- Email: [email protected]
Markers of cartilage and synovial metabolism in joint fluid and serum of patients with chondromalacia of the patella
Osteoarthritis and Cartilage (1998) 6, 115–124 7 1998 Osteoarthritis Research Society
1063–4584/98/020115 + 10 $12.00/0
Markers of cartilage and synovial metabolism in joint fluid and serum of patients with chondromalacia of the patella By Urho Va¨ a¨ ta¨ inen*, L. Stefan Lohmander†, Eugene Thonar‡, Tero Hongisto§, Ulla Ågren¶, Seppo Ro¨ nkko¨ >, Heikki Jaroma*, Veli-Matti Kosma**, Markku Tammi¶, and Ilkka Kiviranta* *Department of Surgery, Kuopio University Hospital, Kuopio, Finland; †Department of Orthopedics, University Hospital, Lund, Sweden; ‡Department of Biochemistry, Rush-Presbyterian-St. Luke’s Medical Center, Chicago, U.S.A.; §Department of Clinical Chemistry, Kuopio University Hospital; ¶Department of Anatomy, University of Kuopio; >Department of Pathology, Kuopio University Hospital, Kuopio, Finland Summary Objective: To further our understanding of the pathogenesis of chondromalacia of the patella (CM), we have studied the release into knee joint fluid and serum, obtained from patients with CM, of molecules associated with the metabolism of joint cartilage matrix and synovium. Methods: Interleukin-1 a (IL-1a), interleukin-1 b (IL-1b), interleukin-6 (IL-6), stromelysin-1 (MMP-3), interstitial collagenase (MMP-1), tissue inhibitor for metalloproteinases-1 (TIMP-1), phospholipase activity A2 (PLA2), hyaluronan (HA), aggrecan fragments (AGN) and antigenic keratan sulfate (KS) were quantified in knee joint lavage fluid from 96 patients with CM; KS and HA also was measured in serum. Chondromalacia was graded on a scale of I to IV according to Outerbridge (1961). The histopathology of the synovial membrane close to the patellofemoral joint was evaluated. Control samples were obtained from nine patients with knee pain presenting with arthroscopically normal knee joints. Results: The concentrations of MMP-3, MMP-1 and TIMP-1 proteins in joint lavage fluid were increased in advanced (grade IV) CM, compared with controls. Levels of MMP-1 in lavage fluid correlated with the severity of CM (r = 0.38, P Q 0.01) and MMP-1 and MMP-3 concentrations correlated with each other (r = 0.45, P Q 0.001). TIMP-1 was elevated in grade IV CM compared with grades II and III CM (P Q 0.02, P Q 0.01). Interleukins (IL-1a, IL-1b and IL-6) showed no significant change in CM. The lavage fluid level of PLA2 increased with the severity of CM (r = 0.40, P Q 0.001). Serum KS was higher in CM IV than in controls (P = 0.05), while lavage fluid KS concentration was elevated in CM I (P = 0.04). There were no differences in the lavage fluid levels of AGN and HA between the different study groups. Synovium showed slight or moderate histological signs of inflammation in 9% of CM patients. Conclusion: The changes in the release and activity of these marker molecules from serum and synovial fluid may reflect changes in the metabolism of articular cartilage and synovium in CM, that are consistent with those observed in early-stage tibiofemoral cartilage changes in OA. Key words: Chondromalacia, Joint fluid, Marker, Osteoarthritis, Knee.
arthritis (OA), but there is no general agreement that CM lesions consistently give rise to OA. In healthy articular cartilage, proteoglycan aggregates entrapped in collagen fibrils in the extracellular matrix enable the tissue to cushion forces applied at the end of long bones. In CM patellar cartilage becomes soft and fibrillated, ultimately degenerating and eroding. The release of cartilage breakdown products may cause synovitis in the knee joint [1], resulting in swelling and knee pain. Metalloproteinases (MMPs) have been suggested to play a role in cartilage matrix degradation [2–3]. These enzymes are able to degrade proteoglycans (PGs) and collagens. High
Introduction Chondromalacia of the patella (CM) is a common condition, but the mechanisms leading to cartilage degeneration and destruction of this joint are poorly understood. In late stages of CM, the cartilage changes resemble that seen in osteoReceived 10 March 1997; accepted 28 August 1997. Correspondence to: Urho Va¨a¨ta¨inen M.D., Department of Surgery, Kuopio University Hospital, Box 1777, FIN-70211 Kuopio, Finland. This study was supported by grants from the Foundation of Finnish Orthopedics and Traumatology, Finnish Medical Foundation, Orion Corporation Research Foundation, The Swedish Medical Research Council, the Medical Faculty and Hospital of Lund University and Merck Research Laboratories. Dr. Michael Lark, Merck Research Laboratories, Rahway, NJ, generously supplied the antibodies for MMP and TIMP assays.
115
116
Va¨a¨ta¨inen et al.: Joint fluid and serum markers of chondromalacia
levels of stromelysin (MMP-3) and collagenase (MMP-1) protein have been detected in the joint fluid of patients with rheumatoid arthritis (RA) and OA [4–5]. MMPs, as well their tissue inhibitors (TIMPs) are synthesized by chondrocytes and synovial cells [6–8]. These cells also produce cytokines, such as interleukin-1 alpha (IL-1a) and interleukin-1 beta (IL-1b) [9–10]. In inflamed joints, IL-1 stimulates the production of MMPs, and also suppresses the synthesis of proteoglycans and TIMP [11–14]. The effect of IL-1 on proteoglycan synthesis may be mediated by IL-6, a cytokine the synthesis of which is upregulated by IL-1 [15]. IL-6 does not appear to influence the production of MMPs, but does stimulate the production of TIMP [16–18]. IL-1 and tumor necrosis factor [TNF] regulate the production of phospholipase A2 (PLA2) [19], an enzyme generating arachidonic acid. In turn, arachidonic acid is converted into prostaglandins, which are involved in inflammatory reactions [20–21]. High levels of PLA2 have been found in joint fluid of RA and OA patients [22-23]. Hyaluronan (HA) is synthesized mainly by the synoviocytes and has an important role in providing lubrication and viscoelasticity in joint fluid [24, 25]. HA synthesis is stimulated by transforming growth factor 1 (TGF-b1), and interleukin-1 [26]. In inflammatory diseases such as RA, the concentration of HA in synovial fluid is decreased [27]. Products of articular cartilage and synovial tissue metabolism diffuse into synovial fluid and have been proposed to reflect joint tissue metabolism [5]. The concentrations of antigenic keratan sulfate (KS), aggrecan fragments (AGN) and certain aggrecan core protein epitopes in joint fluid are elevated in patients with OA and RA [5, 28], as well as after joint trauma [29]. Previous studies have shown that the concentrations of MMPs and TIMPs in joint fluid are likewise elevated in knee OA, after knee injury, [5, 29] and in RA [28]. Some of these markers of specific processes in joint disease also reach the blood circulation, either directly or via the lymphatic system [30], where they can be measured. Antigenic KS, for example, has been proposed to provide a measure of aggrecan metabolism in early OA [31]. In this study, we determined the concentrations in CM joint lavage fluid of the cytokines IL-1a, IL-1b, and IL-6, PLA2, MMPs (MMP-1 and MMP-3) and their inhibitor (TIMP-1), markers of PG degradation (KS and AGN), and HA. We propose that changes in the release of these molecules from joint tissues reflect changes in the metabolism of
joint cartilage and synovium in CM, consistent with those observed in early-stage tibiofemoral cartilage changes in OA.
Patients and Methods patients One hundred and thirty patients with anterior knee pain and strongly suspected CM, but showing no evidence of OA in their knee joint X-rays determined after Ahlba¨ck [32], were selected for study. After arthroscopic examination of the knee joint, the patients showing cartilage degeneration in other joint surfaces than patella (N = 5), or having meniscus tears (N = 20), in addition to CM were excluded from the study group. Nine patients did not have CM. After these exclusions 96 patients remained in the CM group. The CM patients were divided into subgroups according to Outerbridge [33] (Table I): grade I designates softened cartilage, grade II cartilage fibrillation less than 1/2 inch in diameter, grade III cartilage fibrillation greater than 1/2 inch in diameter and grade IV erosion of articular cartilage with exposed bone. The control group was formed from nine knee pain patients exhibiting no signs of CM or OA in any of the joint surfaces at arthroscopy (Table I). Patients with acute knee injuries (Q3 months) were excluded from the control group. Blood samples (20 ml) from CM and control patients were obtained from the cubital vein; centrifuged for 5 min at 500 g at 4°C , and the supernatants frozen and stored at −40°C. In the operating room, before arthroscopy, 20 ml sterile 0.9% saline solution was injected into the knee joint cavity, which was not previously emptied from endogenous joint fluid. To enhance mixing of the synovial fluid with the injected saline the knee joint was moved 10 times from extension to flexion and the suprapatellar region was manipulated before aspiration of the lavage fluid. The total volume of the aspirated fluid was recorded. In all cases, the lavage fluids were obtained using this standardized method between 9 a.m. and 1 p.m. Aliquots of the fluid were centrifuged for 5 min at 500 g, and the supernatants were frozen and stored at −40°C. At the beginning of arthroscopy, a sample of synovium was taken from the patellofemoral region with biopsy forceps from CM patients and from controls.
Osteoarthritis and Cartilage Vol. 6 No. 2
117
determination of il-1a, il-b and il-6
quantification of hyaluronan
Interleukin concentrations were determined by enzyme-linked immunosorbent assays (ELISAs). The IL-1a and IL-6 kits were purchased from Immunotech International (Immunotech S.A. Luminy Case 915, 13288 Marseille Cedex 9, France), while a Quantikine kit (R&D Systems, Inc. Minneapolis, U.S.A.) was used for IL-1b. The ELISA of IL-1a, was a two step assay while those for IL-1b and IL-6 were sandwich ELISAs. All measurements were carried out in duplicate according to the manufacturers’ instructions. Precision within an assay was in IL-1a 5.3%, in IL-1b 4.6% and in IL-6 4.4%. Inter-assay precision was respectively 8.7, 6.5 and 3.6%.
HA was quantified by an ELISA-like method using biotinylated HA-binding complex (bHABC) prepared from PG-link protein complex of bovine articular cartilage [39]. The assay method used is slightly modified from that described by Kongtawelert and Ghosh (1990) [40]. Briefly, Nunc Covalink plates (Covalink, Nunc, Roskilde, Denmark) were coated with HA (Sigma, St. Louis, MO, U.S.A.) and blocked with 1% bovine serum albumin (BSA). The samples were first digested for 24 h at 60°C by papain (250 mg/ml in 5 mm cysteine, 5 mm EDTA, Sigma); the enzyme then was inactivated by boiling for 10 min. The digested samples and a series of standard HA (5–200 ng/ml), diluted in 6% BSA in phosphate-buffered saline (PBS) were mixed with an equal volume of bHABC (450 ng/ml) in small polypropylene tubes (Mekalasi Oy, Kuortane, Finland) and incubated overnight at 4°C. Immediately prior to the assay, the mixtures were incubated for 1 h at 37°C. The samples were then transferred to the HA-coated plates, incubated for 90 min at 37°C, and washed with 0.5% Tween 20 in PBS. Horseradish peroxidase conjugated streptavidin (1:20 000 in PBS, Vector Lab, Burlingame, CA, U.S.A.) was added to the plates and left for 1 h at 37°C. After washing, the substrate-chromogen solution (0-phenylenediamine dihydrochloride; Sigma, 0.03% H2O2 in 0.1 m citrate buffer, pH 5.0) was added and left for 1 h. The reaction was stopped with 8 m H2SO4 and the absorbencies were read at 490 nm using a microtiter plate reader (Molecular Devices Corp, Palo Alto, CA, U.S.A.). All results are reported
pla2 activity assay The PLA2 assay method described by Ro¨nkko¨ and coworkers was used [34]. All measurements were done in duplicate. The intra-assay coefficient of variations (CV) was 3.6% and inter-assay CV was 5.6%.
quantification of aggrecan fragments and ks AGN containing the CS domain was quantified from joint lavage fluid by an ELISA according to Saxne and co-workers [35–36]. KS was determined from serum and joint lavage fluid by an ELISA based on the use of the monoclonal antibody ET-4-A-4 [37–38]. All measurements were performed in duplicate. Precision within an assay was 1.8–3.0% and inter-assay precision was 2.3–4.1%.
Table I Description of the subjects and synovial fluid samples
Group CM CM I CM II CM III CM IV Controls
Age (years) mean 2 s.d. median (range) 34.3 2 9.9 32.0 (40) 29.4 2 7.8 28.7 (29) 30.7 2 9.7 30.0 (37) 37.2 2 8.6 37.0 (33) 34.2 2 11.0 30.0 (31) 27.3 2 8.3 27.0 (22)
Number of patients
Female/male
96
59/37
20
13/7
26
14/12
34
19/15
16
13/3
9
4/5
Volume of the synovial fluid sample (ml)* mean 2 s.d. median (range) 9.8 2 5.5 10.0 (19) 8.6 2 6.0 7.5 (19) 9.4 2 5.8 9.0 (18) 11.4 2 5.2 10.5 (18) 12.0 2 5.5 13.0 (17) 8.7 2 5.5 9.0 (15)
*The amount of aspired synovial fluid from the knee joint after injection of 20 ml saline. CM = chondromalacia of the patella, CM I, CM II, CM III and CM IV refer to the grade of chondromalacia according to Outerbridge (1961).
118
Va¨a¨ta¨inen et al.: Joint fluid and serum markers of chondromalacia
here as the mean of the value obtained for analysis in duplicate. Intra-assay precision was 5.3% and inter-assay precision was 8.1%.
and two-tailed Mann–Whitney U-test and Spearman Rank Correlation test were used. Results
assay of mmp-3, mmp-1 and timp-1 MMP-1, MMP-3 and TIMP-1 concentrations in lavage fluid were measured by sandwich ELISAs, using monoclonal and polyclonal antibodies against the human recombinant proteins, as described previously [28, 41]. Polyclonal antisera against the specific proteins generated in rabbits were used as secondary reagents. The ELISAs for the MMPs detect the proforms of the enzymes, the large molecular active forms and the enzymes complexed to TIMP-1, but not small molecular forms of the enzymes or enzymes complexed to a2-macroglobulin. The assay for TIMP-1 detects only free TIMP-1, but not the inhibitor complexed with MMPs. The ELISAs have a log-linear relationship between 0.5–100 ng/ml (MMP-3), 2– 150 ng/ml (MMP-1), and 0.8–30 ng/ml (TIMP-1). Intra- and interassay coefficients of variation were less than 10% for these assays. The calculation of the molar ratio between MMP-3 and free TIMP was based on the assumption, that the major part (q90%) of MMP-3 is prostromelysin [28]. Similarly, the concentrations of MMP-3, MMP-1 and TIMP-1 are expressed as molarities (nm). histology of the synovium samples Samples of the synovial membrane were taken from 83 CM patients and from nine controls with biopsy forceps from the suprapatellar recess. The samples were fixed in 4% formaldehyde solution in phosphate buffer (pH 7.0, 0.013 m), embedded in paraffin, sectioned at 5 mm and stained with haematoxylin and eosin. The degree of inflammation was evaluated qualitatively in histological sections using a four step grading system. Grade 0 was assigned to samples showing no inflammatory cells (polymorphonuclear leukocytes, lymphocytes or macrophages). Grade 1 was assigned to slight, grade 2 to moderate and grade 3 to extensive number of inflammatory cells in the synovium samples. The samples were screened through and the greatest number of inflammatory cells was recorded. statistical analysis The results are expressed as mean 2 standard deviation and as median (range) values. For statistical analysis BMDP statistical software (BMDP Statistical Software, Los Angeles, U.S.A.)
All the groups were comparable with respect to age distribution (Table I). At the time of operation no macroscopic effusion was noted in the knee joints. The mean amount of aspirated lavage fluid after injection of 20 ml saline was 9.8 2 5.5 ml in CM patients and 8.7 2 5.5 ml in controls.
concentrations of il-1a, il-1b and il-6 in samples of lavage fluid The apparent increase in the level of interleukins (IL-1a, IL-1b and IL-6) in the CM group was not significant (Table II, IL-1a: P = 0.84, IL-1b: P = 0.61, IL-6: P = 0.27). The highest concentrations of IL-1a and IL-6 were in patients with grade II CM. The levels of IL-1b were similar in all CM groups (Table II). There was no correlation between the concentrations of IL-1a and IL-6 and none of these interleukins (IL-1a, IL-1b and IL-6) correlated to the severity of CM.
the concentrations of mmp-1, mmp-3, timp-1 protein, and pla2 activity in samples of lavage fluid The MMP-3, MMP-1 protein concentrations and PLA2 activity did not differ between the CM and control groups, when the whole groups were compared. However, the MMP-3 concentration in grade IV CM was significantly higher than that in the control group (P = 0.007, Table III) and in grade II CM (P = 0.03, Table III). TIMP-1 protein showed greater concentration in grade IV CM versus compared with grades II and III CM, respectively (P = 0.02, P = 0.01, Table III). The molar ratios of MMP-3 to TIMP-1 did not vary significantly among the subgroups (Table III). The concentration of MMP-1 was elevated in grade IV CM compared with the control group (P = 0.04, Table III) and with grades I and II CM (P = 0.03, Table III). The concentration of MMP-1 correlated with the grade of CM (r = 0.38, P = 0.006, N = 66). There was a positive correlation between MMP-3 and MMP-1 concentrations (r = 0.45, P Q 0.001, N = 65). The activity of PLA2 increased with the severity of the disease (r = 0.40, P Q 0.001, N = 79), but did not differ from the control group value when the whole CM group was compared with controls. PLA2 activity was increased in grade IV CM
Osteoarthritis and Cartilage Vol. 6 No. 2
119
Table II Synovial fluid concentrations of interleukin-1a (IL-1a, pg/ml), interleukin-1b (IL-1b, pg/ml) and interleukin-6 (IL-6, pg/ml) in all patients with chondromalacia of the patella (CM) and in subgroups of patients having different grades of CM (CM I–IV, judged according to Outerbridge 1961), and in controls Group CM CM I CM II CM III CM IV Controls
IL-1a (pg/ml)
IL-1b (pg/ml)
IL-6 (pg/ml)
1.9 2 3.7 (77) 0 (22) 1.8 2 3.2 (16) 0 (10) 3.0 2 5.3 (21) 0 (22) 1.5 2 3.0 (29) 0 (15) 1.2 2 1.8 (11) 0 (4) 1.0 2 1.3 (7) 0 (3)
0.7 2 1.8 (77) 0 (9.3) 1.1 2 2.5 (16) 0 (9.3) 0.4 2 1.1 (21) 0 (4.8) 1.0 2 2.1 (29) 0 (8.9) 0.1 2 0.3 (11) 0 (1.1) 0.6 2 1.5 (7) 0 (3.9)
3.9 2 15.2 (74) 0 (126) 4.2 2 8.2 (15) 0 (125) 7.4 2 26.8 (22) 0 (126) 1.5 2 3.2 (26) 0 (14) 2.2 2 3.5 (11) 9 (11) 0.6 2 1.5 (7) 1 (4)
The results are given as mean 2 .. and as median (range). The number of patients is expressed in parenthesis after the mean 2 .. value.
compared with grades I and II CM (P = 0.007, P = 0.02, Table III). concentrations of agn, ks and ha in samples of lavage fluid The concentrations of AGN and HA in the CM patients were not significantly different from those of the control group. The KS concentration in grade I CM was higher than in controls (P = 0.04, Table IV) and the highest concentrations of AGN
and KS were in grade I CM. AGN levels showed a tendency to decrease with increasing disease severity (r = −0.22, P = 0.04, N = 76, Table IV). Not surprisingly, KS levels in lavage fluid correlated closely with the level of AGN in CM patients (r = 0.69, P = 0.000, N = 76). HA concentrations in the joint lavage fluid of CM patients were at the same level in all grades of CM compared with controls (Table IV). A positive correlation was present between HA and AGN (r = 0.33, P = 0.005, N = 73).
Table III The synovial fluid concentrations of stromelysin (MMP-3, nm) tissue inhibitor of metalloproteinases (TIMP-1, nm), the molar ratio of MMP-3/TIMP-1 (MMP-3/TIMP-1), and synovial fluid concentrations of intersitial collagenase (MMP-1, nm) and the activity of phospholipase A2 (PLA2, nmol/min × ml) in all patients with chondromalacia of the patella (CM) and in subgroups of patients having different grades of chondromalacia (CM I–IV, according to Outerbridge 1961), and in controls Group CM CM I CM II CM III CM IV Controls
MMP-3 (nm)
TIMP-1 (nm)
MMP-3/TIMP-1
MMP-1 (nm)
2.3 2 3.2 (76) 1.4 (20.0) 2.6 2 2.9 (17) 1.8 (11.1) 2.1 2 3.3 (21)b 0.8 (14.5) 1.5 2 1.3 (27) 0.9 (5.5) 4.3 2 5.4 (11)* 2.0 (19.1) 1.6 2 1.8 (7) 0.8 (5.0)
2.0 2 2.2 (78) 1.2 (12.4) 1.8 2 1.4 (17) 1.1 (5.1) 2.2 2 2.8 (22)b 1.2 (10.5) 1.6 2 1.4 (28)c 1.0 (5.9) 3.2 2 3.1 (11) 2.3 (11.4) 1.4 2 0.9 (7) 1.0 (2.7)
1.4 2 1.5 (76) 0.9 (9.3) 1.2 2 1.1 (17) 0.9 (4.9) 1.1 2 1.3 (21) 0.9 (5.5) 1.7 2 2.0 (27) 1.0 (9.3) 1.4 2 1.0 (11) 0.9 (2.9) 0.8 2 0.4 (7) 0.8 (1.1)
0.06 2 0.2 (66) 0 (1.3) 0.01 2 0.04 (14)a 0 (0.2) 0.02 2 0.05 (18)b 0 (0.2) 0.04 2 0.08 (25) 0 (0.3) 0.3 2 0.4 (10)* 0.09 (1.3) 0 (9) 0
PLA2 (nmol/min × ml) 0.6 2 0.3 (80) 0.5 (1.5) 0.4 2 0.1 (16)aa 0.4 (0.5) 0.5 2 0.2 (23)b 0.5 (0.6) 0.7 2 0.4 (29) 0.5 (1.3) 0.9 2 0.5 (11) 0.7 (1.2) 0.5 2 0.3 (9) 0.4 (0.8)
The results are given as mean 2 .. and as median (range). The number of patients is expressed in parenthesis after the mean 2 .. value. *P Q 0.05, between the CM-groups and the control group. For statistical analysis two-tailed Mann–Whitney U-test was used. aa = P Q 0.01 between CM I and CM IV, a = P Q 0.05 between CM I and CM IV, b = P Q 0.05 between CM II and CM IV, c = P Q 0.01 between CM II and CM IV.
120
Va¨a¨ta¨inen et al.: Joint fluid and serum markers of chondromalacia Table IV Aggrecan fragments (AGN), keratan sulfate (KS), and hyaluronan (HA) in joint lavage fluid in chondromalacia of the patella (CM) and in different grades of CM (CM I–IV), according to Outerbridge 1961), and in controls Group CM CM I CM II CM III CM IV Controls
AGN (mg/ml)
KS (mg/ml)
HA (mg/ml)
25.8 2 15.3 (79) 20.2 (77.8) 33.3 2 19.2 (17) 36.3 (76.7) 25.2 2 14.6 (22) 21.5 (50.8) 23.2 2 13.8 (29) 17.5 (53.6) 22.5 2 11.2 (11) 19.0 (37.5) 26.6 2 16.5 (9) 16.8 (41.7)
6.3 2 4.6 (76) 5.3 (23.9) 7.4 2 5.6 (16)* 5.9 (22.7) 6.1 2 4.4 (22) 5.0 (15.7) 6.4 2 4.6 (28) 5.3 (15.8) 5.4 2 3.7 (10) 5.1 (10.8) 3.3 2 2.0 (6) 2.7 (5.7)
1.1 2 1.0 (78) 0.8 (4.5) 1.5 2 1.3 (16) 1.6 (4.4) 1.0 2 0.8 (23) 0.5 (2.9) 1.1 2 1.0 (28) 0.8 (3.8) 1.2 2 0.9 (11) 1.0 (2.9) 1.3 2 1.9 (9) 0.6 (6.0)
The number of patients is expressed in parenthesis and results are given as means 2 .. and as median (range). The number of patients is expressed in parenthesis after the mean 2 .. value. *P Q 0.05 between CM I and controls, two-tailed Mann–Whitney U-test.
concentrations of ks and ha in serum The mean KS concentration in grade IV CM was higher than that in the control group or in grade II CM (P Q 0.05; P = 0.02, Table V). A correlation between age and serum KS concentration was established (r = 0.34, P = 0.003, N = 70), and between serum KS and severity of CM (r = 0.22, P = 0.04, N = 70). Serum HA concentrations were equal in all subgroups in CM and no difference was found when compared with controls. A weak correlation Table V The levels of serum keratan sulfate (S-KS) and hyaluronan (S-HA) in all patients with chondromalacia of the patella (CM) and in subgroups of patients having different grades of CM (CM I–IV, according to Outerbridge 1961) and in controls Group CM CM I CM II CM II CM IV Controls
S-KS (ng/ml)
S-HA (ng/ml)
268 2 75 (70) 256 (381) 256 2 57 (15) 252 (215) 247 2 68 (22) 247 (335) 278 2 76 (24) 262 (379) 313 2 101 (9)* 323 (351) 237 2 58 (9) 245 (189)
45.3 2 26.5 (65) 40.0 (113.3) 44.9 2 24.3 (16) 46.4 (89.0) 42.0 2 25.7 (21) 35.8 (107.7) 48.6 2 29.7 (22) 41.7 (111.1) 45.5 2 28.3 (6) 57.5 (66.0) 56.2 2 42.2 (8) 60.1 (113.2)
*P Q 0.05 between CM IV and controls, two-tailed Mann–Whitney U-test. The results are given as mean 2 .. and as median (range). The number of patients is expressed in parenthesis after the mean 2 .. value.
between serum KS and HA was noted (r = 0.25, P = 0.03, N = 54). histology of synovium In CM patients, the synovium samples showed slight inflammation in four patients and moderate inflammation in three patients (one CM I; two CMII; two CMIII and two CMIV). In 76 CM patients, no signs of inflammation could be detected. No inflammatory changes were detected in any of the samples obtained from the control group. Discussion The lavage technique ensures that joint fluid samples can be obtained from all CM and control subjects even in the absence of a clinically evident effusion. Similar techniques have been used previously [42]. However, the lavage dilutes the endogenous synovial fluid and may increase the variation in measured concentrations. To some extent, this increase in variability can be avoided by measuring the ratios of different marker molecules to each other. It is clear, however, that care is needed when interpreting data on synovial fluid and serum markers [43]. We present evidence that concentrations of MMP-3 and MMP-1 are elevated in lavage joint fluid of patients with late stage CM. The concentration of MMP-3 was increased in the most advanced grade of CM, similar to the high levels shown in advanced osteoarthritis, consistent with
Osteoarthritis and Cartilage Vol. 6 No. 2
the possible involvement of MMP-3 in the matrix turnover in CM [5]. The involvement of proteolytic enzymes in the CM pathogenic process was also suggested by our earlier observations of progressive depletion of PGs in the articular cartilage of CM lesions and the fragmentation of those remaining in the matrix [44]. That study showed severe depletion (up to 90% loss) of PGs from patellar cartilage in the grade II lesion sites and almost total loss of PGs in grade IV lesion sites. MMP-3 is known to cleave the core protein between globular domains G1 and G2, and along the glycosaminoglycan-bearing region of the core protein [45–46], producing fragments which rapidly diffuse out of the tissue into synovial fluid. Although MMP-3 may thus play a role in the turnover of cartilage PGs in OA and even in normal cartilage [13, 47], other proteases, notably the putative aggrecanase enzyme, may also be important contributors to the degradation of aggrecan [48]. The assays for MMP used in this investigation recognize both the latent proforms of the MMPs, active MMP, and activated MMP subsequently bound to TIMP. In previous investigations, it was shown that the majority of MMP protein in synovial fluid in similar conditions represents proenzyme [28]. The presence of MMP protein in synovial fluid therefore does not necessarily correlate with MMP activity in synovial fluid or surrounding tissues. The proteolytic activity of MMPs in cartilage is inhibited by their binding to the tissue inhibitor of MMPs (TIMP-1) in a 1:1 ratio [49–50]. It has been suggested that IL-6 stimulates TIMP-1 production [18]. Our finding that the molar ratio of MMP-3 to its inhibitor, TIMP-1, remained virtually unchanged in CM, with TIMP-1 present in excess, suggests that there was no free, active MMP in synovial fluid. However, this does not preclude the presence in cartilage matrix of active MMP at some stage of the disease process. Similar to MMP-3, the level of MMP-1 in joint fluid was significantly higher in patients with CM IV than in the control group (Table III). The level of MMP-1 showed a tendency to increase with the severity of the disease (r = 0.38, P = 0.006, N = 66). In normal conditions, MMP-1 is present in synovial fluid in very small concentrations and we did not find any measurable level of MMP-1 in controls. It is widely believed that cytokines, such as the interleukins, act as inducers of the upregulation in the synthesis of proteolytic enzymes in inflammatory joint disease [14]. We found no significant differences in interleukin levels in the different study groups, perhaps due to the high individual
121
variation or levels of measured interleukins have been at higher level at some phase of the disease. Similarly, we did not find any significant differences in PLA2 levels in synovial fluid (Table III). PLA2 is an important mediator of inflammatory reactions [22] and elevated levels of PLA2 have been found in arthritis [21]. The inflammation may be due to a release of cartilage breakdown products, AGN and KS, as in OA [1, 51]. The joint fluid concentrations of AGN in patients with CM were not significantly different from the control population. Both of these parameters showed their highest concentrations in early phase of CM. These findings are consistent with our previous observation that 90% of the aggrecan molecules already have been lost from tissue showing grade II CM lesions [44]. It is worth noting that the concentration of PG fragments in OA joint fluid also shows a decrease with increasing cartilage damage in OA [5]. After joint injury, such as the experimental transection of the anterior cruciate ligament, the serum level of KS rises by 40%, on average, within 3 weeks after injury [52–53], but in synovial fluid KS levels slowly decrease with time [54]. In this cross-sectional study the patients with CM had had symptoms in their knee for at least 6 months before blood drawing. Therefore, one cannot rule out the possibility that serum KS could have been at a higher level at an earlier time point. Although serum HA concentrations have been shown to increase after traumatic joint injury [55], we did not find any such change in CM patients. The concentrations of HA in serum were equal in all groups (Table IV). This is not surprising since the clearance rate from serum is very fast, 0.3–1.0 g/min/kg body weight and 80–90% of tissue HA is removed by the lymph nodes before re-entering the bloodstream [56]. The concentration of HA in lavage fluid showed similar concentrations in all grades of CM (Table V). Cartilage contains considerable quantities of HA, and actively synthesizes it, but presumably most of the HA in joint fluid is derived from synovium. Remodelling of the tissue promotes HA synthesis and it is possible, that during increased cartilage matrix turnover more HA is synthesized and released from cartilage into synovial fluid [30, 57]. The HA concentration in healthy synovial joints is 2–3 mg/ml [58], but even low-grade inflammation can change its metabolism [59]. We found HA concentrations in lavage fluid of about 1.0 mg/ml, likely reflecting the dilution caused by our standardized sampling procedure. This degree of dilution is also consistent with the measured concentrations of, e.g., MMP, TIMP, and AGN in
122
Va¨a¨ta¨inen et al.: Joint fluid and serum markers of chondromalacia
lavage fluid in the present study, in relation to published data for these markers in undiluted knee synovial fluid from healthy subjects or with OA [5, 28, 29]. The increased concentrations of MMP-3 and MMP-1 in joint lavage fluid during the late stages of CM are consistent with an increase in expression and release of these proteinases into joint tissues and synovial fluid during advanced CM. The active role of these MMPs in the degradation of cartilage matrix in CM or OA, however, still awaits confirmation. An increase in the metabolism of matrix of cartilage and synovium appears to be a feature of early-stage CM, as well as in early phases of OA. The changes in metabolic markers in CM synovial fluid shown in this study, and the changes in aggrecan content and structure shown previously [44] are consistent with similar pathogenic mechanisms being active in both CM and OA.
10.
11.
12.
13.
14. 15.
References 1. Laurin CA, Dussault R, Levesque HP. The tangential X-ray investigation of the patellofemoral joint: X-Ray technique, diagnostic criteria and their interpretation. Clin Orthop 1979;144:16–26. 2. Nguyen Q, Murphy G, Roughley PJ, Mort JS. Proteoglycan aggregate degradation by a cartilage metalloproteinase. Evidence for the involvement of stromelysin in the generation of link protein heterogeneity in situ. Biochem J 1989;259:61–7. 3. Dean DD, Martel-Pelletier J, Pelletier J-P, Howell DS, Woessner JF Jr. Evidence for metalloproteinase and metalloproteinase inhibitor imbalance in human osteoarthritic cartilage. J Clin Invest 1989;84:678–85. 4. Pelletier J-P, Mineau F, Faure M-P, Martel-Pelletier J. Imbalance between the mechanisms of activation and inhibition of metalloproteinases in early lesions of experimental osteoarthritis. Arthritis Rheum 1990;33:1466–76. 5. Lohmander LS, Hoerrner LA, Lark MW. Metalloproteinases, tissue inhibitor, and proteoglycan fragments in knee synovial fluid in human osteoarthritis. Arthritis Rheum 1993;2:181–9. 6. McCachren SS, Haynes BF, Niedel JE. Localization of collagenase mRNA in rheumatoid arthritis synovium by in situ hybridization histochemistry. J Clin Immunol 1990;10:19–27. 7. Lefebvre V, Peeters-Joris C, Vaes G. Modulation by interleukin 1 and tumour necrosis factor of production of collagenase, tissue inhibitor of metalloproteinases and collagen types in differentiated and dedifferentiated articular chondrocytes. Biochim Biophys Acta 1990;1052:366–78. 8. MacNaul K, Chartrain N, Lark M, Tocci MJ, Hutchinson NI. Discoordinate expression of stromelysin, collagenase, and tissue inhibitor of metalloproteinases-1 in rheumatoid human synovial fibroblasts. J Biol Chem 1990;28:17238–45. 9. Wood DD, Ihrie EJ, Hamerman D. Release of
16.
17.
18.
19.
20. 21. 22.
23. 24.
25.
interleukin-1 from human synovial tissue in vitro. Arthritis Rheum 1985;28:853–62. Shinmei M, Masuda K, Kikuchi T, Shimomura Y. The role of cytokines in chondrocyte mediated cartilage degradation. J Rheumatol Suppl 1989;18:32–4. Dayer JM, de Rochemonteix B, Burrus B, Demczuk S, Dinarello CA. Human recombinant interleukin 1 stimulates collagenase and prostaglandin E2 production by human synovial cells. J Clin Invest 1986;77:645–8. Martel-Pelletier J, Zafarullah M, Kodama S, Pelletier JP. In vitro effects of interleukin 1 on synthesis of metalloproteases, TIMP-1, plasminogen activators and inhibitors in human articular cartilage. J Rheum Suppl 1991;27:80–4. Okada Y, Shinmei M, Tanaka O, Naka K, Kimura A, Nakanishi I, et al. Localization of matrix metalloproteinase 3 (stromelysin) in osteoarthritic cartilage and synovium. Lab Invest 1992;6:680–90. Sipe DJ, Martel-Pelletier J, Otterness IG, Pelletier J-P. Cytokine reduction in the treatment of joint conditions. Mediators Inflammation 1994;3:243–56. Nietfeld JJ, Wilbrink B, Helle M, van-Roy JL, den-Otter W, Swaak AJ et al. Interleukin-1 induced interleukin-6 is required for the inhibition of proteoglycan synthesis by interleukin-1 in human articular cartilage. Arthritis Rheum 1990;11:1695–701. Kandel RA, Petelycky M, Dinarello CA, Minden M, Pritzker KPH, Cruz TF. Comparison of the effect of interleukin 6 and interleukin 1 on collagenase and proteoglycan production by chondrocytes. J Rheumatol 1990;17:953–7. Ito A, Itoh Y, Sasaguri Y, Morimatsu M, Mori Y. Effects of interleukin-6 on the metabolism of connective tissue components in rheumatoid synovial fibroblasts. Arthritis Rheum 1992;10:1197– 201. Lotz M, Guerne PA. Interleukin-6 induces the synthesis of tissue inhibitor of metalloproteinases1/erythroid potentiating activity (TIMP-1/EPA). J Biol Chem 1991;266:2017–20. Lyons GB, Pratta MA, Galbraith W, Davis GL, Arner EC. Interleukin-1 differentially modulates chondrocyte expression of cyclooxygenase-2 and phospholipase A2. Exp Cell Res 1993;206:58–62. Vadas P, Stefanski E, Pruzanski W. Characterization of extracellular phospholipase A2 in rheumatoid synovial fluid. Life Sci 1985;36:579–87. Pruzanski W, Vadas P. Phospholipase A2—a mediator between proximal and distal effectors on inflammation. Immunol Today 1991;12:143–6. Nevalainen TJ, Marki F, Kortesuo PT, Grutter MG, Di-Marco S, Schmitz A. Synovial type (group II) phospholipase A2 in cartilage. J Rheumatol 1993;2:325–30. Kortekangas P, Aro HT, Nevalainen TJ. Group II phospholipase A2 in synovial fluid and serum in acute arthritis. Scand J Rheumatol 1994;2:68–72. Ogston AG, Stainer JE. The physiological function of hyaluronic acid in synovial fluid: viscous, elastic and lubricant properties. J Physiol 1953;119:244–52. Goa KL, Benfield P. Hyaluronic acid. A review of its pharmacology and use as a surgical aid in ophthalmology, and its therapeutic potential in
Osteoarthritis and Cartilage Vol. 6 No. 2
26.
27.
28.
29.
30.
31. 32. 33. 34. 35.
36.
37.
38.
39.
40.
41.
joint disease and wound healing. Drugs 1994;47:536–66. Haubeck H-D, Kock R, Fischer D-C, Van De Leur E, Hoffmeister K, Greiling H. Transforming growth factor b. 1, a major stimulator of hyaluronan synthesis in human synovial lining cells. Arthritis Rheum 1995;38:669–77. Castor CW, Prince RK, Hazelton MJ. Hyaluronic acid in human synovial fluid effusions: a sensitive indicator of altered connective tissue cell function during inflammation. Arthritis Rheum 1966;9:783– 94. Walakovits LA, Moore VL, Bhardwaj N, Gallick GS, Lark MW. Detection of stromelysin and collagenase in synovial fluid from patients with rheumatoid arthritis and posttraumatic knee injury. Arthritis Rheum 1992;35:35–42. Lohmander LS, Hoerrner LA, Dahlberg L, Roos H, Bjo¨rnsson S, Lark MW. Stromelysin, tissue inhibitor of metalloproteinases and proteoglycan fragments in human knee joint fluid after injury. J Rheumatol 1993;20:1362–8. Levick JP. Determinants of volume turnover and material concentration. In Kuettner KE, Schleyerbach R, Peyron JG, Hascall VC, Eds. Articular cartilage and osteoarthritis. New York, Raven Press 1992:529–41. Thonar EJ, Glant T. Serum keratan sulfate—a marker of predisposition to polyarticular osteoarthritis. Clin Biochem 1992;3:175–80. Ahlba¨ck S. Osteoarthrosis of the knee: a radiographic investigation. Acta Radiol Suppl 1968;277:7–72. Outerbridge RE. The etiology of chondromalacia patellae. J Bone Joint Surg (Br) 1961;4:752–7. Ro¨nkko¨ S, Lahtinen R, Vanha-Perttula T. Phospholipase A2 in the reproductive system of the bull. Int J Biochem 1991;23:595–603. Saxne T, Heinega˚rd D, Wollheim FA. Therapeutic effects on cartilage metabolism in arthritis as measured by release of proteoglycan structures into the synovial fluid. Ann Rheum Dis 1986;45:491–7. Saxne T, Heinega˚rd D. Synovial fluid analysis of two groups of proteoglycan epitopes distinguishes early and late cartilage lesions. Arthritis Rheum 1992;35:385–90. Thonar EJ-MA, Lenz ME, Klintworth GK et al. Quantification of keratan sulfate in blood as a marker of cartilage catabolism. Arthritis Rheum 1985;28:1367–76. Thonar EJ-MA, Schnitzer TJ, Kuettner KE. Quantification of keratan sulfate in blood as a marker of cartilage catabolism. J Rheumatol Suppl 1987;14:23–4. Wang C, Tammi M, Tammi R. Distribution of HA and its CD44 receptor in the epithelia of human skin appendages. Histochemistry 1992;98:105– 12. Kongtawelert P, Ghosh P. A method for the quantitation of hyaluronan (hyaluronic acid) in biological fluids using a labelled avidin-biotin technique. Anal Biochem 1990;185:313–8. Cooksley S, Hipkiss JB, Tickle SP, Holmes-Ievers E, Docherty AJP, Murphy G et al. Immunoassays for the detection of human collagenase; stromelysin; tissue inhibitor of metalloproteinases (TIMP-1)
42.
43. 44.
45. 46.
47. 48.
49. 50.
51.
52.
53.
54.
55.
56.
123
and enzyme-inhibitor complexes. Matrix 1990;10:285–91. Lohmander LS, Dahlberg L, Ryd L, Heinega˚rd D. Increased levels of proteoglycan fragments in knee joint fluid after injury. Arthritis Rheum 1989;32:1434–42. Simkin PA, Bassett JE. Cartilage matrix molecules in serum and synovial fluid. Curr Opin Rheumatol 1995;7:346–51. Va¨a¨ta¨inen U, Ha¨kkinen T, Kiviranta I, Jaroma H, Inkinen R, Tammi M. Proteoglycan depletion and size reduction in the lesions of early grade chondromalacia of the patella. Ann Rheum Dis 1995;54:831–5. Hughes C, Murphy G, Hardingham TE. Metalloproteinase digestion of cartilage proteoglycan. Biochem J 1991;279:733–9. Fosang AJ, Neame PJ, Hardingham TE, Murphy G, Hamilton JA. Cleavage of cartilage proteoglycan between G1 and G2 domains by stromelysins. J Biol Chem 1991;24:15579–82. Woessner JF Jr, Selzer MG. Two latent metalloproteases of human articular cartilage that digest proteoglycan. J Biol Chem 1984;259:3633–8. Lohmander LS, Neame P, Sandy JD. The structure of aggrecan fragments in human synovial fluid: Evidence that aggrecanase mediates cartilage degradation in inflammatory joint disease, joint injury and osteoarthritis. Arthritis Rheum 1993;36:1214–22. Coulombe B, Ponton A, Kerbel RS, Skup D. Structure and function of murine TIMP gene. Matrix Suppl 1992;1:269–74. Mort JS, Dodge GR, Roughley P, Liu J, Finch SJ, Dipasquale G et al. Direct evidence for active metalloproteinases mediating matrix degradation in interleukin 1-stimulated human articular cartilage. Matrix 1993;13:95–102. Vignon E, Mathieu P, Louisot P, Vilamitjana J, Harmand MF, Richard M. Phospholipase A2 activity in human osteoarthritic cartilage. J Rheumatol Suppl 1989;18:35–8. Manicourt DH, Lenz ME, Thonar EJ. Levels of serum keratan sulfate rise rapidly and remain elevated following anterior cruciate ligament transection in the dog. J Rheumatol 1991;12:1872– 6. Ratcliffe A, Billingham ME, Saed-Nejad F, Muir H, Hardingham TE. Increased release of matrix components from articular cartilage in experimental canine osteoarthritis. J Orthop Res 1992;10:350–8. Ratcliffe A, Beauvais, Saed-Nejad F, Shurety W, Caterson. Synovial fluid analyses detect and differentiate proteoglycan metabolism in canine experimental models of osteoarthritis and disuse atrophy. In van den Berg WB, van der Kraan PM, van Lent PLEM, Eds. Joint Destruction in Arthritis and Osteoarthritis, Basel, Birkha¨user Verlag 1993:63–7. Thonar EJ, Masuda K, Lenz ME, Hauselmann HJ, Kuettner KE, Manicourt DH. Serum markers of systemic disease processes in osteoarthritis. J Rheumatol Suppl 1995;43:68–70. Schiller S, Mathews MB, Cifonelli JA, Dorfman A. The metabolism of mucopolysaccharides in animals. III Further studies on skin utilizing
124
Va¨a¨ta¨inen et al.: Joint fluid and serum markers of chondromalacia
C14-glucose, C14-acetate and S35-sodium sulfate. J Biol Chem 1956;218:139–45. 57. Fraser JRE, Laurent TC. Turnover and metabolism of hyaluronan. In Evered D, Whelan J, Eds. The biology of hyaluronan, CIBA Foundation Symposium 143 New York, John Wiley & Sons 1989:41–53.
58. Laurent UBG, Reed R. Turnover of hyaluronan in the tissues. Adv Drug Del Rev 1991;7:237–56. 59. Fraser JR, Kimpton WG, Pierscionek BK, Cahill RN. The kinetics of hyaluronan in normal and acutely inflamed synovial joints: observations with experimental arthritis in sheep. Semin Arthritis Rheum 6 Suppl 1993;1:9–17.
1063–4584/98/020115 + 10 $12.00/0
Markers of cartilage and synovial metabolism in joint fluid and serum of patients with chondromalacia of the patella By Urho Va¨ a¨ ta¨ inen*, L. Stefan Lohmander†, Eugene Thonar‡, Tero Hongisto§, Ulla Ågren¶, Seppo Ro¨ nkko¨ >, Heikki Jaroma*, Veli-Matti Kosma**, Markku Tammi¶, and Ilkka Kiviranta* *Department of Surgery, Kuopio University Hospital, Kuopio, Finland; †Department of Orthopedics, University Hospital, Lund, Sweden; ‡Department of Biochemistry, Rush-Presbyterian-St. Luke’s Medical Center, Chicago, U.S.A.; §Department of Clinical Chemistry, Kuopio University Hospital; ¶Department of Anatomy, University of Kuopio; >Department of Pathology, Kuopio University Hospital, Kuopio, Finland Summary Objective: To further our understanding of the pathogenesis of chondromalacia of the patella (CM), we have studied the release into knee joint fluid and serum, obtained from patients with CM, of molecules associated with the metabolism of joint cartilage matrix and synovium. Methods: Interleukin-1 a (IL-1a), interleukin-1 b (IL-1b), interleukin-6 (IL-6), stromelysin-1 (MMP-3), interstitial collagenase (MMP-1), tissue inhibitor for metalloproteinases-1 (TIMP-1), phospholipase activity A2 (PLA2), hyaluronan (HA), aggrecan fragments (AGN) and antigenic keratan sulfate (KS) were quantified in knee joint lavage fluid from 96 patients with CM; KS and HA also was measured in serum. Chondromalacia was graded on a scale of I to IV according to Outerbridge (1961). The histopathology of the synovial membrane close to the patellofemoral joint was evaluated. Control samples were obtained from nine patients with knee pain presenting with arthroscopically normal knee joints. Results: The concentrations of MMP-3, MMP-1 and TIMP-1 proteins in joint lavage fluid were increased in advanced (grade IV) CM, compared with controls. Levels of MMP-1 in lavage fluid correlated with the severity of CM (r = 0.38, P Q 0.01) and MMP-1 and MMP-3 concentrations correlated with each other (r = 0.45, P Q 0.001). TIMP-1 was elevated in grade IV CM compared with grades II and III CM (P Q 0.02, P Q 0.01). Interleukins (IL-1a, IL-1b and IL-6) showed no significant change in CM. The lavage fluid level of PLA2 increased with the severity of CM (r = 0.40, P Q 0.001). Serum KS was higher in CM IV than in controls (P = 0.05), while lavage fluid KS concentration was elevated in CM I (P = 0.04). There were no differences in the lavage fluid levels of AGN and HA between the different study groups. Synovium showed slight or moderate histological signs of inflammation in 9% of CM patients. Conclusion: The changes in the release and activity of these marker molecules from serum and synovial fluid may reflect changes in the metabolism of articular cartilage and synovium in CM, that are consistent with those observed in early-stage tibiofemoral cartilage changes in OA. Key words: Chondromalacia, Joint fluid, Marker, Osteoarthritis, Knee.
arthritis (OA), but there is no general agreement that CM lesions consistently give rise to OA. In healthy articular cartilage, proteoglycan aggregates entrapped in collagen fibrils in the extracellular matrix enable the tissue to cushion forces applied at the end of long bones. In CM patellar cartilage becomes soft and fibrillated, ultimately degenerating and eroding. The release of cartilage breakdown products may cause synovitis in the knee joint [1], resulting in swelling and knee pain. Metalloproteinases (MMPs) have been suggested to play a role in cartilage matrix degradation [2–3]. These enzymes are able to degrade proteoglycans (PGs) and collagens. High
Introduction Chondromalacia of the patella (CM) is a common condition, but the mechanisms leading to cartilage degeneration and destruction of this joint are poorly understood. In late stages of CM, the cartilage changes resemble that seen in osteoReceived 10 March 1997; accepted 28 August 1997. Correspondence to: Urho Va¨a¨ta¨inen M.D., Department of Surgery, Kuopio University Hospital, Box 1777, FIN-70211 Kuopio, Finland. This study was supported by grants from the Foundation of Finnish Orthopedics and Traumatology, Finnish Medical Foundation, Orion Corporation Research Foundation, The Swedish Medical Research Council, the Medical Faculty and Hospital of Lund University and Merck Research Laboratories. Dr. Michael Lark, Merck Research Laboratories, Rahway, NJ, generously supplied the antibodies for MMP and TIMP assays.
115
116
Va¨a¨ta¨inen et al.: Joint fluid and serum markers of chondromalacia
levels of stromelysin (MMP-3) and collagenase (MMP-1) protein have been detected in the joint fluid of patients with rheumatoid arthritis (RA) and OA [4–5]. MMPs, as well their tissue inhibitors (TIMPs) are synthesized by chondrocytes and synovial cells [6–8]. These cells also produce cytokines, such as interleukin-1 alpha (IL-1a) and interleukin-1 beta (IL-1b) [9–10]. In inflamed joints, IL-1 stimulates the production of MMPs, and also suppresses the synthesis of proteoglycans and TIMP [11–14]. The effect of IL-1 on proteoglycan synthesis may be mediated by IL-6, a cytokine the synthesis of which is upregulated by IL-1 [15]. IL-6 does not appear to influence the production of MMPs, but does stimulate the production of TIMP [16–18]. IL-1 and tumor necrosis factor [TNF] regulate the production of phospholipase A2 (PLA2) [19], an enzyme generating arachidonic acid. In turn, arachidonic acid is converted into prostaglandins, which are involved in inflammatory reactions [20–21]. High levels of PLA2 have been found in joint fluid of RA and OA patients [22-23]. Hyaluronan (HA) is synthesized mainly by the synoviocytes and has an important role in providing lubrication and viscoelasticity in joint fluid [24, 25]. HA synthesis is stimulated by transforming growth factor 1 (TGF-b1), and interleukin-1 [26]. In inflammatory diseases such as RA, the concentration of HA in synovial fluid is decreased [27]. Products of articular cartilage and synovial tissue metabolism diffuse into synovial fluid and have been proposed to reflect joint tissue metabolism [5]. The concentrations of antigenic keratan sulfate (KS), aggrecan fragments (AGN) and certain aggrecan core protein epitopes in joint fluid are elevated in patients with OA and RA [5, 28], as well as after joint trauma [29]. Previous studies have shown that the concentrations of MMPs and TIMPs in joint fluid are likewise elevated in knee OA, after knee injury, [5, 29] and in RA [28]. Some of these markers of specific processes in joint disease also reach the blood circulation, either directly or via the lymphatic system [30], where they can be measured. Antigenic KS, for example, has been proposed to provide a measure of aggrecan metabolism in early OA [31]. In this study, we determined the concentrations in CM joint lavage fluid of the cytokines IL-1a, IL-1b, and IL-6, PLA2, MMPs (MMP-1 and MMP-3) and their inhibitor (TIMP-1), markers of PG degradation (KS and AGN), and HA. We propose that changes in the release of these molecules from joint tissues reflect changes in the metabolism of
joint cartilage and synovium in CM, consistent with those observed in early-stage tibiofemoral cartilage changes in OA.
Patients and Methods patients One hundred and thirty patients with anterior knee pain and strongly suspected CM, but showing no evidence of OA in their knee joint X-rays determined after Ahlba¨ck [32], were selected for study. After arthroscopic examination of the knee joint, the patients showing cartilage degeneration in other joint surfaces than patella (N = 5), or having meniscus tears (N = 20), in addition to CM were excluded from the study group. Nine patients did not have CM. After these exclusions 96 patients remained in the CM group. The CM patients were divided into subgroups according to Outerbridge [33] (Table I): grade I designates softened cartilage, grade II cartilage fibrillation less than 1/2 inch in diameter, grade III cartilage fibrillation greater than 1/2 inch in diameter and grade IV erosion of articular cartilage with exposed bone. The control group was formed from nine knee pain patients exhibiting no signs of CM or OA in any of the joint surfaces at arthroscopy (Table I). Patients with acute knee injuries (Q3 months) were excluded from the control group. Blood samples (20 ml) from CM and control patients were obtained from the cubital vein; centrifuged for 5 min at 500 g at 4°C , and the supernatants frozen and stored at −40°C. In the operating room, before arthroscopy, 20 ml sterile 0.9% saline solution was injected into the knee joint cavity, which was not previously emptied from endogenous joint fluid. To enhance mixing of the synovial fluid with the injected saline the knee joint was moved 10 times from extension to flexion and the suprapatellar region was manipulated before aspiration of the lavage fluid. The total volume of the aspirated fluid was recorded. In all cases, the lavage fluids were obtained using this standardized method between 9 a.m. and 1 p.m. Aliquots of the fluid were centrifuged for 5 min at 500 g, and the supernatants were frozen and stored at −40°C. At the beginning of arthroscopy, a sample of synovium was taken from the patellofemoral region with biopsy forceps from CM patients and from controls.
Osteoarthritis and Cartilage Vol. 6 No. 2
117
determination of il-1a, il-b and il-6
quantification of hyaluronan
Interleukin concentrations were determined by enzyme-linked immunosorbent assays (ELISAs). The IL-1a and IL-6 kits were purchased from Immunotech International (Immunotech S.A. Luminy Case 915, 13288 Marseille Cedex 9, France), while a Quantikine kit (R&D Systems, Inc. Minneapolis, U.S.A.) was used for IL-1b. The ELISA of IL-1a, was a two step assay while those for IL-1b and IL-6 were sandwich ELISAs. All measurements were carried out in duplicate according to the manufacturers’ instructions. Precision within an assay was in IL-1a 5.3%, in IL-1b 4.6% and in IL-6 4.4%. Inter-assay precision was respectively 8.7, 6.5 and 3.6%.
HA was quantified by an ELISA-like method using biotinylated HA-binding complex (bHABC) prepared from PG-link protein complex of bovine articular cartilage [39]. The assay method used is slightly modified from that described by Kongtawelert and Ghosh (1990) [40]. Briefly, Nunc Covalink plates (Covalink, Nunc, Roskilde, Denmark) were coated with HA (Sigma, St. Louis, MO, U.S.A.) and blocked with 1% bovine serum albumin (BSA). The samples were first digested for 24 h at 60°C by papain (250 mg/ml in 5 mm cysteine, 5 mm EDTA, Sigma); the enzyme then was inactivated by boiling for 10 min. The digested samples and a series of standard HA (5–200 ng/ml), diluted in 6% BSA in phosphate-buffered saline (PBS) were mixed with an equal volume of bHABC (450 ng/ml) in small polypropylene tubes (Mekalasi Oy, Kuortane, Finland) and incubated overnight at 4°C. Immediately prior to the assay, the mixtures were incubated for 1 h at 37°C. The samples were then transferred to the HA-coated plates, incubated for 90 min at 37°C, and washed with 0.5% Tween 20 in PBS. Horseradish peroxidase conjugated streptavidin (1:20 000 in PBS, Vector Lab, Burlingame, CA, U.S.A.) was added to the plates and left for 1 h at 37°C. After washing, the substrate-chromogen solution (0-phenylenediamine dihydrochloride; Sigma, 0.03% H2O2 in 0.1 m citrate buffer, pH 5.0) was added and left for 1 h. The reaction was stopped with 8 m H2SO4 and the absorbencies were read at 490 nm using a microtiter plate reader (Molecular Devices Corp, Palo Alto, CA, U.S.A.). All results are reported
pla2 activity assay The PLA2 assay method described by Ro¨nkko¨ and coworkers was used [34]. All measurements were done in duplicate. The intra-assay coefficient of variations (CV) was 3.6% and inter-assay CV was 5.6%.
quantification of aggrecan fragments and ks AGN containing the CS domain was quantified from joint lavage fluid by an ELISA according to Saxne and co-workers [35–36]. KS was determined from serum and joint lavage fluid by an ELISA based on the use of the monoclonal antibody ET-4-A-4 [37–38]. All measurements were performed in duplicate. Precision within an assay was 1.8–3.0% and inter-assay precision was 2.3–4.1%.
Table I Description of the subjects and synovial fluid samples
Group CM CM I CM II CM III CM IV Controls
Age (years) mean 2 s.d. median (range) 34.3 2 9.9 32.0 (40) 29.4 2 7.8 28.7 (29) 30.7 2 9.7 30.0 (37) 37.2 2 8.6 37.0 (33) 34.2 2 11.0 30.0 (31) 27.3 2 8.3 27.0 (22)
Number of patients
Female/male
96
59/37
20
13/7
26
14/12
34
19/15
16
13/3
9
4/5
Volume of the synovial fluid sample (ml)* mean 2 s.d. median (range) 9.8 2 5.5 10.0 (19) 8.6 2 6.0 7.5 (19) 9.4 2 5.8 9.0 (18) 11.4 2 5.2 10.5 (18) 12.0 2 5.5 13.0 (17) 8.7 2 5.5 9.0 (15)
*The amount of aspired synovial fluid from the knee joint after injection of 20 ml saline. CM = chondromalacia of the patella, CM I, CM II, CM III and CM IV refer to the grade of chondromalacia according to Outerbridge (1961).
118
Va¨a¨ta¨inen et al.: Joint fluid and serum markers of chondromalacia
here as the mean of the value obtained for analysis in duplicate. Intra-assay precision was 5.3% and inter-assay precision was 8.1%.
and two-tailed Mann–Whitney U-test and Spearman Rank Correlation test were used. Results
assay of mmp-3, mmp-1 and timp-1 MMP-1, MMP-3 and TIMP-1 concentrations in lavage fluid were measured by sandwich ELISAs, using monoclonal and polyclonal antibodies against the human recombinant proteins, as described previously [28, 41]. Polyclonal antisera against the specific proteins generated in rabbits were used as secondary reagents. The ELISAs for the MMPs detect the proforms of the enzymes, the large molecular active forms and the enzymes complexed to TIMP-1, but not small molecular forms of the enzymes or enzymes complexed to a2-macroglobulin. The assay for TIMP-1 detects only free TIMP-1, but not the inhibitor complexed with MMPs. The ELISAs have a log-linear relationship between 0.5–100 ng/ml (MMP-3), 2– 150 ng/ml (MMP-1), and 0.8–30 ng/ml (TIMP-1). Intra- and interassay coefficients of variation were less than 10% for these assays. The calculation of the molar ratio between MMP-3 and free TIMP was based on the assumption, that the major part (q90%) of MMP-3 is prostromelysin [28]. Similarly, the concentrations of MMP-3, MMP-1 and TIMP-1 are expressed as molarities (nm). histology of the synovium samples Samples of the synovial membrane were taken from 83 CM patients and from nine controls with biopsy forceps from the suprapatellar recess. The samples were fixed in 4% formaldehyde solution in phosphate buffer (pH 7.0, 0.013 m), embedded in paraffin, sectioned at 5 mm and stained with haematoxylin and eosin. The degree of inflammation was evaluated qualitatively in histological sections using a four step grading system. Grade 0 was assigned to samples showing no inflammatory cells (polymorphonuclear leukocytes, lymphocytes or macrophages). Grade 1 was assigned to slight, grade 2 to moderate and grade 3 to extensive number of inflammatory cells in the synovium samples. The samples were screened through and the greatest number of inflammatory cells was recorded. statistical analysis The results are expressed as mean 2 standard deviation and as median (range) values. For statistical analysis BMDP statistical software (BMDP Statistical Software, Los Angeles, U.S.A.)
All the groups were comparable with respect to age distribution (Table I). At the time of operation no macroscopic effusion was noted in the knee joints. The mean amount of aspirated lavage fluid after injection of 20 ml saline was 9.8 2 5.5 ml in CM patients and 8.7 2 5.5 ml in controls.
concentrations of il-1a, il-1b and il-6 in samples of lavage fluid The apparent increase in the level of interleukins (IL-1a, IL-1b and IL-6) in the CM group was not significant (Table II, IL-1a: P = 0.84, IL-1b: P = 0.61, IL-6: P = 0.27). The highest concentrations of IL-1a and IL-6 were in patients with grade II CM. The levels of IL-1b were similar in all CM groups (Table II). There was no correlation between the concentrations of IL-1a and IL-6 and none of these interleukins (IL-1a, IL-1b and IL-6) correlated to the severity of CM.
the concentrations of mmp-1, mmp-3, timp-1 protein, and pla2 activity in samples of lavage fluid The MMP-3, MMP-1 protein concentrations and PLA2 activity did not differ between the CM and control groups, when the whole groups were compared. However, the MMP-3 concentration in grade IV CM was significantly higher than that in the control group (P = 0.007, Table III) and in grade II CM (P = 0.03, Table III). TIMP-1 protein showed greater concentration in grade IV CM versus compared with grades II and III CM, respectively (P = 0.02, P = 0.01, Table III). The molar ratios of MMP-3 to TIMP-1 did not vary significantly among the subgroups (Table III). The concentration of MMP-1 was elevated in grade IV CM compared with the control group (P = 0.04, Table III) and with grades I and II CM (P = 0.03, Table III). The concentration of MMP-1 correlated with the grade of CM (r = 0.38, P = 0.006, N = 66). There was a positive correlation between MMP-3 and MMP-1 concentrations (r = 0.45, P Q 0.001, N = 65). The activity of PLA2 increased with the severity of the disease (r = 0.40, P Q 0.001, N = 79), but did not differ from the control group value when the whole CM group was compared with controls. PLA2 activity was increased in grade IV CM
Osteoarthritis and Cartilage Vol. 6 No. 2
119
Table II Synovial fluid concentrations of interleukin-1a (IL-1a, pg/ml), interleukin-1b (IL-1b, pg/ml) and interleukin-6 (IL-6, pg/ml) in all patients with chondromalacia of the patella (CM) and in subgroups of patients having different grades of CM (CM I–IV, judged according to Outerbridge 1961), and in controls Group CM CM I CM II CM III CM IV Controls
IL-1a (pg/ml)
IL-1b (pg/ml)
IL-6 (pg/ml)
1.9 2 3.7 (77) 0 (22) 1.8 2 3.2 (16) 0 (10) 3.0 2 5.3 (21) 0 (22) 1.5 2 3.0 (29) 0 (15) 1.2 2 1.8 (11) 0 (4) 1.0 2 1.3 (7) 0 (3)
0.7 2 1.8 (77) 0 (9.3) 1.1 2 2.5 (16) 0 (9.3) 0.4 2 1.1 (21) 0 (4.8) 1.0 2 2.1 (29) 0 (8.9) 0.1 2 0.3 (11) 0 (1.1) 0.6 2 1.5 (7) 0 (3.9)
3.9 2 15.2 (74) 0 (126) 4.2 2 8.2 (15) 0 (125) 7.4 2 26.8 (22) 0 (126) 1.5 2 3.2 (26) 0 (14) 2.2 2 3.5 (11) 9 (11) 0.6 2 1.5 (7) 1 (4)
The results are given as mean 2 .. and as median (range). The number of patients is expressed in parenthesis after the mean 2 .. value.
compared with grades I and II CM (P = 0.007, P = 0.02, Table III). concentrations of agn, ks and ha in samples of lavage fluid The concentrations of AGN and HA in the CM patients were not significantly different from those of the control group. The KS concentration in grade I CM was higher than in controls (P = 0.04, Table IV) and the highest concentrations of AGN
and KS were in grade I CM. AGN levels showed a tendency to decrease with increasing disease severity (r = −0.22, P = 0.04, N = 76, Table IV). Not surprisingly, KS levels in lavage fluid correlated closely with the level of AGN in CM patients (r = 0.69, P = 0.000, N = 76). HA concentrations in the joint lavage fluid of CM patients were at the same level in all grades of CM compared with controls (Table IV). A positive correlation was present between HA and AGN (r = 0.33, P = 0.005, N = 73).
Table III The synovial fluid concentrations of stromelysin (MMP-3, nm) tissue inhibitor of metalloproteinases (TIMP-1, nm), the molar ratio of MMP-3/TIMP-1 (MMP-3/TIMP-1), and synovial fluid concentrations of intersitial collagenase (MMP-1, nm) and the activity of phospholipase A2 (PLA2, nmol/min × ml) in all patients with chondromalacia of the patella (CM) and in subgroups of patients having different grades of chondromalacia (CM I–IV, according to Outerbridge 1961), and in controls Group CM CM I CM II CM III CM IV Controls
MMP-3 (nm)
TIMP-1 (nm)
MMP-3/TIMP-1
MMP-1 (nm)
2.3 2 3.2 (76) 1.4 (20.0) 2.6 2 2.9 (17) 1.8 (11.1) 2.1 2 3.3 (21)b 0.8 (14.5) 1.5 2 1.3 (27) 0.9 (5.5) 4.3 2 5.4 (11)* 2.0 (19.1) 1.6 2 1.8 (7) 0.8 (5.0)
2.0 2 2.2 (78) 1.2 (12.4) 1.8 2 1.4 (17) 1.1 (5.1) 2.2 2 2.8 (22)b 1.2 (10.5) 1.6 2 1.4 (28)c 1.0 (5.9) 3.2 2 3.1 (11) 2.3 (11.4) 1.4 2 0.9 (7) 1.0 (2.7)
1.4 2 1.5 (76) 0.9 (9.3) 1.2 2 1.1 (17) 0.9 (4.9) 1.1 2 1.3 (21) 0.9 (5.5) 1.7 2 2.0 (27) 1.0 (9.3) 1.4 2 1.0 (11) 0.9 (2.9) 0.8 2 0.4 (7) 0.8 (1.1)
0.06 2 0.2 (66) 0 (1.3) 0.01 2 0.04 (14)a 0 (0.2) 0.02 2 0.05 (18)b 0 (0.2) 0.04 2 0.08 (25) 0 (0.3) 0.3 2 0.4 (10)* 0.09 (1.3) 0 (9) 0
PLA2 (nmol/min × ml) 0.6 2 0.3 (80) 0.5 (1.5) 0.4 2 0.1 (16)aa 0.4 (0.5) 0.5 2 0.2 (23)b 0.5 (0.6) 0.7 2 0.4 (29) 0.5 (1.3) 0.9 2 0.5 (11) 0.7 (1.2) 0.5 2 0.3 (9) 0.4 (0.8)
The results are given as mean 2 .. and as median (range). The number of patients is expressed in parenthesis after the mean 2 .. value. *P Q 0.05, between the CM-groups and the control group. For statistical analysis two-tailed Mann–Whitney U-test was used. aa = P Q 0.01 between CM I and CM IV, a = P Q 0.05 between CM I and CM IV, b = P Q 0.05 between CM II and CM IV, c = P Q 0.01 between CM II and CM IV.
120
Va¨a¨ta¨inen et al.: Joint fluid and serum markers of chondromalacia Table IV Aggrecan fragments (AGN), keratan sulfate (KS), and hyaluronan (HA) in joint lavage fluid in chondromalacia of the patella (CM) and in different grades of CM (CM I–IV), according to Outerbridge 1961), and in controls Group CM CM I CM II CM III CM IV Controls
AGN (mg/ml)
KS (mg/ml)
HA (mg/ml)
25.8 2 15.3 (79) 20.2 (77.8) 33.3 2 19.2 (17) 36.3 (76.7) 25.2 2 14.6 (22) 21.5 (50.8) 23.2 2 13.8 (29) 17.5 (53.6) 22.5 2 11.2 (11) 19.0 (37.5) 26.6 2 16.5 (9) 16.8 (41.7)
6.3 2 4.6 (76) 5.3 (23.9) 7.4 2 5.6 (16)* 5.9 (22.7) 6.1 2 4.4 (22) 5.0 (15.7) 6.4 2 4.6 (28) 5.3 (15.8) 5.4 2 3.7 (10) 5.1 (10.8) 3.3 2 2.0 (6) 2.7 (5.7)
1.1 2 1.0 (78) 0.8 (4.5) 1.5 2 1.3 (16) 1.6 (4.4) 1.0 2 0.8 (23) 0.5 (2.9) 1.1 2 1.0 (28) 0.8 (3.8) 1.2 2 0.9 (11) 1.0 (2.9) 1.3 2 1.9 (9) 0.6 (6.0)
The number of patients is expressed in parenthesis and results are given as means 2 .. and as median (range). The number of patients is expressed in parenthesis after the mean 2 .. value. *P Q 0.05 between CM I and controls, two-tailed Mann–Whitney U-test.
concentrations of ks and ha in serum The mean KS concentration in grade IV CM was higher than that in the control group or in grade II CM (P Q 0.05; P = 0.02, Table V). A correlation between age and serum KS concentration was established (r = 0.34, P = 0.003, N = 70), and between serum KS and severity of CM (r = 0.22, P = 0.04, N = 70). Serum HA concentrations were equal in all subgroups in CM and no difference was found when compared with controls. A weak correlation Table V The levels of serum keratan sulfate (S-KS) and hyaluronan (S-HA) in all patients with chondromalacia of the patella (CM) and in subgroups of patients having different grades of CM (CM I–IV, according to Outerbridge 1961) and in controls Group CM CM I CM II CM II CM IV Controls
S-KS (ng/ml)
S-HA (ng/ml)
268 2 75 (70) 256 (381) 256 2 57 (15) 252 (215) 247 2 68 (22) 247 (335) 278 2 76 (24) 262 (379) 313 2 101 (9)* 323 (351) 237 2 58 (9) 245 (189)
45.3 2 26.5 (65) 40.0 (113.3) 44.9 2 24.3 (16) 46.4 (89.0) 42.0 2 25.7 (21) 35.8 (107.7) 48.6 2 29.7 (22) 41.7 (111.1) 45.5 2 28.3 (6) 57.5 (66.0) 56.2 2 42.2 (8) 60.1 (113.2)
*P Q 0.05 between CM IV and controls, two-tailed Mann–Whitney U-test. The results are given as mean 2 .. and as median (range). The number of patients is expressed in parenthesis after the mean 2 .. value.
between serum KS and HA was noted (r = 0.25, P = 0.03, N = 54). histology of synovium In CM patients, the synovium samples showed slight inflammation in four patients and moderate inflammation in three patients (one CM I; two CMII; two CMIII and two CMIV). In 76 CM patients, no signs of inflammation could be detected. No inflammatory changes were detected in any of the samples obtained from the control group. Discussion The lavage technique ensures that joint fluid samples can be obtained from all CM and control subjects even in the absence of a clinically evident effusion. Similar techniques have been used previously [42]. However, the lavage dilutes the endogenous synovial fluid and may increase the variation in measured concentrations. To some extent, this increase in variability can be avoided by measuring the ratios of different marker molecules to each other. It is clear, however, that care is needed when interpreting data on synovial fluid and serum markers [43]. We present evidence that concentrations of MMP-3 and MMP-1 are elevated in lavage joint fluid of patients with late stage CM. The concentration of MMP-3 was increased in the most advanced grade of CM, similar to the high levels shown in advanced osteoarthritis, consistent with
Osteoarthritis and Cartilage Vol. 6 No. 2
the possible involvement of MMP-3 in the matrix turnover in CM [5]. The involvement of proteolytic enzymes in the CM pathogenic process was also suggested by our earlier observations of progressive depletion of PGs in the articular cartilage of CM lesions and the fragmentation of those remaining in the matrix [44]. That study showed severe depletion (up to 90% loss) of PGs from patellar cartilage in the grade II lesion sites and almost total loss of PGs in grade IV lesion sites. MMP-3 is known to cleave the core protein between globular domains G1 and G2, and along the glycosaminoglycan-bearing region of the core protein [45–46], producing fragments which rapidly diffuse out of the tissue into synovial fluid. Although MMP-3 may thus play a role in the turnover of cartilage PGs in OA and even in normal cartilage [13, 47], other proteases, notably the putative aggrecanase enzyme, may also be important contributors to the degradation of aggrecan [48]. The assays for MMP used in this investigation recognize both the latent proforms of the MMPs, active MMP, and activated MMP subsequently bound to TIMP. In previous investigations, it was shown that the majority of MMP protein in synovial fluid in similar conditions represents proenzyme [28]. The presence of MMP protein in synovial fluid therefore does not necessarily correlate with MMP activity in synovial fluid or surrounding tissues. The proteolytic activity of MMPs in cartilage is inhibited by their binding to the tissue inhibitor of MMPs (TIMP-1) in a 1:1 ratio [49–50]. It has been suggested that IL-6 stimulates TIMP-1 production [18]. Our finding that the molar ratio of MMP-3 to its inhibitor, TIMP-1, remained virtually unchanged in CM, with TIMP-1 present in excess, suggests that there was no free, active MMP in synovial fluid. However, this does not preclude the presence in cartilage matrix of active MMP at some stage of the disease process. Similar to MMP-3, the level of MMP-1 in joint fluid was significantly higher in patients with CM IV than in the control group (Table III). The level of MMP-1 showed a tendency to increase with the severity of the disease (r = 0.38, P = 0.006, N = 66). In normal conditions, MMP-1 is present in synovial fluid in very small concentrations and we did not find any measurable level of MMP-1 in controls. It is widely believed that cytokines, such as the interleukins, act as inducers of the upregulation in the synthesis of proteolytic enzymes in inflammatory joint disease [14]. We found no significant differences in interleukin levels in the different study groups, perhaps due to the high individual
121
variation or levels of measured interleukins have been at higher level at some phase of the disease. Similarly, we did not find any significant differences in PLA2 levels in synovial fluid (Table III). PLA2 is an important mediator of inflammatory reactions [22] and elevated levels of PLA2 have been found in arthritis [21]. The inflammation may be due to a release of cartilage breakdown products, AGN and KS, as in OA [1, 51]. The joint fluid concentrations of AGN in patients with CM were not significantly different from the control population. Both of these parameters showed their highest concentrations in early phase of CM. These findings are consistent with our previous observation that 90% of the aggrecan molecules already have been lost from tissue showing grade II CM lesions [44]. It is worth noting that the concentration of PG fragments in OA joint fluid also shows a decrease with increasing cartilage damage in OA [5]. After joint injury, such as the experimental transection of the anterior cruciate ligament, the serum level of KS rises by 40%, on average, within 3 weeks after injury [52–53], but in synovial fluid KS levels slowly decrease with time [54]. In this cross-sectional study the patients with CM had had symptoms in their knee for at least 6 months before blood drawing. Therefore, one cannot rule out the possibility that serum KS could have been at a higher level at an earlier time point. Although serum HA concentrations have been shown to increase after traumatic joint injury [55], we did not find any such change in CM patients. The concentrations of HA in serum were equal in all groups (Table IV). This is not surprising since the clearance rate from serum is very fast, 0.3–1.0 g/min/kg body weight and 80–90% of tissue HA is removed by the lymph nodes before re-entering the bloodstream [56]. The concentration of HA in lavage fluid showed similar concentrations in all grades of CM (Table V). Cartilage contains considerable quantities of HA, and actively synthesizes it, but presumably most of the HA in joint fluid is derived from synovium. Remodelling of the tissue promotes HA synthesis and it is possible, that during increased cartilage matrix turnover more HA is synthesized and released from cartilage into synovial fluid [30, 57]. The HA concentration in healthy synovial joints is 2–3 mg/ml [58], but even low-grade inflammation can change its metabolism [59]. We found HA concentrations in lavage fluid of about 1.0 mg/ml, likely reflecting the dilution caused by our standardized sampling procedure. This degree of dilution is also consistent with the measured concentrations of, e.g., MMP, TIMP, and AGN in
122
Va¨a¨ta¨inen et al.: Joint fluid and serum markers of chondromalacia
lavage fluid in the present study, in relation to published data for these markers in undiluted knee synovial fluid from healthy subjects or with OA [5, 28, 29]. The increased concentrations of MMP-3 and MMP-1 in joint lavage fluid during the late stages of CM are consistent with an increase in expression and release of these proteinases into joint tissues and synovial fluid during advanced CM. The active role of these MMPs in the degradation of cartilage matrix in CM or OA, however, still awaits confirmation. An increase in the metabolism of matrix of cartilage and synovium appears to be a feature of early-stage CM, as well as in early phases of OA. The changes in metabolic markers in CM synovial fluid shown in this study, and the changes in aggrecan content and structure shown previously [44] are consistent with similar pathogenic mechanisms being active in both CM and OA.
10.
11.
12.
13.
14. 15.
References 1. Laurin CA, Dussault R, Levesque HP. The tangential X-ray investigation of the patellofemoral joint: X-Ray technique, diagnostic criteria and their interpretation. Clin Orthop 1979;144:16–26. 2. Nguyen Q, Murphy G, Roughley PJ, Mort JS. Proteoglycan aggregate degradation by a cartilage metalloproteinase. Evidence for the involvement of stromelysin in the generation of link protein heterogeneity in situ. Biochem J 1989;259:61–7. 3. Dean DD, Martel-Pelletier J, Pelletier J-P, Howell DS, Woessner JF Jr. Evidence for metalloproteinase and metalloproteinase inhibitor imbalance in human osteoarthritic cartilage. J Clin Invest 1989;84:678–85. 4. Pelletier J-P, Mineau F, Faure M-P, Martel-Pelletier J. Imbalance between the mechanisms of activation and inhibition of metalloproteinases in early lesions of experimental osteoarthritis. Arthritis Rheum 1990;33:1466–76. 5. Lohmander LS, Hoerrner LA, Lark MW. Metalloproteinases, tissue inhibitor, and proteoglycan fragments in knee synovial fluid in human osteoarthritis. Arthritis Rheum 1993;2:181–9. 6. McCachren SS, Haynes BF, Niedel JE. Localization of collagenase mRNA in rheumatoid arthritis synovium by in situ hybridization histochemistry. J Clin Immunol 1990;10:19–27. 7. Lefebvre V, Peeters-Joris C, Vaes G. Modulation by interleukin 1 and tumour necrosis factor of production of collagenase, tissue inhibitor of metalloproteinases and collagen types in differentiated and dedifferentiated articular chondrocytes. Biochim Biophys Acta 1990;1052:366–78. 8. MacNaul K, Chartrain N, Lark M, Tocci MJ, Hutchinson NI. Discoordinate expression of stromelysin, collagenase, and tissue inhibitor of metalloproteinases-1 in rheumatoid human synovial fibroblasts. J Biol Chem 1990;28:17238–45. 9. Wood DD, Ihrie EJ, Hamerman D. Release of
16.
17.
18.
19.
20. 21. 22.
23. 24.
25.
interleukin-1 from human synovial tissue in vitro. Arthritis Rheum 1985;28:853–62. Shinmei M, Masuda K, Kikuchi T, Shimomura Y. The role of cytokines in chondrocyte mediated cartilage degradation. J Rheumatol Suppl 1989;18:32–4. Dayer JM, de Rochemonteix B, Burrus B, Demczuk S, Dinarello CA. Human recombinant interleukin 1 stimulates collagenase and prostaglandin E2 production by human synovial cells. J Clin Invest 1986;77:645–8. Martel-Pelletier J, Zafarullah M, Kodama S, Pelletier JP. In vitro effects of interleukin 1 on synthesis of metalloproteases, TIMP-1, plasminogen activators and inhibitors in human articular cartilage. J Rheum Suppl 1991;27:80–4. Okada Y, Shinmei M, Tanaka O, Naka K, Kimura A, Nakanishi I, et al. Localization of matrix metalloproteinase 3 (stromelysin) in osteoarthritic cartilage and synovium. Lab Invest 1992;6:680–90. Sipe DJ, Martel-Pelletier J, Otterness IG, Pelletier J-P. Cytokine reduction in the treatment of joint conditions. Mediators Inflammation 1994;3:243–56. Nietfeld JJ, Wilbrink B, Helle M, van-Roy JL, den-Otter W, Swaak AJ et al. Interleukin-1 induced interleukin-6 is required for the inhibition of proteoglycan synthesis by interleukin-1 in human articular cartilage. Arthritis Rheum 1990;11:1695–701. Kandel RA, Petelycky M, Dinarello CA, Minden M, Pritzker KPH, Cruz TF. Comparison of the effect of interleukin 6 and interleukin 1 on collagenase and proteoglycan production by chondrocytes. J Rheumatol 1990;17:953–7. Ito A, Itoh Y, Sasaguri Y, Morimatsu M, Mori Y. Effects of interleukin-6 on the metabolism of connective tissue components in rheumatoid synovial fibroblasts. Arthritis Rheum 1992;10:1197– 201. Lotz M, Guerne PA. Interleukin-6 induces the synthesis of tissue inhibitor of metalloproteinases1/erythroid potentiating activity (TIMP-1/EPA). J Biol Chem 1991;266:2017–20. Lyons GB, Pratta MA, Galbraith W, Davis GL, Arner EC. Interleukin-1 differentially modulates chondrocyte expression of cyclooxygenase-2 and phospholipase A2. Exp Cell Res 1993;206:58–62. Vadas P, Stefanski E, Pruzanski W. Characterization of extracellular phospholipase A2 in rheumatoid synovial fluid. Life Sci 1985;36:579–87. Pruzanski W, Vadas P. Phospholipase A2—a mediator between proximal and distal effectors on inflammation. Immunol Today 1991;12:143–6. Nevalainen TJ, Marki F, Kortesuo PT, Grutter MG, Di-Marco S, Schmitz A. Synovial type (group II) phospholipase A2 in cartilage. J Rheumatol 1993;2:325–30. Kortekangas P, Aro HT, Nevalainen TJ. Group II phospholipase A2 in synovial fluid and serum in acute arthritis. Scand J Rheumatol 1994;2:68–72. Ogston AG, Stainer JE. The physiological function of hyaluronic acid in synovial fluid: viscous, elastic and lubricant properties. J Physiol 1953;119:244–52. Goa KL, Benfield P. Hyaluronic acid. A review of its pharmacology and use as a surgical aid in ophthalmology, and its therapeutic potential in
Osteoarthritis and Cartilage Vol. 6 No. 2
26.
27.
28.
29.
30.
31. 32. 33. 34. 35.
36.
37.
38.
39.
40.
41.
joint disease and wound healing. Drugs 1994;47:536–66. Haubeck H-D, Kock R, Fischer D-C, Van De Leur E, Hoffmeister K, Greiling H. Transforming growth factor b. 1, a major stimulator of hyaluronan synthesis in human synovial lining cells. Arthritis Rheum 1995;38:669–77. Castor CW, Prince RK, Hazelton MJ. Hyaluronic acid in human synovial fluid effusions: a sensitive indicator of altered connective tissue cell function during inflammation. Arthritis Rheum 1966;9:783– 94. Walakovits LA, Moore VL, Bhardwaj N, Gallick GS, Lark MW. Detection of stromelysin and collagenase in synovial fluid from patients with rheumatoid arthritis and posttraumatic knee injury. Arthritis Rheum 1992;35:35–42. Lohmander LS, Hoerrner LA, Dahlberg L, Roos H, Bjo¨rnsson S, Lark MW. Stromelysin, tissue inhibitor of metalloproteinases and proteoglycan fragments in human knee joint fluid after injury. J Rheumatol 1993;20:1362–8. Levick JP. Determinants of volume turnover and material concentration. In Kuettner KE, Schleyerbach R, Peyron JG, Hascall VC, Eds. Articular cartilage and osteoarthritis. New York, Raven Press 1992:529–41. Thonar EJ, Glant T. Serum keratan sulfate—a marker of predisposition to polyarticular osteoarthritis. Clin Biochem 1992;3:175–80. Ahlba¨ck S. Osteoarthrosis of the knee: a radiographic investigation. Acta Radiol Suppl 1968;277:7–72. Outerbridge RE. The etiology of chondromalacia patellae. J Bone Joint Surg (Br) 1961;4:752–7. Ro¨nkko¨ S, Lahtinen R, Vanha-Perttula T. Phospholipase A2 in the reproductive system of the bull. Int J Biochem 1991;23:595–603. Saxne T, Heinega˚rd D, Wollheim FA. Therapeutic effects on cartilage metabolism in arthritis as measured by release of proteoglycan structures into the synovial fluid. Ann Rheum Dis 1986;45:491–7. Saxne T, Heinega˚rd D. Synovial fluid analysis of two groups of proteoglycan epitopes distinguishes early and late cartilage lesions. Arthritis Rheum 1992;35:385–90. Thonar EJ-MA, Lenz ME, Klintworth GK et al. Quantification of keratan sulfate in blood as a marker of cartilage catabolism. Arthritis Rheum 1985;28:1367–76. Thonar EJ-MA, Schnitzer TJ, Kuettner KE. Quantification of keratan sulfate in blood as a marker of cartilage catabolism. J Rheumatol Suppl 1987;14:23–4. Wang C, Tammi M, Tammi R. Distribution of HA and its CD44 receptor in the epithelia of human skin appendages. Histochemistry 1992;98:105– 12. Kongtawelert P, Ghosh P. A method for the quantitation of hyaluronan (hyaluronic acid) in biological fluids using a labelled avidin-biotin technique. Anal Biochem 1990;185:313–8. Cooksley S, Hipkiss JB, Tickle SP, Holmes-Ievers E, Docherty AJP, Murphy G et al. Immunoassays for the detection of human collagenase; stromelysin; tissue inhibitor of metalloproteinases (TIMP-1)
42.
43. 44.
45. 46.
47. 48.
49. 50.
51.
52.
53.
54.
55.
56.
123
and enzyme-inhibitor complexes. Matrix 1990;10:285–91. Lohmander LS, Dahlberg L, Ryd L, Heinega˚rd D. Increased levels of proteoglycan fragments in knee joint fluid after injury. Arthritis Rheum 1989;32:1434–42. Simkin PA, Bassett JE. Cartilage matrix molecules in serum and synovial fluid. Curr Opin Rheumatol 1995;7:346–51. Va¨a¨ta¨inen U, Ha¨kkinen T, Kiviranta I, Jaroma H, Inkinen R, Tammi M. Proteoglycan depletion and size reduction in the lesions of early grade chondromalacia of the patella. Ann Rheum Dis 1995;54:831–5. Hughes C, Murphy G, Hardingham TE. Metalloproteinase digestion of cartilage proteoglycan. Biochem J 1991;279:733–9. Fosang AJ, Neame PJ, Hardingham TE, Murphy G, Hamilton JA. Cleavage of cartilage proteoglycan between G1 and G2 domains by stromelysins. J Biol Chem 1991;24:15579–82. Woessner JF Jr, Selzer MG. Two latent metalloproteases of human articular cartilage that digest proteoglycan. J Biol Chem 1984;259:3633–8. Lohmander LS, Neame P, Sandy JD. The structure of aggrecan fragments in human synovial fluid: Evidence that aggrecanase mediates cartilage degradation in inflammatory joint disease, joint injury and osteoarthritis. Arthritis Rheum 1993;36:1214–22. Coulombe B, Ponton A, Kerbel RS, Skup D. Structure and function of murine TIMP gene. Matrix Suppl 1992;1:269–74. Mort JS, Dodge GR, Roughley P, Liu J, Finch SJ, Dipasquale G et al. Direct evidence for active metalloproteinases mediating matrix degradation in interleukin 1-stimulated human articular cartilage. Matrix 1993;13:95–102. Vignon E, Mathieu P, Louisot P, Vilamitjana J, Harmand MF, Richard M. Phospholipase A2 activity in human osteoarthritic cartilage. J Rheumatol Suppl 1989;18:35–8. Manicourt DH, Lenz ME, Thonar EJ. Levels of serum keratan sulfate rise rapidly and remain elevated following anterior cruciate ligament transection in the dog. J Rheumatol 1991;12:1872– 6. Ratcliffe A, Billingham ME, Saed-Nejad F, Muir H, Hardingham TE. Increased release of matrix components from articular cartilage in experimental canine osteoarthritis. J Orthop Res 1992;10:350–8. Ratcliffe A, Beauvais, Saed-Nejad F, Shurety W, Caterson. Synovial fluid analyses detect and differentiate proteoglycan metabolism in canine experimental models of osteoarthritis and disuse atrophy. In van den Berg WB, van der Kraan PM, van Lent PLEM, Eds. Joint Destruction in Arthritis and Osteoarthritis, Basel, Birkha¨user Verlag 1993:63–7. Thonar EJ, Masuda K, Lenz ME, Hauselmann HJ, Kuettner KE, Manicourt DH. Serum markers of systemic disease processes in osteoarthritis. J Rheumatol Suppl 1995;43:68–70. Schiller S, Mathews MB, Cifonelli JA, Dorfman A. The metabolism of mucopolysaccharides in animals. III Further studies on skin utilizing
124
Va¨a¨ta¨inen et al.: Joint fluid and serum markers of chondromalacia
C14-glucose, C14-acetate and S35-sodium sulfate. J Biol Chem 1956;218:139–45. 57. Fraser JRE, Laurent TC. Turnover and metabolism of hyaluronan. In Evered D, Whelan J, Eds. The biology of hyaluronan, CIBA Foundation Symposium 143 New York, John Wiley & Sons 1989:41–53.
58. Laurent UBG, Reed R. Turnover of hyaluronan in the tissues. Adv Drug Del Rev 1991;7:237–56. 59. Fraser JR, Kimpton WG, Pierscionek BK, Cahill RN. The kinetics of hyaluronan in normal and acutely inflamed synovial joints: observations with experimental arthritis in sheep. Semin Arthritis Rheum 6 Suppl 1993;1:9–17.