SUMMARYWe have tested the effect of administering H22, a hamster neutralizing MoAb to murine interferongamma (IFN-'/) in collagen-induced arthritis. Mice were initTtuni7ed with human type II collagen in adjuvant on day 1 and boosted with soluble collagen on day 21. H22 was administered (250 ng. intraperitoneal!y)either during the induction of arthritis (on daysO, 6, 13 and 20) or around the time ofdiscasc manifestation (on days 21, 28, 35 and 42). Control mice received either an isotypc-matched non-neutralizing MoAb or saline. Both treatment regimes gave sitnilar results. Treattncnt with H22 did not signifieantly affect the incidence of arthritis, time ofonset, degree of oedetiia. histopathological severity, or tcvcl of anti-type II collagen IgG. However, a highly signiticanl increase (P<0'01) in the number of limbs affected by arthritis was observed in the H22-treated group, irrespective of whether the antibody was administered during the induction of arthritis, or during the time of clinical manifcstalion of disease. From these results it was concluded that anti-lFN-y treatment caused an inerease in the niitnber of arthritic lesions, but did not affect the severity of each individual lesion.Keywords interferon-gamma collagen-induced arthritis DBA/1 miee INTRODirCTION Much of the rcsearcb thai has been carried oul into the roles played by cytokines in chronic inflammatory conditions sueh as rheumatoid arthritis (RA) has focused on the identification of these tnolecules in affected tissues, and on the in vitro properties that they exhibit. It is itnportant to establish, however, which of these properties are important in vivo, since this information is critical for the development of therapeutic strategies based on the modulation of cytokine aetivity. For example, on the basis of in vitro studies, a wide range of both pro-and antiinllatnmatory properties have been described for interferongamma (1 FN-v) {1], but it is not known which of these properties is pathologically relevant. One potentially useful approach to the study ofthe pathological importance of eytokines in vivo is to adtninister cytokine-specilie neutralizing antibodies in particular disease situations. For example, the role of tumour necrosis factor in collagen-induced arthritis was established by adtninistering MoAbs specitic for this eytokine [2,3j, Collagen-induced arthritis demonstrates marked similarities to human RA with respect to both elinteal and histological findings [4] and can therefore be used to study the effects of adtninislcring cytokines or agents capable of neutralizing cytokines in vivo-For exatnple. two studies have recently been published on the effect on collagen-induced arthritis of adtTtinistering reeombinant IFN-"/ to collagen-immunized DBA/1 mice.Correspondence: Richard Williams. Kennedy Institute of Rheumatology. 6 Bute Gardens. London W6 7DW. UK.In the first of these studies, higher ineidenec. earlier onset and increased severity of arthritis were demonstrated in mice given footpad injections of recombinant IFN--/ [5]. ...