1. Top of page
  2. Abstract


Hypertrophic chondrocyte differentiation is a key step in endochondral ossification that produces basic calcium phosphates (BCPs). Although chondrocyte hypertrophy has been associated with osteoarthritis (OA), chondrocalcinosis has been considered an irregular event and linked mainly to calcium pyrophosphate dihydrate (CPPD) deposition. The aim of this study was to determine the prevalence and composition of calcium crystals in human OA and analyze their relationship to disease severity and markers of chondrocyte hypertrophy.


One hundred twenty patients with end-stage OA undergoing total knee replacement were prospectively evaluated. Cartilage calcification was studied by conventional x-ray radiography, digital-contact radiography (DCR), field-emission scanning electron microscopy (FE-SEM), and synovial fluid analysis. Cartilage calcification findings were correlated with scores of knee function as well as histologic changes and chondrocyte hypertrophy as analyzed in vitro.


DCR revealed mineralization in all cartilage specimens. Its extent correlated significantly with the Hospital for Special Surgery knee score but not with age. FE-SEM analysis showed that BCPs, rather than CPPD, were the prominent minerals. On histologic analysis, it was observed that mineralization correlated with the expression of type X collagen, a marker of chondrocyte hypertrophy. Moreover, there was a strong correlation between the extent of mineralization in vivo and the ability of chondrocytes to produce BCPs in vitro. The induction of hypertrophy in healthy human chondrocytes resulted in a prominent mineralization of the extracellular matrix.


These results indicate that mineralization of articular cartilage by BCP is an indissociable process of OA and does not characterize a specific subset of the disease, which has important consequences in the development of therapeutic strategies for patients with OA.

Osteoarthritis (OA) is the most common joint disorder and is characterized by cartilage loss, new bone formation at the margins of the joints (osteophytes), changes in subchondral bone, and recurrent synovitis. The incidence of OA increases with age. Calcium pyrophosphate dihydrate (CPPD) crystals are known to cause acute attacks of pseudogout in the joints, but crystal deposition has also been reported to be associated with OA (1). Aside from CPPD crystals, basic calcium phosphates (BCPs), such as carbonate-substituted hydroxyapatite (HA), tricalcium phosphate, and octacalcium phosphate, have been found in the synovial fluid (SF), synovium, and cartilage from patients with OA (2–4). The data concerning the distribution and frequency of their occurrence vary, depending on patient selection and crystal identification methods (5–7). Identification of BCP crystals in OA joints remains problematic because of the lack of simple and reliable methods of detection. The clinical and pathologic relevance of cartilage mineralization in patients with OA is not completely understood. The crystals are associated with aggravated joint degeneration and large joint effusions (4). Aside from its role in OA, pathologic mineralization of articular cartilage has also been described in association with aging and with various genetic and metabolic disorders (8–10).

It is important to note, however, that mineralization of cartilage is not only a pathologic phenomenon. Physiologic mineralization of cartilage matrix occurs during the longitudinal growth of bones through endochondral ossification at the growth plate, accompanied by chondrocyte differentiation. During this process, chondrocytes proliferate rapidly to form a model for the bone. As chondrocytes divide, they build a columnar structure and secrete cartilage-specific extracellular matrix proteins such as type II collagen and aggrecan. Once this is accomplished, chondrocytes stop dividing and increase their volume dramatically, becoming hypertrophic. As part of their hypertrophic differentiation, they alter the matrix by secreting type X collagen, which, along with other factors, promotes matrix mineralization and ossification (11).

In healthy articular cartilage, chondrocytes maintain a stable phenotype and resist proliferation and differentiation (12). In contrast, articular chondrocytes from OA joints develop terminal differentiation and hypertrophy, typically close to areas of mineralized cartilage matrix and near sites of surface lesions (13, 14). Hypertrophic chondrocytes in the growth plate and in OA cartilage share certain features, including synthesis of type X collagen, up-regulated expression of various transglutaminases (15, 16), and the release of matrix vesicles responsible for the initial formation of HA or CPPD crystals (17–19).

A current paradigm involves chondrocyte hypertrophy of OA cartilage in the mineralization process, with characteristics similar to those of the hypertrophic chondrocytes in the growth plate. In this study, we examined the hypothesis that mineralization of articular cartilage is generally present in OA cartilage, and not only as a subentity in certain progressive forms of OA.


  1. Top of page
  2. Abstract


One hundred twenty patients with end-stage primary OA who were consecutively undergoing total knee arthroplasty at one orthopedic center (University Medical Center Hamburg–Eppendorf, Clinic Bad Bramstedt) were prospectively included in this study. All patients in the study gave full written informed consent for participation prior to the operative procedure. The mean ± SD age of the patients was 69.8 ± 8.71 years (range 45–88 years) at the time of surgery, and of these 120 patients, 79 were female and 41 were male. Detailed clinical data on all of the patients are listed in Table 1.

Table 1. Clinical characteristics of the 120 patients with knee osteoarthritis (OA) and 5 control subjects with malignant bone tumors*
  • *

    HSS = Hospital for Special Surgery; K/L = Kellgren/Lawrence.

 No. female/no. male79/41
 Age at time of operation, mean ± SD (range) years69.8 ± 8.7 (45–88)
 Affected knee, no. left/no. right54/66
 Body mass index, mean ± SD (range) kg/m229.4 ± 3.6 (18.1–32.5)
 Duration of disease, mean ± SD (range) years4.4 ± 4.3 (0.8–13.1)
 Preoperative HSS score, mean ± SD (range) points54.9 ± 10.2 (34–73)
 Preoperative K/L radiographic OA grade, no. of patients 
  Grade II31
  Grade III50
  Grade IV39
Controls, diagnosis 
 Female age 14 yearsOsteosarcoma proximal tibia
 Female age 18 yearsOsteosarcoma distal femur
 Female age 58 yearsMetastasis (pulmonal carcinoma) proximal femur
 Male age 14 yearsChondrosarcoma proximal tibia
 Male age 16 yearsChondrosarcoma proximal tibia

Five patients with malignant bone tumors in the lower extremities (3 women ages 14, 18, and 58 years, and 2 men ages 14 and 16 years) were included as controls (Table 1). During total femur or proximal tibia replacement, cartilage was harvested from the unaffected knee joint following the same procedure as used for the patients with OA.

Preoperatively, radiographs of the knees (standard anteroposterior and lateral views) were analyzed for the presence of calcification. The grade of knee OA radiographic severity was classified according to the criteria of Kellgren and Lawrence (K/L) (20). The patients were examined clinically, and the results were documented by the Hospital for Special Surgery (HSS) knee score (21). The HSS score is widely used for objective clinical evaluation of the knee. A maximum of 100 points can be assigned; scores below 60 points indicate a severe functional deficit in the knee joint.

During surgery, the medial and lateral meniscus as well as the distal femoral condyle and the resected tibial plateau were obtained from each patient. In addition, SF samples were obtained from the patients' knee joints.

Analysis of SF

The SF from each patient was analyzed by polarized light microscopy as well as standard light and phase-contrast microscopy. SF samples were studied with a Zeiss AxioPlan 2 (Carl Zeiss Instruments, Oberkochen, Germany) with filters for compensated polarization on light microscopy. At least 40 high-power fields (200× magnification) were analyzed for the presence of crystals. The light microscope, with its limit of resolution of ∼1 μm, can identify CPPD but not BCP crystals.

Digital-contact radiography (DCR)

To assess mineralization, ∼1 cm2 of the articular cartilage from the medial femoral condyle and the medial and lateral meniscus was analyzed using DCR. All specimens were fixed and stored in 10% formalin prior to imaging. DCR was performed using a digital mammography imaging technique (Hologic, Waltham, MA) operating at 25 kV in manual mode, usually at 3.8 mA, and with a film focus distance of 8 cm. These digital images were analyzed for calcification by image analysis software (NIH ImageJ 1.32; National Institutes of Health, Bethesda, MD).

Quantitative measurements of the areas of total cartilage mineralization were performed in a manner as described by Mitsuyama et al (10). This technique marks a given range of grey levels on the contact radiographs, to identify the areas for measurement. By adjusting the upper and lower bounds on a threshold utility, the area of mineralization can be highlighted. These areas are then used for calculating the percentage of calcification in relation to the surrounding whole cartilage area. In some cases, there was a gradient in the cartilage, especially in cartilage from the meniscus, due to varying thickness of the specimens. In this case, the mineralized area was identified using manual methods, such as a spray can tool. Regardless of whether the mineralized area was marked by thresholding or by manual methods, the percentage of calcification was determined by dividing the area of mineralization by the whole area of cartilage. This gives a relative concentration of mineralization for that cartilage specimen. For a subset of 60 randomly selected patients, interobserver and intraobserver analyses were performed to assess variability in measurement reproducibility.

Histologic assessment of cartilage

Histologic assessment of the knee was performed on cartilage plugs from the medial femoral condyle, and a modified Mankin scoring system was used to assess the severity of changes in OA articular cartilage. The Mankin grades (range 0–14 points) for mild, moderate, and severe OA were 2–5, 6–9, and 10–14 points, respectively (22, 23).

Field-emission scanning electron microscopy (FE-SEM).

Electron microscopy studies were performed on specimens of the cartilage after contact radiography had been completed. These FE-SEM studies were performed on a subset of the first 50 consecutive patients, without further selection. This population was identical to the overall population in terms of clinical characteristics.

Specimens were fixed in 10% formalin and critical-point dried. The dishes of cultured chondrocytes and the agarose cultures were also critical-point dried. Mineral characterization was achieved by means of a JEOL 6300F field-emission electron microscope (JEOL, Palo Alto, CA) equipped with an Oxford Inca energy-dispersive x-ray system, as described elsewhere (24). Araldite ultra-thin slides were examined with a JEOL 3010 transmission electron microscope, and x-ray diffraction patterns of the mineral samples were recorded.

Isolation and culture of chondrocytes

Articular cartilage was removed, under sterile conditions, from the femoral condyle and tibial plateau of each patient. The cartilage pieces were cut into 1-mm slices and washed 3–4 times with phosphate buffered saline (PBS) containing 100 units/ml penicillin, 100 μg/ml streptomycin, and 2.5 μg/ml amphotericin B. The slices were incubated for 16–24 hours with calcium-free Dulbecco's modified Eagle's medium (DMEM) at high glucose, along with 1.5 mg/ml collagenase B (Roche, Mannheim, Germany) and 1 mM cysteine. The cell suspension was filtered and washed with PBS. A culture of 1 × 105 cells/cm2 was performed in 24-well plates in DMEM supplemented with 60 μg/ml β-amino-propionitrile fumarate, 25 μg/ml sodium ascorbate, 1 mM cysteine, 1 mM pyruvate, 100 units/ml penicillin, 100 mg/ml streptomycin, and 2.5 μg/ml amphotericin B. The medium was changed every 2 days.

Agarose cultures of chondrocytes

Serum-free suspension cultures of articular chondrocytes stabilized by agarose gels were prepared as described elsewhere (13). Briefly, cell suspensions (3 × 106 cells/ml) were mixed with equal volumes of liquid 1% agarose (low-melting grade, SeaPlaque; FMC Bioproducts, Ratingen, Germany) in DMEM, freshly prepared by autoclaving 2% agarose in H2O before mixing with double-strength DMEM. The cell suspensions were layered on 35-mm dishes precoated with 1% agarose (high-melting grade, SeaKem; FMC Bioproducts) in H2O. The medium was DMEM, containing 60 μg/ml of β-amino-proprionitrile fumarate, 50 μg/ml of sodium ascorbate, 1 mM cysteine, 1 mM pyruvate, 100 units/ml penicillin, and 100 μg/ml streptomycin (complete medium), and was changed every 2 days. For induction of maturation and hypertrophy, 50 ng/ml of L-thyroxine (Sigma, St. Louis, MO) and 10 μg/ml of chymotrypsin (Sigma) were added. For detection of type X collagen, supernatants of the agarose cultures were subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis on 4.5–15% gradient gels, and chemiluminescent immunoblotting with a rabbit antiserum to human type X collagen was performed using an enhanced chemiluminescence kit according to standard procedures (Amersham International, Little Chalfont, UK).

Alizarin red binding assay

Calcification in the culture dishes was measured with alizarin red staining, which was further validated by direct visual observation of alizarin red staining in each plate, as described by Johnson et al (15). Cells were removed by trypsinization and the dishes were washed with PBS 3 times; 250 μl/well of 0.5% alizarin red S was then added, and the mixture was incubated for 10 minutes. After washing with PBS, 250 μl/well cetylpyridium chloride was added, and the mixture was incubated for another 10 minutes. The solution was transferred into standard dilution rows quantifying for micromoles of bound alizarin red S, corrected for protein content. All measurements were performed in triplicate.

Isolation of RNA and quantitative real-time polymerase chain reaction (PCR)

Total RNA from cultured chondrocytes was isolated using the RNeasy Mini Kit (Qiagen, Chatsworth, CA). Samples of total RNA were reverse transcribed into complementary DNA (cDNA) in the presence of 100 pM random hexamers (Life Technologies, Gaithersburg, MD), 10 units of RNase inhibitor, 5× reverse transcriptase buffer, dNTP mix (80 mM), and Moloney murine leukemia virus reverse transcriptase for 120 minutes at 37°C. Amplification of the generated cDNA was performed in a TaqMan 7300, using predesigned TaqMan gene expression assays, with Hs99999905_m1 for GAPDH and Hs00166657_m1 for Col10A1, following the manufacturer's guidelines (Applied Biosystems, Weiterstadt, Germany).

Statistical analysis

Results are expressed as the mean ± SEM. The correlation between continuous numeric data was analyzed by Spearman's rank correlation analysis. The Kruskal-Wallis test was used to determine significant differences between groups of continuous numeric variables and categorical variables. Significant relationships between categorical variables were calculated with Pearson's chi-square test. Student's t-test was used to determine significance in dependent, pairwise comparisons. P values less than 0.05 were considered significant.


  1. Top of page
  2. Abstract

Detection of mineralization of the articular cartilage by DCR, but not by conventional x-ray radiography, in all OA patients.

Analysis of cartilage mineralization by different techniques revealed the low sensitivity of conventional x-ray radiography for the detection of crystal deposition. Figure 1A (first column) shows the data from 4 representative patients whose radiographs showed no evidence of chondrocalcinosis. In spite of their negative conventional radiography results, some mineralization was detected in the menisci by SF and DCR analyses (Figure 1A, second and third columns), and all patients evaluated by DCR displayed a clear mineralization of the articular cartilage (Figure 1A, fourth column). Systematic analysis of the radiographs of the knees of all 120 patients showed mineralization in the medial and/or lateral joint space in 31 patients (25.8%). Analysis of the SF by polarized and phase-contrast light microscopy revealed the presence of crystals in 72 patients (60%). However, when DCR was applied to articular cartilage specimens from the medial femoral condyle and both menisci, mineralization was found in all 120 cartilage specimens. Specifically, mineralization was found in all specimens (100%) of articular cartilage and in 75 specimens (62.5%) of meniscal tissue (Figure 1B).

thumbnail image

Figure 1. Analysis of the menisci and articular cartilage from the knees of patients with end-stage osteoarthritis (OA) for the extent of mineralization, by conventional x-ray radiography, synovial fluid analysis (SFA), and digital-contact radiography (DCR). A, The radiographs from 4 representative patients (first column) did not show any mineralization, whereas SF analysis (second column) and DCR of the meniscus (third column) demonstrated mineralization in some cases. The corresponding DCR of the articular cartilage (fourth column) revealed mineralization in all cases. B, When comparing the percent mineralization by method of detection in all 120 patients with OA, mineralization of the articular cartilage was identified by DCR in 100% of patients, whereas mineralization of the meniscus was shown by DCR in 62.5% of patients, by SF analysis in 60% of patients, and by conventional x-ray radiography in 25.8% of patients. C, The Hospital for Special Surgery (HSS) knee score for each of the 120 patients was calculated preoperatively to document the clinical status of the affected knee joint. The amount of articular cartilage mineralization determined by DCR showed a highly significant inverse correlation (indicated by the red line) with the preoperative HSS score. Patients with worse scores showed a greater area of cartilage mineralization than did patients with better scores. D, The age of the patients at the time of surgery was assessed for any correlation with the area of articular cartilage mineralization as analyzed with DCR, but no correlation was found. NS = not significant.

Download figure to PowerPoint

As part of the preoperative evaluation, clinical data were obtained from all 120 patients, and a clinical HSS score for knee function was calculated for each patient. The mean ± SEM preoperative HSS score was 54.9 ± 10.2 points (range 34–73). The area of articular cartilage mineralization showed a highly significant inverse correlation with the preoperative HSS score (P < 0.01) (Figure 1C). As a result, patients with large areas of mineralized articular cartilage had worse preoperative clinical conditions than did patients who had only a small amount of mineralized cartilage. In contrast, meniscal mineralization and the presence of CPPD crystals in the SF did not correlate with the HSS score, and, as noted in Patients and Methods, analysis of the SF by light microscopy is not capable of detecting BCP crystals.

Most interestingly, there was no correlation between the area of articular cartilage mineralization and age (Figure 1D). In this context, no correlation was found between the age of the patients and the radiologic OA grade according to the K/L classification scale (K/L grade 2 at mean ± SEM age 69.1 ± 1.5 years, K/L grade 3 at age 69.1 ± 1.1 years, and K/L grade 4 at age 71.3 ± 1.7 years). Also, no correlation between age and the histologic OA grade according to the Mankin scoring system was established (Mankin grade 1 at mean ± SEM age 72.6 ± 4.5 years, Mankin grade 2 at age 70.7 ± 1.4 years, and Mankin grade 3 at age 63.2 ± 3.2 years). The control specimens did not show any mineralization when studied by any of the above-described techniques, including DCR. Spearman's rank correlation showed a highly significant intraobserver correlation (P < 0.001, r2 = 0.9628) and a highly significant interobserver correlation (P < 0.001, r2 = 0.8992) for the quantification of mineralization using the DCR technique, as assessed in a subset of 60 patients.

Primary deposition of BCPs and its correlation with the degree of histologic changes as demonstrated by ultrastructural analysis.

Histologic assessment of the articular cartilage showed that the mineralization of articular cartilage increased as the destruction of the cartilage progressed. This is illustrated in Figure 2A, in which representative histologic staining and DCR data from 3 OA cartilage samples (one for each Mankin grade) and a control specimen are shown. Quantitative analysis of all 120 samples revealed a significant increase in matrix mineralization in cartilage specimens as the histologic severity increased according to ascending Mankin grade (Figure 2B).

thumbnail image

Figure 2. A, Histologic assessment was performed on cartilage from the medial femoral condyle of patients with knee osteoarthritis (OA), in the site next to the piece obtained for contact radiography, and a modified Mankin grade was assigned as an indication of the severity of changes in the OA articular cartilage. Three representative histologically stained sections of OA cartilage for each Mankin grade and 1 from the non-OA control group are shown (top), along with their corresponding digital-contact radiography (DCR) results (bottom). Whereas the control shows intact articular cartilage with no mineralization detected, an increase in matrix mineralization in parallel with higher Mankin grades of cartilage destruction was observed in the DCR analysis. B, The percent area of mineralization was compared between each Mankin grade. Results are the mean and SEM for Mankin grade 1 cartilage (0.26 ± 0.11%), for Mankin grade 2 cartilage (1.59 ± 0.40%), and for Mankin grade 3 cartilage (3.87 ± 1.39%). C, On a subset of 50 consecutive specimens from the total of 120, field-emission scanning electron microscopy (FE-SEM) analysis in combination with energy-dispersive x-ray analysis was performed. All areas of mineralization identified with DCR were proven to be calcium phosphate mineral aggregates, comprising either apatite crystals (top) or calcium pyrophosphate dihydrate (CPPD) crystals (bottom). D, Of the 50 specimens, 9 (18%) showed CPPD crystals. A mineral phase containing basic calcium phosphates (BCPs) was detected in all 50 cartilage specimens (100%).

Download figure to PowerPoint

To determine the exact nature of the deposited crystals, FE-SEM analysis was performed on a subset of 50 consecutive specimens. All areas of mineralization identified with DCR were proven to be calcium phosphate aggregates (Figure 2C). In addition, our FE-SEM studies demonstrated that mineral aggregates could be distinguished into at least 2 phases, one nearly amorphous phase and one idiomorphic mineral phase, with different chemical compositions that matched those of CPPD and BCP crystals. On 2 representative specimens for each phase, transmission electron microscopy and x-ray diffraction were performed, and the existence of CPPD and HA-type BCP was confirmed by these techniques. However, of the 50 specimens examined, only 9 cartilage specimens showed CPPD crystals, whereas a mineral phase containing BCPs was detected in all 50 cartilage specimens (Figure 2D).

Relationship between the ability of OA chondrocytes to mineralize matrix in vitro and the mineralization pattern of OA cartilage in vivo.

We next sought to investigate whether the production of BCP crystals is an intrinsic function of OA chondrocytes, and whether the ability of chondrocytes isolated from OA cartilage to produce BCPs in vitro correlates with our in vivo findings. To this end, chondrocytes were isolated from 30 of the 120 OA patients and cultured over a 6-day period. The chondrocyte phenotype was confirmed by demonstrating the expression of type II collagen with real-time PCR (results not shown). Chondrocytes from 4 individuals with malignant bone tumors in the proximal femur but having no involvement of the articular cartilage served as controls.

Alizarin red staining revealed a continuous increase of matrix mineralization by cultured OA chondrocytes over time (Figure 3A, top). Whereas no mineralization by chondrocytes was observed in the control group, we found a significant correlation between the area of cartilage mineralization determined by DCR in vivo (Figure 3A, middle) and the ability of chondrocytes isolated from the respective OA patients to mineralize their surroundings (P = 0.027) (Figure 3B). Specifically, chondrocytes from patients with a large amount of mineralization in the cartilage, as determined in the DCR analysis, showed a greater capability to mineralize the matrix in vitro than did chondrocytes from patients with less mineralization. When FE-SEM was used to study the OA chondrocytes, the crystals found in culture were shown to be BCP crystals in all cases (Figure 3A, bottom), which also corresponds to the in vivo findings. CPPD crystals were found in combination with BCPs in the culture of chondrocytes derived from 1 patient whose articular cartilage also had CPPD and BCP crystals detectable in vivo.

thumbnail image

Figure 3. Changes in articular cartilage mineralization according to increasing histologic severity by Mankin grade and relative increase of absorption by chondrocytes from the second day of chondrocyte culture to the sixth day in relation to mineralization. A, Chondrocytes were isolated and cultured as described in Patients and Methods, and the calcification in the culture dishes was measured with an alizarin red staining assay (ABA) (top). Three representative cultures on day 6 (1 from a control subject and 2 from osteoarthritis patients with small [Mankin grade 1] and large [Mankin grade 3] amounts of articular cartilage mineralization) are shown. A significant positive correlation between articular cartilage mineralization detected by digital-contact radiography (DCR) (middle) and increased mineralization in cultures (bottom) on day 6 was observed. B, The scatterplot shows the significant correlation (indicated by the red line) between the relative chondrocyte culture mineralization determined by ABA and the percent articular cartilage mineralization on day 6.

Download figure to PowerPoint

Association of mineralization of the extracellular matrix with hypertrophic chondrocyte differentiation in OA.

Based on the notion that the deposition of BCPs is a characteristic feature of hypertrophic chondrocytes during endochondral ossification (11), we addressed whether BCP crystal deposition by OA chondrocytes is associated with the hypertrophic differentiation of chondrocytes. Type X collagen staining of cartilage specimens representing each of the Mankin grades of histologic severity, as compared with control specimens, showed increased type X collagen staining in parallel with advanced matrix degeneration according to ascending Mankin grade, as well as increased cartilage mineralization in the DCR analysis (Figure 4A). Using quantitative PCR, we found a significant correlation between the extent of type X collagen messenger RNA expression, as a marker of chondrocyte hypertrophy, with the area of articular cartilage mineralization determined by DCR (P = 0.009) (Figure 4B).

thumbnail image

Figure 4. Results of type X collagen staining and field-emission scanning electron microscopy (FE-SEM) in agarose cultures. A, Type X collagen staining of representative osteoarthritis (OA) cartilage specimens of each Mankin grade, in comparison with a non-OA control case, show increased type X collagen staining (bottom) in parallel with further matrix degeneration according to the Mankin grade, as well as increased cartilage mineralization in the digital-contact radiography (DCR) analysis (top). B, Quantitative real-time polymerase chain reaction was performed to analyze the amount of type X collagen messenger RNA expression as a marker of chondrocyte hypertrophy. The scatterplot demonstrates a significant correlation (indicated by the red line) between increased type X collagen messenger RNA expression and an increase in the corresponding articular cartilage mineralization detected by DCR. C, The increase in the matrix mineralization capacity of articular chondrocytes from articular cartilage over 3 weeks was revealed by FE-SEM studies on long-term suspension cultures in agarose gels. Chondrocytes proliferated and became overtly hypertrophic from day 5 to day 21, and this was enhanced when the medium contained 50 ng/ml of L-thyroxin (T4). With energy-dispersive x-ray analysis, apatite mineralization was detected in these cultures (top). Type X collagen was identified in each culture (cult.) supernatant (supernat.) as an immunoreactive band at 66 kd (bottom). ND = not determined. D, The ability of chondrocytes to mineralize their culture (as detected by alizarin red binding assay) increased with the amount of type X collagen detected in stimulated (T4) compared with unstimulated (unstim.) cultures, which was quantified from the immunoreactive band with pixel analysis. Bars show the mean and SEM.

Download figure to PowerPoint

To test whether healthy chondrocytes start to mineralize the matrix when they become hypertrophic, we tested the mineralization capacity of chondrocytes from healthy articular cartilage in long-term suspension cultures in agarose. Under these conditions, chondrocytes proliferated and became hypertrophic within 3 weeks. The hypertrophy was more pronounced when cells were stimulated with L-thyroxin. Equal amounts of total protein from each culture supernatant were loaded onto a gel, and radioactively labeled type X collagen was identified to different extents in these cells (Figure 4C).

Moreover, the capability of chondrocytes to mineralize their cultures increased with growing hypertrophy. This was seen clearly in the alizarin red binding assay (Figure 4D), in which the results indicated that growing hypertrophy of the chondrocytes was linked to matrix calcification. FE-SEM analysis of the long-term agarose cultures showed that BCP crystals (specifically, apatite crystals) were deposited (Figure 4C).


  1. Top of page
  2. Abstract

These results show that, in contrast to what has been known, mineralization of articular cartilage is a common event in end-stage OA and is associated closely with disease progression. Of note, we found regular mineralization of the hyaline articular cartilage, but not of the menisci, although meniscal cartilage is the area of the joint in which mineralization is usually expected. Furthermore, we demonstrated that BCP, particularly apatite, rather than CPPD is the predominant crystal in OA cartilage and was found in all OA patients. This underlines the hypothesis that articular cartilage mineralization in OA and the classic mineralization of the meniscus, known as chondrocalcinosis, are distinct processes. Conventional x-ray radiographic analysis and SF analysis with both light microscopy and polarized light microscopy were proven to be insensitive measures for the identification of articular cartilage mineralization.

The fact that mineralization with BCP was found in all cartilage specimens from patients with OA has a significant impact on diagnostic strategies, treatment concepts, and our understanding of OA. In the present study we showed that there is a significant correlation between clinical symptoms and the amount of mineralized articular cartilage. Mitsuyama et al (10) recently showed that age, rather than the presence of OA, is the predominant factor driving progressive pathologic calcification in articular cartilage. However, this hypothesis is not supported by the present results obtained in 120 consecutive patients with knee OA, because we did not find any correlation between patients' age and the mineralized condition of the cartilage.

In order to obtain reliable tools for the diagnosis of special entities of OA and, consequently, design special treatment options for these cases, considerable efforts have been undertaken to find simple, usable clinical techniques for the detection of BCP crystals in the joint fluid and articular cartilage of OA patients (25). Our present results suggest that cartilage mineralization is not a variable feature of human OA but occurs as part of the disease, similar to the formation of osteophytes. This suggests that special diagnostic tools for the detection of BCP crystals in the SF of OA patients are not necessary, because these crystals are always present as a part of the disease and do not characterize a specific disease subset.

Total knee arthroplasty is the operative therapy of choice in patients with end-stage OA. The management of the patella articular surface at the time of primary total knee arthroplasty is controversial. As a result of this study, a strong argument for patellar resurfacing at the time of primary total knee arthroplasty can be made. With this technique, the whole OA cartilage is removed from the knee joint and no mineralized cartilage remains. It is possible that mineralized cartilage is responsible for the occasional anterior knee pain that is present after total knee arthroplasty without patella resurfacing.

When developing conservative treatment options, the fact that mineralization is always present should be taken into consideration. Agents that reduce crystal growth, such as phosphocitrate (26) or first-generation bisphosphonates (27), should be considered as new therapeutic strategies to inhibit further BCP crystal formation and to reduce clinical symptoms in OA knees.

In contrast to other studies (2, 5–9), we found mineralization in all of our specimens of articular cartilage when DCR was used for detection. Electron microscopic analysis of 50 specimens revealed that apatite is the prominent mineral, whereas CPPD crystals occur infrequently. CPPD has been associated with many clinical manifestations, including acute pseudogout and OA, and is an important pathogen in joint disease, and was recently also associated with a genetic defect (28). However, according to the findings of this study, the association between BCP crystals and OA is much stronger, since their presence correlated significantly with the severity of cartilage degeneration.

A significant correlation between the expression of type X collagen, a marker for chondrocyte hypertrophy, and the area of cartilage mineralization as determined by DCR was seen, and chondrocyte cultures with increased type X collagen expression showed a significantly greater capability to mineralize matrix than did cultures in which type X collagen was rarely found. It is well established that mineralization of articular cartilage is often found close to hypertrophic chondrocytes (1, 5). In the process of endochondral ossification, the link between hypertrophy and matrix mineralization is particularly well described (12, 13). Hypertrophic chondrocytes in OA cartilage and at the growth line share certain features, comprising not only hypertrophy but also a capability to mineralize the matrix. These data indicate that chondrocyte hypertrophy is a key factor in articular cartilage mineralization that is indissociably linked to OA.


  1. Top of page
  2. Abstract

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. Fuerst 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. Fuerst, Bertrand, Echtermeyer, Schäfer, Steinhagen, Pap, Rüther.

Acquisition of data. Fuerst, Bertrand, Lammers, Dreier, Nitschke, Rutsch, Schäfer, Steinhagen, Lohmann.

Analysis and interpretation of data. Fuerst, Bertrand, Lammers, Dreier, Schäfer, Niggemeyer, Steinhagen, Pap.


  1. Top of page
  2. Abstract
  • 1
    Ryan LM, McCarthy DJ. Calcium pyrophosphate crystal deposition disease, pseudogout and articular chondrocalcinosis. In: KoopmanWJ, editor. Arthritis and allied conditions: a textbook of rheumatology. 13th ed. Baltimore: Williams & Wilkins; 1997. p. 21035.
  • 2
    Derfus BA, Kurian JB, Butler JJ, Daft LJ, Carrera GF, Ryan LM, et al. The high prevalence of pathologic calcium crystals in preoperative knees. J Rheumatol 2002; 29: 5704.
  • 3
    Halverson PB, McCarty DJ. Patterns of radiographic abnormalities associated with basic calcium phosphate and calcium pyrophosphate dihydrate crystal deposition in the knee. Ann Rheum Dis 1986; 45: 6035.
  • 4
    Nalbant S, Martinez JA, Kitumnuaypong T, Clayburne G, Sieck M, Schumacher HR Jr. Synovial fluid features and their relations to osteoarthritis severity: new findings from sequential studies. Osteoarthritis Cartilage 2003; 11: 504.
  • 5
    Boivin G, Lagier R. An ultrastructural study of articular chondrocalcinosis in cases of knee osteoarthritis. Virchows Arch A Pathol Anat Histopathol 1983; 400: 1329.
  • 6
    Sokoloff L, Varma AA. Chondrocalcinosis in surgically resected joints. Arthritis Rheum 1988; 31: 7506.
  • 7
    Abreu M, Johnson K, Chung CB, de Lima JE Jr, Trudell D, Terkeltaub R, et al. Calcification in calcium pyrophosphate dihydrate (CPPD) crystalline deposits in the knee: anatomic, radiographic, MR imaging, and histologic study in cadavers. Skeletal Radiol 2004; 33: 3928.
  • 8
    Mitrovic DR, Stankovic A, Iriarte-Borda O, Uzan M, Quintero M, Miravet L, et al. The prevalence of chondrocalcinosis in the human knee joint: an autopsy survey. J Rheumatol 1988; 15: 63341.
  • 9
    Ishikawa K, Masuda I, Ohira T, Yokoyama M. A histological study of calcium pyrophosphate dihydrate crystal-deposition disease. J Bone Joint Surg Am 1989; 71: 87586.
  • 10
    Mitsuyama H, Healey RM, Terkeltaub RA, Coutts RD, Amiel D. Calcification of human articular knee cartilage is primarily an effect of aging rather than osteoarthritis. Osteoarthritis Cartilage 2007; 15: 55965.
  • 11
    Kronenberg HM. Developmental regulation of the growth plate. Nature 2003; 423: 3326.
  • 12
    Binette F, McQuaid DP, Haudenschild DR, Yaeger PC, McPherson JM, Tubo R. Expression of a stable articular cartilage phenotype without evidence of hypertrophy by adult human articular chondrocytes in vitro. J Orthop Res 1998; 16: 20716.
  • 13
    Von der Mark K, Kirsch T, Nerlich A, Kuss A, Weseloh G, Gluckert K, et al. Type X collagen synthesis in human osteoarthritic cartilage: indication of chondrocyte hypertrophy. Arthritis Rheum 1992; 35: 80611.
  • 14
    Kirsch T, Nah HD, Shapiro IM, Pacifici M. Regulated production of mineralization-competent matrix vesicles in hypertrophic chondrocytes. J Cell Biol 1997; 137: 114960.
  • 15
    Johnson KA, van Etten D, Nanda N, Graham RM, Terkeltaub RA. Distinct transglutaminase 2-independent and transglutaminase 2-dependent pathways mediate articular chondrocyte hypertrophy. J Biol Chem 2003; 278: 1882432.
  • 16
    Johnson KA, Rose DM, Terkeltaub RA. Factor XIIIA mobilizes transglutaminase 2 to induce chondrocyte hypertrophic differentiation. J Cell Sci 2008; 121(Pt 13): 225664.
  • 17
    Thouverey C, Bechkoff G, Pikula S, Buchet R. Inorganic pyrophosphate as a regulator of hydroxyapatite or calcium pyrophosphate dihydrate mineral deposition by matrix vesicles. Osteoarthritis Cartilage 2009; 17: 6472.
  • 18
    Derfus B, Kranendonk S, Camacho N, Mandel N, Kushnaryov V, Lynch K, et al. Human osteoarthritic cartilage matrix vesicles generate both calcium pyrophosphate dihydrate and apatite in vitro. Calcif Tissue Int 1998; 63: 25862.
  • 19
    Derfus BA, Kurtin SM, Camacho NP, Kurup I, Ryan LM. Comparison of matrix vesicles derived from normal and osteoarthritic human articular cartilage. Connect Tissue Res 1996; 35: 33742.
  • 20
    Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis 1957; 16: 494502.
  • 21
    Ranawat CS, Insall J, Shine J. Duo-condylar knee arthroplasty: Hospital for Special Surgery design. Clin Orthop Relat Res 1976; 120: 7682.
  • 22
    Mankin HJ, Dorfman H, Lippiello L, Zarins A. Biochemical and metabolic abnormalities in articular cartilage from osteo-arthritic human hips. II. Correlation of morphology with biochemical and metabolic data. J Bone Joint Surg Am 1971; 53: 52337.
  • 23
    Ostergaard K, Andersen CB, Petersen J, Bendtzen K, Salter DM. Validity of histopathological grading of articular cartilage from osteoarthritic knee joints. Ann Rheum Dis 1999; 58: 20813.
  • 24
    Pritzker KP, Cheng PT, Grynpas MD, Holmyard DP. Crystal associated diseases: role of scanning electron microscopy in diagnosis. Scanning Microsc 1988; 2: 14718.
  • 25
    Yavorskyy A, Hernandez-Santana A, McCarthy G, McMahon G. Detection of calcium phosphate crystals in the joint fluid of patients with osteoarthritis: analytical approaches and challenges. Analyst 2008; 133: 30218.
  • 26
    Cheung HS, Kurup IV, Sallis JD, Ryan LM. Inhibition of calcium pyrophosphate dihydrate crystal formation in articular cartilage vesicles and cartilage by phosphocitrate. J Biol Chem 1996; 271: 280825.
  • 27
    Russell RG, Rogers MJ. Bisphosphonates: from the laboratory to the clinic and back again. Bone 1999; 25: 97106.
  • 28
    Williams CJ, Zhang Y, Timms A, Bonavita G, Caeiro F, Broxholme J, et al. Autosomal dominant familial calcium pyrophosphate dihydrate deposition disease is caused by mutation in the transmembrane protein ANKH. Am J Hum Genet 2002; 71: 98591.