Abstract
Objective
To define the expression pattern of cadherin-11 in destructive pannus tissue of patients with rheumatoid arthritis and to determine if cadherin-11 expression in fibroblast-like synoviocytes controls their invasive capacity.
Methods
Cadherin-11 expression in rheumatoid synovial tissue was evaluated using immunohistochemistry. To examine the role of cadherin-11 in regulating the invasive behavior of fibroblast-like synoviocytes, we generated L-cell clones expressing wild-type cadherin-11, mutant cadherin-11, and empty vector transfected controls. The invasive capacity of L-cell transfectants and cultured fibroblast-like synoviocytes treated with a blocking cadherin-11-Fc protein or control immunoglobulin was determined in Matrigel invasion assays.
Results
Immunohistochemistry revealed that cadherin-11 is abundantly expressed in cells at the cartilage-pannus junction in rheumatoid synovitis. Invasion assays demonstrate a twofold increased invasive capacity of cadherin-11 transfected L-cells compared to L-cells transfected with E-cadherin or control vector. The invasive behavior of the L-cells stably transfected with a cadherin-11 construct that lacked the juxta-membrane cytoplasmic domain (cadherin-11 ΔJMD) was diminished to the level of vector control L-cells. Further, treatment with the cadherin-11-Fc fusion protein diminished the invasive capacity of fibroblast-like synoviocytes.
Conclusion
These in vitro studies implicate a role for cadherin-11 in promoting cell invasion and contribute insight into the invasive nature of fibroblast-like synoviocytes in chronic synovitis and rheumatoid arthritis.
Keywords: Cadherin-11, Fibroblast-like Synoviocytes, Cell Invasion
Rheumatoid arthritis (RA) is a chronic inflammatory disease localized in the synovium that causes joint destruction (1). The normal synovium forms a thin membrane composed of fibroblast-like synoviocytes (FLS) of mesenchymal origin located between the joint cavity and the fibrous joint capsule. The FLS are organized into a condensed lining layer, 1-3 cells thick, that provides nutrients and lubrication for the avascular articular cartilage (2, 3).
During the course of RA, the inflamed synovial tissue undergoes dramatic remodeling with hyperplasia and the formation of an aggressive tissue mass, called pannus, which attaches to, encroaches over, and destroys the articular cartilage (4). The pannus tissue is composed of a greatly expanded number of both FLS and macrophages, the FLS being the prevailing cell type at the erosive interface between pannus tissue and cartilage. Abundant expression of proteases by these FLS may facilitate cartilage destruction (5). However, the mechanisms that govern their aberrant invasive behavior remain incompletely understood.
Recently, we identified cadherin-11 expression on FLS (6). Consistent with the known function of cadherins in mediating homophilic cell-to-cell adhesion, our previous studies demonstrated that the establishment of synovial lining architecture in-vitro and in-vivo is critically dependent on cadherin-11 (7, 8). Cadherins are expressed in most epithelial and mesenchymal tissues and, beyond their adhesive function, may determine the characteristically different behavior of these respective tissues (9). Cadherin-11 is considered a mesenchymal cadherin as its expression is associated with mesenchymal morphogenesis. Importantly, cadherin-11 expression correlates with increased cellular motility. During development, the switch from epithelial cadherin (e.g. E-cadherin) to cadherin-11 expression correlates with cell migration and tissue outgrowth (10). Further, recent evidence indicates that cadherin-11 is aberrantly expressed in epithelial lineage cancer cells that display a more migratory and invasive phenotype (11).
In this study, we demonstrate that cadherin-11 is abundantly expressed in cells at the invasive interface between cartilage and pannus tissue in rheumatoid synovitis. In in-vitro studies, we find that cadherin-11 promotes the invasive capacity of FLS via mechanisms involving its intracellular cytoplasmic juxta-membrane domain.
Materials and Methods
Cell Culture
Discarded synovial tissues from rheumatoid arthritis patients (American College of Rheumatology criteria) were obtained after synovectomy or joint replacement procedures with approval of the Partner’s Health Care Institutional Review Board or the ethical committee of the Vienna General Hospital. Synoviocyte cell suspensions were prepared and cultured as previously described (7). Cultures of FLS between passage 4 and 8 were used for experiments.
Plasmids and generation of L-cell-cadherin-11 stable transfectants
The human cadherin-11 cDNA or E-cadherin cDNA was inserted into the expression plasmid pCEP4 (Invitrogen, Carlsbad, CA). Since L-cells (murine fibroblast cell line L-M; CCL1.3; American Type Culture Collection, Rockville, MD) are known to contain the catenins required for proper cadherin function but no endogenous cadherins, they were used for transfection with pCEP4/cadherin-11, pCEP4/E-cadherin, pCEP4/cadherin-11 ΔJMD, pCEP4/cadherin ΔCBS, or with the empty pCEP4 vector as described before (12). L-cells were grown and transfected cells were selected by culture in 0.8 mg/ml hygromycin B (Invitrogen) (12). Cadherin-11 expression was confirmed by flow cytometry.
Antibodies and other reagents
The mAbs cadherin-11-3H10, cadherin-11-5H6, anti-human E-cadherin (E4.6, IgG1) were developed previously in our laboratory (6). P3 (control mouse IgG1) was purchased from Zymed (San Francisco, CA). Cadherin-11-Fc protein was produced in our laboratory and purified as previously described (6).
Immunohistochemistry
Synovial tissues were fixed with 4% paraformaldehyde in phosphate buffered saline, dehydrated and paraffin-embedded. For immunohistochemistry, sections were deparaffinized, rehydrated and endogenous peroxidase activity was blocked with 3.0 % hydrogen peroxide in methanol. Primary antibodies were incubated on tissues for 1 hour, washed, and developed using the Vectastain ABC kit (Vector Laboratories, Burlingame, CA). Antibodies used: anti-cadherin-11 mAb 3H10 (ref.) and isotype control antibody (P3, Zymed). Sections were counterstained with Meyer’s hematoxyline. Light microscopic images were captured using a Leica DM LB2 light microscope (Leica) with a digital camera (model DFC 300; Leica) using Spot software. Image processing was done using Adobe Photoshop.
Cell invasion assay
To assess the invasive properties of L-cells transfectants or FLS we used Matrigel Invasion Chambers (BioCoat GFR Matrigel Invasion Chambers; Becton Dickinson, Bedford, MA). Matrigel inserts were rehydrated and invasion medium (DMEM, high glucose, supplemented with 0.1% serum albumin , 2 mM L-glutamine, 10 μM non-essential amino acids (Gibco BRL), 100 U/ml penicillin, and 100 μg/ml streptomycin sulfate) containing 10% heat-inactivated fetal bovine serum (Hyclone) as a chemoattractant was added to the bottom wells of the invasion chamber. L-cell transfectants were serum-starved L-cells or 5 × 104 FLS were plated in the top chamber and incubated for 18 hours at 37° C in serum-free medium. Then, cells were fixed and permeabilized with ice-cold methanol. Cells that had not penetrated the filter were removed from the top of the filter using cotton swabs, and cells that penetrated to the underside of the membrane were stained with propidium iodide (1 mg/ml D-glucose in PBS containing 50 μg/ml propidium iodide and 10 Kunitz units/ml ribunuclease A). Membranes were examined by fluorescence microscopy and photographed. Values for invasion were expressed as the average number of migrated cells bound per microscopic field. Three microscopic fields per membrane in triplicate experiments were counted. Student’s t-test was applied to determine the levels of significance for differences in invasion of various L-cell clones or FLS that were treated with cadherin-11-Fc versus control immunoglobulin. Results were considered to be significantly different when the P value was less than 0.05.
For long-term invasion assays, L-cells were plated on top of Matrigel (100 μl in Boyden Chambers; 24-well inserts; Corning Glassworks, Corning, NY) and cultured in regular medium for 10 days. Thereafter, the gels were fixed with 2 % paraformaldehyde in PBS overnight at room temperature, embedded in paraffin, sectioned and stained with hematoxyline and eosin. Invasion depth was assessed from three sections of each individual Matrigel block of triplicate experiments using an Axioskop 2 micorscope (Carl Zeiss Microimaging, Inc.) and OsteoMeasure Analysis System (OsteoMetrics). The unpaired two-tailed Student’s t-test was applied to determine the levels of significance for differences in invasion of the various L-cell clones.
Results
Cadherin-11 is expressed in cells at the cartilage-pannus junction in rheumatoid synovitis
Our previous studies revealed cadherin-11 expression in cells of the lining layer and in a subset of sublining cells of histological sections of human RA synovium (6). However, in RA, the hyperplastic synovium gives rise to pathological pannus tissue that is attached to and invades cartilage. Thus, we sought to determine if cadherin-11 was expressed by synovial cells at the interface of pannus with articular cartilage in RA. Hematoxyline and eosin staining of metacarpophalangeal joint sections from RA patients revealed extensive pannus formation with fibroblast-like cells intimately attached to and invading into the articular cartilage (Figure 1A, upper panel). Strikingly, immunohistochemistry using the anti-cadherin-11 antibody 3H10 showed prominent staining in condensed fibroblast-like cells of the pannus tissue invading into the articular cartilage. By contrast, the isotype control antibody showed no tissue staining (Figure 1A, lower panels).
Figure 1.
Cadherin-11 promotes cell invasion. A. Upper panel : Sections of metacarpophalangeal joints of a RA patient were stained with hematoxyline and eosin. Note pannus eroding into cartilage. Magnification 100X. Lower panels: Sections (see above) stained with anti-cadherin-11 mAb 3H10 or an iso-type control antibody. Note prominent cadherin-11 reactivity (brown) in cells invading into the cartilage. Magnification 400X. B. L-cells expressing cadherin-11, E-cadherin or empty vector were examined using Matrigel Invasion Chambers. Cadherin-11 L-cells were twice as invasive as E-cadherin or control L-cells. Values are means±SD of triplicates. One experiment out of two is shown. Statistical analysis revealed significant differences (*=P<0.05). C. Cadherin-11 L-cells or vector control L-cells were cultured on top of Matrigel for 10 days. Cadherin-11 L-cells invaded as multicellular aggregates. By contrast, the invasive activity of vector control L-cells was very limited. One representative of 4 experiments is shown. Magnification 200X.
Cadherin-11 Promotes Cell Invasion
Besides their role in cell adhesion, cadherins are known to influence cell behavior including participating in morphogenic programs in development and cell migration (13). To gain insight into what elements alter FLS to become invasive in pathological states, we examined the role of cadherin-11 in this process. To model the role of cadherin-11 compared to other cadherins or the absence of any cadherin, we transfected L-cell fibroblasts with expression constructs encoding cadherin-11, E-cadherin or no cadherin and examined their invasive capacity in Matrigel Invasion Chambers. L-cells expressing empty vector or E-cadherin demonstrated limited invasive capacity (190±24 cells/LPF; 208±33 cells/LPF and 143±34 cells/LPF, respectively). In contrast, cadherin-11 L-cells exhibited a twofold increased invasive activity (422±41 cells/LPF and 371±48 cells/LPF) (Figure 1B).
To further visualize the cadherin-11 dependent invasive process, we cultured L-cell transfectants on top of Matrigel for 10 days. Thereafter, the gels were fixed, embedded in paraffin, sectioned and stained with hematoxilin and eosin. Cadherin-11 L-cells formed aggregates on top of the gel and markedly invaded, mainly as multicellular clusters. By contrast, vector-control L-cells did not form aggregates on top of the gel and their capacity to invade the Matrigel was very limited (Figure 1C). These results indicate that cadherin-11 promotes cell invasion.
Cadherin-11 Dependent Cell Invasion Is Determined by its Cytoplasmic Domains
Whereas cadherin-to-cadherin binding in cell adhesion is determined by interactions of the extracellular domains, increasing evidence indicates that cadherins exert their biological effects on the cells that express them through interactions with intracellular molecules via their cytoplasmic domains, either the juxta-membrane domain (JMD) or the β-catenin binding sequence (CBS) (9). Previous comparative characterization of L-cells expressing wild-type cadherin-11 or mutant cadherin-11 revealed striking differences with regard to cell-to-cadherin-11-Fc substrate adhesion, intercellular motility, Rac1 activity, and actin cytoskeletal organization (Table 1) (12). To evaluate the potential role of these domains for cadherin-11 mediated cell invasion, we used L-cells that were stably transfected with cadherin-11 constructs that either lacked CBS (cadherin-11ΔCBS) or JMD (cadherin-11ΔJMD) of the cytoplasmic tail (Figure 2A). We compared the invasive capacity of cadherin-11 L-cells, cadherin-11ΔJMD L-cells, cadherin-11ΔCBS L-cells and vector-control L-cells cultured on top of Matrigel. As expected, the cadherin-11ΔJMD L-cells, which are still capable of cell-to-cell adhesion, formed aggregates on top of the Matrigel. However, we found that their invasive capacity was markedly reduced (Figure 2B, C). Thus, for the cadherin-11ΔJMD mutant, cell-to-cell adhesion was not affected, but invasion was critically dependent upon the cadherin-11 cytoplasmic JMD. By contrast, when cadherin-11ΔCBS L-cells were cultured on top of Matrigel, they appeared mainly as single cells, consistent with their impaired cell-to-cell adhesive function. In contrast to their impaired cell-to-cell adhesion, these cells penetrated into the Matrigel to a similar extent as the wild-type cadherin-11 L-cells (Figure 2B, C). Together, these experiments reveal differential regulation of cadherin-11 cell-to-cell adhesive function and cadherin-11 dependent invasion and indicate a critical role of the cadherin-11 cytoplasmic JMD for the invasive capacity of cells expressing cadherin-11 (Table 1).
Table 1.
Phenotypic characteristics of L-cells stably transfected with wild-type cadherin-11, mutant cadherin-11, or vector-control.
| WT | ΔJMD | ΔCBS | Vector | |
|---|---|---|---|---|
|
*Cell to cadherin-11-Fc substrate adhesion |
+++ | +++ | + | - |
|
*Cadherin-11 dependent intercellular motility |
+++ | + | - | - |
| *Rac1 activity | + | +++ | +++ | +++ |
| *Cadherin-11 dependent | ||||
| F-actin organization | ||||
| - longitudinal fibers | +++ | - | - | - |
| - cortical ring fibers | - | +++ | - | - |
|
+Cadherin-11 dependent cell invasion |
+++ | + | +++ | + |
WT, wild-type cadherin-11 L-cells; ΔJMD, cadherin-11ΔJMD mutant L-cells; ΔCBS, cadherin-11ΔCBS mutant L-cells; Vector, vector-control L-cells.
Phenotypic characteristics from reference 12.
Invasive phenotype described in the present manuscript.
Figure 2.
Cadherin-11 promotes cell invasion. A. Schematic diagram of cadherin-11 constructs. Wild-type cadherin-11 is shown at the top. The cadherin-11 ΔJMD mutant has the juxta-membrane domain deleted; the cadherin-11 ΔCBS mutant has the β-catenin sequence deleted. B. Cadherin mutant L-cells were cultured on top of Matrigel. The cadherin-11 ΔJMD L-cells formed tight aggregates, but their invasive capacity was greatly reduced. By contrast, the cadherin-11 ΔCBS L-cells penetrated into the Matrigel as single cells. One representative of 4 experiments is shown. Magnification 200X. C. Quantification of invasion of L-cell transfectants cultured on top of Matrigel as assessed by measuring the invasive depth in sections of Matrigel blocks. Statistical analysis revealed significant differences (*=P<0.05). D. Human RA FLS were either treated with cadherin-11-Fc protein or control −IgG, and invasion was analyzed using Matrigel Invasion Chambers. Statistical analysis revealed significant differences (*=P<0.05) in the invasive capacity between FLS that were treated with cadherin-11-Fc and FLS treated with control-IgG. Values are means±SD of triplicates. One experiment out of 2 is shown.
Blockade of Cadherin-11 Inhibits Invasion of Human Fibroblast-like Synoviocytes
Next, to determine if cadherin-11 also played a role in the invasion of human FLS we utilized a cadherin-11-Fc fusion protein that we previously demonstrated could bind to cadherin-11 on the cell surface (7). Thus, we determined its ability to inhibit cadherin-11 function in mediating cell invasion. 5 × 104 FLS expressing cadherin-11 were preincubated with the cadherin-11-Fc fusion protein or isotype matched human IgG1 (100 μg/ml, respectively) and placed into invasion chambers in the presence of cadherin-11-Fc protein or human IgG1 (100 μg/ml, respectively). Importantly, cadherin-11-Fc markedly impaired the invasive capacity of human FLS (32±6 cells/LPF). In comparison, when FLS were treated with control IgG1, the number of invasive cells was twice as high (76±6 cells/LPF). Almost no invasion (8±4 cells/LPF) was observed in control chambers lacking fetal bovine serum as a chemoattractant (Figure 2D). Together, the experiments using L-cell transfectants and these using cadherin-11 Fc protein blockade of primary FLS emphasize that cadherin-11 function is critical for the ability of fibroblasts to invade through matrix.
Discussion
In rheumatoid synovitis the inflammatory microenvironment yields the formation of an abnormal pannus tissue that invades into the articular cartilage from the edge of the joint. The prevailing cell type at the cartilage-pannus junction is of mesenchymal origin and presumably derives from FLS (1). In this study, using immunohistochemistry, we demonstrate striking expression of cadherin-11 in cells invading into the cartilage. Further, we establish that cadherin-11 facilitates the invasive behavior of cells expressing cadherin-11. Transfection of cadherin-11 conferred on cells invasive behavior. This was critically dependent on its catenin-binding cytoplasmic JMD. Further, inhibition of cadherin-11 by a cadherin-11-Fc fusion protein diminished the invasive capacity of primary FLS in vitro.
Tissue invasion is a complex process that involves increased motility of invading cells and destruction of the host tissue by proteolytic enzymes. In the context of the pannus tissue in RA, the cells must acquire the capability to reorganize their adhesive cellular interactions and to crawl over one another in order to extend the tissue mass into the adjacent cartilage. Cadherins mediate both the stable homophilic cell-to-cell adhesion that is critical for tissue integrity but at the same time cadherin bonds are not static and continuously dismantle and reform allowing cell rearrangements necessary for tissue extension and invasion (9). Moreover, formation of intercellular junctions involves recruitment of machinery for actin cytoskeletal reorganization that generates the force for movement of cells on other cells (13).
Our results indicate that cadherin-11 function in mediating invasive capability is critically dependent on its cytoplasmic juxta-membrane domain. Cadherin-11ΔJMD L-cells were still able to form tight aggregates, but their invasive behavior was lost. The cytoplasmic juxta-membrane domain of cadherins is highly conserved and known to bind cytoplasmic proteins, including p120-catenin, implicated in modulating cadherin adhesive strength by signaling mechanisms that act on its amino-terminal region (14). Thus, our finding suggests that the cadherin-11 cytoplasmic JMD and its binding partners participate in a signaling circuit that regulates the invasive migratory behavior of these cells.
During development, epithelial cadherins, such as E-cadherin, are responsible for the formation and maintenance of epithelial structures, while expression of cadherin-11 is associated with increased cellular motility and tissue outgrowth (10). Further, recent studies identified the anomalous expression of cadherin-11 on epithelial cancer cells associated with enhanced tumor cell invasiveness (11). Our results are consistent with these observations in tumor cells and demonstrate that cadherin-11 operates in FLS, similarly as in tumor cells, to increase cell invasion. The abundant expression of cadherin-11 at the cartilage-pannus junction suggests that cadherin-11 and its associated intracellular molecular complex may serve as a molecular tool that allows for the pathological extension and invasion of pannus tissue into the articular cartilage.
Acknowledgments
We are grateful to Terry Bowman, David Bowman, and Carl-Walter Steiner for technical assistance.
The authors received grant support from the National Institutes of Health, AR048114-06A1; AI065858-03, and the American College of Rheumatology, Research and Education Foundation (to M.B.B.), the Arthritis Foundation (to H.P.K.) and the EU Framework 6 Integrated Project (Autocure, Project LSHB-CT-2006-018661 to J.S.S.).
References
- 1.Firestein GS. Evolving concepts of rheumatoid arthritis. Nature. 2003;423(6937):356–61. doi: 10.1038/nature01661. [DOI] [PubMed] [Google Scholar]
- 2.Castor CW. The microscopic structure of normal human synovial tissue. Arthritis & Rheumatism. 1960;3:140. doi: 10.1002/art.1780030205. [DOI] [PubMed] [Google Scholar]
- 3.Yielding KL, Tomkins GM, Bunim JJ. Synthesis of hyaluronic acid by human synovial tissue slices. Science. 1957;125:1300. doi: 10.1126/science.125.3261.1300. [DOI] [PubMed] [Google Scholar]
- 4.Schumacher HR, Kitridou RC. Synovitis of recent onset. A clinicopathologic study during the first month of disease. Arthritis & Rheumatism. 1972;15(5):465–85. doi: 10.1002/art.1780150502. [DOI] [PubMed] [Google Scholar]
- 5.Muller-Ladner U, Gay S. MMPs and rheumatoid synovial fibroblasts: Siamese twins in joint destruction? Ann Rheum Dis. 2002;61(11):957–9. doi: 10.1136/ard.61.11.957. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Valencia X, Higgins JM, Kiener HP, Lee DM, Podrebarac TA, Dascher CC, et al. Cadherin-11 provides specific cellular adhesion between fibroblast-like synoviocytes. J Exp Med. 2004;200(12):1673–9. doi: 10.1084/jem.20041545. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kiener HP, Lee DM, Agarwal SK, Brenner MB. Cadherin-11 induces rheumatoid arthritis fibroblast-like synoviocytes to form lining layers in vitro. Am J Pathol. 2006;168(5):1486–99. doi: 10.2353/ajpath.2006.050999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lee DM, Kiener HP, Agarwal SK, Noss EH, Watts GF, Chisaka O, et al. Cadherin-11 in synovial lining formation and pathology in arthritis. Science. 2007;315(5814):1006–10. doi: 10.1126/science.1137306. [DOI] [PubMed] [Google Scholar]
- 9.Gumbiner BM. Regulation of cadherin-mediated adhesion in morphogenesis. Nat Rev Mol Cell Biol. 2005;6(8):622–34. doi: 10.1038/nrm1699. [DOI] [PubMed] [Google Scholar]
- 10.Locascio A, Nieto MA. Cell movements during vertebrate development: integrated tissue behaviour versus individual cell migration. Curr Opin Genet Dev. 2001;11(4):464–9. doi: 10.1016/s0959-437x(00)00218-5. [DOI] [PubMed] [Google Scholar]
- 11.Pishvaian MJ, Feltes CM, Thompson P, Bussemakers MJ, Schalken JA, Byers SW. Cadherin-11 is expressed in invasive breast cancer cell lines. Cancer Res. 1999;59(4):947–52. [PubMed] [Google Scholar]
- 12.Kiener HP, Stipp CS, Allen PG, Higgins JM, Brenner MB. The cadherin-11 cytoplasmic juxtamembrane domain promotes alpha-catenin turnover at adherens junctions and intercellular motility. Mol Biol Cell. 2006;17(5):2366–76. doi: 10.1091/mbc.E05-08-0745. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Uemura T, Oda H, Kraut R, Hayashi S, Kotaoka Y, Takeichi M. Zygotic Drosophila E-cadherin expression is required for processes of dynamic epithelial cell rearrangement in the Drosophila embryo. Genes Dev. 1996;10(6):659–71. doi: 10.1101/gad.10.6.659. [DOI] [PubMed] [Google Scholar]
- 14.Ireton RC, Davis MA, van Hengel J, Mariner DJ, Barnes K, Thoreson MA, et al. A novel role for p120 catenin in E-cadherin function. J Cell Biol. 2002;159(3):465–76. doi: 10.1083/jcb.200205115. [DOI] [PMC free article] [PubMed] [Google Scholar]


