1. Top of page
  2. Abstract
  7. Acknowledgements


To investigate aggrecan degradation in juvenile idiopathic arthritis (JIA).


The pattern and abundance of aggrecan fragments in synovial fluid (SF) aspirates from JIA patients were analyzed and compared with aggrecan fragments in SF from patients with other arthritides, children with knee injury, and a knee-healthy reference group. Concentrations of sulfated glycosaminoglycan (sGAG) in SF were measured by Alcian blue precipitation assay. Aggrecan fragments were purified by dissociative CsCl density-gradient centrifugation, deglycosylated, and analyzed by Western blot using antibodies specific for either aggrecanase-derived ARGS, SELE, and KEEE neoepitopes or the aggrecan G3 domain.


The concentration of sGAG in SF from patients with JIA was significantly lower compared with that in SF from patients with osteoarthritis (OA) (P < 0.001), patients with juvenile knee injury (P = 0.006), and knee-healthy controls (P = 0.022). Western blot analysis revealed KEEE, SELE, and G3 fragments generated by aggrecanase cleavage in the chondroitin sulfate–rich region of aggrecan in patients with JIA. The pattern of aggrecan fragments in JIA patients was not identical to that in pooled OA SF, although there were notable similarities. Surprisingly, aggrecanase-derived ARGS fragments were barely detectable in JIA SF, in marked contrast to levels in OA SF.


Aggrecanases appear to cleave minimally in the interglobular domain of aggrecan in JIA patients despite robust levels of cleavage in the chondroitin sulfate–rich region. These results suggest that in JIA, unlike other arthritides, aggrecanase cleavage in the aggrecan interglobular domain might not be a major pathogenic event.

Juvenile idiopathic arthritis (JIA) is the most commonly diagnosed rheumatic disease in children, with a prevalence of 1–4 per 1,000 children (1, 2). It is a chronic, heterogeneous disorder characterized by persistent synovial inflammation which, if unresolved, can lead to joint destruction and permanent disability. Prior to the advent of the new biologic therapies, the incidence of cartilage erosion, detected radiographically as joint space narrowing in JIA patients with long-term disease, was significant (3–6).

While the new biologic agents have improved the management of synovial inflammation in JIA, there has been little progress in strategies for preventing cartilage damage in JIA or other arthritides. Development of new treatments for limiting cartilage erosion in JIA is hampered by the lack of fundamental information on the mechanisms involved; to date, studies on aggrecanolysis have focused exclusively on adult cartilage, predominantly in osteoarthritis (OA). A recent study by Kim and colleagues suggests that the extent of cartilage damage in JIA might be underrecognized, since new imaging modalities have revealed microstructural changes indicative of cartilage damage in JIA patients by T2 mapping (7). These microstructural changes are thought to be markers of disease progression that develop early, despite clinical improvement and before changes in cartilage morphology can be detected with conventional magnetic resonance imaging. Thus, although there have been major improvements in JIA outcomes brought about by the introduction of biologic therapies, destructive cartilage erosion leading to irreparable joint abnormalities remains a risk for these patients.

The aggrecanases (ADAMTS) have been studied intensively since their discovery in 1999 (8, 9), and it is now clear that ADAMTS-4 and/or ADAMTS-5 are the principal aggrecan-degrading enzymes in OA and inflammatory arthritis in humans and experimental animals (10–16). Cleavage in the interglobular domain (IGD) of aggrecan is the signature activity of this enzyme family and is widely regarded as the defining activity of the aggrecanases, even though cleavage at this site is preceded by cleavage at the SELE1545 and KEEE1714 sites in the chondroitin sulfate–rich region (17–20) (Figure 1). (Numbering is from 1VETS of the mature protein [NCBI accession no. P16112].) The most abundant ADAMTS-derived aggrecan fragments found in OA synovial fluid (SF) are the 374ARGS-SELE1545 fragment (fragment b in Figure 1), its 1546GRGT-G3 counterpart (fragment f in Figure 1), 2 shorter G3-containing fragments (fragments g and h in Figure 1), and a KEEE1714 fragment with an intact N-terminal G1 domain (fragment d in Figure 1) (10, 15, 21). These fragments are readily detected in human OA SF by Western blotting with neoepitope or anti-G3 antibodies.

thumbnail image

Figure 1. Aggrecanase cleavage sites in human aggrecan detected with neoepitope antibodies. Aggrecanase cleavage sites in the human aggrecan core protein are shown, with numbering from the first amino acid (valine) of the mature protein. Circled letters (a–k) identify individual fragments corresponding to bands detected on Western blots (see Figures 3–5). The triangle marks the matrix metalloproteinase cleavage site at IPEN341[DOWNWARDS ARROW]342FFGV. The dotted line indicates regions of the core protein corresponding to the interglobular domain (IGD), keratan sulfate (KS)–rich region, and chondroitin sulfate (CS)–rich region. Boxed numbers 1 and 2 indicate the preferred order of aggrecanase cleavage.

Download figure to PowerPoint

Enzymes other than aggrecanases can also degrade aggrecan in vivo. Members of the matrix metalloproteinase (MMP) family cleave aggrecan in the IGD at IPEN341[DOWNWARDS ARROW]342FFGV, generating G1-IPEN341 and 342FFGV fragments in vivo and in vitro (15, 21, 22). The intracellular, nonlysosomal proteinase calpain (23) and the serine proteinase HtrA1 (24) also cleave aggrecan in vivo, albeit at low levels. It is estimated that in osteoarthritic cartilage ∼75% of aggrecan degradation is due to the action of ADAMTS aggrecanases and ∼25% due to MMP activity (23), with other enzyme families making only minimal contributions to total aggrecanolysis. To date there is no published information on the identity of the proteinases that degrade aggrecan in JIA or the pattern or abundance of aggrecan fragments in JIA SF. Such fragments could potentially be used as biomarkers of early cartilage erosion in children or as markers of the efficacy of cartilage-sparing treatments.

In the present study, we analyzed SF samples from a small group of JIA patients for the presence of aggrecan fragments produced by ADAMTS enzymes. We found that proteolysis by aggrecanases in the aggrecan IGD appears negligible in JIA patients compared with OA patients, despite robust levels of aggrecanase cleavage in the chondroitin sulfate–rich region. These results suggest that ADAMTS cleavage in the aggrecan IGD might not be a pathogenic event in JIA, and that cartilage-sparing drugs designed to block aggrecanase cleavage in the IGD might not be useful for arresting cartilage erosion in this disease.


  1. Top of page
  2. Abstract
  7. Acknowledgements


Chondroitinase ABC (EC, keratanase (EC, and keratanase II (Bacillus sp. Ks36) were from Seikagaku. Neospecific anti-SELE rabbit sera and anti-ARGS monoclonal antibody OA-1 were from GlaxoSmithKline. Neospecific anti-KEEE serum was from Dr. M. Lark (Merck). Monoclonal anti-FFGV antibody AF28 (25) was from Millipore. Polyclonal anti–aggrecan G3 domain antibody recognizing all G3 isoforms was from Affinity BioReagents. Peroxidase-conjugated secondary antibodies were from Cell Signaling Technology (goat anti-mouse IgG) and KPL (goat anti-rabbit IgG). Precision Plus Protein standards (10–250 kd), Criterion Blotter, and PVDF membrane were from Bio-Rad. NuPAGE 3–8% Tris-acetate sodium dodecyl sulfate (SDS) mini or midi gels were from Invitrogen. ECL Plus was from GE Healthcare. X-ray hyperfilm ECL was from Amersham.

Human SF samples and aggrecan SF-D1 preparations.

SF aspirates were collected without lavage from children who attended pediatric rheumatology clinics at the Royal Children's Hospital and were diagnosed as having JIA. SF samples were collected from patients who attended rheumatology clinics at the Royal Melbourne Hospital (Melbourne, Victoria, Australia) and were diagnosed as having rheumatoid arthritis (RA). As comparators, SF samples from patients with OA, children with knee injury, and individuals with healthy knees (from cohorts included in previous cross-sectional studies of knee SF [16, 26–28]) were analyzed. SF collection was approved by the ethics review committees of the Royal Children's Hospital, the Royal Melbourne Hospital, and the Medical Faculty of Lund University.

High-buoyant-density aggrecan fragments present in SF were fractionated by CsCl density-gradient centrifugation (29) in either the bottom half of 2-ml tubes (mini) or the bottom two-fifths of 12.5-ml tubes (normal) under dissociative conditions, yielding SF-D1 fractions (10, 15). The recovery of sulfated glycosaminoglycan (sGAG) was between 73% and 92% with mini SF-D1 preparations and between 82% and 86% with normal SF-D1 preparations, with no significant difference in yield between the methods.

SF-D1 samples from 12 individual JIA patients (23 samples) were prepared, and a JIA SF-D1 pool was prepared by pooling 0.5 ml of SF from 120 patients prior to CsCl density-gradient centrifugation (Table 1). In addition, the SF-D1 RA pool (n = 9), juvenile knee injury pool (n = 9), and knee-healthy reference pool (here, n = 10) were made by mixing equal volumes of SF from each group prior to CsCl density-gradient centrifugation. A previously described OA SF-D1 pool (n = 47) (14, 26) was also included.

Table 1. Characteristics of the patient groups and the knee-healthy reference group*
Study group or patientNo. of patients/samplesSexAge, yearsDisease duration at aspiration, yearsNo. of joints aspiratedDiagnosisMedication
  • *

    NA = not applicable; OA = osteoarthritis; ND = not determined; NSAIDs = nonsteroidal antiinflammatory drugs; RA = rheumatoid arthritis; DMARDs = disease-modifying antirheumatic drugs; sGAG = sulfated glycosaminoglycan; RF = rheumatoid factor.

  • Values for the groups are the percent female.

  • Values for the groups are the mean (range).

  • §

    Mean (range).

  • The 12 juvenile idiopathic arthritis (JIA) patients analyzed in detail are identified as patients A–L, followed by the patient's age (in years) at the time of synovial fluid aspiration from the knee. Patients A, E, F, G, H, and K underwent knee aspiration on 2 or 3 occasions over time. Left and right knees were aspirated from patient B, and from patient F at both time points.

  • #

    Rice bodies were removed prior to analysis.

 Reference12/12835 (19–58)NA1Knee-healthyNA
 OA47/472848 (16–89)ND1Knee OANSAIDs, analgesics
 Juvenile injury9/94414 (12–15)0.3 (0–2.4)§1Juvenile knee injuryAnalgesics
 RA9/94453 (17–75)ND1RANA
 JIA pool120/1207410.5 (0.3–19)NDNAJIAMixed NSAIDs/DMARDs/steroids
 JIA sGAG38/1007110.2 (2.3–17)NDNAJIAMixed NSAIDs/DMARDs/steroids
Individual JIA patient       
 A7F72.41Systemic JIANSAID + DMARD + steroid
 A10F105.31Systemic JIANSAID + DMARD + steroid
 B9F95.72Extended oligoarticular JIADMARD only
 C3F30.51#Extended oligoarticular JIANSAID only
 D15M153.31Oligoarticular JIANone
 E11F113.81Polyarticular, RF− JIANSAID + DMARD
 E12F124.51Polyarticular, RF− JIANSAID + DMARD
 E15F158.51Polyarticular, RF− JIANSAID + DMARD + steroid
 F11F110.12Systemic JIANSAID + DMARD + steroid
 F12F121.22Systemic JIANSAID + DMARD + steroid
 G4F42.31Extended oligoarticular JIANSAID only
 G14F1411.71Extended oligoarticular JIANone
 H6F60.41Polyarticular, RF− JIANSAID only
 H11F115.61Polyarticular, RF− JIANSAID + DMARD
 I16M167.81Oligoarticular JIANone
 J13F136.91Psoriatic arthritisDMARD + steroid
 K8F83.61Psoriatic arthritisDMARD only
 K10F105.41Psoriatic arthritisNSAID + DMARD
 K12F127.61Psoriatic arthritisDMARD only
 L8F81.21Oligoarticular JIANSAID + DMARD

Western blotting.

Deglycosylated SF-D1 samples were electrophoresed on 3–8% SDS gels and transferred to PVDF membranes for immunoblotting using rabbit polyclonal anti-G3 antibody (IgG; 2–5 μg/ml), mouse monoclonal anti-ARGS antibody (IgG; 5.3–7 μg/ml), rabbit polyclonal anti-KEEE antibody (sera; 1:1,000), rabbit polyclonal anti-SELE antibody (sera; 1:2,000), and mouse monoclonal anti-FFGV antibody (IgG; 0.5 μg/ml), followed by peroxidase-conjugated secondary antibodies (horse anti-mouse IgG [6.7–16.7 ng/ml] or goat anti-rabbit IgG [10–20 ng/ml]). Bands were visualized using ECL Plus with exposure to film or a luminescence image analyzer (Fuji Film LAS-1000). The specificity of each antibody used in this study has been confirmed previously (15, 30, 31).

Statistical analysis.

Kruskal-Wallis one-way analysis of variance on ranks was used to avoid mass significance due to multiple group comparisons. If significance was found, the Mann-Whitney rank sum test for analysis of unmatched pairs was done. For correlation analysis, Spearman's rank order correlation was used.


  1. Top of page
  2. Abstract
  7. Acknowledgements


As part of routine patient care, SF was aspirated from the knees of children with JIA. SF aliquots (n = 120) were pooled for preparation of the JIA SF-D1 pool as described above; 100 individual SF samples from 38 patients were assayed for sGAG concentrations, and 23 individual SF samples from 12 patients ages 3–16 years and representing a spectrum of JIA diagnoses including systemic JIA (n = 2), oligoarticular JIA (n = 3), extended oligoarticular JIA (n = 3), rheumatoid factor–negative polyarticular JIA (n = 2), and psoriatic arthritis (n = 2) were analyzed in detail by Western blotting (Table 1). For the Western blot analyses, SF samples from 5 patients were selected at 2 different ages (patients A, F, G, and H) or 3 different ages (patients E and K) in order to observe age-related changes in the pattern or abundance of aggrecan fragments. SF samples aspirated from both the left and right knee joints at the same visit were also included from patients B and F. This selection of patients was appropriate for a hypothesis-forming investigation.

The results presented below first describe the concentration of sGAG in JIA SF compared with sGAG in OA, juvenile injury, and knee-healthy SF; for JIA we also examined the relationship between SF sGAG concentration and age. Thereafter, we describe the pattern of aggrecan fragments in individual JIA patients, analyzing each of the major degradation products that can be uniquely detected by Western blotting; these results provide information about where and to what extent the aggrecan core protein is cleaved by the ADAMTS aggrecanases. To help place these data into a clinical context, we also present the Western blot data sorted according to patient, diagnosis, and age; this presentation helps reveal fragmentation patterns that cluster with disease types and therefore helps identify aggrecanolytic features that would be useful to analyze more extensively in future studies. Finally, to determine the influence of age and inflammation on the unique pattern of aggrecan fragments in JIA, we compare the pooled JIA SF with individual samples from juvenile injury and RA patients.

Concentration of sGAG in JIA SF.

SF concentrations of sGAG, used as a surrogate measure of aggrecan concentration, were determined by Alcian blue assay (32). The median sGAG concentration in JIA SF (31.0 μg/ml; range 3.0–196.3) was 2–3 times lower than in the juvenile knee injury patients (P = 0.006), the OA patients (P < 0.001), and the knee-healthy reference group (P = 0.022) (Figure 2A). We found a negative correlation between JIA SF sGAG concentration and patient age (rS = −0.21, P = 0.039) (Figure 2B); this observation is consistent with the fact that JIA often resolves by adulthood. We also found a positive correlation between the concentrations of sGAG in the left knee and in the right knee among patients in whom both knees were aspirated at the same time (rS = 0.60, P = 0.013) (n = 16 samples from 6 patients).

thumbnail image

Figure 2. Sulfated glycosaminoglycan (sGAG) levels in synovial fluid (SF) and correlation with age. A, SF sGAG concentrations in juvenile idiopathic arthritis (JIA) patients, osteoarthritis (OA) patients, juvenile (Juv) patients with knee injury, and the knee-healthy reference group (Ref). Data are presented as box plots, where the boxes represent the 25th to 75th percentiles, the lines within the boxes represent the median, and the lines outside the boxes represent the 10th and 90th percentiles. Circles indicate outliers. P values are versus the JIA group. B, SF sGAG concentrations in JIA patients plotted against age at the time of sampling. In 6 patients, SF was aspirated from both the left knee and the right knee, on one or more occasions. These patients are represented by solid or shaded squares, diamonds, or triangles; paired symbols at a given time point represent the left and right knees. All other individual patients are represented by circles. A negative correlation between sGAG levels and age was identified (rs = −0.21, P = 0.039).

Download figure to PowerPoint

Western blot analysis of aggrecan fragments in JIA SF.

Aggrecan SELE1545 fragments.

The large aggrecan fragments comprising both 374ARGS-SELE1545 and G1-SELE1545 fragments in the OA pool represent one of the most abundant aggrecan populations in OA SF (15, 21). In contrast, SELE1545 fragments were readily detected in only half the JIA patients (Figure 3A). A high molecular weight SELE1545 band, not present in the OA pool, was detected in several patients and was seen most strongly in patient B (Figure 3A). This high molecular weight SELE1545 fragment might comprise homo- or heterocomplexes of aggrecan fragments. When individual JIA samples on the Western blots were sorted according to patient diagnosis, SELE1514 immunoreactivity was shown to be weak or absent in the patients with seronegative polyarthritis or systemic disease, but more strongly present in patients with oligoarticular, extended oligoarticular, or psoriatic disease (Figure 4C). There was also a tendency toward stronger SELE1514 reactivity in patients with longer disease duration.

thumbnail image

Figure 3. Analysis of aggrecan fragments in the SF of JIA patients by Western blotting. JIA SF-D1 samples (2 μg sGAG per lane) were analyzed by Western blotting using antibodies against SELE1514 neoepitope (A), KEEE1714 neoepitope (B), the G3 globular domain (C), or 374ARGS neoepitope (D). Each lane is labeled with the patient identifier (A–L; Table 1) and the patient's age (in years) at the time of aspiration. Left and right knees (L and R) are identified for patients B and F. Circled letters identify fragments corresponding to those shown in Figure 1. Pooled JIA SF-D1 samples and pooled OA SF-D1 samples were also included on each gel. Intact aggrecan in the OA pool, detected with the anti-G3 antibody and migrating at the top of the gel, is not shown. No bands were present below the portion of the blots shown, and all samples were prepared in normal gradients. Asterisks mark the position of the extra-large SELE1545 and KEEE1714 fragments. Schematic drawings at the left show the position of each fragment with respect to the parent molecule; lettering in the drawings indicates the bands marked on the gels and the fragments shown in Figure 1. See Figure 2 for definitions.

Download figure to PowerPoint

thumbnail image

Figure 4. Western blots of aggrecan fragments sorted by diagnosis, patient, and age. Individual lanes from each of the anti-G3 (B), anti-SELE1514 (C), and anti-KEEE1714 (D) Western blots in Figure 3 were sorted according to JIA type, then patient, then age, to allow visualization of common fragmentation features associated with disease types. Each lane is labeled with the patient identifier (A–L; Table 1) and the patient's age (in years) at the time of aspiration. Left and right knees (L and R) are identified for patients B and F. Circled letters identify fragments corresponding to those shown in Figure 1. Pooled JIA SF-D1 samples and pooled OA SF-D1 samples were also included on each gel. Schematic drawings (A) show the position of each fragment with respect to the parent molecule; lettering in the drawings indicates the bands marked on the gels and the fragments shown in Figure 1. N = nonsteroidal antiinflammatory drug; D = disease-modifying antirheumatic drug; S = steroid (see Figure 2 for other definitions).

Download figure to PowerPoint

Aggrecan KEEE1714 fragments.

Three KEEE1714 fragments previously identified in OA SF (14, 21) were also present in the JIA samples (Figure 3B), but with some differences. Across all samples there were variable proportions of fragments d and e, which in OA SF comprise KEEE1714 fragments with N-terminal G1 domains and MMP-derived 342FFGV N-termini, respectively. The abundance of the 50-kd fragment k, shown previously to comprise 1546GRGT-KEEE1714 (14), was also variable. More importantly, the size variations in fragment k between individual JIA patients suggest marked differences in glycosylation in the chondroitin sulfate–rich region, or the possibility that KEEE1714 fragments in this size range are the products of more than one proteinase. Figure 3B shows the presence of KEEE1714 fragments larger than fragments d/e, as well as smeared KEEE1714 fragments with a markedly delayed migration pattern in patient E at all ages sampled, suggesting an interaction between the KEEE1714 fragment and another molecular moiety.

When individual JIA samples on KEEE1714 Western blots were sorted according to patient diagnosis, the results showed that fragment j was absent or very weak in the 3 patients with oligoarticular disease (Figure 4D). The identity of fragment j is unknown, but since its N-terminus is predicted to be within the chondroitin sulfate–rich region, it could possibly result from MMP cleavage (21, 33). The variability in the abundance and ratio of KEEE1714 fragments between samples from the same patient at different ages shows that aggrecan proteolysis is dynamic over time.

Aggrecan G3 fragments.

With the anti-G3 antibody, fragments f, g, and h were detected in OA SF (10, 14) (Figure 3C); however, an additional fragment, fragment i, was present in a subset of JIA samples (Figure 3C). Fragment f was barely detectable in most JIA samples, consistent with the finding that the SELE1545 epitope was weak or absent in half the JIA samples (Figure 3A), and also consistent with the knowledge that fragment f is further processed to equimolar amounts of the G3 fragment g and the KEEE1714 fragment k. The G3 fragments in the JIA samples were ranked as follows in overall abundance: fragment h > fragment g > fragment i > fragment f. At present the significance of the multiple cleavage events that create the G3 products is unknown.

Aggrecan 374ARGS fragments.

The most interesting result and the most striking differential between the JIA and OA SF samples was seen with the anti-ARGS neoepitope on Western blots (Figure 3D). The large 374ARGS-SELE1545 fragment b and the smaller 374ARGS-CS1 fragment c, seen routinely in SF from adults with OA (Figure 3D), acute injury, and acute inflammation (10, 30), were absent in the pool of 120 JIA SF samples, and only barely detectable in a few individual JIA samples (Figure 3D). These results show that, compared with SF from OA patients, the level of 374ARGS fragments in JIA patients determined by Western blotting is negligible.

Concordant and nonconcordant aggrecan patterns.

There were some consistent observations in this hypothesis-forming investigation of 2–3 patients from each of 5 JIA disease categories. We found that SF samples aspirated from left and right knees of the same patient on the same day were concordant for all aggrecan neoepitopes analyzed (Figures 4B–D) and also for the concentration of sGAG (results not shown). In contrast, there was no concordance in the pattern of aggrecan fragments in the same patient over time (Figures 4B–D).

The aggrecan fragmentation pattern in JIA is not a function of age or inflammation alone.

To explore whether the loss of 374ARGS immunoreactivity in JIA compared with OA was due to differences in age or inflammation status, individual SF-D1 samples from RA patients and children with knee injury were analyzed alongside the SF-D1 pools from OA and JIA patients and the knee-healthy reference group. We found that 374ARGS immunoreactivity with the samples from patients with RA and children with knee injury varied highly, from a strong signal in some patients to no signal in others (Figures 5A and B). In comparison, the 374ARGS epitope in JIA was uniformly below or on the border of detectability (Figure 5C). These results suggest that neither age alone nor inflammation status can account for the striking absence of 374ARGS in JIA. Indeed, as seen in Figure 5B, strong 374ARGS immunoreactivity was found in some juvenile patients with knee injury, indicating that aggrecanase cleavage in the aggrecan IGD occurs in children. The levels of 374ARGS epitope in the knee-healthy reference group were low, as shown previously (10, 30).

Aggrecan 342FFGV fragments.

The exceptionally low levels of 374ARGS epitope in all JIA patients (Figure 3D) and the reduced abundance of SELE1545 fragments in half the JIA patients (Figure 3A) prompted us to question whether proteolysis in the IGD might instead arise from MMP cleavage. Western blotting with the 342FFGV antibody confirmed that MMPs cleaved JIA aggrecan at the IPEN341[DOWNWARDS ARROW]342FFGV site. However, the levels and pattern of the 342FFGV neoepitope in the JIA and OA pools were similar (Figure 5D), suggesting that proteolysis by MMPs does not substitute for proteolysis by ADAMTS enzymes in JIA The presence of 3 342FFGV bands in the knee-healthy reference group was consistent with previous reports that MMPs might have a role in the baseline turnover of aggrecan and normal cartilage homeostasis in vitro (34) and in vivo (35).

thumbnail image

Figure 5. Comparison of 374ARGS epitope and 342FFGV epitope in SF samples from adults and children. Individual (A and B) and pooled (C and D) SF-D1 preparations from rheumatoid arthritis (RA) patients, OA patients, juvenile patients with knee injury, JIA patients, and the knee-healthy reference group were analyzed by Western blotting for 374ARGS neoepitope (A–C) or 342FFGV neoepitope (D). OA samples are pooled in all cases. ∗ = prepared in normal gradients (all other samples were prepared in minigradients). The number below each lane is the quantity (micrograms) of sGAG loaded. Circled letters identify fragments corresponding to those shown in Figure 1. See Figure 2 for other definitions.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  7. Acknowledgements

This is the first study to analyze aggrecan fragments in JIA SF. The apparent absence of 374ARGS fragments in the JIA samples is striking. Possible explanations are 1) that ADAMTS-4 and -5 are less active in JIA cartilage, perhaps as a consequence of processing that removes C-terminal ancilliary domains from these enzymes, 2) that JIA aggrecan is less susceptible to aggrecanase attack, due to different amounts or types of glycosylation near the TEGE373[DOWNWARDS ARROW]374ARGS cleavage site (36, 37), or 3) that an unidentified aggrecanase cleaves in the chondroitin sulfate–rich region, but not in the IGD (38), in JIA cartilage.

Alternatively, it is possible that the 374ARGS epitope is created and then destroyed by either aminopeptidase activity, MMP cleavage at TSED441[DOWNWARDS ARROW]442LVVR in the aggrecan IGD (35), or other processing events that generate low-density fragments that are not recovered in the CsCl gradients. There is currently no evidence to indicate whether 374ARGS fragments are created and then destroyed or not created at all, although we strongly favor the latter interpretation. Another reason 374ARGS fragments might not be created in JIA cartilage relates to enzyme kinetics, since the Km of ADAMTS-4/5 for aggrecan cleavage at C-terminal sites is ∼20-fold lower than at the IGD site. The progressive, age-related shortening (proteolysis) of aggrecan from the C-terminus creates adult aggrecan with fewer preferred cleavage sites. In children, with a higher proportion of preferred (C-terminal) cleavage sites acting as competing sites, the production of 374ARGS fragments from the nonpreferred (IGD) cleavage site will proceed more slowly. One final hypothesis to explain the lack of 374ARGS fragments is that hyaluronidase activity released into the matrix (39) of juvenile but not adult cartilage liberates intact aggrecan from its cartilage anchor into SF, where it is cleaved by aggrecanases at C-terminal sites.

The slowest-migrating SELE1545 and KEEE1714 fragments in some JIA SF samples are much larger than the G1-SELE1545 or G1-KEEE1714 fragments seen in OA SF, suggesting that aggrecan fragments in JIA might interact with one or more molecules present in the cartilage matrix or the SF. The interacting molecule(s) is not an experimental or biologic contaminant because, in patient E for example, the SF samples were collected over a period of 5 years. The strength of the interaction, which is resistant to dissociation by SDS and 4M GuHCl, suggests that it might be stabilized by a molecular crosslinker such as transglutaminase 2 (40). The ubiquitously expressed transglutamase 2 catalyzes transamidation of glutamine residues to lysine residues and is thought to help stabilize tissue against injury or infection. However, in other cases, inappropriately crosslinked protein aggregates may trigger inflammation, and in this context it is interesting to note the association between transglutamase 2 and disease progression in RA (40). SF levels of transglutamase 2 also correlate with knee OA in the Hartley guinea pig model of spontaneous OA (41). Thus, the potential for transglutamase 2 activity to mediate crosslinking of aggrecan fragments in JIA cartilage is intriguing and warrants further investigation.

Given the almost complete absence of 374ARGS bands in JIA samples, it will be interesting in the future to determine whether large SELE1545 and KEEE1714 fragments have an N-terminus other than 374ARGS. The putative 374ARGS N-terminus of JIA SELE1545 fragment b might be destroyed by the activity of aminopeptidases or dipeptidylpeptidases. Alternatively, the N-terminus could be generated by MMP cleavage in the IGD since large molecular weight 342FFGV fragments were indeed present in JIA SF and 342FFGV-SELE1545 fragments have been detected in OA SF (15).

We have observed that the 342FFGV neoepitope is labile in the mouse since, although the IPEN341 and 342FFGV neoepitopes are generated in equimolar amounts, the 342FFGV epitope is extremely difficult to detect compared with IPEN341; this is despite the fact that lower molar amounts of epitope are detected with the anti-FFGV antibody than with the anti-IPEN antibody (34). We suspect that aminopeptidase or dipeptidylpeptidase activity with specificity for N-terminal phenylalanine might be responsible for this apparent loss of 342FFGV immunoreactivity. Similarly, aminopeptidases removing 1 or 2 N-terminal amino acids from an 374ARGS peptide would destroy the antigenicity of the neoepitope. Neutral aminopeptidase activity has been detected in the SF and peripheral blood of RA and JIA patients (42), and inhibitors of aminopeptidase and dipeptidylpeptidase activities have been investigated as treatments for RA (43).

From a clinical perspective, the findings of this study are interesting for two reasons. First, the results suggest that the drivers of aggrecanase activity are uniform between joints during periods of active arthritis; hence, the identical fragmentation patterns in individual patients' left and right knees aspirated at the same time. This is unexpected because systemic treatments for JIA can sometimes show better efficacy in one joint than another, suggesting that the drivers (cytokines, for example) might not be identical between two joints at the one time. The second interesting finding is that in individual patients, the aggrecan fragmentation pattern changes with time. This could be due to intrinsic disease factors or therapies. It is noteworthy that for patients A and F, patient E (at ages 11 and 12 years only), and patient K (at ages 8 and 12 years only) the therapies were the same, even though the fragmentation patterns were different. This suggests that treatment regimens alone are unlikely to be the underlying cause of the changing pattern of aggrecan fragments over time.

Several studies have shown that glycosaminoglycans, and in particular keratan sulfate (KS), may have a role in modulating aggrecanase activity in cartilage explants. For example, aggrecanase cleavage in the IGD is increased in the presence of endogenous KS and reduced when KS is removed by keratanase treatment (44–46). In human aggrecan, the KS content increases from a minimal percentage at birth to more than one-quarter of the GAG content at maturity, and the extent of modification is similarly increased. For example, maturing cartilage (9–18 years) has intermediate and increasing levels of KS sulfation, fucosylation, and sialylation (47). We have shown that in pig aggrecan, KS in the IGD is uniquely undersulfated compared with KS elsewhere on aggrecan (48), and also that KS on recombinant IGD potentiates aggrecanase cleavage in vitro (46). Given the variable patterns of aggrecan fragments in JIA, the proximity of IGD KS to the aggrecanase cleavage site, and evidence that KS potentiates aggrecanase activity in vitro, further studies to explore the role of KS in the pathogenesis of JIA are warranted.

The limitations of this study are the small number of patient SF samples examined in detail and the inability to quantitate 374ARGS bands that cannot be detected on gels. Future studies with more patients to elucidate the mechanisms involved in cartilage erosion in JIA, in conjunction with the use of newer technologies for quantitating the 374ARGS neoepitope at low concentrations (49, 50), are needed for the analysis of aggrecanase fragments in JIA SF.

In summary, this is the first detailed study of aggrecanolysis in JIA. Our results suggest that compared with aggrecanases in OA cartilage, aggrecanases in JIA cleave poorly in the aggrecan IGD. Accordingly, cleavage at the TEGE373[DOWNWARDS ARROW]374ARGS cleavage site in the aggrecan IGD may not be an appropriate therapeutic target for management of cartilage erosion in children with JIA.


  1. Top of page
  2. Abstract
  7. Acknowledgements

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Fosang had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Struglics, Allen, Fosang.

Acquisition of data. Struglics, Last, Akikusa, Allen.

Analysis and interpretation of data. Struglics, Lohmander, Last, Akikusa, Fosang.


  1. Top of page
  2. Abstract
  7. Acknowledgements

We thank Dr. Simon Chatfield for providing synovial fluid specimens from rheumatoid arthritis patients, Drs. Sanjay Kumar and Michael Pratta for the kind gift of anti-ARGS (mAb OA-1) and anti-SELE antibodies, and Dr. Mike Lark for the anti-KEEE antibody. We also thank Maria Hansson for excellent technical assistance.


  1. Top of page
  2. Abstract
  7. Acknowledgements
  • 1
    Manners PJ, Bower C. Worldwide prevalence of juvenile arthritis: why does it vary so much? J Rheumatol 2002; 29: 152030.
  • 2
    Gare BA. Epidemiology. Baillieres Clin Rheumatol 1998; 12: 191208.
  • 3
    Bowyer SL, Roettcher PA, Higgins GC, Adams B, Myers LK, Wallace C, et al. Health status of patients with juvenile rheumatoid arthritis at 1 and 5 years after diagnosis. J Rheumatol 2003; 30: 394400.
  • 4
    Van Rossum MA, Zwinderman AH, Boers M, Dijkmans BA, van Soesbergen RM, Fiselier TJ, et al, on behalf of the Dutch Juvenile Idiopathic Arthritis Study Group. Radiologic features in juvenile idiopathic arthritis: a first step in the development of a standardized assessment method. Arthritis Rheum 2003; 48: 50715.
  • 5
    Wallace CA, Levinson JE. Juvenile rheumatoid arthritis: outcome and treatment for the 1990s. Rheum Dis Clin North Am 1991; 17: 891905.
  • 6
    Mason T, Reed AM, Nelson AM, Thomas KB, Patton A, Hoffman AD, et al. Frequency of abnormal hand and wrist radiographs at time of diagnosis of polyarticular juvenile rheumatoid arthritis. J Rheumatol 2002; 29: 22148.
  • 7
    Kim HK, Laor T, Graham TB, Anton CG, Salisbury SR, Racadio JM, et al. T2 relaxation time changes in distal femoral articular cartilage in children with juvenile idiopathic arthritis: a 3-year longitudinal study. AJR Am J Roentgenol 2010; 195: 10215.
  • 8
    Tortorella MD, Burn TC, Pratta MA, Abbaszade I, Hollis JM, Liu R, et al. Purification and cloning of aggrecanase-1: a member of the ADAMTS family of proteins. Science 1999; 284: 16646.
  • 9
    Abbaszade I, Liu RQ, Yang F, Rosenfeld SA, Ross OH, Link JR, et al. Cloning and characterization of ADAMTS11, an aggrecanase from the ADAMTS family. J Biol Chem 1999; 274: 2344350.
  • 10
    Struglics A, Larsson S, Hansson M, Lohmander LS. Western blot quantification of aggrecan fragments in human synovial fluid indicates differences in fragment patterns between joint diseases. Osteoarthritis Cartilage 2009; 17: 497506.
  • 11
    Stanton H, Rogerson FM, East CJ, Golub SB, Lawlor KE, Meeker CT, et al. ADAMTS5 is the major aggrecanase in mouse cartilage in vivo and in vitro. Nature 2005; 434: 64852.
  • 12
    Glasson SS, Askew R, Sheppard B, Carito B, Blanchet T, Ma HL, et al. Deletion of active ADAMTS5 prevents cartilage degradation in a murine model of osteoarthritis. Nature 2005; 434: 6448.
  • 13
    Song RH, Tortorella MD, Malfait AM, Alston JT, Yang Z, Arner EC, et al. Aggrecan degradation in human articular cartilage explants is mediated by both ADAMTS-4 and ADAMTS-5. Arthritis Rheum 2007; 56: 57585.
  • 14
    Struglics A, Larsson S. A comparison of different purification methods of aggrecan fragments from human articular cartilage and synovial fluid. Matrix Biol 2010; 29: 7483.
  • 15
    Struglics A, Larsson S, Pratta MA, Kumar S, Lark MW, Lohmander LS. Human osteoarthritis synovial fluid and joint cartilage contain both aggrecanase- and matrix metalloproteinase-generated aggrecan fragments. Osteoarthritis Cartilage 2006; 14: 10113.
  • 16
    Lohmander LS, Neame PJ, 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: 121422.
  • 17
    Sandy JD, Verscharen C. Analysis of aggrecan in human knee cartilage and synovial fluid indicates that aggrecanase (ADAMTS) activity is responsible for the catabolic turnover and loss of whole aggrecan whereas other protease activity is required for C-terminal processing in vivo. Biochem J 2001; 358(Pt 3): 61526.
  • 18
    Tortorella MD, Pratta M, Liu RQ, Austin J, Ross OH, Abbaszade I, et al. Sites of aggrecan cleavage by recombinant human aggrecanase-1 (ADAMTS-4). J Biol Chem 2000; 275: 1856673.
  • 19
    Tortorella MD, Liu RQ, Burn T, Newton RC, Arner E. Characterization of human aggrecanase 2 (ADAM-TS5): substrate specificity studies and comparison with aggrecanase 1 (ADAM-TS4). Matrix Biol 2002; 21: 499511.
  • 20
    Sandy JD, Thompson V, Doege K, Verscharen C. The intermediates of aggrecanase-dependent cleavage of aggrecan in rat chondrosarcoma cells treated with interleukin-1. Biochem J 2000; 351(Pt 1): 1616.
  • 21
    Struglics A, Larsson S, Lohmander LS. Estimation of the identity of proteolytic aggrecan fragments using PAGE migration and Western immunoblot. Osteoarthritis Cartilage 2006; 14: 898905.
  • 22
    Fosang AJ, Last K, Maciewicz RA. Aggrecan is degraded by matrix metalloproteinases in human arthritis: evidence that matrix metalloproteinase and aggrecanase activities can be independent. J Clin Invest 1996; 98: 22929.
  • 23
    Struglics A, Hansson M. Calpain is involved in C-terminal truncation of human aggrecan. Biochem J 2010; 430: 5318.
  • 24
    Chamberland A, Wang E, Jones AR, Collins-Racie LA, LaVallie ER, Huang Y, et al. Identification of a novel HtrA1-susceptible cleavage site in human aggrecan: evidence for the involvement of HtrA1 in aggrecan proteolysis in vivo. J Biol Chem 2009; 284: 273529.
  • 25
    Fosang AJ, Last K, Gardiner P, Jackson DC, Brown L. Development of a cleavage site-specific monoclonal antibody for detecting metalloproteinase-derived aggrecan fragments: detection of fragments in human synovial fluids. Biochem J 1995; 310: 33743.
  • 26
    Lohmander LS, Ionescu M, Jugessur H, Poole AR. Changes in joint cartilage aggrecan after knee injury and in osteoarthritis. Arthritis Rheum 1999; 42: 53444.
  • 27
    Larsson S, Englund M, Struglics A, Lohmander LS. Association between synovial fluid levels of aggrecan ARGS fragments and radiographic progression in knee osteoarthritis. Arthritis Res Ther 2010; 12: R230.
  • 28
    Lohmander LS, Hoerrner LA, Lark MW. Metalloproteinases, tissue inhibitor, and proteoglycan fragments in knee synovial fluid in human osteoarthritis. Arthritis Rheum 1993; 36: 1819.
  • 29
    Heinegard D, Sommarin Y. Isolation and characterization of proteoglycans. Methods Enzymol 1987; 144: 31972.
  • 30
    Struglics A, Hansson M, Lohmander LS. Human aggrecanase generated synovial fluid fragment levels are elevated directly after knee injuries due to proteolysis both in the inter globular and chondroitin sulfate domains. Osteoarthritis Cartilage 2011; 19: 104757.
  • 31
    Pratta MA, Su JL, Leesnitzer MA, Struglics A, Larsson S, Lohmander LS, et al. Development and characterization of a highly specific and sensitive sandwich ELISA for detection of aggrecanase-generated aggrecan fragments. Osteoarthritis Cartilage 2006; 14: 70213.
  • 32
    Bjornsson S. Simultaneous preparation and quantitation of proteoglycans by precipitation with alcian blue. Anal Biochem 1993; 210: 28291.
  • 33
    Bonassar LJ, Jeffries KA, Frank EH, Moore VL, Lark MW, Bayne EK, et al. In vivo effects of stromelysin on the composition and physical properties of rabbit articular cartilage in the presence and absence of a synthetic inhibitor. Arthritis Rheum 1995; 38: 167886.
  • 34
    Fosang AJ, Last K, Stanton H, Weeks DB, Campbell IK, Hardingham TE, et al. Generation and novel distribution of matrix metalloproteinase-derived aggrecan fragments in porcine cartilage explants. J Biol Chem 2000; 275: 3302737.
  • 35
    Little CB, Meeker CT, Hembry RM, Sims NA, Lawlor KE, Golub SB, et al. Matrix metalloproteinases are not essential for aggrecan turnover during normal skeletal growth and development. Mol Cell Biol 2005; 25: 338899.
  • 36
    Barry FP, Gaw JU, Young CN, Neame PJ. Hyaluronan-binding region of aggrecan from pig laryngeal cartilage. Biochem J 1992; 286: 7619.
  • 37
    Barry FP, Rosenberg LC, Gaw JU, Koob TJ, Neame PJ. N- and O-linked keratan sulfate on the hyaluronan binding region of aggrecan from mature and immature bovine cartilage. J Biol Chem 1995; 270: 2051624.
  • 38
    Rogerson FM, Stanton H, East CJ, Golub SB, Tutolo L, Farmer PJ, et al. Evidence of a novel aggrecan-degrading activity in cartilage: studies of mice deficient in both ADAMTS-4 and ADAMTS-5. Arthritis Rheum 2008; 58: 166473.
  • 39
    Bastow ER, Last K, Golub SB, Stow JL, Stanley AC, Fosang AJ. Evidence for lysosomal exocytosis and release of aggrecan- degrading hydrolases from hypertrophic chondrocytes, in vitro and in vivo. Biology Open 2012; 1: 31828.
  • 40
    Kim SY. Transglutaminase 2 in inflammation. Front Biosci 2006; 11: 302635.
  • 41
    Huebner JL, Johnson KA, Kraus VB, Terkeltaub RA. Transglutaminase 2 is a marker of chondrocyte hypertrophy and osteoarthritis severity in the Hartley guinea pig model of knee OA. Osteoarthritis Cartilage 2009; 17: 105664.
  • 42
    Riemann D, Schwachula A, Hentschel M, Langner J. Demonstration of CD13/aminopeptidase N on synovial fluid T cells from patients with different forms of joint effusions. Immunobiology 1993; 187: 2435.
  • 43
    Tanaka S, Murakami T, Nonaka N, Ohnuki T, Yamada M, Sugita T. Anti-arthritic effects of the novel dipeptidyl peptidase IV inhibitors TMC-2A and TSL-225. Immunopharmacology 1998; 40: 216.
  • 44
    Pratta MA, Tortorella MD, Arner EC. Age-related changes in aggrecan glycosylation affect cleavage by aggrecanase. J Biol Chem 2000; 275: 39096102.
  • 45
    Tortorella M, Pratta M, Liu RQ, Abbaszade I, Ross H, Burn T, et al. The thrombospondin motif of aggrecanase-1 (ADAMTS-4) is critical for aggrecan substrate recognition and cleavage. J Biol Chem 2000; 275: 257917.
  • 46
    Poon CJ, Plaas AH, Keene DR, McQuillan DJ, Last K, Fosang AJ. N-linked keratan sulfate in the aggrecan interglobular domain potentiates aggrecanase activity. J Biol Chem 2005; 280: 2361521.
  • 47
    Brown GM, Huckerby TN, Bayliss MT, Nieduszynski IA. Human aggrecan keratan sulfate undergoes structural changes during adolescent development. J Biol Chem 1998; 273: 2640814.
  • 48
    Fosang AJ, Last K, Poon CJ, Plaas AH. Keratan sulphate in the interglobular domain has a microstructure that is distinct from keratan sulphate elsewhere on pig aggrecan. Matrix Biol 2009; 28: 5361.
  • 49
    Swearingen C, Carpenter J, Siegel R, Brittain I, Dotzlaf J, Durham TB, et al. Development of a novel clinical biomarker assay to detect and quantify aggrecanase-generated aggrecan fragments in human synovial fluid, serum and urine. Osteoarthritis Cartilage 2010; 18: 11508.
  • 50
    Chockalingam PS, Sun W, Rivera-Bermudez MA, Zeng W, Dufield DR, Larsson S, et al. Elevated aggrecanase activity in a rat model of joint injury is attenuated by an aggrecanase specific inhibitor. Osteoarthritis Cartilage 2011; 19: 31523.