COLLAGEN-INDUCED ARTHRITIS IN RHESUS-MONKEYS - EVALUATION OF MARKERS FOR INFLAMMATION AND JOINT DEGRADATION

The objective of this study was to analyse parameters in rhesus monkey collagen-induced arthritis (CIA) with which the inflammation and dest ruction of the joints can be described in quantitative terms. CIA was induced in genetically susceptible and resistant monkeys, which can be distinguished on the basis of the dominant resistance marker Mamu-A26 . The disease course was monitored daily using a semiquantitative scor ing system. Plasma samples were collected once or twice weekly and ana lysed for C-reactive protein (CRP). Urines were collected overnight on ce a week and analysed for excretion rates of the collagen cross-links hydroxylysylpyridinoline (HP) and lysylpyridinoline (LP). The results show that periods of active CIA are characterized by substantial weig ht loss and increased plasma CRP levels, followed shortly thereafter b y increased excretion rates of the collagen cross-links HP and LP. Rem ission of the disease can be recognized by a decline in plasma CRP lev els and especially an increase in body weight. The highest CRP levels were found in the most severely arthritic monkeys, indicating a possib le relationship of the absolute plasma CRP levels to the severity of i nflammation. During periods of active arthritis, increased excretion r ates of collagen cross-links HP and LP in the urine were found. In par ticular, the major collagen crosslink in articular cartilage, HP, show ed a strong increase (9- to 15-fold). The excretion rates of LP, which is considered as a bone-specific degradation marker, only increased 4 - to 6-fold, thus indicating predominant destruction of cartilage and less of bone. In conclusion, the severity of CIA can be monitored in a quantitative manner using plasma CRP levels, urinary excretion rates of HP and LP, and body weights, superimposed on semiquantitative clini cal scores. The parameters also facilitate a more objective assessment of the effect of anti-arthritic drugs in the model than with the clin ical scores alone.