Osteoarthritis and Cartilage Vol. 15, Supplement C C67 joints (P=0.004). OC injured SF C2C concentrations were posi- tively correlated with radiographic scores (R=0.43; P=0.03) and arthroscopic scores (R=0.52; P=0.03). Arthroscopic scores were positively correlated with radiographic scores (R=0.51; P=0.04). SF C2C concentrations ≥ 64 pmol/mL for MCP/MTP joints and ≥ 75 pmol/mL for carpal joints were arbitrarily chosen to determine predictive value for discriminating OC injured horses from rested or exercised horses. This yielded a positive predictive value of 73% and a negative predictive value of 94% for the MCP/TP joints, and positive predictive value of 79% and negative predic- tive value of 95% for the carpus. Conclusions: OC injury caused a significant increase in SF C2C concentrations in carpal and MCP/MTP joints compared to rested and exercised horses. SF C2C concentrations were correlated to severity of joint injury. Based on these findings, SF C2C analysis may be useful for evaluation of joint injury. Acknowledgment: Study supported in part by the University of Florida Equine Soundness Program. 105 FOLLOW-UP OF COLL2-1, COLL2-1NO 2 AND MYELOPEROXYDASE SERUM LEVELS IN MARATHON RUNNERS M. Deberg 1 , A. Labasse 1 , C. Sanchez 1 , A. Bosseloir 2 , Y. Henrotin 1 1 University of Liège, Liège, Belgium; 2 Zentech, Liège, Belgium Purpose: To determine the influence of marathon on the serum levels of two markers of cartilage degradation, Coll2-1, a peptide of type II collagen triple helix, and its nitrated form, Coll2-1NO 2 , and of a marker of neutrophils activation, the myeloperoxidase (MPO). Methods: Coll2-1, Coll2-1NO 2 and MPO were measured by specific immunoassays in 66 marathon runners without joint pain and aged in mean of 47 (min: 31 - max: 59) years. Sera were taken before and after the marathon. All the subjects were submitted to a questionnaire concerning their physical activity (i.e. training or best performance) and their life style (i.e; diet). The total running distance was 42 km. Their performance in the marathon was ranged from 149 to 270 min. Results: Before the marathon, the Coll2-1 and Coll2-1NO 2 val- ues were not affected by age, body mass index, sex and per- formance [Coll2-1 median: 90.66 (28.77-234.68) nM and Coll2- 1NO 2 median: 0.16 (0.05-0.71) nM], while MPO levels were higher in female [median: 41.3 (29.90-96.50) ng/ml] than in male [median: 33.00 (15.60-84.60) ng/ml] (p<0.05). After the marathon, Coll2-1 levels were slightly decreased [median: 76.63 (28.89-185.06) nM], Coll2-1NO 2 levels were unmodified [median: 0.15 (0.05-0.61) nM] and MPO levels were doubled [median: 71.80 (31.50-172.10) ng/ml] compared to the pre-marathon val- ues [MPO median: 34.20 (15.60-96.50) ng/ml] (p<0.001). The variation of MPO during the marathon was negatively correlated with the training time per week (r=-0.36; p=0.0045). Before the marathon, there was no correlation between the different biolog- ical markers (Coll2-1, Coll2-1NO 2 and MPO). Conclusions: Coll2-1 and Coll2-1NO 2 serum levels were not modified by marathon, suggesting that, at least at short-term, this intensive physical activity does not significantly modified cartilage metabolism. Interestingly, serum MPO levels were in- creased after the marathon and training reduced this elevation. These results suggest that neutrophils were activated during the marathon and that training reduces neutrophils activation during the marathon. 106 QUANTITATIVE WESTERN BLOTTING SHOWS DIFFERENCES IN SYNOVIAL FLUID AGGRECAN FRAGMENT PATTERNS BETWEEN HUMAN JOINT DISEASES A. Struglics, S. Larsson, M. Hansson, S. Lohmander Lund University, Lund, Sweden Purpose: Aggrecan proteolysis is an early key event in arthritis. Aggrecanases cleave the interglobular domain of aggrecan re- leasing N-terminal ARGS fragments, and the chondroitin sulfate rich domain releasing C-terminal G3 domain fragments. Several reports suggest a role for more than one ADAMTS activity and possibly other proteases in the release of aggrecan fragments from joint cartilage. Differences between animal models and hu- man disease, and differences between human joint diseases and disease stages suggest a role for a method to map in detail aggrecan proteolytic fragments in multiple samples. We have developed a quantitative Western blot method for analysis of aggrecan fragment patterns in synovial fluids (SF) from patients with different joint diseases. Methods: Dissociative CsCl gradient centrifugation was used for sample preparation from SF of 5 knee-healthy subjects (ref- erence) and 27 patients with 4 different knee joint diseases. Aggrecan fragments (D1-fractions) separated by SDS-PAGE and transferred to PVDF membranes were screened, using a digi- tal luminescence imager, with an ARGS neo-epitope antibody or an antibody recognizing the G3-domain. ARGS-standards were generated by complete aggrecanase digestion of intact hu- man aggrecan. Quantification was by co-running either ARGS- standards or a control sample (in G3 quantification) on each SDS-gel. Results were compared and expressed as mol (ARGS) or relative AU (G3) per mg glycosaminoglycan (GAG) determined by Alcian Blue precipitation. Results: GAG recovery in D1-fractions was >75%, and West- ern blot CVs between experiments for anti-ARGS and anti-G3 were 19% and 28%. Total ARGS fragment quantity in SF sam- ples by Western blot correlated well with ARGS-ELISA (R 2 =0.87, n=19), while the correlation between ARGS and GAG content was only moderate (R 2 =0.31, n=31). SFs contained two major ARGS-bands, ARGS-SELE and ARGS-GVED, and three major G3-bands (GRGT-G3, GLGS-G3 and AGEG-G3) as a result of aggrecanase proteolysis. Median values of total ARGS fragments in the diagnostic groups were significantly higher in the acute arthritis and acute injury groups compared to the reference, and these SF samples contained the highest total ARGS values per mg GAG (Table 1; *P<0.05; **P<0.01 compared to refer- ence). The differential between reference and disease groups was greatest for the ARGS-SELE fragment. Median values of total G3 per mg GAG were significantly lower in the acute arthri- tis and osteoarthritis groups than in the reference. The ratios of total-ARGS to total-G3 fragments varied markedly between the diagnostic groups. Table 1. Western blot analysis in diagnostic groups. Median values normalized to reference Group n ARGS- ARGS- Total Total Total SELE GVED ARGS G3 ARGS/G3 Reference 5 1 1 1 1 1 Acute arthritis 8 25** 10.1** 10.7* 0.5* 22.3* Acute injury 9 20.2** 10.2** 10* 0.6 9.5** Chronic injury 5 4.2 2.2 2.1 1.2 0.9 Osteoarthritis 5 8.4 2.9 3.4 0.3* 2.6 Conclusions: We have developed a quantitative Western blot method for aggrecan fragment screening of synovial fluids. It provides reliable results as judged by relatively low CV and good correlation between Western and ELISA methods. Anti-ARGS