| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
*
Laboratory of Molecular Genetics, Hellenic Pasteur Institute, Athens, Greece; and
Department of Pathology, G. Gennimatas Hospital, Athens, Greece
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Despite the lack of definitive experimental evidence for an essential role of the T cell compartment in the initiation of CIA, the notion that disease onset is dependent upon the MHC class II-restricted antigenic properties of type II collagen remains the prevalent concept. However, due to the conflicting evidence both for and against B or T lymphocyte involvement in the induction of CIA, we tested the possibility that a lymphocyte-independent mechanism may be involved. We challenged DBA/1 mice lacking functional mature T and B lymphocytes with rat type II collagen in the presence of adjuvant. These mice were generated by backcrossing recombination activation gene (RAG-1)-deficient mice into the DBA/1 background. This generated disease-susceptible (H-2q) mice lacking both mature B and T lymphocytes as a result of the inactivation of the RAG-1 enzyme that catalyzes the V(D)J-recombination reaction of Ig and TCR genes (18). Our investigations showed that arthritis can be initiated in the absence of mature, functional B and T lymphocytes. Analysis of the disease profile revealed a delay in clinical disease onset and a reduction in severity, confirming a positive contribution by T and B lymphocytes to disease progression. However, similar pathologic features and normal incidence suggest that lymphocyte-independent mechanisms of disease induction also operate in the standard CIA model. We conclude that adaptive immune responses are not the only mechanism of CIA initiation and hypothesize that the nonantigenic properties of type II collagen can also lead to arthritis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RAG-1-deficient mice (kindly provided by Dr. Spanopoulou, Rockefeller University, New York, NY) were backcrossed into the disease-susceptible DBA/1 background (H-2q) for at least five generations; a change in MHC haplotype was confirmed by FACScan analysis (Abs were obtained from PharMingen, San Diego, CA). Genotype was determined by Southern blot hybridization as described previously (18) and was confirmed by FACScan analysis for CD4+ and CD8+ lymphocytes (Abs were obtained from PharMingen). Mixed groups of virgin male littermates, which had been separated from females at 3 wk of age, were housed at low numbers of mice per cage. Experimental mice comprised males of between 8 and 10 wk of age at the time of immunization. Mice were fed water and food ad libitum in accordance with European Union guidelines on animal welfare.
Induction of CIA
Rat type II collagen from a rat chondrosarcoma (19) (kindly donated by Dr. R. Holmdahl, Lund, Sweden) was prepared by pepsin digestion and subsequent purification as described previously (20). Native type II collagen was dissolved in 0.1 M acetic acid at a concentration of 1 mg/ml. Mice were immunized intradermally at the base of the tail with a single injection of 100 µg of collagen emulsified in an equal volume of CFA containing 100 µg of H37Ra Mycobacterium tuberculosis (Sigma, St. Louis, MO). Where stated, chicken egg OVA (Sigma, USA) was dissolved in 0.1 M acetic acid at a concentration of 1 mg/ml and used as the immunogen instead of type II collagen.
Clinical analysis of disease
Arthritis was scored three times per week in each of four paws based on the following arthritis severity index: grade 0, normal appearance and flexion; grade 1, mild swelling and/or erythema; grade 2, moderate swelling and erythema; grade 3, severe swelling and erythema. Scores were compiled for each mouse at each timepoint by adding the score for each of the four paws, thereby giving a maximum potential score of +12 and a minimum score of 0. To derive the arthritic index for each experiment, maximal arthritic scores were combined and divided by the number of arthritic mice in the group.
Histopathologic examination
Joint tissues from freshly dissected mice were immersion-fixed in 10% formalin in 0.15 M PBS (pH 7.4) overnight, decalcified in 30% formic acid/0.28 M sodium citrate for 48 h, and subsequently dehydrated in graded alcohol and embedded in paraplast (BDH, Dorset, U.K.). Serial sections throughout the joint were cut at 4–6 µm on a microtome and stained with hematoxylin and eosin. Sections were evaluated in a blinded manner. The sections were studied for histologic signs of arthritis and scored as follows: 1, synovial cell proliferation, synovial hypertrophy with villus formation and/or fibrin deposition; 2, inflammation, synovitis and/or generalized inflammation; 3, cartilage disruption, chondrocyte degeneration and/or ruffling of cartilage surface and/or dystrophic cartilage; and 4, joint destruction, cartilage erosion with abundant inflammation and pannus formation with bone erosion.
Evaluation of serum Ab levels
Serum Ab levels against type II collagen were measured by a standard ELISA assay. Briefly, a 96-well Immuno-Maxisorp Plate (Nunc, Roskilde, Denmark) was coated with rat type II collagen (10 µg/ml) overnight at 4°C and blocked with 1% BSA in PBS. Plates were washed, and sample sera (at a dilution of 1/200) were incubated for 3 h at 37°C. Subsequently, plates were washed and incubated with biotin-conjugated polyclonal rabbit anti-mouse IgG (heavy and light chain) or IgM Abs (Sigma, St. Quentin Fallavier, France) followed by incubation with horseradish peroxidase-streptavidin (Sigma). The ELISA was developed with o-phenyldiamine dihydrochloride (Sigma) containing 0.03% H2O2, and the reaction was terminated with 50 µl of 2 M H2SO4. OD was read at 490 nm using an MRX microplate reader (Dynatech Labs, Chantilly, VA). Data generated were used to confirm the presence of Ab response to the Ag.
Statistical analysis
Data were analyzed using the unpaired Student t test. p values of <0.05 were considered statistically significant. Data are presented as the mean ± SEM where appropriate.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To assess whether the initiation of CIA is dependent upon
mature T and B cells, RAG-1-deficient mice in a DBA/1 background were
challenged with rat type II collagen emulsified in adjuvant.
Clinically, we found that, whereas lymphocyte-positive
RAG+/- DBA/1 littermate controls presented severe
swelling with peak incidence around week 5 postchallenge,
RAG-/- DBA/1 mice developed mild swelling, with
peak incidence around week 7 (Fig. 1
;
Table I). In the RAG-1-deficient mice,
both the delay in disease onset (p = 0.0001)
and the reduction in disease severity (p =
0.0077) were deemed to be statistically significant. However, the
overall degree of incidence between RAG+/- and
RAG-/- DBA/1 mice was not significantly different
(p = 0.8614; Table I). No arthritis was
observed in the control, nonimmunized DBA/1 group. In all mice, the
functional deletion of the RAG-1 gene product was confirmed by the
absence of anticollagen Abs in the sera of mice after challenge with
type II collagen, in comparison with positive control RAG-1
heterozygous littermates, as determined by ELISA analysis (Table II). Additional characterization by
FACScan analysis (anti-CD4 and anti-CD8 Abs by
PharMingen) was conducted on selected RAG-1-deficient and RAG-1
heterozygous DBA/1 mice to confirm the presence or absence of
CD4+ and CD8+ single-positive T lymphocytes
(Table II).
|
|
|
). Arthritic lesions were observed in 11
of 18 RAG-/- DBA/1 mice, of which 4 were characterized as
suffering from mild focal arthritis. In this group of mice, synovial
hyperplasia initiating at the cartilage/synovium junction was apparent,
in some cases extending down the shaft of the bone cortex (Fig. 2A). The synovium showed signs of activation characterized
by fibrin deposition along the hypertrophic synovium with occasional
villus formation. There was no evidence of inflammation in these mice,
with the soft tissue remaining free of cellular infiltration (Fig. 2B). The articular cartilage showed areas of proteoglycan
loss as determined by toluidine blue and safranin O staining (data not
shown). In addition, histologic examination of the articular cartilage
showed a variable degree of cartilage disruption characterized by
chondrocyte degeneration and necrosis. Ruffling and fragmentation of
the articular cartilage was also obvious. In areas of greatest synovial
expansion, erosion of the subchondral bone was observed.
|
C). Subchondral bone erosion was also more
pronounced. One interesting feature found in four of seven mice was the
development of dystrophic cartilage that was characterized by the
deposition of extracellular matrix proteins within the
hyperplastic synovium (Fig. 2D) and confirmed by
toluidine blue and safranin O staining (data not shown). In two of
seven mice with advanced arthritis, an inflammatory infiltrate was
observed. Histologic analysis of this inflammatory lesion revealed
major differences in the inflammatory component compared with normal
CIA. In lymphocyte-positive control mice with CIA, heavy inflammation
in both the subsynovium and the soft tissue was observed (Fig. 2E). This inflammation consisted of lymphocytes interspersed
with polymorphonuclear cells. In RAG-1-deficient DBA/1 mice with mild
focal arthritis, there was no inflammatory component in the
hyperplastic synovium (Fig. 2B); only synovial cells were
detectable. In RAG-1-deficient DBA/1 mice with advanced arthritis that
showed inflammation, the inflammatory component contained virtually no
lymphocytes as determined by histologic examination; although
polymorphonuclear cells were present, they were greatly reduced in
number (Fig. 2F). Of the mice that were scored as clinically nonarthritic, three of seven showed signs of synovial activation (characterized by mild hyperplasia with either villus formation or fibrin deposition). Consequently, there remains the possibility that disease would have developed in these mice over a longer period. Studies into the chronicity of CIA in the absence of T and B lymphocytes are now underway.
Disease induction in DBA/1 mice lacking functional mature lymphocytes is collagen-specific
Single-blinded induction experiments were conducted to investigate
the specificity of disease. Grouped RAG-/- DBA/1 males
were injected with either rat type II collagen, chicken egg OVA, or
diluent (0.1 M acetic acid) emulsified in adjuvant. These results
(Table I) showed an induction of CIA only in those mice challenged with
type II collagen. Histologic analysis confirmed the clinical scores and
showed characteristic features of arthritis only in those mice
challenged with heterologous type II collagen. Mice challenged with
either OVA or diluent showed normal histologic features in the joint
(data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Taken together, our results imply that the arthritis-triggering properties of collagen can act independently of the adaptive arm of host immunity. However, the reduced severity argues that the pathologic changes observed in the joint are accentuated by the action of lymphocytes. Therefore, it is possible that there are additional mechanisms of disease induction at work in CIA. This conclusion would explain the partial protection from CIA induction and severity observed in a diverse range of mice deficient in gene products that are known to play immunomodulatory roles. For example, in mice lacking major components of the immune response such as CD4/CD8 (17), ICAM-1 (23), IL-12 (24), IFN regulatory factor-1 (25), 5-lipoxygenase-activating protein (26), or the p55 TNF receptor (27), protection from CIA was never complete. Interestingly, the only study so far that reported total protection from CIA concerned IL-6 knockout mice (28), but the exact mechanism for this protection remains unclear. In contrast to our report, recent experiments have shown that B cell-deficient mice do not develop clinical signs of CIA (29). This contradictory finding may be due to the difference in genetic background between the DBA/1 and the B10 mouse strains used by Svensson et al. or to a difference in the immunizing protocol. An alternative explanation may be that B cells are important in eliciting an inflammatory response, the absence of which may mask the clinical signs of disease. As shown in our study, a histopathologic analysis of CIA should be helpful in defining disease states, especially in immunodeficient strains in which clinical/inflammatory lesions may be dampened.
There is a wealth of experimental evidence describing a pivotal role for the adaptive immune response in CIA; however, the data reported here suggest that additional mechanisms may also be present. Several hypotheses can be proposed that rationalize the presence of such additional mechanisms. First, CIA could be a complex disease consisting of two distinct stages, with stage 1 being synovial activation and stage 2 consisting of lymphocytic involvement and leading to readily observable clinical manifestations. Evidence supporting this theory comes from the observations of Caulfield et al., who noted that synovial hyperplasia appears before inflammatory cell influx (21). Second, it is also possible that other mechanisms in disease induction become important in the absence of lymphocytes, and that an arthritic disease distinct from standard DBA/1 CIA is observed in the RAG-1-deficient DBA/1 mouse. However, the similar pathologic features and normal incidence of disease between lymphocyte-deficient and normal DBA/1 arthritic mice do not support this theory. Therefore, to resolve this issue, it is important to define the parameters leading to disease induction in the RAG-1-deficient model. For example, it will be interesting to assess whether disease induction in RAG-deficient mice requires the DBA/1 background and whether it remains specific for type II and not type I or denatured type II collagen. Experiments are now underway to address these questions.
We conclude that the observed pathologic events upon collagen immunization in RAG-deficient mice depend upon the nonantigenic properties of type II collagen. One possible mechanism may use the recently reported capacity of the native triple helical fibrilar collagens (types I, II, III, V, and XI) to act as the specific ligands for the discoidin domain receptor tyrosine kinases DDR1 and DDR2 (30, 31). These receptors are widely expressed in several fetal and adult organs and tissues, but their specific biologic function in mammals remains unclear. Remarkably, activation of the DDR2 receptor by collagens is found to specifically induce the expression and secretion of matrix metalloproteinases, which are molecules that are implicated in the pathogenesis and perpetuation of destructive inflammatory arthritis (32). In general, receptor tyrosine kinases control a wide variety of cellular responses, including the regulation of cell growth, differentiation, migration, metabolism, and survival. Therefore, it is tempting to speculate that upon the administration of type II collagen, specific collagen receptor-mediated cellular events act as the triggers for an arthritogenic response that could then be exacerbated by the involvement of the adaptive immune response. This hypothesis may explain the unaltered incidence but reduced severity and delayed time of onset of CIA as documented in this study in RAG-1-deficient DBA/1 mice and is further supported by the finding that the activation of macrophages or monocytes can occur in vitro with exogenously added type II collagen or collagen-derived peptides in a dose-dependent manner (33, 34, 35, 36).
In the absence of an active T and B lymphocyte population, it is
necessary to consider which other cellular types could be involved in
disease initiation. It is known that synovial macrophages are activated
in the arthritic joints of experimental animal models, and that
macrophage numbers correlate with disease severity (37). It has also
been reported that the transfer of CIA into SCID mice can be blocked by
treatment with macrophage-specific anti-CD11b (Mac-1) mAbs (38),
and that a clodronate-mediated depletion of phagocytic synovial
macrophages ameliorates localized inflammation in CIA (39). In
addition, synovial fibroblasts from arthritic patients possess
autologous invasive behavior (40). Thus, resident synovial cells are
clearly involved in established disease and may also have the potential
to act as the pathogenic trigger upon activation. There is much
evidence to suggest that the cytokines TNF
and IL-1 play diverse but
important roles in arthritis (41). In agreement with its assumed
importance, anti-TNF treatment is one of the most effective
therapies developed both for CIA and rheumatoid arthritis alleviation
(42, 43), suggesting that pathogenic mechanisms are governed by this
cytokine. The activated synovial fibroblast, which is highly responsive
to TNF (44) and shows the properties of a transformed (45) and
aggressive cell (40), occupies a central role in attempts to set up a
mechanistic paradigm of disease pathogenesis in rheumatoid arthritis
(46). Our evidence of synoviocyte proliferation in the absence of
synovitis occurring after collagen administration in
DBA/1-immunodeficient mice indicates an immediate responsiveness of
this specific cell type to collagen. Therefore, it is conceivable that
sequestered collagen resulting from diverse insults affecting cartilage
integrity may contribute directly to the chronic activation of the
synoviocyte in susceptible genetic backgrounds. With further
investigation, the nonantigenic properties of collagen may offer new
mechanistic clues and therapeutic targets to combat rheumatoid
arthritis.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. George Kollias, Laboratory of Molecular Genetics, Hellenic Pasteur Institute, 127 Vas. Sofias Avenue, 11521 Athens, Greece. E-mail address:
3 Abbreviations used in this paper: CIA, collagen-induced arthritis; RAG; recombination activation gene.
Received for publication July 22, 1998. Accepted for publication September 22, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and interleukin-1ß release from human monocytes by cyanogen bromide peptides of type II collagen. Arthritis Rheum. 31:1508.
(TNF) blockade in rheumatoid arthritis. Clin. Exp. Immunol. 101:207.[Medline]
This article has been cited by other articles:
|
|
 |
|
  N. K. Banda, J. M. Thurman, D. Kraus, A. Wood, M. C. Carroll, W. P. Arend, and V. M. Holers Alternative Complement Pathway Activation Is Essential for Inflammation and Joint Destruction in the Passive Transfer Model of Collagen-Induced Arthritis J. Immunol., August 1, 2006; 177(3): 1904 - 1912. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Das, J. Cho, I. Lambertz, M. A. Kelliher, A. G. Eliopoulos, K. Du, and P. N. Tsichlis Tpl2/Cot Signals Activate ERK, JNK, and NF-{kappa}B in a Cell-type and Stimulus-specific Manner J. Biol. Chem., June 24, 2005; 280(25): 23748 - 23757. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Svendsen, C. B. Andersen, N. Willcox, A. J. Coyle, R. Holmdahl, T. Kamradt, and L. Fugger Tracking of Proinflammatory Collagen-Specific T Cells in Early and Late Collagen-Induced Arthritis in Humanized Mice J. Immunol., December 1, 2004; 173(11): 7037 - 7045. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A.-S. Hansson, M. Johannesson, L. Svensson, K. S. Nandakumar, D. Heinegard, and R. Holmdahl Relapsing Polychondritis, Induced in Mice with Matrilin 1, Is an Antibody- and Complement-Dependent Disease Am. J. Pathol., March 1, 2004; 164(3): 959 - 966. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Corr and B. Crain The Role of Fc{gamma}R Signaling in the K/B x N Serum Transfer Model of Arthritis J. Immunol., December 1, 2002; 169(11): 6604 - 6609. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.-G. Zhang, P. Yang, J. Xie, Z. Liu, D. Liu, L. Xiu, T. Zhou, Y. Wang, H.-C. Hsu, and J. D. Mountz Depletion of Collagen II-Reactive T Cells and Blocking of B Cell Activation Prevents Collagen II-Induced Arthritis in DBA/1j Mice J. Immunol., April 15, 2002; 168(8): 4164 - 4172. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. M. Johansson, A.-S. Hansson, K. S. Nandakumar, J. Backlund, and R. Holmdahl IL-10-Deficient B10.Q Mice Develop More Severe Collagen-Induced Arthritis, but Are Protected from Arthritis Induced with Anti-Type II Collagen Antibodies J. Immunol., September 15, 2001; 167(6): 3505 - 3512. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. S. K. de Hooge, F. A. J. van de Loo, O. J. Arntz, and W. B. van den Berg Involvement of IL-6, Apart from Its Role in Immunity, in Mediating a Chronic Response during Experimental Arthritis Am. J. Pathol., December 1, 2000; 157(6): 2081 - 2091. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Thornton, G. P. Boivin, K. N. Kim, F. D. Finkelman, and R. Hirsch Heterogeneous Effects of IL-2 on Collagen-Induced Arthritis J. Immunol., August 1, 2000; 165(3): 1557 - 1563. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Vermeire, L. Thielemans, P. Matthys, and A. Billiau The effects of NO synthase inhibitors on murine collagen-induced arthritis do not support a role of NO in the protective effect of IFN-{gamma} J. Leukoc. Biol., July 1, 2000; 68(1): 119 - 124. [Abstract] [Full Text]
|
|
|
|
|
|
P. Matthys, K. Vermeire, T. Mitera, H. Heremans, S. Huang, D. Schols, C. De Wolf-Peeters, and A. Billiau Enhanced Autoimmune Arthritis in IFN-{gamma} Receptor-Deficient Mice Is Conditioned by Mycobacteria in Freund's Adjuvant and by Increased Expansion of Mac-1+ Myeloid Cells J. Immunol., September 15, 1999; 163(6): 3503 - 3510. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Seetharaman, A. L. Mora, G. Nabozny, M. Boothby, and J. Chen Essential Role of T Cell NF-{kappa}B Activation in Collagen-Induced Arthritis J. Immunol., August 1, 1999; 163(3): 1577 - 1583. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Corthay, A. Johansson, M. Vestberg, and R. Holmdahl Collagen-induced arthritis development requires {alpha}{beta} T cells but not {gamma}{delta} T cells: studies with T cell-deficient (TCR mutant) mice Int. Immunol., July 1, 1999; 11(7): 1065 - 1073. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
,
n = 18; the total incidence of arthritis from two
experiments is shown. Disease onset is significantly delayed;
p = 0.0001.