If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
To quantify the urinary concentration of cartilage oligomeric matrix protein (COMP), and to evaluate the relationship between urinary COMP concentration and the catabolic activity of synovial fluid (SF) in diseased horses.
Methods
COMP in horse urine was detected by immunoblotting with a monoclonal antibody (mAb; 14G4) raised against equine COMP from articular cartilage. Urine and serum samples were obtained from 83 Thoroughbred horses with aseptic joint diseases (AJD, 79 horses) or septic joint diseases (SJD, four horses) at the time of anesthesia induction, and samples of SF were obtained during surgery. Control samples of urine (n=111) were collected from normal horses free of any orthopedic diseases after they had been racing. COMP concentration was determined in all samples using inhibition enzyme-linked immunosorbent assay (ELISA) with mAb 14G4. SF samples were also used for the quantification of gelatinase activity.
Results
Positive bands of COMP fragments were determined on the immunoblots with mAb 14G4. The urinary COMP concentrations in AJD and SJD horses (1.02±0.75 and 1.55±1.17 μg/100 mg creatinine, respectively) were significantly higher than normal (0.57±0.29 μg/100 mg creatinine). In 55 horses with fractures in the AJD group there was a logarithmic relationship (r=−0.45, P<0.001) between the urinary and SF COMP measurements, while the urinary COMP level was positively correlated with matrix metalloproteinase (MMP)-2 and -9 activities (r=0.30, P<0.05 and r=0.51, P<0.001, respectively) in SF.
Conclusions
The urinary COMP assay with mAb 14G4 is useful for discriminating horses with osteoarthritis. The higher COMP levels in urine from such horses would be indicative of enhanced proteolytic activity, in addition to the increased COMP levels in the diseased joints.
Cartilage oligomeric matrix protein (COMP) in synovial fluid (SF) and serum has been universally recognized as a sensitive marker of cartilage degradation in clinical studies of osteoarthritis (OA) and rheumatoid arthritis (RA)
. The COMP level in SF correlates positively with antigenic keratan sulfate, and also changes depending on differences in cartilage deterioration in idiopathic osteonecrosis as well as hip OA
Increased pentosidine, an advanced glycation end product, in serum and synovial fluid from patients with knee osteoarthritis and its relation with cartilage oligomeric matrix protein.
. The serum level of COMP is significantly correlated with the Western Ontario and McMaster Universities index pain scale for the lower limbs, and higher in patients with bilateral knee OA than in unilaterally affected patients
. Another study has suggested that serum COMP concentration at the baseline could be useful for predicting the course of OA, progressors showing higher average values than non-progressors
Suggestion of nonlinear or phasic progression of knee osteoarthritis based on measurements of serum cartilage oligomeric matrix protein levels over five years.
. As just above, so far, the serum COMP level could be well indicative of the changeable pathology in human OA, whereas no report has demonstrated whether the higher serum COMP represents the increased COMP molecules in diseased SF or not.
In horses with OA, we have demonstrated augmented degradation of COMP in SF
, and suggested that COMP fragmentation in SF could be useful for monitoring equine OA in combination with enzyme-linked immunosorbent assay (ELISA) measurements
. Thereafter, we investigated the clinical usefulness of the serum baseline COMP level as a diagnostic marker of OA in racehorses; however, we did not obtain any clear-cut evidence such as a significant correlation of the COMP level in serum to that in SF, as well as to joint pathology or prognosis. There may be certain reasons why serum COMP may be a less sensitive indicator of joint deterioration in equine OA. Exercise-induced up-regulation of COMP turnover might be one factor that would complicate the detection of changes in serum COMP concentration due to OA pathology in horses. Neidhart et al.
pointed out that the baseline level of COMP in human marathon runners (before exercise) was significantly higher than in non-running healthy volunteers (controls who had never run in any competitive race, and who were age- and sex-matched). We have also obtained evidence that the baseline level of COMP in the serum of Thoroughbreds can increase significantly as a result of more intensive workouts (unpublished data). Our finding that the serum COMP level was increased 1 h after strenuous exercise on a racecourse and then recovered to the baseline within 24 h is also in agreement with previous studies
. The rapid clearance of COMP from serum might also make it difficult to determine any increase in the serum level specific to joint diseases.
Attenuation of COMP outflow from joint fluid into the blood circulation might be another factor that would obscure any change in the serum COMP concentration due to OA pathology. In horses with OA, increased catabolic activity reflected in the levels of matrix metalloproteinases (MMPs) and inflammatory cytokines has been demonstrated in SF
, but so far not in serum. Thus, the attenuation of COMP outflow from affected joints might not be due to proteolytic degradation in the blood compartment. We have another hypothesis that the increased amount of COMP fragments released from affected joints could be attenuated principally through dilution with a large volume of blood, and so would not be detectable unless a blood concentrate was examined. If it was possible to quantify COMP using samples of urine, which is a concentrated filtrate of blood, then measurements might be correlated with changes in COMP turnover in response to joint pathology. The urinary level of C-terminal cross-linking telopeptide of type II collagen (U-CTX-II) determined using a monoclonal antibody (mAb) raised against a less-proteolytic linear amino acid peptide has been demonstrated to be the most reliable of several biomarkers of OA
Cross-sectional association of 10 molecular markers of bone, cartilage, and synovium with disease activity and radiological joint damage in patients with hip osteoarthritis: the ECHODIAH cohort.
. Hypothesizing that urinary COMP would be a more reliable marker of equine OA than serum COMP, we devised a system for assay of urinary COMP using mAb 14G4. The aims of this study were to evaluate the usefulness of urinary COMP measurement for discriminating horses with OA from normal horses, and the relationship between the level of urinary COMP and joint pathology in OA assessed by sampling of SF.
Subjects, materials and methods
Sample collection
Individual urine and serum samples were obtained from 83 Thoroughbred horses with aseptic or septic joint diseases (AJD and SJD, 79 and four horses, respectively) at the time of anesthesia induction, and samples of SF were obtained at the time of arthroscopic surgery. All the horses were determined to be free of any medical diseases including renal failure and uropathy, based on the preoperative assessment of blood and urine. The surgical cases of AJD comprised 55 cases of intra-articular fracture including secondary degenerative joint disease (DJD) and 24 cases of osteochondrosis (OC; including osteochondritis dissecans (OCD) and bone cyst). Control samples of urine (n=111) were collected from normal horses judged to be free of any orthopedic diseases on the basis of clinical, radiological and ultrasound examinations, after the horses had won a race. The laboratory tests for the urine samples also showed no abnormalities suggesting any renal disorder. The samples were centrifuged to remove debris or cells, and the supernatants were stored at −70°C until assay. Prior to assay, all SF samples were pretreated with hyaluronidase according to Neidhart et al.
(characteristics of the horses are summarized in Table I).
Table ICharacteristics of the horses, and diagnosis of the affected joints
Normal horses (n=111)
Age (years, mean±1 SD)
2.5±1.1
Gender
Male
71
Female
40
Joint diseased horses (n=83)
Age (years, mean±1SD)
2.2±1.2
Gender
Male
52
Female
37
AJD (n=79)
Fracture
Carpus
37
Hind fetlock
14
Fore fetlock
3
Tarsus
1
OC
Tarsus
19
Stifle
5
SJD (n=4)
Shoulder
1
Fore fetlock
1
Hind fetlock
1
Tarsus
1
The control samples were obtained from the normal horses without any joint diseases. Diseased samples were divided into two categories such as AJD and SJD.
, the gels were electrotransferred onto polyvinyl difluoride membranes. After blocking, mAb 14G4, diluted 1:10,000 in 5% skimmed milk in Tris-buffered saline containing 0.05% Tween 20 (TBS/Tween), was applied to the membrane. Positive binding to COMP was detected using alkaline phosphatase (AP)-conjugated goat anti-mouse IgG antibody (A36882) diluted 1:10,000 in 2% skimmed milk in TBS/Tween, and demonstrated by development of the reaction with a substrate (BCIP/NBT, Sigma Fast/B-56552).
Inhibition ELISA
COMP concentration in urine, SF, and serum was analyzed by inhibition ELISA with mAb 14G4 directed against equine COMP designed according to previous protocols
with some modifications. In brief, 50 μl of purified horse COMP antigen in a coating buffer (20 mM sodium carbonate, 20 mM sodium bicarbonate, 0.002% sodium azide, pH 10) was placed in each well at 5 μg/ml, and then incubated for 2 h at room temperature followed by overnight at 4°C. Seventy microliters of diluted standards (range: 13.5–0.01 μg/ml) and the samples (final dilutions: urine 1/5, SF 1/70, serum 1/20) were mixed with the same volume of mAb 14G4 (final dilution: 1/100,000) in phosphate-buffered saline containing 0.05% Tween 20 (PBS/Tween), and then incubated overnight at 4°C. Coated wells were washed with PBS, blocked, and then incubated with 100 μl/well of the COMP-antibody mixture for 1 h at 37°C and then for 1 h at 4°C. Binding of the primary antibody to COMP on the plate was detected using goat anti-mouse IgG conjugated to AP and addition of an AP chromophore. One hour later, production of the chromophore was stopped by addition of 2 M H2CO3 (30 μl/well), and the absorbance was read at 405 nm (OD405). To determine the intra- and inter-assay variability, we thawed fresh aliquots of a urine sample and measured 10 repeats on one plate for 10 consecutive working days (10 plates). Considering that more concentrated urine would show a higher COMP level, measurements of urinary COMP (μg/ml) were normalized to the urinary creatinine concentration (mg/dl), which was quantified by the Jaffe method (Creatinine Test Wako, 277-10501), and expressed as μg COMP/100 mg creatinine.
Analysis of SF gelatinase activity
Gelatin zymography was performed using a protocol described previously
with some modifications. Two microliters of gelatinase standard [supernatant of medium used for culture of LPS-stimulated equine peripheral blood mononuclear cells (PBMC)] and 1.5 μl of SF were added to 20 μl of zymography sample buffer containing 7.5% SDS, 2.5% glycerol, 1% bromophenol blue, and 1 M Tris–HCl (pH 6.8). Then 10 μl of the mixture was poured into each well of a 7.5% polyacrylamide gel containing 2.5 mg/ml gelatin, followed by electrophoresis at room temperature. The gel was placed in 2.5% Triton X-100, dissolved in distilled water for 1 h in order to remove the SDS, and then rinsed repeatedly in distilled water to remove the Triton. The gelatinolytic reaction was allowed to proceed at 37°C for 18 h in the gelatin refolding buffer including 50 mM Tris–HCl, pH 7.6, 0.2 M NaCl, and 50 mM CaCl2. After staining the gels with 1% Coomassie brilliant blue for 30 min and immersion in distilled water overnight, gelatinases were visible in the gels as fine decolorized bands on a blue background. A mixture of gelatinase standard from a LPS-stimulated PBMC culture was loaded into one well per plate as the internal standard for the gelatinase activity.
The enzyme activity of the samples was quantified as the band density of pro- and active MMP-2 and -9 in each lane, which was determined on the basis of molecular size, using the software packages Image Gauge Ver. 4.0 and L Process (Science Lab. 2001, Fuji Film, Tokyo, Japan). The actual measurements for the samples were converted to values relative to the internal gelatinase standard.
Statistics
All quantitative group data obtained by ELISA were analyzed for normality, and expressed as mean±1 standard deviation (SD). As the data showed a normal distribution, we selected the parametric means by statistical analysis. The differences in urinary COMP concentrations between the groups (vs control) were analyzed by factorial analysis of variance (ANOVA), and Scheffé's method was used for simultaneous multiple comparisons. Differences at P<0.01 were considered statistically significant. Relationships of COMP measurements among the urine, SF, and serum samples from diseased horses, and the correlation between urinary COMP and MMP activity in SF were analyzed using the linear regression model. Correlations were considered statistically significant at P<0.05.
Results
As shown in the immunoblots using the AP reaction (Fig. 1), mAb 14G4 bound to the fragments of equine COMP molecules in urine. The sizes of the positive bands were determined to be approximately 150, 100, and 60 kDa. The urine inhibition curves are given in Fig. 2. The linear portion of the inhibition curve obtained by diluting the horse urine (dilution ratio between 0.5 and 0.03125) paralleled that of the standard curve for purified COMP (between 0.8 and 6.75 μg of equine cartilage COMP/ml) using mAb 14G4. As in urine samples from other species such as mouse, rat, dog, and human, the inhibition curves were evident for equine articular COMP, although the linear portions for the other species differed slightly in parallel with that for horse urine. The intra- and inter-assay variabilities of horse urinary COMP measurements were calculated to be 6.0 and 8.0%, respectively.
Fig. 1Immunoblot analysis of urine using mAb 14G4 and AP reaction; mAb 14G4 bound to fragments of equine COMP in urine. The sizes of the positive bands were determined to be approximately 150, 100, and 60 kDa.
Fig. 2Inhibition curves of COMP in urine. The linear portion of the curve obtained by dilution of horse urine (dilution ratio between 0.5 and 0.03125) paralleled that of the standard curve obtained using purified COMP (between 0.8 and 6.75 μg of equine cartilage COMP/ml) and mAb 14G4.
The measured levels of COMP in urine from horses with AJD (1.02±0.75 μg/100 mg creatinine) and SJD (1.55±1.17 μg/100 mg creatinine) were significantly (P<0.001 and P<0.01, respectively) higher than those in normal horses (0.57±0.29 μg/100 mg creatinine) (Fig. 3). The levels of urinary COMP in horses with fractures (1.01±0.70 μg/100 mg creatinine) and OC (1.04±0.87 μg/100 mg creatinine) were also significantly (P<0.001 and P<0.01, respectively) higher than those in the controls.
Fig. 3COMP concentration in urine from control and diseased horses. Urinary COMP levels in horses with AJD (including fractures and OC) and SJD were significantly higher than those in the controls.
The relationships of COMP measurements among the urine and SF samples are shown in Fig. 4. The relationship between urinary and SF COMP levels in the 55 horses with fractures in the AJD group was expressed by a negative logarithmic function (r=−0.45, P<0.001), but there was no significant correlation in the OC and SJD cases. Urinary COMP level was also not significantly correlated with the serum levels of COMP in any cases of AJD and SJD (data not shown). In cases of fracture, the urinary COMP levels were positively correlated with MMP-2 and -9 activity (r=0.30, P<0.05 and r=0.51, P<0.001, respectively) in SF (Fig. 5).
Fig. 4Relationship of measured COMP levels among urine, SF, and serum samples from diseased horses. There was a significant negative logarithmic correlation between the urinary and SF COMP levels in horses with fracture, but no significant correlation between OC and SJD cases.
Fig. 5Correlation of urinary COMP levels with MMP-2 and -9 activities in SF from horses with fracture. The urinary COMP levels were positively correlated with MMP-2 and -9 activities.
The mAb 14G4, which was raised against monomeric COMP from equine articular cartilage, recognizes a specific peptide sequence, C134FPRVRCINTSPGF147, in the EGF-like repeat portion of the molecule
. Since positive binding with mAb 14G4 was well conserved in spite of the increased fragmentation of COMP in osteoarthritic SF, the epitope appears to be relatively undamaged during the proteolysis that occurs in OA. The resistance of the epitope recognized by mAb 14G4 to COMP catabolism might account for the detectability of the fragments in urine. Our immunoblots of equine urine revealed positive bands of approximately 150, 100, and 60 kDa. This molecular size, which was larger than we expected, was approximately similar to that of the bands representing degradation products in SF we reported previously
. In OA patients it has been reported that most of the MMPs that play a critical role in joint destruction show evidently higher proteolytic activity in SF, while not so high in serum, unlike the situation in RA
. COMP molecules released from the extra cellular matrix (ECM) could be fragmented enzymatically in the SF, whereas the fragments flowing out of the SF might be degraded to a lesser extent during blood circulation. Although at present we have no explanation for how large COMP fragments can be transferred through the glomerular basement membrane, COMP fragments might be degraded to a lesser extent during filtration from the blood compartment into urine.
The linear portions of inhibition curves for not only equine urine but also samples from other species such as mouse, rat, dog, and human, contained portions that paralleled the standard curve obtained using purified COMP. In preliminary studies (data not shown), we investigated whether any other factors such as circadian rhythm, exercise, and diuretics might compromise the reproducibility of urinary COMP data. Although we found no evidence of any effect of circadian rhythm on the measurements, the urine level of COMP in the afternoon was slightly higher than that in the morning (not significant difference). In addition, urine samples taken within 30 min after strenuous exercise showed a slightly higher concentration of COMP, which was also not significantly different from that before exercise. The large variance of the urine levels in our normal horses, as the percentage of SD divided by the mean value in the controls is 51%, could be partially due to the circadian rhythm. In rat urine obtained at the same time of a day, the SD is lower, that is around 10% of the average of the COMP baseline. Urinary COMP became undetectable in samples taken after diuresis with furosemide. We obtained the control samples from horses on one day, and intravenously administered 80 mg furosemide to them just around the same time of the next day. The urine samples were obtained on urination after a few minutes of the injection, and analyzed by immunoblotting and ELISA. The positive bands of COMP seen in the control samples disappeared in the sample after the furosemide injection. Identically, the urine COMP level after the injection fell below measurable limits of the inhibition ELISA. On the basis of the inhibition curve for the equine cartilage COMP standard, the urinary COMP levels for horses with AJD and SJD were significantly higher than those for normal horses. Nevertheless, the increased urine COMP levels in the diseased horses were not positively correlated with the levels in either their SF or serum. Taken as a whole, the statistical analysis demonstrated that the higher COMP level in urine might not simply reflect the increase of COMP in diseased joints, even though COMP accumulation has been demonstrated around cartilage lesions and sites showing mild fibrillation, severe fibrillation and fibrocartilage
The distribution of cartilage oligomeric matrix protein (COMP) in equine carpal articular cartilage and its variation with exercise and cartilage deterioration.
In the 55 horses with fracture in the AJD group, the negative logarithmic relationship between the urinary and SF levels of COMP suggested further possibilities about the clinical significance of urinary COMP as a biomarker. Based on the dispersal of sample values in the scattergram (Fig. 4), it appeared that cases could be sorted approximately into three groups: (1) a high SF COMP level with a low urinary COMP level, (2) a high urinary COMP level with low SF COMP level, and (3) a group that lay outside the other two groups. This might correspond to three stages in joint pathology secondary to intra-articular fracture: an acute phase during which mechanical injury of the bone and cartilage is limited to the area adjacent to the fracture site, a chronic phase (or OA phase) during which moderate to severe cartilage degradation and bone remodeling (deformity) could expand around the wound, and a transitional phase in which the OA (or DJD) pathology would be mild. It has been reported previously that MMP activity in SF increases in relation to the severity of cartilage damage in equine osteoarthritic joints
Relationship between synovial fluid levels of glycosaminoglycans, hydroxyproline and general MMP activity and the presence and severity of articular cartilage change on the proximal articular surface of P1.
, we evaluated the correlation of urinary COMP measurements with MMP-2 and -9 activities in SF. As expected, the urinary COMP level was significantly correlated with the activities of MMP-2 and -9 in SF, especially the latter. COMP macromolecules originating from injured or degenerative ECM would be enzymatically fragmented in SF prior to lymphatic/blood outflow, while subsequently the fragments would not be degraded any more in blood circulation, and filtered and concentrated into urine. Thus the catabolic activity in the SF of horses with OA would be predictable by analysis of urinary COMP.
Again, considering the characteristic correlation between measured COMP levels in SF and urine in horses with fracture, with reference to enhanced enzymatic degradation rather than overproduction (or accumulation) of COMP during the development of OA, the partitioning of the data into the three groups mentioned above could be explained as follows. A lower urinary COMP level indicates lower gelatinolytic activity in SF, where a large amount of COMP would be degraded to a lesser extent and increasingly retained (the cases represented in the first group on the scattergram). On the other hand, in the cases with a high level of urinary COMP, strong catabolic activity in SF would reduce the molecular size of COMP, and the resulting fragments would easily flow out into the synovial lymphatic system (the cases represented in the second group of the scattergram). The transitional cases might be those distributed in the intermediate part of the scattergram. It has been reported that increased MMP-9 activity may be associated with the severity of cartilage deterioration in OCD
, and another study of OC lesions has demonstrated an increase of catabolic enzymes and lower regeneration of matrix molecules in response to mechanical forces
. In the present study there was no significant correlation between urinary COMP levels and SF gelatinase activity in OC cases. Although the COMP levels in urine and SF from OC cases were equivalent to those of fracture cases, the gelatinase levels in SF from OC horses were lower than the fracture cases (Table II, Fig. 3). It has been reported that the MMP-2 and -9 can be overgenerated by chondrocytes at the OCD lesions
. The MMP-2 and -9 activities could be more increased at the OC lesions of cartilage rather than in the SFs. We speculate in our OC cases that the COMP molecules could be principally degenerated by the gelatinase increasing at the OC lesions in cartilage, therefore that the increased urine level of COMP fragments might correlate to the increased gelatinase activity in OC cartilage rather than that in the SF. Re-evaluation should be also done using a larger number of OC samples.
Table IIMMP-2, -9, and COMP levels in SFs from the diseased horses
MMP-2
MMP-9
COMP (μg/ml)
Fracture
4.13±2.12
0.20±0.26
7.31±3.97
OC
2.83±0.47
0.12±0.21
5.93±3.15
SJD
3.14±1.46
12.04±4.91
6.06±2.01
* and † indicate the statistical significance at P<0.025 and P<0.001, respectively.
This study represents the first attempt to quantify the concentration of COMP in urine, and is the first to demonstrate increased urinary COMP levels in horses with AJD and SJD. Assay of urinary COMP using mAb 14G4 is useful for discriminating horses with OA. The higher COMP levels in urine from horses with OA are indicative of augmented proteolytic activity, in addition to increased COMP levels in the diseased joints themselves.
Acknowledgments
This study was supported by The Japan Society for the Promotion of Science (grant No. 16380212).
References
Skoumal M.
Kolarz G.
Klingler A.
Serum levels of cartilage oligomeric matrix protein. A predicting factor and a valuable parameter for disease management in rheumatoid arthritis.
Increased pentosidine, an advanced glycation end product, in serum and synovial fluid from patients with knee osteoarthritis and its relation with cartilage oligomeric matrix protein.
Suggestion of nonlinear or phasic progression of knee osteoarthritis based on measurements of serum cartilage oligomeric matrix protein levels over five years.
Cross-sectional association of 10 molecular markers of bone, cartilage, and synovium with disease activity and radiological joint damage in patients with hip osteoarthritis: the ECHODIAH cohort.
The distribution of cartilage oligomeric matrix protein (COMP) in equine carpal articular cartilage and its variation with exercise and cartilage deterioration.
Relationship between synovial fluid levels of glycosaminoglycans, hydroxyproline and general MMP activity and the presence and severity of articular cartilage change on the proximal articular surface of P1.
00133-6&id=gr1.jpg){kind=link}
00133-6&id=gr2.jpg){kind=link}
00133-6&id=gr3.jpg){kind=link}
00133-6&id=gr4.jpg){kind=link}
00133-6&id=gr5.jpg){kind=link}