Muir,1995; Buckwalter and Mankin,1998a; Wong and Carter,2003). In the current study, we have described zonal variation in the incidence and length of primary cilia of normal cartilage, which both increased with distance from the articular surface (see Table 1). These variations may result from the different stress, strain, and fluid flow characteristics of each of the zones. In the superficial zone, fluid flow, tensile strain, and compressive strain are maximal, with compressive strains up to 25% (O'Connor et al.,1988; Guilak et al,1995; Wong and Carter,2003). In contrast, compressive strains in the middle and deep zones of normal cartilage range between 0 and 15% (O'Connor et al.,1988; Guilak et al,1995; Wong and Carter,2003). We believe that the mechanical forces generated during joint motion influence cilia incidence and length via intraflagellar transport, a process involved in the assembly, maintenance, and resorption of cilia (Wheatley and Bowser,2000; Pazour and Rosenbaum,2002; Rosenbaum and Witman,2002). For example, several studies of endothelial cells have shown that continuous fluid flow results in cilia disassembly, suggesting that the level of shear stress regulates cilia number (Iomini et al.,2004; Van der Heiden et al.,2006,2007). Our own preliminary studies of articular chondrocytes cultured in three-dimensional gels show that prolonged supraphysiological exposure to compressive strain causes a dramatic reduction in cilia number and length (McGlashan et al.,2006b), thus supporting the hypothesis that mechanical forces regulate the incidence and length of cilia in mechanically responsive tissues.

During OA, there are marked changes in the mechanical properties of both the ECM and the pericellular microenvironment of the chondron due to the catalytic breakdown of the ECM (Poole et al.,1991,1997a; Alexopoulos et al.,2003; Guilak et al.,2006; Zhang et al.,2006). Significantly, we found that with increased severity of OA, there was also an increase in both cilia incidence and length at the eroding articulating surface (see Table 1). This finding could be due to the loss of superficial zone tissue, leaving the middle and/ or deep zone tissue remaining, both of which contain a higher proportion of ciliated cells with longer cilia. Alternatively, the increase in cilia incidence may be a direct response to the ongoing disease processes. If primary cilia do have a mechanosensory function in cartilage, then a population of cells with an increased proportion of cilia could ultimately affect mechanotransduction signalling pathways within the tissue. Further studies are required to investigate these differences and are beyond the scope of the current study.

The orientation and position of primary cilia confer a specific polarity to the cell that provides essential positional cues for tissue morphogenesis and maintenance (Park et al.,2006; Ascenzi et al.,2007; Song et al.,2007; Blacque et al.,2008). Although chondrocytes do not have apical and basal polarity in the conventional sense of epithelial cell polarity, they show a clear orientation in three ways: (1) within the chondron, (2) with respect to the articular surface (and/ or the tidemark) and, (3) in relation to the direction of mechanical loading. In this study, superficial zone cilia in normal cartilage were almost always oriented away from the articular surface on the inferior aspect of the cell, concurring with studies in equine, canine, and murine articular cartilage (Wilsman et al.,1980; Poole et al.,1985; McGlashan et al.,2007). This may provide some degree of stress shielding in an area where both tensile and compressive forces are maximal (Alexopoulos et al.,2003). In contrast, cilia within the cell clusters at the eroding articulating surface in OA projected toward the center of the cluster (see Fig. 3). The majority of middle and deep zone cilia of both normal and OA cartilage also projected away from the articular surface, but the significance of this orientation remains unclear, although it may be related to the polarized secretion of ECM macromolecules.

Unlike chondrocytes of the growth plate, articular chondrocytes do not undergo terminal differentiation and, unless perturbed, remain in a postmitotic quiescent state throughout life (Horton et al.,2006). However, the clonal expansion of cells within an osteoarthritic chondron (see Fig. 3) is thought to be a recapitulation of the proliferation and terminal differentiation observed in the developing growth plate (Tchetina et al.,2005; Appleton et al.,2007). Because there is increasing evidence supporting a critical role of the primary cilium in growth plate development (Zhang et al.,2003; Gouttenoire et al.,2007; Haycraft et al.,2007; Koyama et al.,2007; McGlashan et al.,2007; Ruiz-Perez et al.,2007; Song et al.,2007), we believe that cilia-mediated signalling may be involved in the activation of the terminal differentiation of the chondrocytes observed in chondron remodeling during OA progression.

In summary, this is the first study to examine primary cilia in a defined model of OA. We have shown that primary cilia are present on chondrocytes throughout OA and the overall percentage of ciliated cells within the degenerating cartilage increases with OA severity. Although OA has long been thought of as a condition affecting the ECM of articular cartilage, it is now well accepted that the chondrocytes also play a pivotal role in the catabolic and anabolic processes that drive the degenerative process. However, just how the process of degeneration in OA is initiated (by means of the ECM or the cell or both) remains unknown. We speculate that changes in cilia-mediated signalling may play a role in the onset of OA and research in this area will be the focus of future functional studies.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Bovine Patellae

Twelve bovine patellae were obtained from the local abattoir and graded based on the extent of fibrillation, yielding normal (grade 0; N = 4), mild (grade 1; N = 2), and severe (grade 2; N = 6) samples. The patellae were fixed in 4% paraformaldehyde for 2 hr and washed in phosphate buffered saline (PBS; Invitrogen, Auckland, New Zealand). The osteoarthritic lesions were present on the superiolateral surface of the patella and tissue from normal patellae was sampled from similar weight-bearing positions. Three-millimeter-thick slices perpendicular to the cartilage surface were cut using a bandsaw in a mediolateral orientation for each patella. A chisel was then used to separate the cartilage from the bone, yielding a full-thickness cartilage sample with limited subchondral bone attached.

Histochemical and Immunohistochemical Staining

For histological staining, cartilage samples were decalcified in Gooding and Stewart's solution containing 5% formaldehyde/25% formic acid for 10 days at room temperature, dehydrated through a series of alcohols, and impregnated with paraffin wax. Sections 4 μm thick were stained using Safranin O/Fast Green and examined using brightfield microscopy.

For immunohistochemical detection of primary cilia, nondecalcified cartilage samples were embedded in OCT compound and snap-frozen in methyl butane vapor and liquid nitrogen. Sections 20 μm thick were cut perpendicular to the surface of the cartilage using a cryostat and placed into individual wells of a 24-well plate containing PBS. The “free-floating” sections were subsequently incubated in testicular hyaluronidase (2 mg/ml in Tris-HCl, pH 5.5; Sigma-Aldrich, Auckland, New Zealand) for 2 hr at 37°C followed by washing in PBS containing 0.1% bovine serum albumin (BSA; Serva, Heidelberg, Germany). Sections were then permeabilized using 0.5% Triton X-100 (v/v) in PBS for 5 min, followed by 5% (v/v) goat serum in PBS (Sigma-Aldrich, Auckland, New Zealand) for 30 min at room temperature. Sections were labeled overnight at 4°C with a primary antibody raised against acetylated α-tubulin (1:500; Sigma) followed by goat anti-mouse Alexa 488 secondary antibody (1:500; Invitrogen, Auckland, New Zealand) for 2 hr at room temperature. Sections were mounted onto slides with ProLong Gold antifade (Invitrogen).

Microscopy and Analysis of Primary Cilia

Fluorescently labeled primary cilia were imaged using a Nikon epifluorescent microscope and a ×100 oil immersion objective lens (Nikon, Japan). To quantify primary cilia incidence for full-depth cartilage sections, the number of cells expressing a single primary cilium was recorded from each field of view from the articular surface to the tidemark in five separate regions of each patella. The mean (± SE) percentage of ciliated cells was calculated.

Normal articular cartilage is divided into four clear zones—superficial, middle, deep, and calcified zones. In this study, cilia incidence was measured in the superficial, middle, and deep zones, and 20 randomly selected fields of view, each containing between 4 and 10 cells, were examined in each zone. Similarly, the percentage of cells expressing a primary cilium was recorded within each field of view and the mean (± SE) percentage value was calculated. Images of the same area were also captured using differential interference contrast (DIC) microscopy. The orientation of cilia with respect to the articular surface was also recorded during this analysis.

For cilia length measurement, sections were imaged using a Leica TCS SP2 confocal laser scanning microscope (Leica, Heidelberg, Germany) with a ×100 oil immersion objective lens and ×4 optical zoom. Due to the differences in the z resolution of the microscope compared to the x and y planes, only cilia that were approximately 90° to the incident light were selected. This ensured that the maximum z depth measured was 1.5 μm. Serial optical z-sections (z = 0.35 μm thick) were collected for each image. Two-dimensional projections of cilia were acquired from a minimum of 40 cells for each zone and cilia length was measured using Image J software (NIH Image, Bethesda, MD).

Statistical Analysis

Data are expressed as mean (± SE). Differences in cilia incidence and length were analyzed using a one-way analysis of variance followed by post hoc Bonferroni corrected t-tests.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We thank Professor Neil Broom and Dr. Rene Flachsmann (University of Auckland) for providing tissue and advice on OA grading for this study. We also thank the Biomedical Imaging Research Unit, University of Auckland, for technical support and Sarah Kennedy for her technical assistance. EC Cluett's studentship was supported by the HOPE Foundation for Research on Ageing, New Zealand.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES