Introduction
Collagen-induced arthritis (CIA) was first described by Trentham et al (1) in both outbred and inbred rat strains following immunization with type II collagen (CII), the major constituent protein of articular cartilage. CIA, an experimental model of autoimmune arthritis, has been shown to exhibit similar histological, immunological and clinical characteristics and genetic linkage to human rheumatoid arthritis (RA) (1,2). CIA has therefore been widely used in laboratories worldwide in studies focusing on the pathogenesis of RA and the preclinical evaluation of novel therapeutics for the disease (2,3).
The immune response to CII and subsequent development of arthritis in rats is linked to the RT1 locus of the rat major histocompatibility complex (4). The arthritogenic potency of CII from different species used for immunization varies; in general, porcine CII is the most potent, followed by bovine and chicken CII (2). As described in numerous studies from Eastern and Southern Asia, CIA is induced in rats predominantly by immunizing the animals with either chicken or bovine CII emulsified in complete Freund's adjuvant (CFA) containing heat-killed Bacillus Calmette-Guérin in liquid paraffin (5–8). The injection is administered in either the subplantar tissues of one of the hind paws (5), at the base of the tail (7,8) or at both the base of the tail and the back (6). These immunization protocols, however, are associated with several problems, one of which is that CFA alone can induce adjuvant arthritis (AIA) in rats (9,10). The use of CFA for the preparation of CII emulsion could therefore produce an immune response to antigens other than CII in the immunized rats. In addition, the injection of the rat footpads with a high concentration of the CII/CFA emulsion puts the injected paws at risk of a severe infection from bacteria contained in the CFA, which could eventually result in the failure of the experiments. Furthermore, intradermal injection at multiple locations at once, such as the base of the tail and the back of the rat, is not only time-consuming and inconvenient to perform, but also requires larger amounts of CII emulsion (0.5–1 ml) for the induction of arthritis.
The aim of the present study was to determine a more simple, specific and efficient method for the induction of CIA in rats. Incomplete Freund's adjuvant (IFA) was used instead of CFA to emulsify bovine CII, and the CII/IFA emulsion was intradermally injected at the base of the tail, with a booster injection administered on day 7 after primary immunization. Using this immunization method, different strains of rats of both genders were compared according to the susceptibility they exhibited to CIA. Finally, the most successfully established CIA model was characterized in terms of clinical, hematological, histopathological and radiological features of human RA, in order to provide scientific evidence for the reliability of the model and for its specific use in studies on the etiopathogenetic mechanisms of RA and for the development of novel anti-arthritic drugs.
Discussion
CIA is an experimental model of autoimmune-mediated arthritis that has been used worldwide in a wide variety of studies, including studies focusing on the understanding of the immunopathogenic mechanisms of arthritis, the analysis of genetic susceptibility factors and the preclinical development of novel therapeutic approaches. CIA can be induced in susceptible strains of rodents (rats and mice) and non-human primates by CII immunization (1,2,14). With regard to the establishment of CIA in rats, which are generally susceptible to AIA, the use of potent immunoadjuvants (e.g. CFA) should be obviated. A more simple and efficient immunization method, which requires smaller amounts of CII emulsion, could be particularly advantageous for research purposes.
In the present study, the susceptibility of different rat strains to immunization with bovine CII emulsified in IFA at the base of the tail was in the following order: SD, Wistar Furth and Wistar (from lowest to highest, respectively) (Table I). SD rats were relatively resistant to CIA and Wistar Furth rats were exhibited a moderate response (Table I). By contrast, female Wistar rats were highly responsive to CIA and exhibited a high incidence and severity of arthritis (Table I). Following primary and booster immunization with a total of 0.2 ml CII/IFA emulsion at the base of the tail, female Wistar rats developed pronounced arthritis with a high incidence and a low variability in clinical signs (Figs. 1 and 2).
In association with the development of arthritis, the ESR in the CIA rats was found to be markedly elevated between days 6 and 24 after CII/IFA immunization, as compared with that in the control rats (Fig. 2D). ESR is a common laboratory test used to aid the definitive diagnosis of RA, the assessment of the disease activity and its response to therapeutics. Studies have reported that the ESR is affected by increases in the plasma levels of acute-phase reactant proteins (e.g. fibrinogen, α- and β-globulin); consequently, the ESR is considered to reflect the acute-phase response to inflammatory diseases (15,16). In this CIA model of female Wistar rats, the ESR peaked on day 12 and underwent a sharp decrease thereafter, which indicated a relatively early production and stimulation of acute-phase proteins during the progression of the disease (Fig. 2D).
RA is initially characterized by the presence of an immune-mediated inflammatory response of the synovial membrane, accompanied by diffuse infiltration of a variety of mononuclear cells and the formation of new vessels. Chronic inflammation and the abnormal proliferation of the synovial membrane ultimately form a destructive pannus, which progressively invades and destroys the cartilage and bone in joints over the course of the disease (17,18). In patients with RA, distinct patterns of bone loss are observed, such as juxta-articular osteopenia adjacent to inflamed joints and the presence of focal erosion within the subchondral bone and at the joint margins in areas of direct pannus invasion (19,20). Bone remodeling also occurs in the arthritic joints (21,22). In the present CIA model of female Wistar rats, histological sections of the ankle joints revealed diffuse cellular infiltration, extensive synovial proliferation, the formation of erosive panni, notable narrowing of the joint space and marked cartilage and bone erosion (Fig. 4). The radiological examination also revealed distinct patterns of bone loss and erosion associated with periosteal new bone formation in the tarsal, metatarsal and interphalangeal regions (Fig. 3). The combination of pronounced bone matrix resorption and heterotopic new bone formation was strongly indicative of the presence of a disordered pattern of bone remodeling in the CIA rats. Thus, the findings in the female Wistar rat CIA model closely resemble the disease characteristics of human RA.
In conclusion, a more specific, simple and efficient method of establishing a CIA rat model delineating the development of human RA has been determined in the present study. An intradermal injection of only 200 µg bovine CII emulsified in IFA at the base of the tail can successfully induce a pronounced autoimmune-mediated polyarthritis in female Wistar rats. This CIA rat model shows a high disease incidence and low variability in clinical signs and closely resembles its human counterpart, RA. This well-developed and well-characterized CIA model in female Wistar rats could therefore be specifically and efficiently used to study the pathological and immunological processes of RA, as well as to test and develop novel anti-arthritic agents.