Induction of Experimental Arthritis in Balb/C ... - Wiley Online Library

doi: 10.1111/j.1365-3083.2007.02069.x ..................................................................................................................................................................

Induction of Experimental Arthritis in Balb⁄C Mice by Inclusion of a Foreign Protein in the Collagen Inoculum N. F. Ba¨ckstro¨m* & U. I. H. Dahlgren*,

Abstract *Faculty of Odontology, Section of Oral Microbiology and Immunology and Department of Rheumatology and Inflammation Research, The Sahlgrenska Academy at Go¨teborg University, Go¨teborg, Sweden Received 2 October 2007; Accepted in revised form 29 November 2007 Correspondence to: Ulf Dahlgren, Department of Rheumatology and Inflammation Research, Guldhedsgatan 10, 413 46 Go¨teborg, Sweden. E-mail: [email protected]

The susceptibility of mice to collagen-induced arthritis (CIA) has, among other things, been linked to the major histocompatibility complex class II genes as well as other genes. This study was designed to examine the possibilities to establish CIA in low susceptible I-Ad (Balb ⁄ C) mice. Balb ⁄ C mice were immunized twice with bovine type II collagen (BCII) in complete Freund’s adjuvant (CFA) containing the amount of Mycobacterium tuberculosis needed to induce CIA in low susceptible I-Ab (C57BL ⁄ 6) mice. Some mice received the conceivable arthritogenic inoculum mixed with ovalbumin (OVA). Clinical arthritis was monitored. Antibody activity and T-cell reactivity to BCII were determined. Unexpectedly, only mice that were immunized with the BCII– OVA mixture developed arthritis. Combining BCII with another foreign protein, keyhole limpet hemocyanin, but not the self-protein mouse serum albumin, also triggered arthritis. Prior to the appearance of arthritis the serum levels of IgG autoantibodies to BCII were higher in the coimmunized mice than in the mice that were immunized with BCII alone. Yet, splenocytes stimulated in vitro with BCII did not proliferate or produce interferon-c. Immunization of Balb ⁄ c mice with an emulsion-containing CFA and BCII mixed with a foreign body, but not a self-protein, elevates the level of circulating autoantibodies to CII and subsequently induces arthritis.

Introduction The most widely used model to study the pathogenesis of rheumatoid arthritis (RA) is collagen-induced arthritis (CIA) in mice. It is histologically comparable to human RA and elicited in genetically susceptible strains of mice by immunization with native type II collagen (CII) emulsified in complete Freund’s adjuvant (CFA) [1]. Although the precise pathogenesis of murine CIA is unknown, induction of arthritis requires both B- and T-cell responses to CII [2]. In accordance with human RA, susceptibility to CIA has been associated with the expression of certain major histocompatibility complex (MHC) class II molecules. Mouse strains expressing the class II molecule I-Aq (e.g., DBA ⁄ 1 and B10.Q) or I-Ar (e.g., B10.RIII) have been shown to be the most susceptible strains [3]. Mice expressing the class II molecule I-Ad (e.g., Balb ⁄ C) or I-Ab (e.g., C57BL ⁄ 6) are regarded as resistant strains [2].

However, a study made by Campbell et al. showed that arthritis can be induced in C57BL ⁄ 6 mice by an increase of the bacterial content in the adjuvant [4]. The present study aimed to trigger CIA in low susceptible Balb ⁄ C mice by an immunization protocol similar to the protocol used for arthritis induction in C57BL ⁄ 6 mice. Simultaneously, we wanted to examine the impact on the presumed arthritis development of inclusion of a foreign protein in the CII inoculum. The rationale for this was that we have recently shown that immunization with a mixture of ovalbumin (OVA) and CII modulated the CII-specific B- and T-cell responses and to some extent reduced the severity of CIA in DBA ⁄ 1 mice [5].

Materials and methods Animals. Seven- or 9-week-old male Balb ⁄ C mice were purchased from M&B (Balb ⁄ CJ, Ry, Denmark) or 2008 The Authors

322

Journal compilation 2008 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 67, 322–328

N. F. Ba¨ckstro¨m & U. I. H. Dahlgren Induction of Experimental Arthritis in Balb ⁄ C Mice 323 ..................................................................................................................................................................

Scanbur BK (Balb ⁄ C ⁄ Bkl, Sollentuna, Sweden). Nineweek-old male DBA ⁄ 1 mice were purchased from Charles River (DBA ⁄ 1JCrl, Sulzfeld, Germany). The mice were housed under standard conditions in the animal facilities of the laboratory of Experimental Biomedicine (Go¨teborg University, Go¨teborg, Sweden). The Go¨teborg ethical committee on animal experiments approved the experimental protocol. Antigen preparations and immunization protocols. Bovine type II collagen (BCII, Chondrex, Redmond, WA, USA) was solubilized in 0.1 M acetic acid (HAc) by gentle stirring over night, at 4 C. Ovalbumin (OVA, Grade V), mouse serum albumin (MSA, Fraction V), keyhole limpet hemocyanin (KLH, lyophilized with stabilizing buffer), and lipopolysaccharide (LPS, from Escherichia coli O127:B8) (all from Sigma, St Louis, MO, USA) were dissolved in phosphate-buffered saline (PBS). The endotoxin level in the OVA preparation was quantified by the Limulus amoebocyte lysate assay, and estimated to be 0.05% of the OVA weight (by the Department of Clinical Bacteriology, at the Sahlgrenska Academy). Complete Freund’s adjuvant (CFA) was prepared by mixing grinded heat-killed Mycobacterium tuberculosis H37Ra (Difco, Detroit, MI, USA) in incomplete Freund’s adjuvant (IFA; Sigma) to a final concentration of 5 mg ⁄ ml. Nine-week-old mice were subcutaneously (s.c.) immunized at the tail base with 50 ll PBS containing BCII (100 lg) or OVA (100 lg) or KLH (100 lg) or BCII (100 lg) mixed with OVA (100 lg), MSA (100 lg), KLH (100 lg) or LPS (5 lg or 0.05 lg), and an equal volume of CFA. The same emulsion was injected at day 21and at day 45, 70 or 72 but in IFA instead of CFA at day 70. Because of toughening of the skin around the primary injection site, booster injections were located s.c. at the lower portion of the back. One group of mice was immunized s.c. with BCII and KLH at separate sites. In that group of mice, the primary injection of BCII was located at the tail base and the booster injections at the lower portion of the back while both KLH injections were located at the nape. All primary injections were done under isofluran anaesthesia. Clinical and histological evaluation of arthritis. After the booster injection at day 21, the incidence and severity of arthritis were determined at several time points, as indicated in the figures, by visual examination of each paw. The severity was scored from 0 to 3 (no disease, mild swelling or redness, swelling and redness, severe swelling and redness). Each paw was graded, so that a mouse could achieve a maximum score of 12. Hind limbs were dissected, fixed and decalcified 1 week after the second booster at day 70. Tissue sections were cut and stained with haematoxylin and eosin. The joints (tarsal, ankles and knees) were blindly examined and scored from 0 to 3 (normal, mild, moderate and severe) with regard to

synovial hypertrophy and cartilage ⁄ subchondral bone destruction. Measurement of splenocyte proliferation and interferon-c secretion. Splenocytes were prepared 1 week after the booster immunization at day 21 and the BCII-specific proliferation was measured by [3H]-thymidine incorporation, as previously described [5]. In brief, 100 ll individual splenocyte suspensions (5 · 106 cells ⁄ ll) were plated in triplicate microtitre plate wells in the presence of 100 ll medium or 100 ll medium containing heat denatured BCII (200 lg ⁄ ml). The culture medium consisted of Dulbecco’s Modified Eagle’s Medium, supplemented with 5% fetal calf serum, penicillin (100 U ⁄ ml), streptomycin (100 lg ⁄ ml) and gentamycin (50 lg ⁄ ml; all from Gibco BRLTM, Life Technologies, Invitrogen AB, Lidingo¨, Sweden). Cells were incubated at 37 C in 5% CO2 for 4 days, and 1 lCi ⁄ well of [3H]-thymidine was added to the culture for the last 24 h. Incorporated radioactivity was counted in a scintillation beta counter (MicroBeta TriLux, Wallac, Turku, Finland). Supernatants were collected after 72 h and stored at )20 C until the interferon (IFN)-c level was determined with a commercial sandwich enzyme-linked immunosorbent assay (ELISA) kit (DuoSet ELISA kit; R&D Systems Inc, Minneapolis, MN, USA). Splenocytes obtained from BCII primed DBA ⁄ 1 mice were run in a parallel experiment as positive controls. Measurement of serum levels of antigen-specific antibodies. Sera were collected 1 week after the booster immunization at day 21. Antibodies to BCII were measured by an ELISA as previously described [5]. Briefly, serially diluted serum samples were added to BCII-coated microtitre plates and incubated overnight in 4 C. After washings, biotin-conjugated goat anti-mouse IgG (Jackson Immuno Research Laboratories, Inc., West Grove, PA, USA), rat anti-mouse IgG1, IgG2a or IgG2b antibodies (BD Pharmingen, Heidelberg, Germany) were added. Plates were incubated for 2 h, washed and incubated with extravidin-alkaline phosphatase (Sigma E2636) for another 2 h. Then the plates were washed and the substrate para-nitrophenyl (1 mg ⁄ ml; 104 Sigma) dissolved in diethanolamine buffer (pH 9.8) was added. The absorbance was read on a Spectra MAX 340 at 405 nm (Molecular Devices, Sunnyvale, CA, USA). All incubations were performed in humid atmosphere. The antibody activity was expressed as arbitrary ELISA units calculated by interpolation in the linear range of a standard curve made of a pool of positive serum samples run on each plate. Statistical analysis. The Mann–Whitney U-test was used for statistical analyses of differences in antibody activities. The student’s t-test was used for comparison of spleen cell proliferation and IFN-c secretion in vitro. STATVIEW for Macintosh was used for statistical calculation. P-values less than 0.05 were considered significant.

2008 The Authors Journal compilation 2008 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 67, 322–328

324 Induction of Experimental Arthritis in Balb ⁄ C Mice N. F. Ba¨ckstro¨m & U. I. H. Dahlgren ..................................................................................................................................................................

Results

3

Balb ⁄ C mice were immunized twice, three weeks apart, with BCII or BCII mixed with OVA in CFA containing 5 mg killed M. tuberculosis per ml. The mice that were immunized with BCII alone failed to develop clinical arthritis (Fig. 1). Instead, animals immunized with BCII mixed with OVA developed progressive arthritis (Fig.1A) with high incidence (Fig.1B). Control mice receiving the adjuvant with PBS or OVA alone did not show any signs of clinical arthritis (Fig. 1). Histological analysis of the joints in the hind limbs showed typical features of arthritis in all coimmunized mice but not in mice that were immunized with BCII alone (Fig. 2). The histological signs of arthritis were characterized by marked synovial and periarticular inflammation with extensive leukocyte infiltration, synovial hyperplasia, pannus formation and erosion of cartilage and bone (Fig. 3). Leukocytes together with degenerative products of the bone structure were also found in the articular cavity (Fig. 3). Coimmunized animals developed clinical arthritis if the mixture of BCII and OVA was emulsified in CFA but not if the same mixture was emulsified in IFA (Table 1, Expt. 1).

Mean arthritic score

A

7 6 5 4 3 2 1 0 21

25

Vehicle

BCII

BCII + OVA

OVA

29

33

39

43

47

53

70

Day after primary immunization

Incidence %

B 100

80 60 40 20 0 21 25

29 33

39

43 47

53

70

Day after primary immunization Figure 1 (A) Clinical progression of arthritis in Balb ⁄ C mice. The mice were immunized with PBS, ovalbumin (OVA), bovine type II collagen (BCII) or BCII mixed with OVA, emulsified in complete Freund’s adjuvant. The same injection was given 21 days later. Each symbol represents the mean arthritic score (n = 7–8) at the indicated day. (B) Incidence of arthritis in the same mice and the same time point as in (A).

Histopathology score

2.5

Induction of arthritis in Balb ⁄ C mice by inclusion of a foreign protein in the BCII inoculum

Synovitis Cartilage erosion Bone erosion

2 1.5 1 0.5 0 Vehicle

BCII BCII + OVA Immunization

OVA

Figure 2 Histological analysis of synovitis, cartilage erosion and bone erosion in Balb ⁄ C mice immunized with PBS, bovine type II collagen (BCII), ovalbumin (OVA) or BCII mixed with OVA. The antigen preparations were emulsified in complete Freund’s adjuvant and injected twice, 3 weeks apart. A second booster injection was given at day 70, with the same antigen(s), but at that time emulsified in incomplete Freund’s adjuvant. One week after the second booster immunization, hind joints were obtained and stained with hematoxylin and eosin. Signs of joint histopathology were blindely scored on all animals on a scale of 0–3. Each symbol represents the individual median score of the examined joints (tarsal, ankles and knees).

Commercially available OVA is contaminated with endotoxins. Table 1 (Expt. 1) shows that addition, to the BCII emulsion, of an amount of lipopolysaccharide (LPS) corresponding to the endotoxin activity measured in the OVA preparation, or even hundred times more, did not provoke arthritis in the animals. The animals immunized with any combination of OVA responded with OVA-specific T-cell proliferation in vitro as well as anti-OVA antibody production in vivo (not shown). Next we examined the outcome of immunizations with BCII mixed with KLH or a self-protein, MSA. Mice that were immunized with a mixture of BCII and KLH developed clinical arthritis with high incidence, whereas immunization with BCII and MSA did not lead to arthritis (Table 1, Expt. 2). The arthritis in the KLH and BCII coimmunized mice appeared earlier than in the OVA and BCII coimmunized animals (not shown). Mice that were immunized with BCII and KLH at different sites failed to develop clinical arthritis (not shown). The animals immunized with any combination of KLH responded with KLH-specific T-cell proliferation in vitro as well as anti-KLH antibody production in vivo (not shown). Increased levels of anti-BCII antibodies but no BCII-specific splenocyte proliferation or IFN-c secretion in coimmunized Balb ⁄ C mice

Mice were immunized twice, three weeks apart, with an emulsion containing CFA and either PBS, BCII, KLH or 2008 The Authors


N. F. Ba¨ckstro¨m & U. I. H. Dahlgren Induction of Experimental Arthritis in Balb ⁄ C Mice 325 .................................................................................................................................................................. B

A

C

Figure 3 Representative histology of knee joints in the Balb ⁄ C mice that are described in Fig. 2. No signs of histopathology were observed in mice immunized with BCII alone (A). Extensive cellular infiltration and, synovial hyperplasia were observed in mice coimmunized with BCII mixed with ovalbumin (B). Pannus formation and erosion of cartilage and bone were detected as well (C). Degenerative products of the articular structure together with leukocytes were also found in the articular cavity (B and C). F, femur; m, meniscus; t, tibia.

Treatment protocol Expt.

Adjuvant

Antigen(s)

Incidence of arthritis

1

CFA

PBS BCII BCII BCII BCII BCII PBS BCII BCII BCII BCII

0⁄8 0⁄8 5⁄8 0⁄8 0⁄8 0⁄8 0⁄8 0⁄8 5⁄8 0⁄8 7⁄8

2

IFA CFA + LPS (0.05 lg) CFA + LPS (5 lg) CFA

+ OVA + OVA

+ OVA + MSA + KLH

Dayb 90

80

5.5

AntiBCII antibodies (log arbitrary units)

Table 1 Incidence of clinical arthritis in Balb ⁄ C mice after different immunization protocols.a

A

IgG

B IgG1

C IgG2a

D IgG2b

**

**

5 4.5 4 3.5 3 2.5 2 BCII

BCII + KLH

45

BCII

BCII + KLH

BCII

BCII + KLH

BCII

BCII + KLH

Immunization

a

Incidence of clinical arthritis in Balb ⁄ C mice after immunizations with either PBS, bovine collagen type II (BCII), BCII + ovalbumin (OVA), BCII + mouse serum albumin (MSA) or BCII + keyhole limpet hemocyanin (KLH). The antigen(s) were emulsified in complete Freund’s adjuvant (CFA), CFA + various concentrations of lipopolysaccharide (LPS), or incomplete Freund’s adjuvant (IFA). The mice were booster immunized 3 weeks after the primary immunization. In experiment (Expt.) 1, all groups of mice received a second booster injection at day 72. In Expt. 2, all groups of mice except mice coimmunized with BCII + KLH received a second booster injection at day 45. Each mouse was injected with an equally prepared antigen-adjuvant emulsion at all immunization occasions. b The day the arthritis was scored.

BCII mixed with KLH. One week after the booster immunization, i.e. prior to the appearance of clinical arthritis, mice were killed and serum and splenocytes were recovered. Anti-BCII antibody activities in serum were measured and BCII-specific splenocyte proliferation and IFN-c secretion in vitro was determined. The levels of IgG, IgG1, IgG2a and IgG2b anti-BCII antibodies were significantly higher in coimmunized mice than in mice immunized with BCII alone (Fig. 4). If BCII and KLH were injected at separate sites the serum IgG anti-BCII antibody activity remained at the

Figure 4 Increased serum (A) IgG, (B) IgG1, (C) IgG2a and (D) IgG2b anti-bovine type II collagen (BCII) antibody activity in Balb ⁄ C mice, 1 week after the booster immunization. The mice were immunized with BCII or BCII mixed with keyhole limpet hemocyanin (KLH), emulsified in complete Freund’s adjuvant. The same injection was given 21 days later. Data are presented as box plots showing median, interquartile range, 10th and 90th percentiles and extremes. Mann–Whitney U test was used for statistical analyses. **P < 0.01.

same low level as in mice that were immunized with just BCII (not shown). Control mice receiving the adjuvant emulsified with PBS or KLH alone did not show any signs of BCII-specific serum antibody activities (not shown). The BCII-specific proliferation was similar, regardless of immunization protocol, in cultures of splenocytes collected from Balb ⁄ C mice (Fig. 5A). The splenocytes did not secret IFN-c in response to BCII stimulation in vitro (Fig. 5B). In contrast, splenocytes collected from BCII immunized DBA ⁄ 1 mice showed a significantly increased BCII-specific proliferation and IFN-c secretion compared with control immunized DBA ⁄ 1 mice or Balb ⁄ C mice immunized with BCII alone (Fig. 5A,B). The KLH-specific splenocyte proliferation was significantly increased in coimmunized Balb ⁄ C


326 Induction of Experimental Arthritis in Balb ⁄ C Mice N. F. Ba¨ckstro¨m & U. I. H. Dahlgren .................................................................................................................................................................. A

Discussion

3.5 Immunization

BCll specific proliferation (proliferation index)

3 2.5

***

Vehicle BCII BCII + KLH

2

KLH

1.5 1 0.5 0 Balb/C

B

DBA/1

700 BCll specific IFN-γ secretion (pg/ml)

Immunization

600 500

***

Vehicle BCII BCII + KLH

400

KLH

300 200 100 0 Balb/C

DBA/1

Figure 5 Undetectable BCII-specific spleen cell (A) proliferation and (B) IFN-c secretion in vitro in Balb ⁄ C mice, 1 week after the booster immunization. The Balb ⁄ C mice (left side of dashed line) were immunized twice, 3 weeks apart, with PBS, bovine type II collagen (BCII), keyhole limpet hemocyanin (KLH) or BCII mixed with KLH, emulsified in complete Freund’s adjuvant. DBA ⁄ 1 mice (right side of dashed line) were immunized in the same manner with either PBS or BCII. The proliferation results are presented as proliferation indexes (mean cpm in triplicate cultures stimulated with BCII ⁄ mean cpm in triplicate control cultures). The BCII-specific IFN-c secretion is determined by subtracting the IFN-c level measured in supernatants of the control culture from the IFN-c level measured in supernatants of the BCII stimulated culture. The bars represent mean ± SEM (n = 6–8). Statistical analysis was done with the Student’s t-test. ***P < 0.001 versus PBS immunized DBA ⁄ 1 mice and BCII immunized Balb ⁄ C mice.

mice compared with splenocytes obtained from negative control immunized animals (not shown). Identically treated Balb ⁄ C mice kept alive, coimmunized with BCII and KLH, started to develop arthritis about 1 week after detection of the enhanced serum level of anti-BCII antibodies (not shown).

In the present study, Balb ⁄ C mice were immunized with BCII in CFA containing the amount of M. tuberculosis needed to induce arthritis in C57BL ⁄ 6 mice [4]. Both Balb ⁄ C and C57BL ⁄ 6 mice belong to strains of mice regarded to have low susceptibility to CIA [2]. Our results demonstrated that if a foreign protein antigen was included in the BCII inoculum, Balb ⁄ C mice developed arthritis with high incidence. BCII and a high content of M. tuberculosis were not sufficient to induce arthritis in the Balb ⁄ C mice. This result was unexpected as previous studies have shown that inclusion of OVA in an arthritogenic BCII inoculum delays the progression of arthritis in CIA susceptible DBA ⁄ 1 mice [5] and protects DA rats from arthritis [6]. The arthritis induction in the coimmunized Balb ⁄ C mice was not caused by endotoxin activities contaminating the commercially available OVA preparation, as mice that were immunized with BCII mixed with LPS of corresponding endotoxin activity did not develop clinical arthritis. Yet, emulsifying the antigens in CFA was required to induce clinical arthritis in the mice as no signs of clinical arthritis were observed in animals that were coimmunized with the antigens emulsified in IFA. Compared to arthritis in DBA1 mice, the main difference in the Balb ⁄ C mice was that the arthritis appeared a somewhat later and the arthritis score was a little lower than in the DBA1 mice [5]. The onset and incidence of CIA after BCII and OVA immunization varied between experiments. In one experiment, mice developed arthritis with high incidence after just one booster injection while in two other experiments the mice needed two booster injections. Yet, none of the mice that received BCII alone developed arthritis. It is known that in the highly susceptible DBA ⁄ 1 mice, the time of onset, severity and incidence of CIA can vary even though the immunization protocol is identical in repeating experiments. Environmental factors such as health status, stress and behaviour beside genetic factors such as the MHC haplotype contribute to CIA susceptibility. For instance, inter-male aggressiveness enhance responsiveness to CIA [7]. Thus, differences in the stability of the social structure might explain the observed temporal discrepancy between our experiments. We further observed that inclusion of another foreign protein KLH, but not the self-protein MSA, elicited clinical arthritis in the mice as well. Mixing KLH and BCII was necessary to induce arthritis as immunization with BCII and KLH at separate sites failed to produce clinical arthritis. The group of mice that was immunized with BCII and KLH developed arthritis with high incidence after just one booster injection whereas mice that were immunized with BCII and OVA most often needed two booster injections. KLH is regarded as a strong immunogen, more so than OVA, which suggests that this 2008 The Authors


N. F. Ba¨ckstro¨m & U. I. H. Dahlgren Induction of Experimental Arthritis in Balb ⁄ C Mice 327 ..................................................................................................................................................................

property of the added antigen may be related to the arthritis induction. Thus, KLH seems to be more efficient in terms of arthritis induction than OVA. It has been well documented that autoantibodies to CII play an important role during arthritis induction in CIA susceptible mouse strains. For example, B-cell deficient mice are resistant to CIA [8] and a severe course of murine arthritis is correlated with high anti-CII antibody levels in sera [9]. In addition, anti-CII-specific antisera and monoclonal anti-CII antibodies induce arthritis in naive recipients, independently of the recipients MHC haplotype [10, 11]. Furthermore, transfer of CII-specific monoclonal antibodies to Balb ⁄ C mice induce arthritis in a dose-dependent manner that eventually reaches an upper limit [12]. These observations fits with our results showing an increased IgG anti-BCII antibody activity prior to the appearance of arthritis in mice immunized with BCII together with KLH. The observed enhancement in the levels of both Th1 (IgG2a and IgG2b) and Th2 (IgG1) associated IgG anti-BCII antibody subclasses imply that a non-polarizing mechanism causes the high IgG anti-BCII antibody response. Injections of BCII and KLH at separate sites failed to boost the IgG anti-BCII response. A potential involvement of autoreactive T cells has been at focus in several studies concerning the pathogenesis of CIA. The induction phase of the disease has been shown to be dependent on CD4+ T cells because depletion of these cells around the time of CII immunization prevents arthritis induction [13, 14]. CII primed T cells show a Th1 cytokine-secreting pattern after CII restimulation in vitro, with a marked increase of IFN-c production that peaks around onset of the disease [15, 16]. Conversely in a study on Balb ⁄ C mice, it was shown that inhibition of IFN-c production was a prerequisite for arthritis development in this mouse strain [17]. In the present study, the Balb ⁄ C splenocytes did not proliferate or produce IFN-c in response to BCII, irrespective of the immunization protocol. This was in sharp contrast to splenocytes obtained from BCII-primed DBA ⁄ 1 mice. Whether this reflects an actual absence of BCII-reactive T-cell clones in Balb ⁄ C mice or a downregulated BCIIspecific T-cell response needs to be further examined by refined BCII-specific T-cell assays. It is known that production of arthritogenic antibodies in CIA is dependent on the help from T cells [8]. This implies that BCII-specific T cells in BCII-primed Balb ⁄ C mice are fewer and ⁄ or proliferate with different kinetics than T cells in BCII-primed DBA ⁄ 1 mice after restimulation or that they supply the help without proliferation. Another explanation could be that non-BCII-specific T cells promoted the autoantibody production. One can speculate that T cells that reacted with KLH peptides directly or indirectly promoted production of autoantibodies to CII and that this triggered the arthritis

development. It might be that the KLH and BCII molecules aggregate or are in such close vicinity to each other that they are endocytosed together by the same BCII-specific B cell. Hence, both molecules are degraded and presented as peptides on class II MHC molecules on the same anti-BCII-specific B cell. This would make it possible for KLH-specific T cells to help the anti-BCII-specific B cell to produce anti-CII antibodies. The hypothesis that KLH and BCII are engulfed together by the same B cell, agrees with a previous study on Lewis rats. Immunization with rat CII associated with porcine pepsin activates anti-CII-producing B cells and induces arthritis in Lewis rats [18]. The rats showed a minute immune response to CII after immunization with pepsin-free rat CII. Since it has been shown that pepsin and collagen form a stable complex [19], the authors concluded that the activation of the anti-CII-producing B cells depended on heteroreactive T cells recognizing pepsin peptides presented by the same B cell. However, in contrary to our results the study further demonstrated that T cells proliferated and produced IFN-c in response to CII prior to clinical arthritis in the rats [18]. In summary, herein we describe a novel model of experimental arthritis in Balb ⁄ C mice. The induction of the disease seemed to be linked to increased autoantibody activities to CII but not to increased T-cell anti-CII responses. Further investigations in this model may lead to the discovery of genes other than MHC that participate in the disease. Such knowledge could be helpful in the attempts to understand the complexity of human RA.

Acknowledgment This work was supported by grants from the Swedish Research Council; grant number K2003-74X-1405403A, the Swedish Rheumatism Association, the Go¨teborg Rheumatism Association and King Gustaf V’s 80-year Foundation.

References 1 Myers LK, Rosloniec EF, Cremer MA, Kang AH. Collagen-induced arthritis, an animal model of autoimmunity. Life Sci 1997;61:1861– 78. 2 Seki N, Sudo Y, Yoshioka T et al. Type II collagen-induced murine arthritis. I. Induction and perpetuation of arthritis require synergy between humoral and cell-mediated immunity. J Immunol 1988;140:1477–84. 3 Brand DD, Kang AH, Rosloniec EF. Immunopathogenesis of collagen arthritis. Springer Semin Immunopathol 2003 Aug;25:3–18. 4 Campbell IK, Hamilton JA, Wicks IP. Collagen-induced arthritis in C57BL ⁄ 6 (H-2b) mice: new insights into an important disease model of rheumatoid arthritis. Eur J Immunol 2000;30:1568–75. 5 Backstrom NF, Dahlgren UI. Bystander suppression of collageninduced arthritis in mice fed ovalbumin. Arthritis Res Ther 2004;6:R151–60.


328 Induction of Experimental Arthritis in Balb ⁄ C Mice N. F. Ba¨ckstro¨m & U. I. H. Dahlgren .................................................................................................................................................................. 6 Lorentzen JC, Erlandsson H, Mussener A et al. Specific and longlasting protection from collagen-induced arthritis and oil-induced arthritis in DA rats by administration of immunogens. Scand J Immunol 1995;42:82–9. 7 Holmdahl R, Bockermann R, Backlund J, Yamada H. The molecular pathogenesis of collagen-induced arthritis in mice–a model for rheumatoid arthritis. Ageing Res Rev. 2002 Feb;1:135–47. 8 Svensson L, Jirholt J, Holmdahl R, Jansson L. B cell-deficient mice do not develop type II collagen-induced arthritis (CIA). Clin Exp Immunol 1998;111:521–6. 9 Wooley PH, Luthra HS, Stuart JM, David CS. Type II collageninduced arthritis in mice. I. Major histocompatibility complex (I region) linkage and antibody correlates. J Exp Med 1981;154:688–700. 10 Stuart JM, Dixon FJ. Serum transfer of collagen-induced arthritis in mice. J Exp Med 1983;158:378–92. 11 Terato K, Hasty KA, Reife RA, Cremer MA, Kang AH, Stuart JM. Induction of arthritis with monoclonal antibodies to collagen. J Immunol 1992 Apr 1;148:2103–8. 12 Nandakumar KS, Svensson L, Holmdahl R Collagen type II-specific monoclonal antibody-induced arthritis in mice: description of the disease and the influence of age, sex, and genes. Am J Pathol 2003 Nov;163:1827–37.

13 Ranges GE, Sriram S, Cooper SM. Prevention of type II collageninduced arthritis by in vivo treatment with anti-L3T4. J Exp Med 1985;162:1105–10. 14 Chiocchia G, Boissier MC, Fournier C Therapy against murine collagen-induced arthritis with T cell receptor V beta-specific antibodies. Eur J Immunol 1991 Dec;21:2899–905. 15 Mauri C, Williams RO, Walmsley M, Feldmann M Relationship between Th1 ⁄ Th2 cytokine patterns and the arthritogenic response in collagen-induced arthritis. Eur J Immunol 1996 Jul;26:1511–8. 16 Doncarli A, Stasiuk LM, Fournier C, Abehsira-Amar O. Conversion in vivo from an early dominant Th0 ⁄ Th1 response to a Th2 phenotype during the development of collagen-induced arthritis. Eur J Immunol 1997;27:1451–8. 17 Ortmann RA, Shevach EM. Susceptibility to collagen-induced arthritis: cytokine-mediated regulation. Clin Immunol 2001 Jan;98:109–18. 18 Vingsbo C, Larsson P, Andersson M, Holmdahl R Association of pepsin with type II collagen (CII) breaks control of CII autoimmunity and triggers development of arthritis in rats. Scand J Immunol 1993 Mar;37:337–42. 19 Zlabinger GJ, Menzel EJ, Steffen C Induction of anti-pepsin antibodies after immunization with pepsin-extracted collagen. Matrix 1989 Mar;9:135–9.
