Expression of extracellular matrix metalloproteinase inducer and enhancement of the production of matrix metalloproteinases in rheumatoid arthritis

Authors


Abstract

Objective

To investigate the expression of extracellular matrix metalloproteinase inducer (EMMPRIN) at sites of joint destruction in rheumatoid arthritis (RA) and to correlate it with the production of matrix metalloproteinases (MMPs).

Methods

Reverse transcription-polymerase chain reaction was performed to study the existence of EMMPRIN in synovial tissue derived from RA and osteoarthritis (OA) patients. In situ hybridization with a human complementary DNA specific for EMMPRIN and immunohistochemistry were performed to characterize the EMMPRIN-expressing cells at sites of joint destruction, including bone. Northern blot analysis was performed to detect the level of expression of EMMPRIN messenger RNA (mRNA) in synovial tissue. The production of MMP-1 and MMP-3 by synovial tissue from RA patients was examined by enzyme-linked immunosorbent assay.

Results

Expression of EMMPRIN mRNA was detected in synovium from 9 of 11 patients with RA and 1 of 5 patients with OA. The presence of mRNA encoding EMMPRIN was recognized in the invasive synovium at sites of joint destruction in RA but not OA. Fibroblast-like synovial cells and granulocytes were demonstrated to express EMMPRIN mRNA. MMP-1 and MMP-3 production by synovial tissue was correlated with levels of expression of EMMPRIN mRNA, as detected by Northern blotting.

Conclusion

The expression of EMMPRIN stimulates the production of MMP-1 and MMP-3 in the synovial tissue of affected joints in RA. The results of this study suggest that EMMPRIN may be one of the important factors in progressive joint destruction in RA.

Rheumatoid arthritis (RA) is a chronic inflammatory disease characterized by progressive joint destruction. Previous reports have demonstrated that at sites of joint destruction, abnormally expressed matrix metalloproteinases (MMPs) are involved (1–4). Joint degradation involves damage to articular cartilage caused by inflammatory cells, activated fibroblasts in the synovial membrane, and chondrocytes. Several cytokines have been shown to play a role in activating the cells that produce MMPs (5).

MMPs degrade collagens, proteoglycans, and other matrix macromolecules in bone as well as in articular cartilage. Notably, it has been shown that collagenase 1 (MMP-1) and stromelysin 1 (MMP-3) are produced by fibroblasts and macrophage-like cells in the synovium and pannus and have been found to be important in the pathologic destruction of joints in patients with RA (3, 4). Suppression of MMPs may be a potential alternative therapeutic target for the treatment of joint destruction in RA. However, the precise pathomechanism of MMP production at the site of joint destruction remains partly unknown.

Extracellular matrix metalloproteinase inducer (EMMPRIN; formerly called tumor cell-derived collagenase stimulatory factor) is a 57-kd transmembrane glycoprotein that is a member of the immunoglobulin superfamily located on the surface of human tumor cells and normal keratinocytes (6). EMMPRIN interacts with fibroblasts to stimulate the expressions of several MMPs associated with tissue degradation and remodeling during tumor invasion and wound healing. The expression of EMMPRIN has been shown to be up-regulated in the synovial membrane of RA patients (7). However, the contribution of EMMPRIN to joint destruction in RA is still unknown.

In the present study, we investigated the expression of EMMPRIN and characterized EMMPRIN-expressing cells at sites of joint destruction in RA. We also analyzed the correlation of EMMPRIN expression with the activities of MMP-1 and MMP-3 in synovial tissue.

PATIENTS AND METHODS

Patients and specimens. Joint specimens were obtained from 11 patients with RA who were undergoing joint reconstruction surgery at Osaka University Hospital. All RA patients satisfied the 1987 revised diagnostic criteria of the American College of Rheumatology (formerly, the American Rheumatism Association) (8). Joint specimens were also obtained from 5 patients with osteoarthritis (OA) for use as controls. All study patients gave their informed consent.

Reverse transcription—polymerase chain reaction (RT-PCR). Using a Fast Track messenger RNA (mRNA) isolation kit (Invitrogen, San Diego, CA), RNA was extracted from synovial tissue. The isolated total RNA (3 mg) was reverse transcribed to complementary DNA (cDNA) with a Ready to Go T-Primed First-Strand kit (Amersham Pharmacia Biotech, Buckinghamshire, UK). The completed first-strand cDNA was amplified by PCR with primers specific for EMMPRIN (sense 5′-ACATCAACGAGGGGGAGACG-3′ and antisense 5′-GGCTTCAGACAGGCAGGACA-3′). As a control, β-actin mRNA was used.

Amplification was performed by adding 28.7 ml of sterile diethyl pyrocarbonate-water, 8 ml of 1.25 mM of each dNTP mix, 0.25 ml of 10× PCR buffer, and 0.25 ml of AmpliTaq DNA polymerase at 5 units/ml to 3 ml of each completed first-strand reaction. The reaction tubes were heated to 94°C for 5 minutes and incubated in a GeneAmp 1000 PCR system (Perkin Elmer, Foster City, CA) thermal cycler using 28 cycles of 94°C for 30 seconds, 62°C for 30 seconds, and 72°C for 60 seconds. Finally, the samples were incubated at 72°C for 10 minutes. PCR products were electrophoresed on 1.5% agarose gels and visualized by ethidium bromide staining.

Tissue samples. Tissue samples were prepared as previously described (9). Tissue samples with invasive synovium with bone were fixed in 4% paraformaldehyde (Merck, Darmstadt, Germany) in phosphate buffered saline (PBS), pH 7.4 (Sigma, St. Louis, MO), decalcified in 20% EDTA, dehydrated in an ethanol series, and embedded in paraffin. Serial sections (5 μm) were cut with a microtome, and some were stained with hematoxylin and eosin. The remaining sections were prepared for in situ hybridization (ISH) and immunohistochemistry.

In situ hybridization. ISH was carried out as previously described (9). The sections were dewaxed, incubated with 1 mg of proteinase K in TE (0.1M Tris, pH 8.0, 50 μM EDTA, pH 8.0) at 37°C for 10 minutes, and fixed with 4% paraformaldehyde in 0.1M PBS for 20 minutes at room temperature. The sections were then treated with 0.2N HCl to inactivate endogenous alkaline phosphatase, acetylated with 0.25% acetic anhydride in 0.1M triethanolamine (pH 8.0), dehydrated again with an ethanol series, and air-dried. Hybridization was performed at 50°C for 16 hours in a moist chamber with 50 μl of hybridization solution (50% deionized formamide, 10% Dextran sulfate, 1× Denhardt's solution), 4× SSC (0.15M NaCl, 0.015M sodium citrate), and 150 mg of Escherichia coli transfer RNA containing ∼0.5 mg/ml of RNA probe on each section.

After hybridization, the sections were washed briefly in 5× SSC and in 50% formamide, 2× SSC at 50°C for 30 minutes and then rinsed in 1× TES (10 mM Tris HCl, pH 7.6, 1 mM EDTA, 0.5M NaCl) at 37°C for 15 minutes. The sections were then treated with RNase A (10 mg/ml of 1× TES) at 37°C for 30 minutes, rinsed in 1× TES for 15 minutes at 37°C, and washed twice at 50°C for 20 minutes each with 2× SSC and 0.2× SSC.

Hybridized digoxigenin (DIG)-labeled probes were detected with the aid of a nucleic acid detection kit (Boehringer Mannheim, Mannheim, Germany). DIG-11-UTP-labeled single-stranded complementary RNA was prepared with a DIG RNA labeling kit (Boehringer Mannheim). A human EMMPRIN cDNA was obtained by PCR, and was subcloned into pGEM-T plasmid. The plasmid was linearized by Sac II and transcribed by SP6 RNA polymerase to generate a 0.851-kb-long antisense probe. The plasmid was also linearized by Spe I and transcribed by T3 RNA polymerase to generate a sense probe.

Immunohistochemical staining. To identify the cells expressing EMMPRIN mRNA on ISH, immunohistochemical staining was performed by the streptavidin-peroxidase technique using Histofine streptavidin-biotin-peroxidase kits (Nichirei, Tokyo, Japan). Monoclonal antibody (mAb) against CD3 (Nichirei) was used to detect T cells, mAb against CD15 (Dako, Santa Barbara, CA) was used to detect granulocytes, mAb against CD20 (Dako) was used to detect B cells, mAb against CD68 (Dako) was used to detect macrophages, and mAb against prolyl-4-hydroxylase (Fuji Chemical, Toyama, Japan) was used to detect fibroblasts.

The serial sections used for ISH were incubated with 10% normal rabbit nonimmune serum for minimizing background staining, and then were incubated with primary antibody for 2 hours at room temperature. Normal mouse serum was used as a control for the primary antibody. After washing in PBS, pH 7.2, the sections were incubated with secondary antibody for 20 minutes at room temperature in a humidified chamber. They were placed in 3% H2O2 in methanol to block endogenous peroxidase, then incubated with peroxidase-conjugated streptavidin for 20 minutes at room temperature in a humidified chamber. Finally, sections were washed with PBS, and substrate reagent (3,3′-diaminobenzidine tetrahydrochloride; Dojindo, Tokyo, Japan) was added. Counterstaining was performed with hematoxylin, and the sections were mounted.

Tartrate-resistant acid phosphatase (TRAP) staining. TRAP staining was performed as described previously (10). Briefly, the consecutive sections used for ISH were incubated for 30–90 minutes with medium containing 11.5 mg of disodium tartrate (Wako, Osaka, Japan) and 7 mg of naphthol-AS-TR phosphate and Fast Red TR (Sigma) in 5 ml of 0.2M acetate buffer, pH 5.0. Counterstaining was performed with hematoxylin, and the sections were mounted.

Measurement of MMP levels produced by synovial tissue. Synovial tissue samples from 4 patients with RA that expressed EMMPRIN mRNA and from 2 patients with RA that did not express EMMPRIN mRNA were studied. Synovial tissue was cut into small pieces (20 mg wet volume), and cultured in 24-well microtiter plates (Becton Dickinson Labware, Franklin Lakes, NJ) with Dulbecco's modified Eagle's medium containing 10% fetal bovine serum (Gibco BRL, Grand Island, NY). After 48 hours, culture medium was collected for the measurement of MMP-1 and MMP-3. The assays for human MMP-1 in the supernatant were performed with an enzyme-linked immunosorbent assay (ELISA) kit (Amersham) according to the manufacturer's protocol. The assays for human MMP-3 in the supernatant were performed with an ELISA using a 1-step sandwich method (Fuji Chemical), as described previously (11).

Northern blot and densitometric analysis. Thirty micrograms of total RNA was subjected to electrophoresis on a 1.5% agarose-formaldehyde denaturing gel and transferred to a nitrocellulose membrane (Amersham). The filter was backed, prehybridized, and hybridized; EMMPRIN oligonucleotide was used as a probe, and the RNA was labeled by 3′-end labeling for Northern blotting. The filter was then washed and exposed to radiographic film. The relative intensities of the bands of interest were analyzed with the use of an NSF-300G scanner (Microtek, Anaheim, CA) and scan analysis software (Biosoft, Palo Alto, CA). Results are expressed as the ratio of band intensities of the sample relative to the band intensities of the GAPDH mRNA as determined by Northern blot analysis.

Statistical analysis. All values are expressed as the mean ± SD. Statistical analyses were performed with the Mann-Whitney U test, and P values less than 0.05 were considered significant. Analysis of statistical correlations was performed using Pearson's correlation coefficient.

RESULTS

Expression of EMMPRIN mRNA in synovium. We first studied the expression of EMMPRIN mRNA in synovial tissue. RT-PCR revealed EMMPRIN mRNA expression in synovial samples from 9 of the 11 RA patients and only 1 of the 5 OA patients (Figure 1

Figure 1.

Representative reverse transcription-polymerase chain reaction showing extracellular matrix metalloproteinase inducer bands (492 bp) in synovial tissue samples from 11 patients with rheumatoid arthritis (RA) and 5 patients with osteoarthritis (OA). Lane M contains DNA markers.

Localization of EMMPRIN mRNA expression at the site of joint destruction. The results of ISH demonstrated marked expression of EMMPRIN mRNA at the site of joint destruction in RA (Figures 2A and C). This expression was detected in 28.0 ± 7.4% (mean ± SD) of the mononuclear cells of the invasive synovium (Figure 2B) and in 36.3 ± 8.1% of fibroblast-like cells at the site of bone destruction (Figure 2D) in joint specimens from RA patients. However, little expression of EMMPRIN mRNA was detected in the samples of synovium and bone obtained from the OA patients (Figure 2E).

Figure 2.

In situ hybridization of extracellular matrix metalloproteinase inducer (EMMPRIN) mRNA in synovium and bone from patients with rheumatoid arthritis (RA) and osteoarthritis (OA). Expression of EMMPRIN mRNA was confirmed in both A, the invasive synovium (arrows) and C, the eroded bone (arrows) in sections from a patient with RA and was detected in B, mononuclear cells of the invasive synovium (arrows) as well as in D, fibroblast-like cells at the site of bone erosion (arrows). E, Little expression of EMMPRIN mRNA was noted in the sample from the patient with OA. (Original magnification × 100 in A, B, and E; × 400 in C and D.)

Characterization of cells expressing EMMPRIN. To characterize EMMPRIN-expressing cells, immunohistochemical staining was performed. The cells expressing EMMPRIN mRNA on ISH were negative for CD3, CD20, and CD68 (data not shown). Of the cells positive for CD15, 28.2 ± 5.8% (mean ± SD) expressed EMMPRIN mRNA on ISH (Figure 3A). Of the cells positive for prolyl-4-hydroxylase, 23.3 ± 9.1% expressed EMMPRIN mRNA on ISH (Figure 3B).

Figure 3.

Immunohistochemical (IH) staining for CD15 and prolyl-4-hydroxylase and in situ hybridization (ISH) of synovial tissue sections from a patient with rheumatoid arthritis. Cells that expressed extracellular matrix metalloproteinase inducer mRNA on ISH (arrows) were also positive for CD15 and prolyl-4-hydroxylase (arrows). (Original magnification × 400 for CD15; × 200 for prolyl-4-hydroxylase.)

The results of TRAP staining showed that the cells that expressed EMMPRIN mRNA on ISH were negative for TRAP both in the synovium (Figure 4A) and in the bone (Figure 4B). These results demonstrated that the cells expressing EMMPRIN mRNA were synovial fibroblast-like cells and granulocytes, not infiltrating lymphocytes or synovial macrophages, and that neither osteoclasts nor osteoclast-like cells in the synovium and bone at the site of joint destruction expressed EMMPRIN.

Figure 4.

Tartrate-resistant acid phosphatase (TRAP) staining and in situ hybridization (ISH) of serial sections of synovium and bone from a patient with rheumatoid arthritis. Cells in both the synovium and bone that expressed extracellular matrix metalloproteinase inducer mRNA on ISH were negative for TRAP. (Original magnification × 400.)

Enhanced production of MMP-1 and MMP-3 by synovial tissue expressing EMMPRIN. To study the relationship between EMMPRIN expression and the production of MMP-1 and MMP-3 in synovial tissue from patients with RA, we performed Northern blot analysis and used an ELISA kit to measure the levels of these metalloproteinases in the culture medium from synovial tissues obtained from 6 patients. In 2 synovial tissue samples, EMMPRIN mRNA was under the detection limit; however, in 4 samples, gene expression of EMMPRIN was confirmed (Figure 5A).

Figure 5.

Relationship between the expression of extracellular matrix metalloproteinase inducer (EMMPRIN) mRNA and the production of matrix metalloproteinases (MMPs) 1 and 3 by synovial tissue from 6 patients with rheumatoid arthritis. A, Gene expression of EMMPRIN in synovial tissue. B and C, Relationship between synovial tissue expression of EMMPRIN mRNA and the production of MMP-1 and MMP-3, respectively, by Pearson's correlation coefficient. EMMPRIN mRNA signals were measured by densitometry and were normalized relative to those of GAPDH.

In the culture medium of synovial tissue from an RA patient expressing EMMPRIN mRNA, the average level of MMP-1 was 3,559.0 ± 677.8 pg/ml/mg and the average level of MMP-3 was 588.8 ± 132.5 pg/ml/mg (mean ± SD). In the culture medium of synovial tissue from an RA patient expressing no EMMPRIN mRNA, the average level of MMP-1 was 2,164.2 ± 1319 pg/ml/mg and the average level of MMP-3 was 208.3 ± 125.6 pg/ml/mg. Production of MMP-1 and MMP-3 by synovial tissue correlated with levels of EMMPRIN mRNA expression detected by Northern blotting (r = 0.850, P = 0.0295 for MMP-1 and r = 0.903, P = 0.0098 for MMP-3) (Figures 5B and C).

DISCUSSION

In this study, we clearly showed the presence of mRNA encoding EMMPRIN in synovium at the site of joint destruction in tissue from RA patients. However, little EMMPRIN expression was found in synovium from OA patients. These results are consistent with the previous report by Konttinen et al (7). By RT-PCR, only 1 of the 5 synovial samples from OA patients revealed a very weak band indicating the presence of EMMPRIN mRNA. We studied the clinical background and laboratory data of the patient from whom this synovial sample was taken and found no striking difference between this patient and the other 4 OA patients examined. Previous studies have demonstrated that most normal adult tissues, including epidermis, retinal pigment epithelium, and breast lobules and ductules, express very low levels of EMMPRIN, which suggests that EMMPRIN may play a physiologic role in tissue remodeling by inducing stromal MMPs (6, 12–14). The EMMPRIN mRNA found in the synovium of the patient with OA may reflect tissue remodeling, localized inflammation, or wound healing occurring in that patient.

To characterize the EMMPRIN-expressing cells, we performed immunohistochemical staining of serial sections of tissues. EMMPRIN expression was found in the synovial fibroblast-like cells and granulocytes, but not in the infiltrating lymphocytes or macrophage-like cells. In tumors, EMMPRIN has been detected on the surface of tumor cells, but not on adjacent fibroblasts (15–17). In RA, however, fibroblast-like synovial cells are the primary cells that express EMMPRIN. In addition to a paracrine stimulation by granulocytes expressing EMMPRIN, there may be an autocrine stimulation of fibroblasts to produce MMPs in the synovium. A recent study has demonstrated that EMMPRIN expression was present at the surface of both tumor epithelial cells and peritumor stromal cells from breast and lung tissues (18). Granulocytes from the peripheral blood of patients with RA express higher levels of EMMPRIN compared with healthy donors (19). Granulocytes that have migrated to affected joints are peripheral granulocytes, and our histologic results confirmed their increased expression of EMMPRIN.

Although the precise function of EMMPRIN remains unknown, recent experimental data have demonstrated that EMMPRIN stimulates the production of interstitial collagenase, stromelysin 1, and gelatinase A in fibroblasts (20, 21). These results suggest a connection between EMMPRIN and progressive joint destruction, because synovial cells from patients with RA have been demonstrated to be the principal source of several MMPs. The results of the present study demonstrated that synovial tissue that expressed EMMPRIN mRNA produced significantly more MMP-1 and MMP-3 than did synovial tissue that did not express EMMPRIN mRNA, and the findings strongly suggested that MMP-1 and MMP-3 production was stimulated by EMMPRIN in affected RA joints.

In this study, we showed a correlation between the expression of EMMPRIN and the production of MMP-1 and MMP-3 on Northern blot analysis. It would be difficult, however, to clearly demonstrate regulation of the expression and activity of MMPs by EMMPRIN. Previous studies have shown that EMMPRIN stimulates fibroblasts to produce 3–10 times the usual level of MMP-1 and MMP-3 in tumor cells and Chinese hamster ovary cells (20–22). The present study showed a 1.5–3 times increase in the production of MMP-1 and MMP-3. Of course, EMMPRIN expression alone does not appear to regulate MMP expression; the results of this study suggest that EMMPRIN may be involved in the regulation of MMP expression in the RA synovium. In addition, EMMPRIN may play a role in the degradation of bone and cartilage associated with synovium invasion by stimulating the synthesis of several MMPs by synovial fibroblast-like cells.

Acknowledgements

We wish to thank Kaori Izumi and Shouko Kuroda for excellent technical assistance.

Ancillary