Objective. To examine the role of endogenous interleukin‐4 (IL‐4) and interleukin‐10 (IL‐10) and the therapeutic effect of the addition of IL‐4 and IL‐10 in early and established murine collagen‐induced arthritis (CIA). Methods. Murine recombinant IL‐4, IL‐10, or the combination was given intraperitoneally twice daily from the day of arthritis onset up to 7–10 days of CIA in DBA/1 mice. Anti‐IL‐4, anti–IL‐10, or both antibodies were given intraperitoneally before or after the onset of CIA. The effect of cytokine or anticytokine treatment was monitored visually by macroscopic scoring. Histology and reverse transcription‐polymerase chain reaction (RT‐PCR) analyses were performed at the end of the treatment period. Results. IL‐4 alone did not provoke any effect, IL‐10 slightly suppressed the arthritis, but a more pronounced amelioration was found with the combination. This cooperative effect was noted after early treatment but also occurred when the start of treatment was delayed until 1 week after onset. Apart from suppression of macroscopic signs of inflammation, combined treatment with IL‐4/IL‐10 also reduced cellular infiltrates in the synovial tissue and caused pronounced protection against cartilage destruction. Moreover, levels of mRNA for tumor necrosis factor α (TNFα) and IL‐1 were highly suppressed both in the synovial tissue and in the articular cartilage. In contrast, levels of IL‐1 receptor antagonist (IL‐1Ra) mRNA remained elevated, which suggests that the mechanism of protection may be related to suppressed production of TNFα and IL‐1, with concomitant up‐regulation of the IL‐1Ra/IL‐1 balance. However, accelerated onset of CIA and increased severity could be achieved with neutralizing anti–IL‐10 antibodies. This expression could be further optimized with a combination of anti–IL‐4 and anti–IL‐10 antibodies, although anti–IL‐4 alone was without effect. Conclusion. Our data are consistent with a dominant role of IL‐10 in the natural suppression of arthritis expression, whereas combined treatment with IL‐4 and IL‐10 appears of potential therapeutic value, not only at the onset, but also in established arthritis.

In this study, the major endosomal/lysosomal proteases cathepsin D and cathepsin B were tested on their ability to release T cell stimulatory peptides from hen egg white lysozyme (HEL) in vitro. Whereas neither enzyme could cleave unreduced HEL under mild conditions, reduced HEL was readily cleaved by cathepsin D but not by cathepsin B. Instead, cathepsin B was found to be very active in the trimming of HEL peptides after their release by cathepsin D. Following high-performance liquid chromatography (HPLC) fractionation, cathepsin D-released HEL fragments were screened for recognition by HEL-specific T cells from three strains of mice, i.e. B10.A (H-2a), C57BL/6 (H-2b) and BALB/c (H-2d). Peptides in a large number of different HPLC fractions triggered significant T cell responses in all three strains. Interestingly, the response profiles of T cells from the three different strains showed marked similarities. Also, several individual synthetic HEL sequences corresponding to selected cathepsin D-released fragments were recognized by murine T cells in the context of all three major histocompatibility complex (MHC) haplotypes tested. Our data suggest that cathepsin D rather than cathepsin B may play a central role in the initial release of HEL fragments during endosomal/lysosomal processing. The relatively long HEL fragments released by cathepsin D, containing about 20-30 amino acid residues, are significantly more promiscuous in murine class II MHC binding than the shorter synthetic HEL sequences previously employed by others for the delineation of HEL epitopes. Extensive documentation of HEL epitopes in previous investigations indicate that this promiscuity cannot be explained by simply assuming that longer peptides contain additional epitopes. Rather, an increased peptide length by itself appears to promote promiscuous MHC binding.

We studied the induction of tolerance in female C57B1/6 mice by oral and intravenous (i.v.) administration of protein antigens before immunization. Native proteins [chicken egg albumin (OVA), bovine serum albumin (BSA)] and their cationic derivatives [amidated (a)OVA, aBSA, methylated (m)BSA] were compared in their capacity to suppress cell-mediated immunity (CMI), as measured by a delayed type hypersensitivity test (DTH). Oral feeding of 0.5 mg negative proteins gave a clear suppression. By contrast, cationic derivatives (50 micrograms to 50 mg) did not suppress CMI. Data from cross-experiments, where aOVA was fed in OVA-immune mice, showed no suppression at all. When OVA was fed in aOVA-immune mice, only a partial suppression was achieved. The potency to induce tolerance differed also when the antigens were administered i.v.: 25 micrograms OVA was sufficient to induce a clear suppression, whereas a much higher amount of aOVA (500 micrograms) caused marginal suppression in OVA- and aOVA-immune mice, respectively. Nevertheless, when concentrations of aOVA up to 2.5 mg were tested in a chronic model of arthritis, a significant suppression was achieved. The differences in inducing a CMI suppression may have implications for the feasibility of immunointervention. Successful modulation of human arthritis may be highly dependent on the nature of the antigen involved.

SUMMARYThe induction of tolerance, particularly by intervention before established immunity, is widely accepted. We studied the effects of intravenous (i.v.) administration of hen egg lysozyme (HEL). before as well as after immunization, on a HEL-induced arthritis. Arthritis and also cartilage destruction were almost completely suppressed when 100 ^i% HEL was injected before immunization. Antigen-specific proliferative T cell responses and IL-2 production in vitro were inhibited. Antigenspecific immunogiobulin and IgGl titres were equal in control and tolerized mice, in contrast to lowered IgG2a litres in tolerized animals. Detailed hisiological studies showed that the immune complex-dependent poiymorphonuclear cell phase {<24 h after arthritis induction) was equal for control and HEL-injected mice. Only in the T cell-dependent phase of the arthritis (> 24 h), did suppression become pronounced in tolerized mice. I.V. administration of 100 ^g HEL after immunization could only marginally reduce infiltrate and exudate. and no reduction of cartilage destruction was seen. An elegant way to interfere in an established immunity can be offered by creation of bystander suppresson. We show that i.v. administration of HEL followed by triggering with HEL, at the moment either of immunization or of arthritis induction., does not reduce a methylated bovine serum albumin (BSA)-arthritis. We conclude that arthritis can be suppressed almost totally when HEL is injected intravenously before immunization. Treatment after immunization is less efTeciivc. The i.v. induced suppression is T cell-mediated and antigen-specific: no bystander suppression circuit can be generated.Keywords tolerance intravenous arthritis antigen-specific antigen-driven bystander suppression