Anti-inflammatory effects of a new tumour necrosis factor-alpha (TNF-alpha) inhibitor (CNI-1493) in collagen-induced arthritis (CIA) in rats

Abstract

A recently developed compound, a multivalent guanylhydrazone (CNI-1493) that inhibits TNF-alpha production by suppressing TNF-alpha translational efficiency, was administered in an experimental model of collagen type II-induced arthritis in DA rats. CNI-1493 was injected daily intraperitoneally either before the onset of arthritis or after the establishment of clinical disease. Prophylactic treatment with CNI-1493 significantly prevented or delayed the onset and suppressed the severity of arthritis in a dose-dependent manner. Therapeutic intervention with CNI-1493 in established joint disease also resulted in a significant reduction of clinical signs of arthritis in treated animals. No severe side-effects were noted when animals were treated with daily CNI-1493 doses up to 5 mg/kg. An immunohistochemical study was performed which demonstrated that CNI-1493 led to a reduced expression of TNF-alpha at the site of disease activity. Thus, CNI-1493 with documented inhibitory effects on TNF-alpha synthesis, has proven successful in ameliorating the course of arthritis in CIA. We believe that the use of a compound such as CNI-1493 with a defined mode of action provides a useful tool for dissecting and understanding important pathogenic mechanisms operating in the development of chronic arthritis.

Fig. 1

Clinical expression of CIA. Each paw was examined daily for periarticular erythema and oedema and was designated a daily score of 0, 1, 2, 3 or 4 as described in Materials and Methods. Photographs illustrate representative examples of different scores: (A) unaffected forepaw (score 0); (B) forepaw with one type of joint affected (score 1); (C) unaffected hind paw (score 0); and (D) hind paw with maximal score (score 4) with three types of joints affected and maximal oedema.

Fig. 2

Effects in two prophylactic trials of CNI-1493 on the incidence (a,c) and on clinical expression (b,d) of CIA. CNI-1493 was given as a prophylactic treatment administered from day 0. Rats were injected intraperitoeneally once daily with either CNI-1493 or placebo (vehicle) for 30 days. The studied doses of CNI-1493 were 2 mg/kg per day in the first trial and in the subsequent trial two different doses were tried (1.5 mg/kg per day and 5 mg/kg per day). (a,c) Incidence of CIA in the two different trials, which is expressed as percentage of all animals in each group (n = 8 in each group in the first trial and n = 7 in each group in the second trial). All treatment groups differed significantly when compared with placebo. P values compare group means of percentages of animals with arthritis on a daily basis over a 30-day observation period by Wilcoxon signed rank test. (a) P value was 0.0001. (c) P value 0.0002 (placebo versus 1.5 mg/kg per day) and 0.0001 (placebo versus 5 mg/kg per day). Signs of clinical arthritis were scored daily as described in Materials and Methods, with a score from 0 to 4 for each paw. The sum of the scores for the paws was calculated as an arthritis index with a maximum possible value of 16 per animal. (b,d) Mean arthritis indices in the prophylactic trials. Data are expressed as means of seven to eight rats per group. P values compare group means of arthritis indices on a daily basis over a 30-day observation period. (b) P value was 0.0001. (d) P values were 0.0003 (placebo versus 1.5 mg/kg per day) and 0.0001 (placebo versus 5 mg/kg per day), respectively, by Wilcoxon signed rank test. P.i., Post-immunization.

Fig. 3

Effects of CNI-1493 on clinical expression of CIA in two different trials when introduced as a treatment after clinical onset of arthritis. CNI-1493 was administered to animals with a minimal arthritis index of 2. Rats were injected intraperitoneally daily with CNI-1493 or placebo (vehicle). Arthritis indices were scored daily as described in Materials and Methods, with a maximum value of 16 per animal. The graphs represent the mean arthritis indices of all animals in the different treatment groups (n = 8 in the placebo group and n = 7 in each CNI group).(a) CNI-1493 (5 mg/kg per day) was administered either once daily (5 mg × 1) or divided into two injections (2.5 mg × 2). Significant differences were found between the CNI-treated and non-treated groups (P < 0.0002 when placebo was compared with CNI 5 mg × 1 and P < 0.0007 placebo versus CNI 2.5 mg × 2 as judged by Wilcoxon signed rank test). P values compare group means of arthritis indices on a daily basis over a 30-day observation period. (b) CNI-1493 was given intraperitoneally in two different doses of 5 mg/kg per day and 7.5 mg/kg per day, respectively, over a 25-day period. Data are expressed as means of arthritis indices of seven rats per group. P values were 0.0067 (placebo versus 5 mg/kg per day) and 0.0067 (placebo versus 7.5 mg/kg per day), respectively, by Wilcoxon signed rank test. P.i., Post-immunization.

Fig. 4

Immunostaining of synovial tissue in CNI-1493-treated and non-treated DA rats with CIA. Immunostaining of TNF-α (A,C) and MHC II (B,D) in knee joint tissue from rats with CIA. Synovitis, articular cartilage and bone (dark blue staining) are evident in all figures. (A,B) Stainings in consecutive sections from an animal treated daily with CNI-1493 from start of immunization (rat V in Table 3). (C,D) Consecutive sections in tissue from a placebo-treated animal (rat I in Table 3). Both animals were killed on day 21 post-immunization (p.i.), when maximal signs of inflammation occurred in the placebo-treated group of animals. A massive infiltration of MHC II⁺ macrophages was recorded in the synovitis of both the CNI-treated (B) and the placebo-treated (D) rat, Yet TNF-α-expressing cells were only abundant in the placebo-treated animal (C), while the number of these cells was dramatically reduced in the CNI-treated rat (A).