homologous type ii collagen—induced arthritis in rats

693

HOMOLOGOUS TYPE I1 COLLAGEN-INDUCED ARTHRITIS IN RATS

Characterization of the Disease and Demonstration of Clinically Distinct Forms of Arthritis in Two Strains of Rats after Immunization with the Same

Collagen Preparation

PER LARSSON. SANDRA KLEINAU, RIKARD HOLMDAHL. and LARS KLARESKOG

Arthritis was induced in Lewis and DA rat strains after immunization with native homologous type I1 collagen (CII). Differences were noted in the clinical signs of arthritis in the 2 rat strains, which were immunized with the same arthritogenic preparation of CII. DA rats showed disease onset characterized by symmetric involvement of the interphalangeal joints of the hind feet. Lewis rats showed disease onset characterized by involvement of only the ankle or knee joints. Moreover, the arthritis tended to be chronic (i.e., persistent and variable redness and swelling seen in interphalangeal joints) in DA rats, but not in Lewis rats. Analysis of the delayed-type hypersensitivity reaction to CII and anti-CII autoantibody production demonstrated the presence of T cell, as well as B cell, reactivity to rat CII in both strains of rat. A spontaneous T cell reactivity to CII, as measured by rat CII-induced ear swelling, was observed in nonimmunized DA rats, but not in nonimmunized Lewis rats. The demonstration that clinical features of arthritis induced with identical arthritogenic stimuli are different depending on the genetic background of the affected animal may be relevant in

From the Department of Medical and Physiological Chem- istry, University of Uppsala, and the Department of Inflammation Research, Pharmacia Leo Therapeutics AB. Uppsala. Sweden.

Supported by grants from the Swedish Medical Research Council. the Swedish Agency for Technical Development, King Gustaf V 80-Year Foundation, the Professor Nanna Svartz Research Foundation, and the Swedish Medical Association.

Per Larsson, PhD. MD: University of Uppsala; Sandra Kleinau. MS: University of Uppsala and Pharmacia Leo Therapeu- tics AB; Rikard Holmdahl. PhD. MD: University of Uppsala; Lars Klareskog, PhD. MD: University of Uppsala.

Address reprint requests to Per Larsson. PhD, MD. Depart- ment of Medical and Physiological Chemistry, Box 575, Uppsala S-751 23. Sweden.

Submitted for publication March 13. 1989; accepted in revised form October 16. 1989.

understanding the heterogeneity of the clinical features of human inflammatory joint disease.

Collagen-induced arthritis is the designation for a group of experimental arthritides that can be induced in several animal species, including mice, rats. and primates, after immunization with either heterologous or homologous type I1 collagen (CII) (1-6). Certain variants of this disease, especially the arthritides induced in rats or mice with heterologous CII (1-3.6-1 I ) , have often been used as models for rheumatoid arthritis (RA). These disease variants have been used in attempts to understand basic pathophysiologic mechanisms that may be similar in collagen-induced arthritis and RA, and to evaluate the potential value of new therapeutic principles that may be considered for use in the treatment of RA. The value of the collagen- induced arthritis model, however, has been questioned on several grounds. One fundamental problem is the obvious lack of knowledge of whether immunity to CII is of pathogenetic importance in RA. Since anti- CII-reactive lymphocytes and anti-CII antibodies can be demonstrated in the joints of a majority of RA patients ( 1 2,13), the most reasonable way of determining whether anticollagen immunity is of pathogenetic importance is to selectively eliminate this reactivity in RA patients, and analyze the effect of its absence on the disease state.

Homologous CII-induced arthritis has previously been described in DBNI mice, in which the arthritis was characterized by a variable and delayed onset, and a slow and progressive development (3.14.15). However, the study of homologous CII- induced arthritis in mice is complicatcd by major problems concerning the preparation and characteri-

Arthritis and Rheumatism, Vol. 33, No. 5 Way 1990)

694 LARSSON ET AL

zation of the mouse CII. Therefore, we decided to investigate homologous CII-induced arthritis in rats. This disease should be inducible in rats with the use of chondrosarcoma-derived CII, and therefore, should provide a reproducible model for studies using well- characterized rat collagen preparations. Homologous CII-induced arthritis in rats has been briefly described ( I ) , but no extensive clinical or immunologic characterization of this disease has been reported.

We demonstrated that homologous CII- induced arthritis in the DA rat strain displays a number of clinical characteristics, such as early symmetric involvement of proximal interphalangeal (PIP) joints as well as a chronic and remitting course, that has not been previously described for heterologous CII- induced arthritis in rats. In addition, we observed that immunization of Lewis rats with the same collagen preparation gave rise to a clinical picture quite distinct from that seen in DA rats, i.e., the involvement of primarily the ankles and a lack of chronicity.

Our investigation focused on the clinical and immunologic characterization of homologous CII- induced arthritis in the DA and Lewis strains of rat. The study was conducted with the view that homologous CII-induced arthritis in rats, as well as homologous CII-induced arthritis in mice, may prove to be better models of RA than the previously used variants of heterologous collagen-induced arthritis in rats and mice, in terms of both clinical characteristics and potential development of methods for regulating anti- CII autoimmunity.

MATERIALS AND METHODS Animals. Lewis and DA rats, originally obtained

from Mollegaards Laboratories (Roskilde. Denmark), were kept and bred at the animal unit of the Biomedical Center, University of Uppsala. The experiments were performed with female rats, since they are substantially more suscep- tible than males to the induction of collagen-induced arthritis (Larsson P: unpublished observations). Rats used in the experiments were between 8 and 12 weeks old and were age-matched in each experiment.

Collagens. Native rat CII from the Swarm rat chondrosarcoma (16), native bovine CII from calf nasal septa, and native rat type I and type I11 collagen (CYCIII) from the skin of rats were prepared by pepsin digestion and subsequent purification as described by Miller (17), with the modification that the rat CII preparation was not passed over a DEAE-cellulose column (Pharmacia, Uppsala, Sweden). Native lathyritic rat CII from rat chondrosarcoma was prepared by neutral salt extraction, followed by salt precip- itation and DEAE-cellulose chromatography (16).

Induction and evaluation of arthritis. Native rat or bovine CII was dissolved in 0.1M acetic acid at a concentration of 1 rnglml in Freund’s incomplete adjuvant (Difco, Detroit, MI), at 4°C. Rats were immunized intradermally with 250 pl of this emulsion around the base of the tail. Scoring of arthritis severity was made on each of the distal interphalangeal (DIP), PIP, and metatarsophalangeal (MTP) joints. A similar evaluation protocol has been used to describe the development of adjuvant-induced arthritis in rats (18). Since it was not possible to discern the individual intracarpal joints and wrist joints. these were regarded as 1 site of inflammation. The knee joints were also examined for signs of inflammation. The severity of each site (joint) of inflammation was graded according to a 1-3 scale ( 1 = detectable redness and swelling: 2 = moderate swelling and redness; 3 = severe swelling). Signs of ankylosis were graded as 0. The maximum score for each paw was 48 points (16 joints x 3).

Histologic techniques. Whole ankle joints and toes from arthritic DA and Lewis rats were fixed in 4% formal- dehyde (Sigma, St Louis, MO), decalcified in EDTA (Sig- ma), embedded in paraffin, sectioned, and stained with hematoxylin and eosin.

Collection of sera. Rats were bled by retroorbital puncture, and all sera were collected individually and stored at -20°C until assayed.

Enzyme-linked immunosorbent assay (ELISA). The quantification of the anti-type I1 collagen antibodies in sera was performed as previously described (19). Briefly, micro-ELISA plates (Dynatech, Plochingen, FRG) were coated with native rat CII. Anti-CII antibodies were purified from sera of CII-immunized rats, further purified on a rat CII-Sepharose column, and used to calibrate a previously described standard preparation of affinity-purified rat anti- CII antibodies (19). where bovine CII was used as a standard. This calibrated standard was used throughout and was diluted, in 10-fold dilution steps, in parallel with unknown serum samples in the ELISA plates. AU tests were canied out in duplicate. The amount of bound antibody was estimated after incubation with a goat anti-rat IgG polyclonal antibody conjugated to alkaline phosphatase (Jackson Im- munoresearch, West Grove, PA). The subsequent quantification was performed with a p-nitrophenol substrate buffer in a Titertek Multiscan spectrophotometer (Flow Laborato- ries, Eflab Oy, Helsinki, Finland). Absorbance was read at 405 nm.

For each serum sample, the dilution that gave 50% absorbance compared with the maximum value obtained for the standard antibodies was related to the concentration (in &ml) of the standard anti-rat CII antibodies required to give the same 50% absorbance value. Using this method, all calculations were performed with data from the steep por- tion of the slope, where the sample titration curves were parallel to the standard titration curves.

Delayed-type hypersensitivity (DTH) reactions. DTH reactions were induced at different times after immunization of rats with CII. For immunologic challenge, 20 pl of lathyritic rat CII dissolved in O.05M acetic acid (final concentration 1 mglml) was injected intradermally into the right ear, and 20 pI of rat CYCIII dissolved in O.05M acetic acid ( 1 mglrnl) was injected intradermally into the left ear. Forty-

COLLAGEN-INDUCED ARTHRITIS IN RATS 695

Table 1. Incidence of acute (25 days postimmunization) and chronic (4 months postimmunization) arthritis in Lewis and DA rats after immunization with 250 pg of native pepsin-modified rat type I 1 collagen

25 days 4 months

No. of No. of arthritic Total no. anhritic Total no.

rats of rats rats of rats ~~~~~ ~ ~

First experiment Lewis 10 10 0 10 DA 10 10 7 10

Lewis 4 10 0 10 DA 8 10 6 10

Second experiment

eight hours after the CII challenge, the change in ear thickness was measured with a constant-pressure caliper, and the increase in ear thickness (the thickness of the right ear minus the thickness of the left ear) was estimated. The presence of a DTH reaction to rat CII was also investigated in nonimmunized female Lewis and DA rats. Challenges and measurements were performed as described above. The thickness of the ears injected intradermally with CI/CIII was comparable with that of ears injected intradermally with physiologic buffered saline. and no increase in ear thickness was observed when these control ears were compared with normal ears (data not shown).

Statistical analysis. Continuous variables (i.e., antibody levels and DTH reactions) were analyzed in terms of their group mean by Student's r-test. Dichotomous variables (i.e., incidence of arthritis) were analyzed by their propor- tionate group frequencies (chi-square test 1. The significance of arthritis scores and the significance of differences in anti-CII antibody levels between arthritic and nonarthritic rats were analyzed by the Mann-Whitney U-test.

RESULTS Macroscopic evaluation of arthritis development.

Immunization of both Lewis and DA rats with homologous CII gave rise to severe and rapidly developing polyarthritis (Table 1). Clinical onset was seen 13-15 days after immunization with CII. The course of arthritis was characterized separately in female rats from the 2 strains. During an observation period of 6 months, 3 clinically different stages could be dis- cerned. In DA rats. the first stage covered the first 2-4 days of the arthritis and was characterized by the onset of joint inflammation in the PIP joints, with symmetric involvement of both hindpaws (Figure 1 A). In only 1 case was an MTP joint the first joint to become inflamed. During this stage, no inflammation was seen in the ankle joints. Lesions developed in the hindpaws first. The second stage occurred 3-30 days

after the first signs of disease and showed a pattern of a progressive inflammation that spread to the remain- ing joints of the hindpaws (Figure 1B). During this stage, joints in the front paws also became inflamed. The third stage ranged from day 30 until the end of the experiment and was characterized by a regression of joint inflammation, the development of ankylosis, and by chronic inflammation in some of the interphalangeal joints (Figure IC). In the joints that developed chronic arthritis (i.e., where the swelling and the redness did not subside after 30 days). arthritis was seen in 1 or more of the PIP joints of the hindpaws or the front paws, and in 1 case, was observed in an MTP joint in the hindpaw. but not in the ankle or wrist joints. Chronicity was observed in 7 of 10 DA rats for more than 4 months (Table I ) .

The first stage in Lewis rats was characterized by inflammation that started in the ankle joints of the hindpaws (Figure 1D). During the second stage, arthritis had spread to both the interphalangeal and the MTP joints (Figure IE). At this stage, the disease had become progressive, and in some rats, joints in the front paws were also affected. At the third stage of disease, the inflammation had subsided in all the previously affected joints, and no signs of chronic arthritis were observed, but signs of ankylosis were apparent (Figure IF).

Microscopic evaluation of arthritis development. Joint specimens from 2 DA and 2 Lewis rats were obtained at the direrent stages of arthritis development to establish how the observed macroscopic features could be correlated with observations at the microscopic level. In DA rats, histopathologic exami- nation of biopsy specimens taken during the early disease stage showed infiltration of polynuclear and mononuclear cells and hyperplasia of the synovial tissue (Figures 2A and B). Evidence of cartilage and bone destruction was apparent at the site where the synovial membrane adheres to cartilage and bone. Sections from distal and proximal joints of the hindpaws of DA rats were compared. and mild synovitis was observed in the proximal joints (ankle joints) in only a few cases (Figure 2B). Inflammation was wide- spread in the DIP joints (Figure 2A). Large numbers of inflammatory cells and erosion of cartilage and bone were seen at the marginal zones. The second stage was characterized by large numbers of predominantly polymorphonuclear cells infiltrating the synovial membrane and the soft tissue, and gross cartilage and bone destruction. Inflammation and cartilage and bone destruction could be detected in biopsy specimens from

696 LARSSON ET AL

D I P

P I P

MTP

Ankle

D I P

P I P

MTP

Ankle

D I P

P I P

MTP

Ankle

le f t hind Daw right hind paw le f t hind paw right hind paw A A

D I P

P I P

MTP

A D

D I P

P I P

MTP

Ankle

B E

C

D I P

P I P

MTP

Ankle

F Figure 1. Schematic drawings showing the development of the 3 stages of arthritis in the hindpaws of DA and Lewis rats immunized with homologous type II collagen. Circles represent the various involved joints. The joints of the ankle and wrist and intracarpal joints are regarded as 1 site of inflammation, since it was not possible to clinically discern the location of inflammation to single joints at these sites. The numbers represent the intensity of inflammation, graded according to a I-3-point scale (see Materials and Methods). A, First stage in DA rats ( I day after onset of arthritis). B, Second stage in DA rats (10 days after onset of arthritis). C, Third stage in DA rats (4 months after onset of arthritis). D, First stage in Lewis rats ( I day after onset ofarthritis). E, Second stage in Lewis rats (10 days after onset of arthritis). F, Third stage in Lewis rats (4 months after onset of arthritis). DIP = distal interphalangeal; PIP = proximal interphalangeal; MTP = metatarsophalangeal.


A B C

D E F Figure 2. Photomicrographs showing the histopathologic characteristics of the different stages of collagen-induced arthritis in DA rats (A-D) and Lewis rats (E and F). A, Early stage ( I day after onset of arthritis) in a hindpaw proximal interphalangeal (PIP) joint, showing infiltration by mononuclear and polynuclear cells. B, Early stage ( I day after onset of arthritis) in a hindpaw ankle joint. showing mild synovitis. C, Late stage (4 months after onset ofarthritis) in a hindpaw PIPjoint. showing polymorphonuclear cell infiltration of the synovial membrane. and bone and cartilage destruction. D, Higher magnification view of C, showing local infiltration of mononuclear cells in the synovial membrane and in areas of bone and joint destruction. E, Early stage ( 1 day after onset of arthritis) in a hindpaw ankle joint, showing gross inflammatory changes. F. Early stage ( I day a!ler onset of arthritis) in a hindpaw PIP joint. showing less inflammation. Two hindpaws were obtained from each rat at each time point. There were no distinct histopathologic differences between the 2 specimens for any rat. (Hematoxylin and eosin stained; original magnification x 100 in A. B, C, E. and F. original magnification ~ 4 0 0 in D.)

DA rats during the chronic phase of disease (Figure 2C). At this time (4 months postimmunization). mononuclear cells were seen more frequently than polymorphonuclear cells at sites near regions of cartilage and bone destruction (Figure 2D). Fibrotic changes were observed concomitant with chronic joint destruction.

The early phase of arthritis in Lewis rats was characterized by gross inflammatory changes in the ankle joints (Figure 2E). Inflammatory cells were observed infrequently in the distal joints (Figure 2F). At the second disease stage, inflammatory changes and cartilage and bone erosion were prominent in both proximal and distal joints (results not shown). The late stage of disease was characterized by fibrosis of many proximal and distal joints (results not shown). Al- though it was not possible to detect any clinical signs of joint inflammation at this stage, minimal signs of synovitis were observed by microscopy in a few joints.

Analysis of production of autoantibodies to rat CII. Levels of CII autoantibodies from rats immunized with rat CII were measured in 2 different experiments at 2 times, once at 25 days and again at 4 months after immunization with CII (Table 2). All the immunized rats developed serum anti-CII antibodies. Lewis rats had a tendency to develop higher titers of anti-CII autoantibodies than did DA rats. At 4 months postimmunization, anti-CII antibodies were still detectable, and Lewis rats had significantly higher antibody levels than did DA rats. Levels of anti-CII autoantibodies did not correlate with the development of arthritis, i.e., arthritic rats did not develop higher mean anti-CII antibody levels than nonarthritic rats in any of the experiments. No anti-CII antibodies could be detected in nonimmunized rats of either group.

DTH reactions to homologous CII. T cell reactivity to homologous rat CII was measured in immu-

698 LARSSON ET AL

Table 2. Analysis of anti-rat type I 1 collagen (CII) autoantibody levels in Lewis and DA rats by an enzyme-linked immunosorbent assay at 25 days and at 4 months postimmunization with 250 pg of native pepsin-modified homologous rat type I1 collagen'

Antibody levels, d m l

Total

First experiment (25 days) Lewis DA

Lewis DA

Lewist DA

Lewist DA

Second experiment (25 days)

First experiment (4 months)

Second experiment (4 months)

280 2 250 (10) 150 2 110(10)

60 2 40 (10) 62 2 27 (10)

100 -t 65 (8) 32 ? 32 (8)

45 f 45 (10) 10 2 6 (10)

Arthritic rats

260 2 270 (8) 150 2 110 (10)

38 2 16 (4) 66 2 25 (7)

- (0) 38 2 35 (6)

- (0) 6 2 3 (4)

Nonarthritic rats

310 2 240 (2) - (0)

76 2 45 (6) 50 2 34 (3)

100 2 65 (8) 16 -t 16 (2)

45 2 45 (10) 12 % 6 (6)

* Values are the mean 2 SD (no. of rats). Concentrations of anti431 autoantibodies were determined by comparison with an affinity-purified rat anti-CII antibody of known activity. t No macroscopic signs of arthritis were observed at 4 months postimmunization in Lewis rats.

nized rats that were challenged with lathyritic rat CII in the right ear and with rat CI/CIII in the left ear. The change in ear thickness was measured after 48 hours, in 3 separate experiments. It was possible to induce anti-CII DTH reactions during both the acute stage (20-50 days postimmunization) and the chronic stage (4 months postimmunization). During this latter stage, active disease was observed only in the DA rat strain.

All rats of both strains developed positive DTH reactions (Table 3). The increase in ear thickness was significantly greater in DA rats than in Lewis rats in 1 experiment. In both Lewis and DA rats, the DTH reactions were found to be antigen specific, since the increase in ear thickness was induced only by rat C11, and not by rat CUCIII. Nonimmunized Lewis and DA rats were also tested for their DTH response to lathy-

Table 3. Measurement of delayed-type hypersensitivity reactions in Lewis and DA rats immunized with homologous type I1 collagen (CII) that were challenged with homologous CII (first 3 experiments) and in nonimmunized rats that were challenged with homologous and heterologous CII (fourth experiment)*

~

Increase in ear thickness (mm)

Postimmunization time Controls

12 days 40 days 4 months Rat CII Bovine CI1

First experiment Lewis 0.3 ? 0.4 DA 0.35 C 0.2

Lewis 0.3 5 0.05 0.25 C 0.1 DA 0.35 -c 0.2 0.2 2 0.07

Lewis 0.3 ? 0.06 0.1 2 0.08 DA 0.5 2 0.07t 0.2 5 0.06.t

Lewis 0.01 z 0.02 0.01 ? 0.025 DA 0.1 2 0.04t 0.002 2 0.025

Second experiment

Third experiment

Fourth experiment

~ ~

* Values are the mean -t SD. Ear thickness measurements were made 48 hours after challenge. Five rats per group were used in each experiment. t P < 0.01 versus Lewis rats, by Student's r-test. .t P < 0.05 versus Lewis rats, by Student's r-test.


ritic rat CII. A positive response was detected only in DA rats (Table 3). The DTH response of nonimmunized DA rats was restricted to homologous rat CII: no response could be elicited by heterologous bovine CII (Table 3).

DISCUSSION We have reported the induction of arthritis by

homologous CII in 2 rat strains. This CII-induced arthritis is an erosive arthritis with a rapid onset, and it is characterized by T cell- and B cell-mediated immunity to homologous CII in 2 strains of rats that have distinct major histocompatibility complex haplo- types (20). These results indicate that T cells capable of responding to “self’ collagen are present in the T cell repertoire and are not eliminated during T cell ontogeny in the thymus of both rat strains (for review, see refs. 21-24). This could not be concluded from earlier studies on heterologous collagen-induced arthritis, where T cells recognizing foreign epitopes on immunizing heterologous collagens may have aug- mented production of pathogenic anti-CII autoantibodies (25-27).

In addition, profound differences both in initial clinical symptoms and in the subsequent clinical course were observed in Lewis and DA rats after immunization with the same preparation of rat collagen. These differences are of considerable interest, both in understanding how anti-CII autoimmunity is regulated and in determining which factors are involved in the development of various clinical signs of arthritis. Thus, the observation of chronic arthritis in DA rats, but not in Lewis rats, might indicate the presence in Lewis rats of factors that have the capac- ity to down-regulate a manifested collagen-induced arthritis. Identification of such regulating factors should be useful in efforts to develop immunotherapy in these animals. The development of different and reproducible forms of arthritis in the 2 rat strains immunized with the same collagen preparation is com- pelling from the point of view of the clinical rheuma- tologist, who is obliged to define different human rheumatic diseases primarily on the basis of different clinical symptoms (28). It may not be improper to assume that a putative clinical classification of the 2 presently studied arthritides (in DA and Lewis rats) may have resulted in the description of 2 different “criterion-based’’ diseases.

We conducted a limited study of the histopathologic symptoms in collagen-induced arthritis. This

study confirmed the clinical observation that the distal joints are preferentially affected during the early phase of arthritis in DA rats. However, histopathologic ex- amination also revealed sites of inflammation in the proximal joints that had not been detected by clinical inspection. Similarly, during the late stage in Lewis rats, a few inflammatory foci were detected that were not clinically apparent. The histopathologic features of the joint inflammation induced in rats by immunization with heterologous CII have been previously described by Caulfield et al (6) and Terato et a1 (29). Our results in the present study are consistent with those reports in terms of the characteristics of inflammation in single joints.

From the present data, it is difficult to define precisely the role of T and B cell activities in the pathogenetic mechanisms responsible for the development of homologous CI1-induced arthritis. The presently observed lack of correlation between total serum levels of anti-CII antibodies and arthritis development has been noted in several earlier studies of both heterologous collagen-induced arthritis in rats and mice and in homologous collagen-induced arthritis in mice (3.25.30,31). In this study, however, we did observe longstanding T cell reactivity to CII in the form of the ability to mount a DTH response in both DA and Lewis rats. This response was present whether or not the rats were prone to develop chronic collagen-induced arthritis. This indicates that mechanisms that do not affect T helper cells directly may be important in preventing the development of chronic arthritis in Lewis rats, as well as in many DA rats.

The differences in the appearance of chronic disease in Lewis and DA rats may be explained by the observation that only DA rats show a spontaneous DTH reaction to “self’ CII. This suggests that Lewis rats contain T cell-regulatory factors that are absent in DA rats. An in vitro T cell reactivity to foreign CII has been previously described in Lewis rats immunized with Freund’s complete adjuvant, i.e., T cells obtained from non-CII-immunized rats (32). In this case, the in vitro cultured cells were shown to produce an arthritogenic factor capable of inducing joint inflammation when injected locally into the knee joint of normal rats. This observation is further evidence that certain rat strains harbor CII-specific T cells that may partic- ipate in the pathogenetic processes that result in the onset of arthritis.

There is currently little knowledge concerning factors that may be responsible for the different pat- terns of joint involvement observed in Lewis and DA

LARSSON ET AL

rats. However, we have observed that arthritis trans- ferred by serum precipitates preferentially affects large joints such as the ankles (33), even in mouse strains such as DBNI , where immunization with homologous collagen also affects smaller peripheral joints (3,14,15). Thus, it is reasonable to suggest that the involvement of anti-CII autoantibodies and/or immune complexes may be more important in Lewis rats than in DA rats.

It is apparent that the clinical features of homologous collagen-induced arthritis in DA rats are more similar to those seen in RA than those observed in other variants of collagen-induced arthritis (28). One notable feature of DA rats was that this rat strain spontaneously showed a significant DTH response to homologous CLI. In addition, this DTH reaction of nonimmunized DA rats was restricted to homologous CII, since no reaction could be elicited by injections of heterologous CII into the ears of either nonimmunized DA or Lewis rats. If such a spontaneous autoimmunity to a self antigen that is known to induce experimental autoimmunity is indicative of susceptibility to pathogenic autoimmunity, similar methods may also be useful in efforts to screen human populations to iden- tify those at risk for developing specific autoimmune reactions. Thus, it is possible that homologous collagen-induced arthritis may be important in efforts to understand and modify arthritogenic anti-CI1 autoimmunity.

In conclusion, the availability of models for homologous collagen-induced arthritis in both mice (3,14,15) and rats should permit more detailed map- ping of arthritogenic epitopes on homologous collagen molecules. Structural comparisons of such arthritogenic epitopes in these 2 d8erent species may also provide a basis for studies o n whcther there exist similar arthritogenic epitopes on CII in humans.

REFERENCES

1. Trentham DE, Townes AS, Kang AH: Autoimmunity to type I1 collagen: an experimental model of arthritis. J Exp Med 77:857-868, 1977

2. Courtenay JS. Dallman MJ, Dayan AD, Martin A, Mosedale B: Immunization against heterologous type I1 collagen induces arthritis in mice. Nature 283:666468, I980

3. Holmdahl R , Jansson L. Gullberg D, Forsberg PO, Rubin K, Klareskog L: Incidence of arthritis and auto- reactivity of anti-collagen antibodies after immunization of heterologous and autologous collagen 11. Clin Exp Immunol 62:63%646, 1985

4. Cathcart ES, Hayes KC. Gonnerman WA. Lazam AA, Franzblau C: Experimental arthritis in a nonhuman primate. 1. Induction by bovine type I1 collagen. Lab Invest 54:2&31, 1986

5. Yo0 TJ, Sung-Yoon Kim, Stuart JM. Floyd RA, Olson GA, Cremer MA, Kang AH: Induction of arthritis in monkeys with type I1 collagen. J Exp Med 168:777-782, I988

6. Caulfield JP, Hein A, Dynesius-Trcntham R, Trentham DE: Morphologic demonstration of two stages in development of type I1 collagen-induced arthritis. Lab Invest

7. Klareskog L. Holmdahl R, Larsson E, Wigzell H: Role of T lymphocytes in collagen 11-induced arthritis in rats. Clin Exp Immunol 51:117-125. 1983

8. Holmdahl R, Rubin K. Klareskog L, Gustafsson G, Larsson E: Appearance of different lymphoid cells in synovial tissue and peripheral blood during the coursc of collagen 11-induced arthritis. Scand J Immunol 21: 197-204. 1985

9. Stuart JM: Collagen-induced arthritis: a new animal model for the evaluation of antiarthritic drugs. Adv lnflamm Res 7:26>275, 1984

10. Trentham DE: Collagen arthritis as a relevant model for rheumatoid arthritis: evidence pro and con. Arthritis Rheum 25:911-916, 1982

11. Stuart JM, Kang AH: Monkeying around with collagen autoimmunity and arthritis. Lab Invest 54: 1-3, 1986

12. Tarkowski A, Klareskog L, Carlsten H, Herberts P, Koopman WJ: Secretion of antibodies to types I and I1 collagen by synovial tissue cells in patients with rheumatoid arthritis. Arthritis Rheum 32: 1087-1092, 1989

13. Jasin HE: Autoantibody specificities of immune complexes sequestered in articular cartilage of patients with rheumatoid arthritis and osteoarthntis. Arthritis Rheum

14. Holmdahl R, Jansson L. IArsson E, Rubin K, Klareskog L: Homologous type I1 collagen induces chronic and progressive arthritis in mice. Arthritis Rheum 29: 106-113, 1986

15. Boissier MC, Feng X-Z, Callioz A, Roudier R, Fournier C: Experimental autoimmune arthritis in mice. I. Ho- mologous type 11 collagen is responsible for self- perpetuating chronic polyarthritis. Ann Rheum Dis 46: 691-700. 1987

16. Smith BD, Martin GR, Dorfman A, Swarm R: Nature of the collagen synthesized by a transplanted chondrosarcoma. Arch Biochem Biophys 166: 181-186. 1975

17. Miller EJ: Structural studies on cartilage collagen em- ploying limited cleavage and solubilization with pepsin. Biochemistry I1:4903-4909, 1972

18. Pearson CM, Wood FD: Studies of polyarthritis and other lesions induced in rats by injection of mycobac- terial adjuvant. I. General clinical and pathologic char-

461321-343, 1982

28~241-248, 1985

COLLAGEN-INDUCED ARTHRITIS IN RATS 70 1

acteristics and some modifying factors. Arthritis Rheum 2:440-459, 1959

19. Larsson P, Holmdahl R: Oestrogen-induced suppression of collagen arthritis. 11. Treatment of rats suppresses development of arthritis but does not affect anti-type I1 collagen humoral response. Scand J Immunol 26: 579-583, 1987

20. Gunther E, Stark 0: The Major Histocompatibility Complex in Man and Animals. Edited by D Gotze. Berlin, Springer-Verlag, 1977

21. Von Boehmer H, Kajalainen K, Pelkonen J. Borgulya P, Rammensee H-G: The T-cell receptor for antigen in T-cell development and repertoire selection. Immunol Rev 101:21-38, 1988

22. Sprent J, Lo D, Gao E-K. Ron Y: T cell selection in the thymus. Immunol Rev 101:17>190. 1988

23. Marrack P. Kappler J: The T-cell repertoire for antigen and MHC. Immunol Today 9:308-315. 1988

24. MacDonald HR. Lees RK, Schneider R. Zinkernagel R M , Hengartner H: Positive selection of CD4+ thymo- cytes controlled by MHC class I1 gene products. Nature 336:471473, 1988

25. Wooley PH. Luthra HS, Stuart JM, David CS: Type I1 collagen-induced arthritis in mice. I. Major histocompatibility complex (I region) linkage and antibody corre- lates. J Exp Med 154:688-700, 1981

26. Stuart JM, Dixon FJ: Serum transfer of collagen- induced arthritis in mice. J Exp Med 158:378-393. 1983

27. Kenvar SS, Englert ME, McReynolds RA, Landes MJ, Lloyd JM. Oronsky AL, Wilson FJ: Type I1 collagen-

induced arthritis: studies with purified anticollagen im- munoglobulin. Arthritis Rheum 26: 1120-1 131, 1983

28. Arnett FC. Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, Healey LA, Kaplan SR. Liang MI, Luthra HS, Medsger TA Jr. Mitchell DM, Neus- tadt DH, Pinals RS, Schaller JG. Sharp JT, Wilder RL, Hunder GG: The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 31:315-324, 1988

29. Terato K , Hashida R, Miyamoto K, Morimoto T. Kato Y, Kobayashi S. Tajima T, Otake S. Hori H, Nagai Y: Histological, immunological and biochemical studies on type I1 collagen-induced arthritis in rats. Biochem Res 3:495-505, 1982

30. Grifiith MM. Dewitt CW: Immunogenetic control of experimental collagen-induced arthritis in rats. 11. ECIA susceptibility and immune response to type I1 collagen (calf) are linked to RTI. J Immunogenet 8:463470, 1981

31. Carlson RP, Blazek EM, Datko LJ, Lewis AJ: Humoral and immunologic responses in collagen-induced arthritis in rats: their correlation with severity of arthritis. J Immunopharmacol 6:379-388, 1984

32. Helfgott SM, Kieval RI, Breedveld FC, Brahn E. Young CT. Dynesius-Trentham R, Trentham DE: Detection of arthritogenic factor in adjuvant arthritis. J Immunol 140:1838-1843, 1988

33. Holmdahl R, Anderson M, Enander I, Goldschmidt T, Jansson L. Larsson P, Mo J , Nordling C, Klareskog L: Nature of type I1 collagen autoimmunity in mice suscep- tible to collagen induced arthritis. Int Rev Immunol 4:49-64. 1988

homologous type ii collagen—induced arthritis in rats

Documents