The roles of interleukin-18 in collagen-induced arthritis in the BB rat

Abstract

Interleukin (IL)-18 is a member of the IL-1 cytokine family. Its expression is increased in rheumatoid arthritis synovium, and its proinflammatory effects have been demonstrated in experimental models of murine arthritis. Here, we investigate the actions of varying doses of recombinant rat IL-18 (rIL-18) on the course of type II collagen-induced arthritis (CIA) in BB rats, including clinical and immune events, plus splenic cytokine production. Small doses of rIL-18 (10 and 50 microg/rat) administered intraperitoneally (i.p.) increased arthritis incidence and severity (P < 0.01) when a low-potency CII preparation was used for immunization. IgG1 and IgG2a anti-CII antibody levels were significantly greater in rats given 10 and 50 microg rIL-18 doses than controls. rIL-18 significantly increased levels of proinflammatory cytokines [interferon (IFN)-gamma, IL-2, tumour necrosis factor (TNF)-alpha and IL-6] produced by splenocyte cultures. Larger doses of rIL-18 (300 microg/rat) suppressed arthritis and immunity. To ascertain whether the pro-arthritic effects of IL-18 could be attenuated, rats were treated with neutralizing rabbit anti-rIL-18 IgG before immunization with a high-potency CII preparation. When given serially for 3 weeks, the incidence and severity of CIA, in addition to anti-CII IgG2a and splenic IL-6 and IFN-gamma production, were all significantly reduced. Similar results were noted when antibody was given twice, just before arthritis onset. These results demonstrate that IL-18 plays an important proinflammatory role in the pathogenesis of CIA which is achieved, in part, by an immunostimulatory action. Neutralizing endogenous IL-18 with antibodies attenuated CIA, CII immunity and cytokine responses. These studies support the use of IL-18 antagonists as treatments for inflammatory arthritis.

Fig. 1

Expression of mRNA for IL-18 in synovium as a function of time as measured by RT-PCR. The data shown are the ratios of mRNA for IL-18 to β-actin. Arthritis scores for each rat constituting one time-point given in the figure are: 0 (normal rat, day 0), 2+, 4+, 4+, 4+, 3+, 3+ and 3+, respectively. Synovium was obtained from the arthritic hind-paws of one rat and pooled for each time point, except for the normal rat where four hind-paws were pooled from two animals.

Fig. 2

(a) Ability of increasing doses of rIL-18 to stimulate splenocytes to produce IFN-γ and TNF-αin vitro. (b) Production of IFN-γ could also be inhibited in a dose-dependent manner by adding rabbit anti-rIL-18 IgG to the culture system, either 1 h before or with rIL-18. The supernatants were obtained after 48 h of culture and assayed for cytokine by ELISA. Average values are shown; variation between duplicate samples was less than 5% of the average.

Fig. 3

Actions of rIL-18 on the incidence (a) and severity (b) of CIA in the BB rat. Rats (n = 7) given rIL-18 at 10 and 50 µg doses on the day shown (arrows) developed arthritis at a significantly higher incidence of arthritis than controls (n = 8) given PBS (P < 0·001 and <0·0001, respectively); both doses worsened arthritis severity similarly (P < 0·002). In contrast, 300 µg of rIL-18 abrogated arthritis (n = 6). Arrows represent the days of rIL-18 administration; values are expressed as mean ± s.e.m. per group.

Fig. 4

Actions of rIL-18 on the production of CII-specific antibodies. (a) Levels of anti-CII IgG measured in the sera of rats treated with different doses of rIL-18 as shown in Fig. 3a and b. Increased levels were significantly greater at the 50 µg dose, whereas changes with 10 µg were minor and 300 µg was suppressive (P < 0·009). (b) IgG subclass distribution of anti-CII antibodies in rats treated with 50 µg doses of rIL-18 or controls given PBS. The greatest increase was found in the IgG2a subclass (P < 0·002), but increases were also noted in IgG1 (P < 0·02) and IgG2b (P = n.s.). Values are expressed as mean ± s.e.m. per group.

Fig. 5

Actions of rIL-18 pretreatment on the production of cytokines by splenocytes obtained from CII-immune rats. The spleens were harvested 4 weeks after immunization as described in Fig. 3 and splenocytes cultured with CII for 48 h. Cytokine proteins were quantified in the supernatants by ELISA. IL-18-treated rats produced significantly greater amounts of proinflammatory cytokines (TNF-α and IL-6), Th1 cell cytokines (IL-2 and INF-γ) than controls (P < 0·04 to < 0·02), did but not produce the Th2 cell cytokine IL-10. Values are expressed as mean ± s.e.m. per group.

Fig. 6

Neutralizing antibodies to rIL-18 attenuate arthritis. Both the incidence (a) and severity (b) of arthritis were significantly reduced in BB rats treated with rabbit anti-rIL-18 IgG versus controls given normal rabbit IgG at the time-points shown above by the arrows. Both groups contained six rats. Values are expressed as mean ± s.e.m. per group.

Fig. 7

Neutralizing antibodies to rIL-18 decrease splenic cytokine production in CII-immunized rats. Splenocytes from CII-immune rats shown in Fig. 6 were harvested 4 weeks post-immunization and cultured with CII for 48 h. Cytokine proteins were quantified by ELISA. Anti-rIL-18 IgG-treated rats produced less IFN-γ and IL-6 than controls (P < 0·05 to < 0·03). Values are expressed as mean ± s.e.m. per group.