| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Rheumatology, University Hospital Nijmegen, Nijmegen, The Netherlands; and
Merck Research Laboratories, Rahway, NJ 07065
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In synovial fluid, as well as in cartilage of arthritis patients, neoepitopes resulting from both enzyme activities were found (7, 8, 9). Also, in animal models of arthritis, both neoepitopes were induced in cartilage (10, 11). We recently studied the occurrence of VDIPEN during Ag-induced arthritis (AIA) (12). IL-1 was found to be a pivotal mediator in VDIPEN expression, becaue IL-1 receptor antagonist prevented the formation of the VDIPEN epitope. IL-1 is a known inducer of pro-MMPs, but these latent MMPs need further activation by other enzymes to become active. Involvement of VDIPEN-inducing MMPs was only found during the late phase of cartilage degradation, and not during early proteoglycan (PG) depletion, which suggests linkage of MMP activity to advanced cartilage damage (12, 13). Furthermore, we have found that the collagenase-induced cleavage site in type II collagen fully colocalizes with VDIPEN epitopes in this model (14, 15). This further suggests that VDIPEN is linked to severe and maybe even irreversible cartilage damage because collagen damage is difficult to repair (16). Stromelysin-1 (SLN-1) is pivotally involved in this colocalization of MMP neoepitopes in aggrecan and collagenase neoepitopes in type II collagen (14, 15), because SLN-1-deficient (SLN-1KO) mice showed no VDIPEN expression during AIA, nor did they show COL2–3/4C (Gly-Pro-Hyp-Gly-Pro-Gln-Gly) staining. This link is probably through activation of collagenases by SLN-1, because in vitro studies have shown that SLN-1 is capable of activating collagenase-1, -2, and -3 (17, 18, 19).
The linkage of VDIPEN occurrence to severe cartilage damage was further substantiated by studies in two other murine arthritis models (13). On the one hand, zymosan-induced arthritis (ZIA), which is induced by local injection of yeast particles, showed almost no expression of VDIPEN neoepitopes. On the other hand, collagen-induced arthritis (CIA), which is based on an autoimmune response to type II collagen, showed widespread and persistent VDIPEN expression. Cartilage damage that is induced during ZIA is fully reversible, whereas full destruction of cartilage is noted in CIA.
These data suggest that activation of latent MMPs in cartilage is correlated to induction of irreversible cartilage damage and is possibly of importance in destruction of cartilage. The question remains which mechanisms are involved in activation of pro-MMPs leading to VDIPEN expression. There are two major differences between the mechanisms of inflammation in ZIA on the one hand and AIA/CIA on the other hand. In both AIA and CIA, T cells and immune complexes (ICs) are involved, whereas these are not present in ZIA.
The objective of this study was to establish the role of ICs in the induction of VDIPEN epitopes. This was done in a model of passive IC arthritis (ICA). In this model, arthritis is induced by injecting lysozyme coupled to poly-L-lysine (PLL) in the knee joint of mice that previously were given anti-lysozyme Abs systemically. If induced in the knee joints of mice, which are prone to develop collagen type II autoimmune arthritis, a severe arthritis develops resulting in PG depletion and surface erosion of the cartilage matrix. This model has been described before and was shown to be dependent on IL-1 (20, 21). Furthermore, the role of SLN-1 in IC-mediated VDIPEN expression was studied with use of SLN-1KO mice. The activation mechanism of MMPs was further studied in vitro.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
SLN-1KO mice were generated in congenic 129SvEv mice and
outcrossed to C57BL/6 mice as described earlier (22).
Homozygous C57BL/6/129SvEv KO mice were outcrossed to B10RIII mice.
F1 outcross mice were twice backcrossed to
B10RIII mice. KO mice showed no gross histopathological abnormalities.
Age-matched congenic B10RIII and C57BL/6 mice were used as wild-type
controls. Mice were 
10 wk of age. They were given a standard
laboratory diet and tap water ad libitum.
Induction of ICA
PLL (Sigma, St. Louis, MO) was coupled to lysozyme as described earlier (20). PLL-lysozyme (3 µg) was injected into the right knee joint of mice that had previously been given specific antisera directed against lysozyme (0.2 ml, i.v.). The antisera, raised in rabbits, were decomplemented by heating at 56°C for 30 min.
Histology
Knee joints were dissected and fixed in 4% paraformaldehyde in phosphate buffer (pH 7.4) for 3 days. Subsequently, joints were decalcified in 10% EDTA (Merck, Rahway, NJ) in phosphate buffer (pH 7.4) and embedded in paraffin wax. Coronal whole knee joint sections were prepared. Sections were stained for PG with safranin O and counterstained with fast green.
Determination of PG content and amount of inflammation
Histologic parameters were scored in a blinded way by two independent observers. Cartilage PG content is reflected in safranin O staining intensity. PG depletion was visualized by diminished staining of the matrix. Safranin O staining was scored arbitrarily; scoring was from 0 to 3 (0, no PG depletion; 1, minimal; 2, moderate; and 3, complete PG depletion). Erosion of the cartilage matrix was scored by expressing the amount of cartilage loss as a percentage of the cartilage surface up to the tide mark and was graded on a scale from 0 to 3 (0, 0%; 1, 0–30%; 2, 30–60%; and 3, 60–100%). Inflammation was scored for the amount of exudated and infiltrated cells. The amount of cells were graded on a scale from 0 to 3 (0, no cells; 1, minimal; 2, moderate; and 3, maximal number of cells).
Immunohistochemical VDIPEN staining
For immunohistochemical analysis, sections were deparaffinized, rehydrated, and digested with proteinase-free chondroitinase ABC (0.25 U/ml 0.1 M Tris-HCl, pH 8.0; Sigma) for 1 h at 37°C to remove chondroitin sulfate from the PG. Subsequently, sections treated with 1% H2O2 in methanol for 20 min and 5 min with 0.1% TritonX-100 in PBS. After incubation with 1.5% normal goat serum for 20 min, sections were incubated with affinity-purified rabbit anti-VDIPEN IgG overnight at 4°C. This Ab has been extensively characterized before (7, 10, 11). Then sections were incubated with biotinylated goat anti-rabbit IgG, the binding of which was detected using avidine-streptavidin-peroxidase staining (Elite kit; Vector Laboratories, Burlingham, CA). Development of the peroxidase staining was done using nickel enhancement. Counterstaining was done with 2% orange G for 5 min.
Quantitative analysis of immunohistochemical staining
Areas of immunostaining were measured using image analysis. Images of histologic sections were recorded with a Sony CCD color video camera (Tokyo, Japan) and displayed on a computer monitor. Immunostained areas were marked by hand, and marked areas were quantified in µm2 by the computer. For every knee joint, three sections were measured, and the mean of these measurements was recorded.
In vitro cartilage assay
Whole murine patellae were isolated with a minimal amount of surrounding tissue. Patellae were cultured in RMPI 1640 (Life Technologies, Gaithersburg, MD) supplemented with 0.25 µg/ml insulin-like growth factor and with or without 10 ng/ml IL-1ß. Subsequently, patellae were incubated with several MMP-activating agents. Patellae specimens were then frozen in liquid nitrogen. Sagittal cryosections were prepared and stained with safranin O and immunostained anti-VDIPEN.
Polymorphonuclear neutrophil (PMN) isolation
PMN were isolated from the blood of healthy volunteers. After dextran sedimentation, PMNs were separated from mononuclear cells on Ficol-Paque (Pharmacia BV, Uppsala, Sweden). The isolation procedure yielded 95% pure PMN. The cells were suspended in RPMI 1640 and directly used for experiments.
PMN cartilage incubations
Purified PMNs were suspended in RPMI 1640 and added to patella specimens in 96-wells microtiter plates (1 patella per well, 5 x 105 cells per well). PMNs were pelleted on the cartilage by centrifugation (5 min at 1200 rpm). PMA (100 ng/ml) was added to the culture, and the plate was incubated for 3 h. Subsequently the patellae were washed in 0.7% NaCl and then frozen in liquid nitrogen.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The involvement of ICs in the formation of VDIPEN epitopes was
studied in passive ICA. Anti-lysozyme Abs were injected i.v. 16 h
before arthritis induction. Arthritis was induced by injecting 3 µg
PLL intraarticularly into the right knee joint. At day 2 of arthritis,
knee joints were dissected and processed for histology. Safranin O
staining showed that PG depletion was pronounced (Fig. 1
A). Furthermore, many
exudating cells were present in the synovial cavity, mainly constisting
of PMNs. Some PMN adherence to the cartilage was consistently seen
(Fig. 1B).
|
, C and D). VDIPEN was found mainly in the femur
and sometimes at the edges of the patella. Typically, these sites were
also more depleted of PG. SLN-1 determines VDIPEN expression and cartilage destruction
Our earlier study identified SLN-1 as a pivotal enzyme in
expression of VDIPEN epitopes in AIA (13, 14). We studied
whether this was also true in ICA. Special attention was given to
occurrence of cartilage destruction. ICA was induced in SLN-1KO mice
and the wild-type strains B10.RIII and C57BL/6. At day 2 of arthritis,
histology was taken and studied for inflammation and PG depletion.
Synovial infiltrate and exudate were comparable in all strains (Fig. 2). PG depletion, as studied with
safranin O staining, was identical in SLN-1KO mice and the wild-type
strain B10.RIII. However, C57BL/6 mice showed less PG depletion
compared with the other two strains (Fig. 2). Strikingly, a clear
difference between B10.RIII and the SLN-1KO was found with respect to
erosions (Table I). B10.RIII mice showed
beginning erosions of the cartilage (see Fig. 4A), whereas
in SLN-1KO mice no cartilage destruction was found (see Fig. 4B). VDIPEN expression was studied by immunohistochemistry.
Profound staining was detected in both C57BL/6 and B10RIII mice, but
remarkably no VDIPEN epitopes were found in SLN-1KO mice (Fig. 3). At day 5 of arthritis, erosions had
progressed in B10.RIII (illustrated in Fig. 4C and Table I), but were
still absent in SLN-1KO mice (Fig. 4D). Erosions were most
often found in the femur and the edges of the patella. This
localization correlated with VDIPEN expression in knee joints of
B10.RIII at day 2 of arthritis. At day 5 of arthritis, SLN-1KO mice
still did not show expression of the VDIPEN neoepitope.
|
|
|
|
To further evaluate the activation mechanism of MMPs, and possible involvement of PMNs in this process, in vitro studies were conducted.
In vitro induction of MMPs was performed by culturing cartilage with
IL-1 for 24 h to induce latent MMPs and subsequent activation of
these latent enzymes. Activation of latent MMPs was first done by
incubating the cartilage with aminophenylmercuric acetate (APMA) (1 mM
APMA in RPMI 1640), a well-known chemical activator of MMPs. This
resulted in profound expression of VDIPEN (Fig. 5A) in IL-1 exposed cartilage,
which was absent when cartilage was incubated with IL-1 alone or APMA
alone (data not shown). When the same treatment was done with cartilage
of SLN-1KO mice, no VDIPEN neoepitopes were found (Fig. 5B),
which identified SLN-1 as a pivotal enzyme in IL-1-induced VDIPEN
expression.
|
Next, we studied whether PMNs are able to activate latent MMPs in
cartilage from wild-type mice. Cartilage was first incubated with IL-1
for 24 h and subsequently with PMA-activated PMNs for 3 h.
This resulted in profound VDIPEN expression (Fig. 6C), which did not occur when
cartilage was not previously incubated with IL-1 (Fig. 6D).
Full PG depletion was seen in both groups treated with PMNs (Figs. 6, A and B), showing the great potency of PMNs to
degrade these cartilage components.
|
C) and VDIPEN expression
(Fig. 7D), identifying elastase as a possible activator for
VDIPEN-inducing MMPs. Cartilage of SLN-1KO mice again did not show
expression of the MMP neoepitope (Fig. 7B), despite the PG
depletion also seen in SLN-1KO cartilage (Fig. 7A).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The role of ICs in the formation of VDIPEN epitopes could involve PMN activation. ICs can provide an anchorage and a trigger for PMN activation (27). Igs, adherent to cartilage surfaces, have been identified in rheumatoid joints (28, 29). As a consequence, reactive oxygen species and proteolytic enzymes present in the PMNs will be released directly onto the surface of the cartilage, thereby escaping inhibitors present in the synovial fluid. PMNs need this close contact to the cartilage to inflict damage to the cartilage (30, 31). In the present study, stimulated PMNs were able to directly activate latent MMPs present in the cartilage after IL-1 exposure. This indicates that PMNs might be involved in the activation of latent MMPs through IC-targetted PMNs. This would fit with the absence of VDIPEN epitopes in ZIA, because no IC is found in this model. In agreement with this hypothesis is the presence of latent MMPs in the cartilage during ZIA (our unpublished observations, data not shown), which shows that MMPs are produced in the cartilage during ZIA, but are not activated.
ICA in SLN-1KO mice resulted in the absence of VDIPEN epitopes and erosions, but PG depletion was not inhibited. This implies that VDIPEN-inducing MMPs are not important in the PG degradation in this model. A likely mediator of early PG degradation is the putative enzyme activity aggrecanase (32, 33, 34). Earlier in vitro studies showed that IL-1-induced PG degradation is mediated by aggrecanase (34, 35). In line with these studies, we also found that in vitro incubation of murine cartilage with IL-1 only results in aggrecanase neoepitopes in cartilage, and not MMP-induced epitopes. When we studied the appearance of aggrecanase neoepitopes in vivo, we detected an increase of neoepitopes during early PG depletion (13). This suggests that aggrecanase is involved in the early phase of PG degradation during arthritis.
Earlier studies by our group revealed that the major PMN proteinase involved in PG degradation is elastase (30, 36). In this study, a specific inhibitor of leukocyte elastase completely blocked PMN-mediated PG degradation. Besides inflicting direct damage to the cartilage, elastase is also capable of activating latent MMP in vitro using recombinant enzymes (37). The present study showed that elastase is able to activate latent MMPs present in the cartilage, identifying elastase as a possible activator for MMPs in vivo.
In ICA, VDIPEN epitopes were only detected on sites extensively depleted from PGs, which suggests a correlation between VDIPEN expression and advanced cartilage damage, which had been shown in other models (13). In AIA, VDIPEN epitopes completely colocalized with collagenase-induced cleavage sites (COL2–3/4C) in type II collagen (14, 15). Degradation of type II collagen is thought to be a key event in cartilage degradation, which eventually leads to irreversible cartilage damage, such as surface erosions. Collagenase activity is considered to be of pivotal importance in the primary cleavage and subsequent denaturation of type II collagen in cartilage (38, 39). Several reports indicate that SLN-1 is able to activate several latent collagenases (17, 18) and might even be nessecary for full activation of collagenase in vitro (19, 40). Our studies in SLN-1KO mice showed that SLN-1 is pivotally involved in induction of both VDIPEN and COL2–3/4C epitopes during AIA (14), which suggests that SLN-1 activates collagenase in vivo during AIA. In the present study, we showed that SLN-1 is also important in VDIPEN epitope induction during ICA. Moreover, no cartilage erosions were found in SLN-1KO mice, whereas B10.RIII wild-type mice showed widespread loss of cartilage. These results suggest an important role for SLN-1 in cartilage erosions in this arthritis model.
An earlier study of CIA in SLN-1KO mice showed clear VDIPEN expression and cartilage erosion in these mice (22). This seems contradictory with our present study in ICA. During ICA, only ICs that are not related to cartilage components are involved in pathology (20). In contrast, during CIA both cellular and humoral immune responses are directed against collagen type II, forming the main component of the cartilage matrix (22). This cartilage-directed attack by the immune system against the large depot of collagen type II may lead to a substantial overkill of other enzymes, overruling the prominent role of SLN-1 as is found in milder forms of arthritis like ICA, which lacks the T cell component in the inflammation. In the early phase of cartilage destruction during CIA, collagen damage might be SLN-1-dependent, but because of its severity its contribution might be overruled by other enzymes in a later phase.
In summary, this study indicates that ICs are involved in the activation of latent MMPs present in the cartilage. This occurs in the context of IL-1-mediated induction of latent MMPs and subsequent activation through PMN triggering. In addition, SLN-1 is a pivotal enzyme in overall VDIPEN-inducing MMP activity and cartilage erosions in this model.
|
Footnotes |
|---|
2 Abbreviations used in this paper: RA, rheumatoid arthritis; MMP, matrix metalloproteinases; PG, proteoglycan; SLN-1, stromelysin-1; KO, knockout; ZIA, zymosan-induced arthritis; CIA, collagen-induced arthritis; AIA, Ag-induced arthritis; IC, immune complex; ICA, IC arthritis; PLL, poly-L-lysine; PMN, polymorphonuclear neutrophil; APMA, aminophenylmercuric acetate.
Received for publication June 15, 1999. Accepted for publication September 1, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-1-proteinase inhibitor. Br. J. Exp. Pathol. 68:81.[Medline]
This article has been cited by other articles:
|
|
 |
|
  S. N. Lauder, S. M. Carty, C. E. Carpenter, R. J. Hill, F. Talamas, J. Bondeson, P. Brennan, and A. S. Williams Interleukin-1{beta} induced activation of nuclear factor-{kappa}b can be inhibited by novel pharmacological agents in osteoarthritis Rheumatology, May 1, 2007; 46(5): 752 - 758. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. East, H. Stanton, S. B. Golub, F. M. Rogerson, and A. J. Fosang ADAMTS-5 Deficiency Does Not Block Aggrecanolysis at Preferred Cleavage Sites in the Chondroitin Sulfate-rich Region of Aggrecan J. Biol. Chem., March 23, 2007; 282(12): 8632 - 8640. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Fukuda, Y. Fujitsu, N. Kumagai, and T. Nishida Inhibition of matrix metalloproteinase-3 synthesis in human conjunctival fibroblasts by interleukin-4 or interleukin-13. Invest. Ophthalmol. Vis. Sci., July 1, 2006; 47(7): 2857 - 2864. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Poon, A. H. Plaas, D. R. Keene, D. J. McQuillan, K. Last, and A. J. Fosang N-Linked Keratan Sulfate in the Aggrecan Interglobular Domain Potentiates Aggrecanase Activity J. Biol. Chem., June 24, 2005; 280(25): 23615 - 23621. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. B. Little, C. T. Meeker, R. M. Hembry, N. A. Sims, K. E. Lawlor, S. B. Golub, K. Last, and A. J. Fosang Matrix Metalloproteinases Are Not Essential for Aggrecan Turnover during Normal Skeletal Growth and Development Mol. Cell. Biol., April 15, 2005; 25(8): 3388 - 3399. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Mehrzad, C. Desrosiers, K. Lauzon, G. Robitaille, X. Zhao, and P. Lacasse Proteases Involved in Mammary Tissue Damage During Endotoxin-Induced Mastitis in Dairy Cows J Dairy Sci, January 1, 2005; 88(1): 211 - 222. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K Nissler, D Pohlers, M Huckel, J Simon, R Brauer, and R W Kinne Anti-CD4 monoclonal antibody treatment in acute and early chronic antigen induced arthritis: influence on macrophage activation Ann Rheum Dis, November 1, 2004; 63(11): 1470 - 1477. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. J. McCawley, H. C. Crawford, L. E. King Jr., J. Mudgett, and L. M. Matrisian A Protective Role for Matrix Metalloproteinase-3 in Squamous Cell Carcinoma Cancer Res., October 1, 2004; 64(19): 6965 - 6972. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. van Lent, K. C. Nabbe, P. Boross, A. B. Blom, J. Roth, A. Holthuysen, A. Sloetjes, S. Verbeek, and W. van den Berg The Inhibitory Receptor Fc{gamma}RII Reduces Joint Inflammation and Destruction in Experimental Immune Complex-Mediated Arthritides Not Only by Inhibition of Fc{gamma}RI/III but Also by Efficient Clearance and Endocytosis of Immune Complexes Am. J. Pathol., November 1, 2003; 163(5): 1839 - 1848. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I Tchetverikov, L R Lard, J DeGroot, N Verzijl, J M TeKoppele, F C Breedveld, T W J Huizinga, and R Hanemaaijer Matrix metalloproteinases-3, -8, -9 as markers of disease activity and joint damage progression in early rheumatoid arthritis Ann Rheum Dis, November 1, 2003; 62(11): 1094 - 1099. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G Fossati, R C Bucknall, and S W Edwards Insoluble and soluble immune complexes activate neutrophils by distinct activation mechanisms: changes in functional responses induced by priming with cytokines Ann Rheum Dis, January 1, 2002; 61(1): 13 - 19. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. L. E. M. Van Lent, K. Nabbe, A. B. Blom, A. E. M. Holthuysen, A. Sloetjes, L. B. A. Van De Putte, S. Verbeek, and W. B. Van Den Berg Role of Activatory Fc{gamma}RI and Fc{gamma}RIII and Inhibitory Fc{gamma}RII in Inflammation and Cartilage Destruction during Experimental Antigen-Induced Arthritis Am. J. Pathol., December 1, 2001; 159(6): 2309 - 2320. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. B van den Berg Arguments for interleukin 1 as a target in chronic arthritis Ann Rheum Dis, November 1, 2000; 59(90001): i81 - 84. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
