Sirolimus (rapamycin, Rapamune) and combination therapy with cyclosporin A in the rat developing adjuvant arthritis model: correlation with blood levels and the effects of different oral formulations.

OBJECTIVE AND DESIGN: To determine whole blood levels of sirolimus, a macrolide antibiotic in the rat developing adjuvant arthritis (AA) model after dosing orally with two different vehicles, and whether combinational doses of sirolimus and cyclosporin A (CsA) produced additive or synergistic inhibitory effects in this model. MATERIAL: Male Lewis rats (150-180g). TREATMENT: Arthritis was induced by the injection (0.5 mg/ rat) of heat-killed Mycobacterium butyricum suspended in light mineral oil. Drugs were administered orally either in fine suspension (0.5% Tween 80) or in emulsion (phosal 50 PG in 1% Tween 80) at doses of 0.1 to 5 mg/kg in a 7 day, MWF or daily regimen. METHOD: Paw volumes (ml) were measured by automated mercury plethysmograph and sirolimus concentrations in whole blood were quantitated by liquid chromatography/ mass spectroscopy. RESULTS: At 72h (7 days after adjuvant) after receiving the third oral dose (4.5 mg/kg p.o.), the phosal vehicle resulted in higher sirolimus blood levels (2.5 ng/ml) than in Tween 80 (1.6 ng/ml). After the rats received the last oral dose on day 14, (7 total doses of sirolimus at 4.5 mg/kg) the sirolimus blood levels (2h after the last dose) were about 2 times higher for the phosal dosed rats (9.8 ng/ml) compared to Tween 80 dosed rats (4.6ng/ml). Even 24h after the last dose, sirolimus blood levels were still elevated in the phosal dosed rats (0.8 ng/ml) relative to 0.5% Tween 80 dosed rats (0.5 ng/ml). At day 16 in the rat developing model, sirolimus, when given in phosal vehicle, produced an ED50 of 0.28 mg/ kg (i.e. inhibition of uninjected paw edema) that was about 5.5 times lower than using 0.5% Tween 80 as the suspending agent (ED50 = 1.6mg/kg). When combining sirolimus and CsA using precalculated doses for producing an additive effect in this adjuvant model, an additive inhibitory effect on uninjected paw edema was observed at equal combinational doses of 0.5 and 2 mg/kg, respectively. CONCLUSIONS: The phosal vehicle used in administering sirolimus increases the absorption and whole blood levels in the rat and the elevated blood levels correlated positively with the therapeutic effect in the rat developing AA model. In addition, combination therapy using sirolimus and CsA produced an additive effect in rat developing AA.

Experimental induction of arthritis in LEW rats and antibody response to four Mycoplasma arthritidis strains.

Four Mycoplasma arthritidis strains were examined for differences in virulence for LEW rats and elicitation of antibody responses in the immunoglobulin (Ig) M and G classes and in the four IgG subclasses. Two strains were highly arthritogenic and two were relatively avirulent. When the latter strains did induce arthritis, it was significantly less severe (P less than 0.05) and developed significantly later (P less than 0.001) than in rats injected with the two virulent strains, suggesting that the low-virulence organisms are able to persist asymptomatically in rats for several weeks. None of the M. arthritidis-injected rats developed metabolism-inhibiting (MI) antibodies at any time during the 6-week observation period. Responses to other M. arthritidis antigens from all four strains were measured by enzyme immunoassay (ELISA); they were similar qualitatively but differed quantitatively. Rats injected with the two avirulent strains showed significantly lower titers of IgM antibodies (P less than 0.01) throughout the 6-week observation period and significantly lower early titers of IgG antibodies (P less than 0.05) than rats injected with the two virulent strains. In addition, peak IgM antibody titers, IgM titers measured 1 and 6 weeks after injection and IgG antibody titers measured 1 week after injection all correlated significantly with peak arthritis scores (P less than 0.05). The IgG antibody response against all four strains appeared mostly in the IgG2a and IgG2b fractions, with very little in the IgG1 and IgG2c fractions. Using immunoblotting, the immunodominant antigens of the two virulent strains appeared very similar, but the avirulent strains differed slightly from each other and from the other two. This study indicates that immune responses of rats to virulent and avirulent strains are similar but not identical and that immunogenicity for LEW rats may be a strain-specific characteristic for M. arthritidis.

Development of an experimental animal model for reactive arthritis induced by Yersinia enterocolitica infection.

We attempted to induce experimental arthritis in rats by systematically testing the effect of Yersinia infections in five strains of rats, using the intragastric, intraperitoneal, and intravenous routes of inoculation. We observed that Lewis rats which were given 10(4) to 10(5) Yersinia enterocolitica WA organisms via the intravenous route consistently developed arthritis. The arthritis was most severe at 3 weeks and subsided at 6 weeks. No arthritis was observed when this bacterial strain was administered to Buffalo, Fisher, DA, and LDA rats. No replicable bacteria were detected in the joints. This aseptic characteristic parallels that seen in the human condition and establishes this as an animal model of Yersinia-induced arthritis. The probable reason for arthritis development in only the Lewis rats became apparent when we analyzed the numbers of live bacteria in the spleens and livers of these infected animals. The arthritis-susceptible Lewis rats harbored 10-fold more bacteria than the arthritis-resistant rat strains, and this systemic infection also persisted for a significantly longer period. Speculations as to why human subjects who develop Yersinia-induced arthritis are genetically predisposed have been centered principally around the role of the HLA-B27 histocompatibility antigens. The present study reveals a heretofore unrecognized critical factor: the susceptibility of the hosts to the virulence of the infectious organisms. In addition, the present animal model will provide the necessary tool for the investigation of this and other important host and bacterial factors.