Oxidized Low-Density Lipoproteins Stimulate Extracellular Matrix Metalloproteinase Inducer (EMMPRIN) Release by Coronary Smooth Muscle Cells
http://www.100md.com
《动脉硬化血栓血管生物学》
From the Central Department of Clinical Chemistry (C.H., C.L., F.D., M.G.B.), University Hospital Ulm, Germany; and the Laboratory of Molecular Biology (F.D.), Faculty of Medicine, Catholic University of Santisima. Concepción, Chile.
ABSTRACT
Objective— Matrix metalloproteinases (MMPs) seem to play a prominent role in atherogenesis. Extracellular MMP inducer (EMMPRIN), a cell surface glycoprotein which stimulates MMP synthesis, has recently been detected in human atheroma. We have investigated the influence of oxidized low-density lipoproteins (oxLDLs) on EMMPRIN expression in human coronary artery smooth muscle cells (HCA-SMCs).
Methods and Results— OxLDL induced a significant increase of EMMPRIN release into HCA-SMC supernatants and a concomitant decrease of cell-associated EMMPRIN. These effects were antagonized by antioxidants as well as by EDTA and the MMP inhibitor GM6001. Western blot analysis demonstrated that MMP-1 and MMP-2 induce the cleavage of the extracellular domain from cell-associated EMMPRIN. MMP-1 and MMP-2 synthesis was upregulated by oxLDL, and, in addition, we have shown that soluble EMMPRIN, isolated from macrophage supernatants, increased MMP-1 and MMP-2 synthesis in HCA-SMC.
Conclusion— Our data suggest that oxLDLs stimulate the release of soluble EMMPRIN, at least in part, by MMP-dependent shedding from the cell surface. Additionally, oxLDLs might induce a circular upregulation of matrix degradation because, in turn, soluble EMMPRIN stimulates MMP synthesis in HCA-SMC.
This study demonstrates that oxidized low-density lipoproteins stimulate the release of soluble extracellular matrix metalloproteinase inducer (EMMPRIN) by human coronary artery smooth muscle cells, an effect which seems to result from enhanced EMMPRIN shedding. In addition, we have shown that isolated purified EMMPRIN stimulates MMP synthesis in coronary smooth muscle cells.
Key Words: smooth muscle cells ? low density lipoproteins ? matrix metalloproteinases ? extracellular MMP inducer ? atherosclerosis
Introduction
Deposition of low-density lipoproteins (LDLs) in the vessel wall and their oxidative modification seem to initiate, or at least accelerate, the atherosclerotic process by several mechanisms, including promotion of foam cell formation, chemotactic effects on monocytes, and mitogenic effects on smooth muscle cells. Recent studies have demonstrated that oxidized LDLs (oxLDLs) increase the expression of matrix metalloproteinases (MMPs) in endothelial cells, monocyte-derived macrophages, and smooth muscle cells1–3 and that MMP expression is enhanced in atherosclerotic plaque tissue.4 MMPs increase smooth muscle cell migration and proliferation5,6 and also seem to promote plaque instability by inducing extracellular matrix degradation.7 Expression of extracellular MMP inducer (EMMPRIN) in human atheroma has been described recently.8,9 EMMPRIN, also called CD147 or basigin, is a highly glycosylated plasma membrane protein (58 kDa) belonging to the immunoglobulin superfamily. EMMPRIN is expressed on several cell types like leukocytes and different tumor cells10 and stimulates MMP production in fibroblasts and various tumor cells.11,12 Several studies have shown that cancer cells express significantly higher EMMPRIN levels than normal cells, and it has been proposed that elevated EMMPRIN expression in tumor cells may promote tumor progression by inducing MMP expression in peritumoral stromal cells. In this study, we have demonstrated that EMMPRIN is expressed in cultured human coronary artery smooth muscle cells (HCA-SMCs) and that oxLDL significantly enhanced the release of soluble EMMPRIN as well as MMP-1 and MMP-2 release. In addition, we have shown that MMP-1 and MMP-2 seem to promote the cleavage of soluble EMMPRIN from the cell surface and that soluble EMMPRIN stimulates MMP synthesis in HCA-SMC. Thus our data suggest that oxLDL might induce a circular upregulation of matrix degradation.
Materials and Methods
For more detail, please see the online Methods section, available at http://atvb.ahajournals.org.
Cell Culture
HCA-SMCs (Clonetics, San Diego, Calif) were cultured in SmGM-2 BulletKit medium (BioWhittaker). Experiments were performed with cells from passages 5 to 8 in culture medium containing 0.1%, 1.5%, or 5% fetal calf serum (FCS), respectively.
Human monocytes were isolated from buffy coats of healthy donors by density gradient separation and cultured for 8 to 10 days before supernatants were collected for EMMPRIN isolation.
Isolation and Oxidation of LDL
LDLs were isolated from blood samples of healthy volunteers by sequential ultracentrifugation. For preparation of native LDL (nLDL), plasma samples (4 μmol EDTA per mL blood) were supplemented with butylated hydroxytoluene (BHT, 20 μmol/L) to prevent oxidation. LDLs isolated from serum samples of the same donors were oxidized by exposure to 5 μmol/L CuSO4 and oxygen for 24 hours at 37°C. For controls, "native" LDL was also prepared from serum samples without addition of EDTA and BHT. Lipoprotein concentrations are expressed in terms of their protein content. The degree of oxidation was quantified by absorption measurement at 234 nm and fluorescence measurement at 430 nm.13
Immunometric Measurement of EMMPRIN and MMP
For EMMPRIN quantification, microtitre plates coated with rabbit-anti-mouse IgG and monoclonal anti-CD147 antibody (extracellular domain; R&D Systems) were incubated with standard (R&D Systems) or sample, biotinylated polyclonal anti-CD147 antibody (extracellular domain; R&D Systems), streptavidin-Europium (Delfia Wallac), and enhancer solution. Time-resolved fluorescence of the Europium-chelate was measured with a Victor Multilabel Counter (Wallac). All measurements were done in duplicate, intraassay coefficient of variation was 8.7% (n=30).
Total MMP-1 was measured with an ELISA kit (Amersham Biosciences), which recognizes free MMP-1 and MMP-1 complexed with inhibitors, such as tissue inhibitor of metalloproteinases-1. For MMP-2 quantification, microtitre plates coated with rabbit-anti-mouse IgG and monoclonal anti-MMP-2 antibody (recognizes pro–MMP-2 and active MMP-2; R&D Systems) were incubated with standard (Chemicon) or sample, polyclonal biotinylated anti-MMP-2 antibody (R&D Systems), streptavidin-Europium and enhancer solution, followed by time-resolved fluorescence measurement. Measurements were done in duplicate, and intraassay coefficient of variation was 7.9% (n=30).
Total MMP-9 was measured with an ELISA kit (Oncogene), which detects free MMP-9 and MMP-9 bound to tissue inhibitor of metalloproteinase-1. EMMPRIN, MMP-1, MMP-2, and MMP-9 concentrations in supernatants were referred to the DNA content in the corresponding wells. DNA content was measured by fluorescent DNA staining with bisbenzimide using calf thymus DNA as standard.14
RNA Isolation and RT-PCR
For detection of EMMPRN mRNA, HCA-SMC were incubated for 2, 3, 4, 6, 8, and 12 hours with oxLDL and nLDL (1 to 20 μg/mL) in medium containing 0.1% FCS. After extraction of total RNA and reverse transcription, DNA amplification (human EMMPRIN15,16: Acc NM_001728; GAPDH17: Acc J02642) was performed using the LightCycler technology (Idaho Technology; SYBR Green I). Reactions were cycled 34 to 40x, melting curves were performed and polymerase chain reaction (PCR) products quantified with the LightCycler software using purified sequenced PCR product as standard. EMMPRIN mRNA was referred to GAPDH mRNA. No detectable PCR products were present in water controls and in controls, amplified without prior reverse transcription. For visualization, PCR products were applied to 1% agarose gel.
EMMPRIN Isolation and Purification
Macrophage supernatant and HCA-SMC supernatant were applied at 2 mL/min to a 20HQ anion exchange column (Applied Biosystems). Bound proteins were eluted with a linear gradient of NaCl (0 to 1 mol/L), 1 mL fractions were collected, and EMMPRIN was identified by time-resolved fluorescence immunoassay. EMMPRIN-containing fractions were applied to an immunoaffinity column, where agarose protein G was coupled to monoclonal anti-human EMMPRIN antibody (Pierce). The column was washed, EMMPRIN was eluted, and the eluted protein was dialyzed and concentrated. HCA-SMC lysate was first purified by immunoaffinity chromatography, then applied to a 20HQ anion exchange column, eluted with a linear gradient of NaCl, and concentrated. EMMPRIN concentrations were measured by fluorescence immunoassay. For investigation of MMP-associated EMMPRIN cleavage, EMMPRIN isolated from HCA-SMC lysate was incubated for 2 hours with 0.1 μg/mL recombinant human MMP-1 or MMP-2 (R&D Systems). Thereafter, EMMPRIN was deglycosylated with a deglycosylation kit (Roche Diagnostics).
Western Blot
For detection of cell-associated EMMPRIN, cellular proteins were solubilized with lysis buffer. Sample volumes of cell lysates and supernatants were adjusted according to the protein content. SDS-PAGE was performed (cell lysate and supernatant, nonreducing conditions; isolated EMMPRIN, reducing conditions, sample volume adjusted to EMMPRIN content), and separated proteins were electroblotted to polyvinylidene difluoride membranes. Western blots were incubated with monoclonal anti-human EMMPRIN antibody (extracellular domain; R&D Systems), and for detection of MMP-associated EMMPRIN cleavage, an additional antibody against the C-terminus of EMMPRIN was used (Santa Cruz). After washing, blots were incubated with biotinylated secondary antibody and HRP-streptavidin, and finally chemiluminescence staining was performed.
Zymography
For detection of MMP activity, cells were incubated for 48 hours with oxLDL or nLDL (1 to 20 μg/mL) in medium containing 0.1% FCS. Supernatants were adjusted according to the protein content. Nonreducing SDS-PAGE was performed in 7.5% polyacrylamide gels containing 0.2% gelatin. After electrophoresis, gels were washed, incubated in developing buffer, stained with 0.34% Coomassie Blue, and destained with acetic acid and methanol.
Statistical Analysis
Results are expressed as mean±SEM and were evaluated by 1-way ANOVA, followed by the Dunnett test (comparison with the control group) or the Newman-Keuls test (comparison between all groups), respectively.
Results
EMMPRIN Release by HCA-SMC
OxLDL induced a significant concentration-dependent increase of EMMPRIN release (Figure 1). In contrast, EMMPRIN release was significantly reduced after incubation with higher concentrations of nLDL (1 μg/mL, 95.7±5.3% of control; 10 μg/mL, 94.1±4.4%; 20 μg/mL, 85.6±3.7%**; 50 μg/mL, 60.3±5.6%***; control, 100.0±1.7%, n=15; ** indicates P<0.01; ***, P<0.001 versus control). Because the nLDL preparation contained EDTA and BHT to prevent spontaneous LDL oxidation, controls were performed with EDTA and BHT containing medium. A similar decrease of EMMPRIN release was observed (EDTA and BHT concentrations corresponding to the final concentrations in 20 and 50 μg/mL nLDL, 91.9±7.9% and 69.6±3.3%*** of antioxidant-free control; ***, P<0.001, n=6), whereas nLDL, which was prepared without EDTA and BHT and underwent a slight spontaneous oxidation during the preparation procedure, induced a slight increase of EMMPRIN release (20 μg/mL, 112.7±6.4%; 50 μg/mL, 114.1±8.1%, n=6).
Figure 1. Effect of oxLDL on EMMPRIN release by HCA-SMC. Cells were incubated for 48 hours with oxLDL in medium containing 0.1% FCS. EMMPRIN concentrations in supernatants were measured by fluorescence immunoassay and referred to the DNA content in the corresponding wells. Mean±SEM (n=12), expressed as relative EMMPRIN release compared with controls; *P<0.05 and ***P<0.001 vs control.
Hydrogen peroxide, which was used to mimic oxidative stress, also induced a significant increase of EMMPRIN release (100 μmol/L, 154.2±3.4%** of control; 500 μmol/L, 182.1±15.1%***; control, 100.0±4.0%, n=6; ** indicates P<0.01; ***, P<0.001 versus control), whereas EDTA, an unspecific protease inhibitor, reduced EMMPRIN release (20 μmol/L, 70.5±9.1%*** of control; 200 μmol/L, 59.8±4.3%***; 500 μmol/L, 55.1±1.1%***; control, 100.0±4.0%, n=6; ***, P<0.001 versus control). In accordance with these findings, the oxLDL-induced increase of EMMPRIN release was significantly antagonized by the antioxidants trolox and probucol as well as by EDTA, and the basal EMMPRIN release was also slightly reduced by trolox and probucol (Figure 2). In addition, preincubation with vitamin C for 24 hours also significantly reduced the oxLDL-stimulated EMMPRIN release (oxLDL 50 μg/mL, 272.5±30.3% of control; oxLDL 50 μg/mL+vitamin C 50 μg/mL, 162.6±11.6%, n=6, P<0.01), whereas the basal EMMPRIN release was not significantly reduced by vitamin C. The results shown above were obtained in experiments performed in culture medium containing 0.1% FCS. Because oxidative stress might be decreased with higher FCS concentrations in the culture medium, most experiments were also performed with higher FCS concentrations (1.5% and 5%). However, similar results were obtained (data not shown).
Figure 2. Reduction of basal and oxLDL-induced EMMPRIN release by trolox, probucol, and EDTA. HCA-SMC were incubated for 48 hours with oxLDL (20 μg/mL and 50 μg/mL, n=6), 500 μmol/L trolox (trol., n=6), 200 μmol/L probucol (pro., n=6), 200 μmol/L EDTA (n=5), or a combination of oxLDL (20 μg/mL and 50 μg/mL) and trolox, probucol, or EDTA, respectively (n=6). EMMPRIN concentrations were measured by fluorescence immunoassay and referred to the DNA content in the corresponding wells. Mean±SEM, expressed as relative EMMPRIN release compared with controls: *P<0.05, **P<0.01, ***P<0.001 vs control (C.), +++P<0.001 vs oxLDL 20 μg/mL, and ###P<0.001 vs oxLDL 50 μg/mL.
Investigation of EMMPRIN mRNA expression by quantitative real-time RT-PCR using the LightCycler technology showed no relevant changes of EMMPRIN mRNA after incubation with oxLDL or nLDL (2 to 24 hours, 1 to 20 μg/mL).
Western blot analysis showed a decrease of cell-associated EMMPRIN and an increase of soluble EMMPRIN in the corresponding supernatants after incubation with oxLDL, and to a much lesser extent also after incubation with antioxidant-free nLDL (Figure 3). NLDL containing EDTA and BHT as well as EDTA alone inhibited the decrease of cell-associated EMMPRIN and in parallel reduced the EMMPRIN release into cell culture supernatants. The oxLDL-induced decrease of cell-associated EMMPRIN and the increase of soluble EMMPRIN in HCA-SMC supernatants was partly antagonized by trolox, probucol, and EDTA (Figure 3).
Figure 3. Western blot analysis (nonreducing conditions) of EMMPRIN expression in cell lysates (C) and corresponding supernatants (S) of HCA-SMC, incubated for 48 hours with the following substances. Control (Con.), medium with 0.1% FCS; ox50, 50 μg/mL oxidized LDL; n50, 50 μg/mL nLDL containing EDTA and BHT; EDTA, 500 μmol/L EDTA; n20+EDTA/BHT, 20 μg/mL nLDL containing EDTA and BHT; n20–EDTA/BHT, 20 μg/mL nLDL without EDTA and BHT; ox20, 20 μg/mL oxLDL; Con.,: medium with 0.1% FCS; ox20, 20 μg/mL oxLDL; ox20+trol., 20 μg/mL oxLDL+500 μmol/L trolox; ox20+pro., 20 μg/mL oxLDL+200 μmol/L probucol; and ox20+EDTA, 20 μg/mL oxLDL+500 μmol/L EDTA.
Quantitative EMMPRIN measurement in HCA-SMC supernatants demonstrated that the MMP inhibitor GM6001 (Chemicon; Ki values: human MMP-1, 0.4 nmol/L; MMP-2, 0.5 nmol/L; MMP-3, 27 nmol/L; MMP-8, 0.1 nmol/L; and MMP-9, 0.2 nmol/L) reduced the basal EMMPRIN release and significantly antagonized the oxLDL-induced EMMPRIN release (Figure 4A). In accordance with the quantitative measurement, EMMPRIN Western blot showed that the oxLDL-induced reduction of cell-associated EMMPRIN and the concomitant increase of EMMPRIN release was effectively antagonized by the MMP inhibitor (Figure 4B).
Figure 4. A, Influence of the MMP inhibitor GM6001 (GM, 15 μmol/L and 25 μmol/L) on the basal and oxLDL-induced EMMPRIN release by HCA-SMC. EMMPRIN concentrations in supernatants were measured by fluorescence immunoassay and referred to the DNA content in the corresponding wells. Mean±SEM, expressed as relative EMMPRIN release compared with controls: n=6, +P<0.05, +++P<0.001 vs control, and ***P<0.001 vs oxLDL 50 μg/mL. B, Western blot analysis (nonreducing conditions) of EMMPRIN expression in cell lysates (C) and corresponding supernatants (S) of HCA-SMC, incubated for 48 hours with Con., medium with 0.1% FCS; GM, GM6001 15 μmol/L; ox50+GM, oxLDL 50 μg/mL+GM6001 15 μmol/L; and ox50, oxLDL 50 μg/mL. C, Western blot analysis (reducing conditions, EMMPRIN antibody against the extracellular domain) of EMMPRIN, isolated from HCA-SMC supernatant (S) or cell lysate (C) by anion exchange and immunoaffinity chromatography. Isolated purified EMMPRIN from cell lysate was incubated for 2 hours with recombinant human MMP-1 or MMP-2; thereafter, isolated EMMPRIN from cell lysate and HCA-SMC supernatant was deglycosylated with N-Glycosidase F (N-Glyc.).
To further investigate whether soluble EMMPRIN is cleaved from the cell surface and whether MMP might be involved in EMMPRIN cleavage, EMMPRIN isolated from HCA-SMC lysate was incubated with recombinant human MMP-1 and MMP-2. Because the transmembrane and cytoplasmic domain exhibit a putative molecular weight of only 7 kDa, such a shift in molecular weight is not clearly visible with glycosylated EMMPRIN, which appears as broad band in Western blot analysis because of different degrees of glycosylation. Therefore, Western blots were performed with glycosylated and deglycosylated samples (deglycosylation was performed after incubation with MMP). Deglycosylated EMMPRIN without MMP pretreatment exhibited a molecular weight of 27 kDa, corresponding to the reported molecular weight of the native protein; preincubation with MMP-1 or MMP-2 resulted in the appearance of an additional band with a molecular weight of 20 kDa (Figure 4C). EMMPRIN isolation from HCA-SMC supernatant collected after incubation with oxLDL yielded a hydrophilic (73%) and a hydrophobic (27%) EMMPRIN-containing fraction in anion exchange chromatography. Purified EMMPRIN from the hydrophilic fraction exhibited molecular weights of 50 kDa (glycosylated form) and 20 kDa (after deglycosylation), respectively. Thus, the additional band of MMP-treated deglycosylated EMMPRIN from cell lysate exhibited the same molecular weight (20 kDa) as deglycosylated EMMPRIN from the hydrophilic fraction of HCA-SMC supernatant. The size difference between these fractions and deglycosylated EMMPRIN from untreated cell lysate corresponds to the putative molecular weight of the cytoplasmic and transmembrane domain of EMMPRIN (7 kDa). Glycosylated and deglycosylated EMMPRIN from the hydrophilic fraction of HCA-SMC supernatant as well as the 20 kDa form of MP-treated deglycosylated EMMPRIN from cell lysate were not detectable with an EMMPRIN antibody against the intracellular domain (data not shown).
Effect of oxLDL on MMP Release by HCA-SMC
Zymography of HCA-SMC supernatants showed predominant expression of MMP-2 and an upregulation of active MMP-2 after incubation with oxLDL (Figure 5). Quantitative MMP measurement showed an oxLDL-induced upregulation of MMP-1 and MMP-2 release (MMP-1, oxLDL 20 μg/mL 125.5±8.7% of control; 50 μg/mL 167.5±10.7%***, control 100.0±2.6%, n=6; MMP-2, oxLDL 20 μg/mL 119.4±6.4%, 50 μg/mL 190.8±17.6%***, control 100.0±2.4%, n=15; ***, P<0.001 versus control) and only minor changes after incubation with nLDL. Hydrogen peroxide also induced a significant concentration-dependent increase of MMP-2 secretion (100 μmol/L, 118.5±5.8% of control ; 500 μmol/L, 179.1±13.9% P<0.001 versus control, n=5) and a less pronounced increase of MMP-1 release (500 μmol/L, 136.9±2.2% of control , P<0.001 versus control, n=3). The oxLDL(50 μg/mL)-induced MMP release was significantly antagonized by preincubation with 50 μg/mL vitamin C (MMP-1, 206.4±8.8% of control versus 149.7±11.5%, P<0.01, n=3; MMP-2, 240.9±24.5% of control versus 175.9±8.3%, P<0.01, n=3).
Figure 5. Zymography of HCA-SMC supernatants after 48 hours incubation with oxLDL (1 μg/mL, 10 μg/mL, and 20 μg/mL: ox1, ox10, and ox20) or native LDL (1 μg/mL, 10 μg/mL, and 20 μg/mL: n1, n10, and n20).
Effect of Isolated Purified EMMPRIN on MMP Release by HCA-SMC
EMMPRIN isolated from macrophage supernatants exerted a significant stimulatory effect on MMP-1 and MMP-2 release and a slight stimulatory effect on MMP-9 release (Figure 6), whereas commercially obtained recombinant human EMMPRIN (R&D Systems) reduced MMP-1 release (rhEMMPRIN 1 ng/mL, 71.8±10.5%* of control; 10 ng/mL, 68.2±12.9%**; 20 ng/mL, 76.6±0.9%; 40 ng/mL, 81.0±2.3%, control 100.0±2.7%, n=3; *, P<0.05; **, P<0.01 versus control) and did not significantly alter MMP-2 release (rhEMMPRIN 1 ng/mL, 78.8±6.5%%; 10 ng/mL, 109.3±17.2%; 20 ng/mL, 91.8±7.8%; 40 ng/mL, 114.7±0.3%, n=3, control 100.0±3.0%). Western blot analysis demonstrated that the recombinant human EMMPRIN mainly contained a lower molecular weight form of EMMPRIN of 30 kDa (data not shown), corresponding to unglycosylated or only slightly glycosylated EMMPRIN.
Figure 6. A, MMP-1, MMP-2, and MMP-9 release by HCA-SMC after 48 hours incubation with EMMPRIN, isolated from supernatants of human monocyte-derived macrophages. MMP concentrations were measured with immunoassays and referred to the corresponding DNA content. Mean±SEM (n=3), relative MMP release compared with controls: **P<0.01 vs control. B, Zymography of HCA-SMC supernatants after 48 hours incubation with isolated EMMPRIN. 1 indicates control, and 2, 20 ng/mL EMMPRIN, isolated from macrophage supernatants.
Discussion
MMPs seem to play an important role in atherosclerotic plaque growth, neointima formation, and plaque disruption by inducing smooth muscle cell migration and proliferation and by enhancing extracellular matrix degradation.5–7 Several factors, such as cytokines, reactive oxygen species, and cholesterol have been shown to induce MMP expression in endothelial cells, monocyte-derived macrophages, and smooth muscle cells.1–3 EMMPRIN has been shown to regulate MMP release and activity in fibroblasts, endothelial cells, and tumor cells.11,12,18 Recent studies have demonstrated that EMMPRIN is expressed in human atherosclerotic plaque and is induced on monocyte differentiation.8,9 EMMPRIN was initially characterized as a tumor cell surface glycoprotein, comprising 2 extracellular immunoglobulin domains, a transmembrane domain and a cytoplasmic domain.15,19 The reported molecular weight of EMMPRIN is variable because of different degrees of glycosylation of the native 27-kDa protein; however, it is generally observed to have a molecular weight of 58 kDa.10 Previous results have suggested that EMMPRIN exerts its effects mainly by cell-to-cell contact; however, EMMPRIN release into cell culture supernatants has been observed in different tumor cells.20–23
In this study, we demonstrate that EMMPRIN is expressed in HCA-SMC and that oxLDLs enhance EMMPRIN release. The oxLDL-induced increase of EMMPRIN release was significantly antagonized by the antioxidants trolox and probucol and by preincubation with vitamin C. Hydrogen peroxide, which was used to mimic oxidative stress, also induced a significant increase of EMMPRIN release. In contrast, a significant concentration-dependent reduction of EMMPRIN release was observed after incubation with nLDL, which contained BHT and EDTA to prevent spontaneous oxidation. NLDL prepared without BHT and EDTA exhibited a slight spontaneous oxidation during the isolation procedure and elicited a slight increase of EMMPRIN release. Controls containing BHT and EDTA or EDTA alone significantly reduced the EMMPRIN release. These findings suggest that the oxLDL-induced EMMPRIN release is mainly related to oxLDL-associated oxidative stress and that the release of soluble EMMPRIN might result from proteolytic cleavage.
By Western blot analysis we have demonstrated that the molecular weight of soluble EMMPRIN was similar to the molecular weight of cell-associated EMMPRIN and that the increase of EMMPRIN release into cell culture supernatants was accompanied by a decrease of cell-associated EMMPRIN. This effect was antagonized by antioxidants, the unspecific protease inhibitor EDTA, and by the MMP inhibitor GM6001. To further investigate whether MMP might promote the cleavage of soluble EMMPRIN from the cell surface, EMMPRIN isolated from HCA-SMC lysate was incubated with MMP-1 and MMP-2. To identify a possible MMP-induced shift in molecular weight, Western blots were performed with glycosylated and deglycosylated samples (deglycosylation was performed after MMP treatment). Because the transmembrane and cytoplasmic segment of EMMPRIN exhibit a molecular weight of only 7 kDa (24 and 39 amino acids, respectively),10 such a shift in molecular weight is not clearly visible with glycosylated EMMPRIN, which appears as broad band in Western blots because of different degrees of glycosylation. Deglycosylated EMMPRIN without MMP pretreatment exhibited a molecular weight of 27 kDa, corresponding to the reported molecular weight of the native protein.10 After incubation with MMP-1 or MMP-2, an additional band with a molecular weight of 20 kDa was detectable. EMMPRIN isolation from supernatant of HCA-SMC incubated with oxLDL yielded a hydrophilic (73%) and a hydrophobic (27%) EMMPRIN-containing fraction in anion exchange chromatography. Deglycosylated EMMPRIN from the hydrophilic fraction of HCA-SMC supernatant exhibited a molecular weight of 20 kDa, corresponding to the size of the additional band of MMP-treated deglycosylated EMMPRIN from cell lysate. Glycosylated and deglycosylated EMMPRIN from the hydrophilic fraction of HCA-SMC supernatant as well as the 20-kDa form of MMP-treated deglycosylated EMMPRIN from cell lysate were not detectable with an EMMPRIN antibody against the intracellular domain.
These findings suggest that MMP-1 and MMP-2 promote shedding of soluble EMMPRIN from the cell surface. Fitting with our results, very recently published data also have suggested MMP-dependent proteolytic cleavage of soluble EMMPRIN in human breast cancer cells.22 Another recently published study has demonstrated release of full-length EMMPRIN via budding of microvesicles in a lung carcinoma line.21 This finding might appear contradictory to the presented results; however, we also have observed that supernatant of HCA-SMC incubated without oxLDL yielded mainly a hydrophobic EMMPRIN-containing fraction in anion exchange chromatography (data not shown). Thus, the presently available data suggest that under variable conditions different mechanisms might be involved in EMMPRIN release. In addition, it must be considered that results from in vitro studies do not necessarily reflect the in vivo situation.
As mentioned above, oxLDL-induced upregulation of MMP production in endothelial cells and monocyte-derived macrophages has been described in several studies, and it has been suggested that the oxLDL-induced MMP expression involves extracellular signal regulated kinase activation.24 In this study, we have demonstrated predominant MMP-2 activity in HCA-SMC. MMP-1 and MMP-2 release was significantly increased by oxLDL and hydrogen peroxide, and the oxLDL-induced MMP-1 and MMP-2 release was partly antagonized by antioxidants. In addition, we have demonstrated that soluble EMMPRIN, isolated from macrophage supernatants, significantly enhanced MMP-1 and MMP-2 release by HCA-SMC, whereas recombinant human EMMPRIN slightly reduced MMP-1 release and did not significantly alter MMP-2 release. In contrast to isolated EMMPRIN, recombinant human EMMPRIN mainly contained a lower molecular weight form of EMMPRIN of 30 kDa (data not shown), corresponding to unglycosylated or only slightly glycosylated EMMPRIN. Similar observations have been made in previous studies, where bacterially produced recombinant EMMPRIN with a molecular weight of 29 kDa was inactive in stimulating MMP production by human fibroblasts25, and deglycosylated EMMPRIN abolished the stimulatory effect of glycosylated EMMPRIN on MMP-1 production in dermal fibroblasts.12 These studies have also demonstrated that the degree and pattern of MMP stimulation by EMMPRIN varies in different fibroblast preparations25 and that EMMPRIN may be a counterreceptor for itself and might induce MMP production by homophilic interaction.12 The signaling pathways by which EMMPRIN regulates MMP production still remain unclear; however, recent studies have suggested involvement of mitogen-activated protein kinase p38 and phospholipase A2/5-lipoxygenase activation, respectively.23,26
To our knowledge, this is the first study that demonstrates EMMPRIN expression in HCA-SMC. Our data provide evidence for a pathophysiological role of EMMPRIN in the progression of coronary atherosclerosis, and on the basis of our findings and recently published results from tumor cells we propose the following mechanism: oxLDL and oxidative stress stimulate the release of soluble EMMPRIN as well as MMP-1 and MMP-2 synthesis in HCA-SMC. EMMPRIN release results from enhanced shedding, which, at least in part, seems to be MMP-dependent. In turn, soluble EMMPRIN might enhance MMP-1 and MMP-2 synthesis by autocrine or paracrine mechanisms. Thus oxLDL might induce a circular cascade of increased MMP activity, enhanced MMP-dependent shedding of soluble EMMPRIN, and EMMPRIN-induced upregulation of MMP production. This cascade might accelerate extracellular matrix degradation in atherosclerotic plaques and thereby promote plaque growth and plaque destabilization.
Acknowledgments
We thank Martina de Groot, Martina Adam-J?ger, and Gisela Sailer for expert technical assistance.
This work was supported by a grant from Deutsche Forschungsgemeinschaft (SFB 451, Teilprojekt B3 to C.H. and M.G.B.)
References
Rajavashisth TB, Liao JK, Galis ZS, Tripathi S, Laufs U, Tripathi J, Chai NN, Xu XP, Jovinge S, Shah PK, Libby P. Inflammatory cytokines and oxidized low density lipoproteins increase endothelial cell expression of membrane type 1-matrix metalloproteinase. J Biol Chem. 1999; 274: 11924–11929.
Rajavashisth TB, Xu XP, Jovinge S, Meisel S, Xu XO, Chai NN, Fishbein MC, Kaul S, Cercek B, Sharifi B, Shah PK. Membrane type 1 matrix metalloproteinase expression in human atherosclerotic plaques: evidence for activation by proinflammatory mediators. Circulation. 1999; 99: 3103–3109.
Xu XP, Meisel SR, Ong JM, Kaul S, Cercek B, Rajavashisth TB, Sharifi B, Shah PK. Oxidized low-density lipoprotein regulates matrix metalloproteinase-9 and its tissue inhibitor in human monocyte-derived macrophages. Circulation. 1999; 99: 993–998.
Matsumoto S, Kobayashi T, Katoh M, Saito S, Ikeda Y, Kobori M, Masuho Y, Watanabe T. Expression and localization of matrix metalloproteinase-12 in the aorta of cholesterol-fed rabbits: relationship to lesion development. Am J Pathol. 1998; 153: 109–119.
Cho A, Reidy MA. Matrix metalloproteinase-9 is necessary for the regulation of smooth muscle cell replication and migration after arterial injury. Circ Res. 2002; 91: 845–851.
Mason DP, Kenagy RD, Hasenstab D, Bowen-Pope DF, Seifert RA, Coats S, Hawkins SM, Clowes AW. Matrix metalloproteinase-9 overexpression enhances vascular smooth muscle cell migration and alters remodeling in the injured rat carotid artery. Circ Res. 1999; 85: 1179–1185.
Shah PK, Falk E, Badimon JJ, Fernandez-Ortiz A, Mailhac A, Villareal-Levy G, Fallon JT, Regnstrom J, Fuster V. Human monocyte-derived macrophages induce collagen breakdown in fibrous caps of atherosclerotic plaques. Potential role of matrix-degrading metalloproteinases and implications for plaque rupture. Circulation. 1995; 92: 1565–1569.
Choi EY, Kim D, Hong BK, Kwon HM, Song YG, Byun KH, Park HY, Whang KC, Kim HS. Upregulation of extracellular matrix metalloproteinase inducer (EMMPRIN) and gelatinases in human atherosclerosis infected with Chlamydia pneumoniae: the potential role of Chlamydia pneumoniae infection in the progression of atherosclerosis. Exp Mol Med. 2002; 34: 391–400.
Major TC, Liang L, Lu X, Rosebury W, Bocan TM. Extracellular matrix metalloproteinase inducer (EMMPRIN) is induced upon monocyte differentiation and is expressed in human atheroma. Arterioscler Thromb Vasc Biol. 2002; 22: 1200–1207.
Toole BP. Emmprin (CD147), a cell surface regulator of matrix metalloproteinase production and function. Curr Top Dev Biol. 2003; 54: 371–389.
Li R, Huang L, Guo H, Toole BP. Basigin (murine EMMPRIN) stimulates matrix metalloproteinase production by fibroblasts. J Cell Physiol. 2001; 186: 371–379.
Sun J, Hemler ME. Regulation of MMP-1 and MMP-2 production through CD147/extracellular matrix metalloproteinase inducer interactions. Cancer Res. 2001; 61: 2276–2281.
Cominacini L, Garbin U, Davoli A, Micciolo R, Bosello O, Gaviraghi G, Scuro LA, Pastorino AM. A simple test for predisposition to LDL oxidation based on the fluorescence development during copper-catalyzed oxidative modification. J Lipid Res. 1991; 32: 349–358.
Labarca C, Paigen K. A simple, rapid and sensitive DNA assay procedure. Anal Biochem. 1980; 102: 344–352.
Biswas C, Zhang Y, DeCastro R, Guo H, Nakamura T, Kataoka H, Nabeshima K. The human tumor cell-derived collagenase stimulatory factor (renamed EMMPRIN) is a member of the immunoglobulin superfamily. Cancer Res. 1995; 55: 434–439.
Miyauchi T, Masuzawa Y, Muramatsu T. The basigin group of the immunoglobulin superfamily: complete conservation of a segment in and around transmembrane domains of human and mouse basigin and chicken HT7 antigen. J Biochem. 1991; 110: 770–774.
Allen RW, Trach KA, Hoch JA. Identification of the 37-kDa protein displaying a variable interaction with the erythroid cell membrane as glyceraldehyde-3-phosphate dehydrogenase. J Biol Chem. 1987; 262: 649–653.
Caudroy S, Polette M, Nawrocki-Raby B, Cao J, Toole BP, Zucker S, Birembaut P. EMMPRIN-mediated MMP regulation in tumor and endothelial cells. Clin Exp Metastasis. 2002; 19: 697–702.
Guo H, Majmudar G, Jensen TC, Biswas C, Toole BP, Gordon MK. Characterization of the gene for human EMMPRIN, a tumor cell surface inducer of matrix metalloproteinases. Gene. 1998; 220: 99–108.
Ellis SM, Nabeshima K, Biswas C. Monoclonal antibody preparation and purification of a tumor cell collagenase-stimulatory factor. Cancer Res. 1989; 49: 3385–3391.
Sidhu SS, Mengistab AT, Tauscher AN, LaVail J, Basbaum C. The microvesicle as a vehicle for EMMPRIN in tumor-stromal interactions. Oncogene. 2004; 23: 956–963.
Tang Y, Kesavan P, Nakada MT, Yan L. Tumor-stroma interaction: positive feedback regulation of extracellular matrix metalloproteinase inducer (EMMPRIN) expression and matrix metalloproteinase-dependent generation of soluble EMMPRIN. Mol Cancer Res. 2004; 2: 73–80.
Taylor PM, Woodfield RJ, Hodgkin MN, Pettitt TR, Martin A, Kerr DJ, Wakelam MJO. Breast cancer cell-derived EMMPRIN stimulates fibroblast MMP2 release through a phospholipase A2 and 5-lipoxygenase catalyzed pathway. Oncogene. 2002; 21: 5765–5772.
Song L, Xu M, Lopes-Virella MF, Huang Y. Quercetin inhibits matrix metalloproteinase-1 expression in human vascular endothelial cells through extracellular signal-regulated kinase. Arch Biochem Biophys. 2001; 391: 72–78.
Guo H, Zucker S, Gordon MK, Toole BP, Biswas C. Stimulation of matrix metalloproteinase production by recombinant extracellular matrix metalloproteinase inducer from transfected Chinese hamster ovary cells. J Biol Chem. 1997; 272: 24–27.
Lim M, Martinez T, Jablons D, Cameron R, Guo H, Toole B, Li JD, Basbaum C. Tumor-derived EMMPRIN (extracellular matrix metalloproteinase inducer) stimulates collagenase transcription through MAPK p38. FEBS Lett. 1998; 441: 88–92.(Cornelia Haug; Christina )
ABSTRACT
Objective— Matrix metalloproteinases (MMPs) seem to play a prominent role in atherogenesis. Extracellular MMP inducer (EMMPRIN), a cell surface glycoprotein which stimulates MMP synthesis, has recently been detected in human atheroma. We have investigated the influence of oxidized low-density lipoproteins (oxLDLs) on EMMPRIN expression in human coronary artery smooth muscle cells (HCA-SMCs).
Methods and Results— OxLDL induced a significant increase of EMMPRIN release into HCA-SMC supernatants and a concomitant decrease of cell-associated EMMPRIN. These effects were antagonized by antioxidants as well as by EDTA and the MMP inhibitor GM6001. Western blot analysis demonstrated that MMP-1 and MMP-2 induce the cleavage of the extracellular domain from cell-associated EMMPRIN. MMP-1 and MMP-2 synthesis was upregulated by oxLDL, and, in addition, we have shown that soluble EMMPRIN, isolated from macrophage supernatants, increased MMP-1 and MMP-2 synthesis in HCA-SMC.
Conclusion— Our data suggest that oxLDLs stimulate the release of soluble EMMPRIN, at least in part, by MMP-dependent shedding from the cell surface. Additionally, oxLDLs might induce a circular upregulation of matrix degradation because, in turn, soluble EMMPRIN stimulates MMP synthesis in HCA-SMC.
This study demonstrates that oxidized low-density lipoproteins stimulate the release of soluble extracellular matrix metalloproteinase inducer (EMMPRIN) by human coronary artery smooth muscle cells, an effect which seems to result from enhanced EMMPRIN shedding. In addition, we have shown that isolated purified EMMPRIN stimulates MMP synthesis in coronary smooth muscle cells.
Key Words: smooth muscle cells ? low density lipoproteins ? matrix metalloproteinases ? extracellular MMP inducer ? atherosclerosis
Introduction
Deposition of low-density lipoproteins (LDLs) in the vessel wall and their oxidative modification seem to initiate, or at least accelerate, the atherosclerotic process by several mechanisms, including promotion of foam cell formation, chemotactic effects on monocytes, and mitogenic effects on smooth muscle cells. Recent studies have demonstrated that oxidized LDLs (oxLDLs) increase the expression of matrix metalloproteinases (MMPs) in endothelial cells, monocyte-derived macrophages, and smooth muscle cells1–3 and that MMP expression is enhanced in atherosclerotic plaque tissue.4 MMPs increase smooth muscle cell migration and proliferation5,6 and also seem to promote plaque instability by inducing extracellular matrix degradation.7 Expression of extracellular MMP inducer (EMMPRIN) in human atheroma has been described recently.8,9 EMMPRIN, also called CD147 or basigin, is a highly glycosylated plasma membrane protein (58 kDa) belonging to the immunoglobulin superfamily. EMMPRIN is expressed on several cell types like leukocytes and different tumor cells10 and stimulates MMP production in fibroblasts and various tumor cells.11,12 Several studies have shown that cancer cells express significantly higher EMMPRIN levels than normal cells, and it has been proposed that elevated EMMPRIN expression in tumor cells may promote tumor progression by inducing MMP expression in peritumoral stromal cells. In this study, we have demonstrated that EMMPRIN is expressed in cultured human coronary artery smooth muscle cells (HCA-SMCs) and that oxLDL significantly enhanced the release of soluble EMMPRIN as well as MMP-1 and MMP-2 release. In addition, we have shown that MMP-1 and MMP-2 seem to promote the cleavage of soluble EMMPRIN from the cell surface and that soluble EMMPRIN stimulates MMP synthesis in HCA-SMC. Thus our data suggest that oxLDL might induce a circular upregulation of matrix degradation.
Materials and Methods
For more detail, please see the online Methods section, available at http://atvb.ahajournals.org.
Cell Culture
HCA-SMCs (Clonetics, San Diego, Calif) were cultured in SmGM-2 BulletKit medium (BioWhittaker). Experiments were performed with cells from passages 5 to 8 in culture medium containing 0.1%, 1.5%, or 5% fetal calf serum (FCS), respectively.
Human monocytes were isolated from buffy coats of healthy donors by density gradient separation and cultured for 8 to 10 days before supernatants were collected for EMMPRIN isolation.
Isolation and Oxidation of LDL
LDLs were isolated from blood samples of healthy volunteers by sequential ultracentrifugation. For preparation of native LDL (nLDL), plasma samples (4 μmol EDTA per mL blood) were supplemented with butylated hydroxytoluene (BHT, 20 μmol/L) to prevent oxidation. LDLs isolated from serum samples of the same donors were oxidized by exposure to 5 μmol/L CuSO4 and oxygen for 24 hours at 37°C. For controls, "native" LDL was also prepared from serum samples without addition of EDTA and BHT. Lipoprotein concentrations are expressed in terms of their protein content. The degree of oxidation was quantified by absorption measurement at 234 nm and fluorescence measurement at 430 nm.13
Immunometric Measurement of EMMPRIN and MMP
For EMMPRIN quantification, microtitre plates coated with rabbit-anti-mouse IgG and monoclonal anti-CD147 antibody (extracellular domain; R&D Systems) were incubated with standard (R&D Systems) or sample, biotinylated polyclonal anti-CD147 antibody (extracellular domain; R&D Systems), streptavidin-Europium (Delfia Wallac), and enhancer solution. Time-resolved fluorescence of the Europium-chelate was measured with a Victor Multilabel Counter (Wallac). All measurements were done in duplicate, intraassay coefficient of variation was 8.7% (n=30).
Total MMP-1 was measured with an ELISA kit (Amersham Biosciences), which recognizes free MMP-1 and MMP-1 complexed with inhibitors, such as tissue inhibitor of metalloproteinases-1. For MMP-2 quantification, microtitre plates coated with rabbit-anti-mouse IgG and monoclonal anti-MMP-2 antibody (recognizes pro–MMP-2 and active MMP-2; R&D Systems) were incubated with standard (Chemicon) or sample, polyclonal biotinylated anti-MMP-2 antibody (R&D Systems), streptavidin-Europium and enhancer solution, followed by time-resolved fluorescence measurement. Measurements were done in duplicate, and intraassay coefficient of variation was 7.9% (n=30).
Total MMP-9 was measured with an ELISA kit (Oncogene), which detects free MMP-9 and MMP-9 bound to tissue inhibitor of metalloproteinase-1. EMMPRIN, MMP-1, MMP-2, and MMP-9 concentrations in supernatants were referred to the DNA content in the corresponding wells. DNA content was measured by fluorescent DNA staining with bisbenzimide using calf thymus DNA as standard.14
RNA Isolation and RT-PCR
For detection of EMMPRN mRNA, HCA-SMC were incubated for 2, 3, 4, 6, 8, and 12 hours with oxLDL and nLDL (1 to 20 μg/mL) in medium containing 0.1% FCS. After extraction of total RNA and reverse transcription, DNA amplification (human EMMPRIN15,16: Acc NM_001728; GAPDH17: Acc J02642) was performed using the LightCycler technology (Idaho Technology; SYBR Green I). Reactions were cycled 34 to 40x, melting curves were performed and polymerase chain reaction (PCR) products quantified with the LightCycler software using purified sequenced PCR product as standard. EMMPRIN mRNA was referred to GAPDH mRNA. No detectable PCR products were present in water controls and in controls, amplified without prior reverse transcription. For visualization, PCR products were applied to 1% agarose gel.
EMMPRIN Isolation and Purification
Macrophage supernatant and HCA-SMC supernatant were applied at 2 mL/min to a 20HQ anion exchange column (Applied Biosystems). Bound proteins were eluted with a linear gradient of NaCl (0 to 1 mol/L), 1 mL fractions were collected, and EMMPRIN was identified by time-resolved fluorescence immunoassay. EMMPRIN-containing fractions were applied to an immunoaffinity column, where agarose protein G was coupled to monoclonal anti-human EMMPRIN antibody (Pierce). The column was washed, EMMPRIN was eluted, and the eluted protein was dialyzed and concentrated. HCA-SMC lysate was first purified by immunoaffinity chromatography, then applied to a 20HQ anion exchange column, eluted with a linear gradient of NaCl, and concentrated. EMMPRIN concentrations were measured by fluorescence immunoassay. For investigation of MMP-associated EMMPRIN cleavage, EMMPRIN isolated from HCA-SMC lysate was incubated for 2 hours with 0.1 μg/mL recombinant human MMP-1 or MMP-2 (R&D Systems). Thereafter, EMMPRIN was deglycosylated with a deglycosylation kit (Roche Diagnostics).
Western Blot
For detection of cell-associated EMMPRIN, cellular proteins were solubilized with lysis buffer. Sample volumes of cell lysates and supernatants were adjusted according to the protein content. SDS-PAGE was performed (cell lysate and supernatant, nonreducing conditions; isolated EMMPRIN, reducing conditions, sample volume adjusted to EMMPRIN content), and separated proteins were electroblotted to polyvinylidene difluoride membranes. Western blots were incubated with monoclonal anti-human EMMPRIN antibody (extracellular domain; R&D Systems), and for detection of MMP-associated EMMPRIN cleavage, an additional antibody against the C-terminus of EMMPRIN was used (Santa Cruz). After washing, blots were incubated with biotinylated secondary antibody and HRP-streptavidin, and finally chemiluminescence staining was performed.
Zymography
For detection of MMP activity, cells were incubated for 48 hours with oxLDL or nLDL (1 to 20 μg/mL) in medium containing 0.1% FCS. Supernatants were adjusted according to the protein content. Nonreducing SDS-PAGE was performed in 7.5% polyacrylamide gels containing 0.2% gelatin. After electrophoresis, gels were washed, incubated in developing buffer, stained with 0.34% Coomassie Blue, and destained with acetic acid and methanol.
Statistical Analysis
Results are expressed as mean±SEM and were evaluated by 1-way ANOVA, followed by the Dunnett test (comparison with the control group) or the Newman-Keuls test (comparison between all groups), respectively.
Results
EMMPRIN Release by HCA-SMC
OxLDL induced a significant concentration-dependent increase of EMMPRIN release (Figure 1). In contrast, EMMPRIN release was significantly reduced after incubation with higher concentrations of nLDL (1 μg/mL, 95.7±5.3% of control; 10 μg/mL, 94.1±4.4%; 20 μg/mL, 85.6±3.7%**; 50 μg/mL, 60.3±5.6%***; control, 100.0±1.7%, n=15; ** indicates P<0.01; ***, P<0.001 versus control). Because the nLDL preparation contained EDTA and BHT to prevent spontaneous LDL oxidation, controls were performed with EDTA and BHT containing medium. A similar decrease of EMMPRIN release was observed (EDTA and BHT concentrations corresponding to the final concentrations in 20 and 50 μg/mL nLDL, 91.9±7.9% and 69.6±3.3%*** of antioxidant-free control; ***, P<0.001, n=6), whereas nLDL, which was prepared without EDTA and BHT and underwent a slight spontaneous oxidation during the preparation procedure, induced a slight increase of EMMPRIN release (20 μg/mL, 112.7±6.4%; 50 μg/mL, 114.1±8.1%, n=6).
Figure 1. Effect of oxLDL on EMMPRIN release by HCA-SMC. Cells were incubated for 48 hours with oxLDL in medium containing 0.1% FCS. EMMPRIN concentrations in supernatants were measured by fluorescence immunoassay and referred to the DNA content in the corresponding wells. Mean±SEM (n=12), expressed as relative EMMPRIN release compared with controls; *P<0.05 and ***P<0.001 vs control.
Hydrogen peroxide, which was used to mimic oxidative stress, also induced a significant increase of EMMPRIN release (100 μmol/L, 154.2±3.4%** of control; 500 μmol/L, 182.1±15.1%***; control, 100.0±4.0%, n=6; ** indicates P<0.01; ***, P<0.001 versus control), whereas EDTA, an unspecific protease inhibitor, reduced EMMPRIN release (20 μmol/L, 70.5±9.1%*** of control; 200 μmol/L, 59.8±4.3%***; 500 μmol/L, 55.1±1.1%***; control, 100.0±4.0%, n=6; ***, P<0.001 versus control). In accordance with these findings, the oxLDL-induced increase of EMMPRIN release was significantly antagonized by the antioxidants trolox and probucol as well as by EDTA, and the basal EMMPRIN release was also slightly reduced by trolox and probucol (Figure 2). In addition, preincubation with vitamin C for 24 hours also significantly reduced the oxLDL-stimulated EMMPRIN release (oxLDL 50 μg/mL, 272.5±30.3% of control; oxLDL 50 μg/mL+vitamin C 50 μg/mL, 162.6±11.6%, n=6, P<0.01), whereas the basal EMMPRIN release was not significantly reduced by vitamin C. The results shown above were obtained in experiments performed in culture medium containing 0.1% FCS. Because oxidative stress might be decreased with higher FCS concentrations in the culture medium, most experiments were also performed with higher FCS concentrations (1.5% and 5%). However, similar results were obtained (data not shown).
Figure 2. Reduction of basal and oxLDL-induced EMMPRIN release by trolox, probucol, and EDTA. HCA-SMC were incubated for 48 hours with oxLDL (20 μg/mL and 50 μg/mL, n=6), 500 μmol/L trolox (trol., n=6), 200 μmol/L probucol (pro., n=6), 200 μmol/L EDTA (n=5), or a combination of oxLDL (20 μg/mL and 50 μg/mL) and trolox, probucol, or EDTA, respectively (n=6). EMMPRIN concentrations were measured by fluorescence immunoassay and referred to the DNA content in the corresponding wells. Mean±SEM, expressed as relative EMMPRIN release compared with controls: *P<0.05, **P<0.01, ***P<0.001 vs control (C.), +++P<0.001 vs oxLDL 20 μg/mL, and ###P<0.001 vs oxLDL 50 μg/mL.
Investigation of EMMPRIN mRNA expression by quantitative real-time RT-PCR using the LightCycler technology showed no relevant changes of EMMPRIN mRNA after incubation with oxLDL or nLDL (2 to 24 hours, 1 to 20 μg/mL).
Western blot analysis showed a decrease of cell-associated EMMPRIN and an increase of soluble EMMPRIN in the corresponding supernatants after incubation with oxLDL, and to a much lesser extent also after incubation with antioxidant-free nLDL (Figure 3). NLDL containing EDTA and BHT as well as EDTA alone inhibited the decrease of cell-associated EMMPRIN and in parallel reduced the EMMPRIN release into cell culture supernatants. The oxLDL-induced decrease of cell-associated EMMPRIN and the increase of soluble EMMPRIN in HCA-SMC supernatants was partly antagonized by trolox, probucol, and EDTA (Figure 3).
Figure 3. Western blot analysis (nonreducing conditions) of EMMPRIN expression in cell lysates (C) and corresponding supernatants (S) of HCA-SMC, incubated for 48 hours with the following substances. Control (Con.), medium with 0.1% FCS; ox50, 50 μg/mL oxidized LDL; n50, 50 μg/mL nLDL containing EDTA and BHT; EDTA, 500 μmol/L EDTA; n20+EDTA/BHT, 20 μg/mL nLDL containing EDTA and BHT; n20–EDTA/BHT, 20 μg/mL nLDL without EDTA and BHT; ox20, 20 μg/mL oxLDL; Con.,: medium with 0.1% FCS; ox20, 20 μg/mL oxLDL; ox20+trol., 20 μg/mL oxLDL+500 μmol/L trolox; ox20+pro., 20 μg/mL oxLDL+200 μmol/L probucol; and ox20+EDTA, 20 μg/mL oxLDL+500 μmol/L EDTA.
Quantitative EMMPRIN measurement in HCA-SMC supernatants demonstrated that the MMP inhibitor GM6001 (Chemicon; Ki values: human MMP-1, 0.4 nmol/L; MMP-2, 0.5 nmol/L; MMP-3, 27 nmol/L; MMP-8, 0.1 nmol/L; and MMP-9, 0.2 nmol/L) reduced the basal EMMPRIN release and significantly antagonized the oxLDL-induced EMMPRIN release (Figure 4A). In accordance with the quantitative measurement, EMMPRIN Western blot showed that the oxLDL-induced reduction of cell-associated EMMPRIN and the concomitant increase of EMMPRIN release was effectively antagonized by the MMP inhibitor (Figure 4B).
Figure 4. A, Influence of the MMP inhibitor GM6001 (GM, 15 μmol/L and 25 μmol/L) on the basal and oxLDL-induced EMMPRIN release by HCA-SMC. EMMPRIN concentrations in supernatants were measured by fluorescence immunoassay and referred to the DNA content in the corresponding wells. Mean±SEM, expressed as relative EMMPRIN release compared with controls: n=6, +P<0.05, +++P<0.001 vs control, and ***P<0.001 vs oxLDL 50 μg/mL. B, Western blot analysis (nonreducing conditions) of EMMPRIN expression in cell lysates (C) and corresponding supernatants (S) of HCA-SMC, incubated for 48 hours with Con., medium with 0.1% FCS; GM, GM6001 15 μmol/L; ox50+GM, oxLDL 50 μg/mL+GM6001 15 μmol/L; and ox50, oxLDL 50 μg/mL. C, Western blot analysis (reducing conditions, EMMPRIN antibody against the extracellular domain) of EMMPRIN, isolated from HCA-SMC supernatant (S) or cell lysate (C) by anion exchange and immunoaffinity chromatography. Isolated purified EMMPRIN from cell lysate was incubated for 2 hours with recombinant human MMP-1 or MMP-2; thereafter, isolated EMMPRIN from cell lysate and HCA-SMC supernatant was deglycosylated with N-Glycosidase F (N-Glyc.).
To further investigate whether soluble EMMPRIN is cleaved from the cell surface and whether MMP might be involved in EMMPRIN cleavage, EMMPRIN isolated from HCA-SMC lysate was incubated with recombinant human MMP-1 and MMP-2. Because the transmembrane and cytoplasmic domain exhibit a putative molecular weight of only 7 kDa, such a shift in molecular weight is not clearly visible with glycosylated EMMPRIN, which appears as broad band in Western blot analysis because of different degrees of glycosylation. Therefore, Western blots were performed with glycosylated and deglycosylated samples (deglycosylation was performed after incubation with MMP). Deglycosylated EMMPRIN without MMP pretreatment exhibited a molecular weight of 27 kDa, corresponding to the reported molecular weight of the native protein; preincubation with MMP-1 or MMP-2 resulted in the appearance of an additional band with a molecular weight of 20 kDa (Figure 4C). EMMPRIN isolation from HCA-SMC supernatant collected after incubation with oxLDL yielded a hydrophilic (73%) and a hydrophobic (27%) EMMPRIN-containing fraction in anion exchange chromatography. Purified EMMPRIN from the hydrophilic fraction exhibited molecular weights of 50 kDa (glycosylated form) and 20 kDa (after deglycosylation), respectively. Thus, the additional band of MMP-treated deglycosylated EMMPRIN from cell lysate exhibited the same molecular weight (20 kDa) as deglycosylated EMMPRIN from the hydrophilic fraction of HCA-SMC supernatant. The size difference between these fractions and deglycosylated EMMPRIN from untreated cell lysate corresponds to the putative molecular weight of the cytoplasmic and transmembrane domain of EMMPRIN (7 kDa). Glycosylated and deglycosylated EMMPRIN from the hydrophilic fraction of HCA-SMC supernatant as well as the 20 kDa form of MP-treated deglycosylated EMMPRIN from cell lysate were not detectable with an EMMPRIN antibody against the intracellular domain (data not shown).
Effect of oxLDL on MMP Release by HCA-SMC
Zymography of HCA-SMC supernatants showed predominant expression of MMP-2 and an upregulation of active MMP-2 after incubation with oxLDL (Figure 5). Quantitative MMP measurement showed an oxLDL-induced upregulation of MMP-1 and MMP-2 release (MMP-1, oxLDL 20 μg/mL 125.5±8.7% of control; 50 μg/mL 167.5±10.7%***, control 100.0±2.6%, n=6; MMP-2, oxLDL 20 μg/mL 119.4±6.4%, 50 μg/mL 190.8±17.6%***, control 100.0±2.4%, n=15; ***, P<0.001 versus control) and only minor changes after incubation with nLDL. Hydrogen peroxide also induced a significant concentration-dependent increase of MMP-2 secretion (100 μmol/L, 118.5±5.8% of control ; 500 μmol/L, 179.1±13.9% P<0.001 versus control, n=5) and a less pronounced increase of MMP-1 release (500 μmol/L, 136.9±2.2% of control , P<0.001 versus control, n=3). The oxLDL(50 μg/mL)-induced MMP release was significantly antagonized by preincubation with 50 μg/mL vitamin C (MMP-1, 206.4±8.8% of control versus 149.7±11.5%, P<0.01, n=3; MMP-2, 240.9±24.5% of control versus 175.9±8.3%, P<0.01, n=3).
Figure 5. Zymography of HCA-SMC supernatants after 48 hours incubation with oxLDL (1 μg/mL, 10 μg/mL, and 20 μg/mL: ox1, ox10, and ox20) or native LDL (1 μg/mL, 10 μg/mL, and 20 μg/mL: n1, n10, and n20).
Effect of Isolated Purified EMMPRIN on MMP Release by HCA-SMC
EMMPRIN isolated from macrophage supernatants exerted a significant stimulatory effect on MMP-1 and MMP-2 release and a slight stimulatory effect on MMP-9 release (Figure 6), whereas commercially obtained recombinant human EMMPRIN (R&D Systems) reduced MMP-1 release (rhEMMPRIN 1 ng/mL, 71.8±10.5%* of control; 10 ng/mL, 68.2±12.9%**; 20 ng/mL, 76.6±0.9%; 40 ng/mL, 81.0±2.3%, control 100.0±2.7%, n=3; *, P<0.05; **, P<0.01 versus control) and did not significantly alter MMP-2 release (rhEMMPRIN 1 ng/mL, 78.8±6.5%%; 10 ng/mL, 109.3±17.2%; 20 ng/mL, 91.8±7.8%; 40 ng/mL, 114.7±0.3%, n=3, control 100.0±3.0%). Western blot analysis demonstrated that the recombinant human EMMPRIN mainly contained a lower molecular weight form of EMMPRIN of 30 kDa (data not shown), corresponding to unglycosylated or only slightly glycosylated EMMPRIN.
Figure 6. A, MMP-1, MMP-2, and MMP-9 release by HCA-SMC after 48 hours incubation with EMMPRIN, isolated from supernatants of human monocyte-derived macrophages. MMP concentrations were measured with immunoassays and referred to the corresponding DNA content. Mean±SEM (n=3), relative MMP release compared with controls: **P<0.01 vs control. B, Zymography of HCA-SMC supernatants after 48 hours incubation with isolated EMMPRIN. 1 indicates control, and 2, 20 ng/mL EMMPRIN, isolated from macrophage supernatants.
Discussion
MMPs seem to play an important role in atherosclerotic plaque growth, neointima formation, and plaque disruption by inducing smooth muscle cell migration and proliferation and by enhancing extracellular matrix degradation.5–7 Several factors, such as cytokines, reactive oxygen species, and cholesterol have been shown to induce MMP expression in endothelial cells, monocyte-derived macrophages, and smooth muscle cells.1–3 EMMPRIN has been shown to regulate MMP release and activity in fibroblasts, endothelial cells, and tumor cells.11,12,18 Recent studies have demonstrated that EMMPRIN is expressed in human atherosclerotic plaque and is induced on monocyte differentiation.8,9 EMMPRIN was initially characterized as a tumor cell surface glycoprotein, comprising 2 extracellular immunoglobulin domains, a transmembrane domain and a cytoplasmic domain.15,19 The reported molecular weight of EMMPRIN is variable because of different degrees of glycosylation of the native 27-kDa protein; however, it is generally observed to have a molecular weight of 58 kDa.10 Previous results have suggested that EMMPRIN exerts its effects mainly by cell-to-cell contact; however, EMMPRIN release into cell culture supernatants has been observed in different tumor cells.20–23
In this study, we demonstrate that EMMPRIN is expressed in HCA-SMC and that oxLDLs enhance EMMPRIN release. The oxLDL-induced increase of EMMPRIN release was significantly antagonized by the antioxidants trolox and probucol and by preincubation with vitamin C. Hydrogen peroxide, which was used to mimic oxidative stress, also induced a significant increase of EMMPRIN release. In contrast, a significant concentration-dependent reduction of EMMPRIN release was observed after incubation with nLDL, which contained BHT and EDTA to prevent spontaneous oxidation. NLDL prepared without BHT and EDTA exhibited a slight spontaneous oxidation during the isolation procedure and elicited a slight increase of EMMPRIN release. Controls containing BHT and EDTA or EDTA alone significantly reduced the EMMPRIN release. These findings suggest that the oxLDL-induced EMMPRIN release is mainly related to oxLDL-associated oxidative stress and that the release of soluble EMMPRIN might result from proteolytic cleavage.
By Western blot analysis we have demonstrated that the molecular weight of soluble EMMPRIN was similar to the molecular weight of cell-associated EMMPRIN and that the increase of EMMPRIN release into cell culture supernatants was accompanied by a decrease of cell-associated EMMPRIN. This effect was antagonized by antioxidants, the unspecific protease inhibitor EDTA, and by the MMP inhibitor GM6001. To further investigate whether MMP might promote the cleavage of soluble EMMPRIN from the cell surface, EMMPRIN isolated from HCA-SMC lysate was incubated with MMP-1 and MMP-2. To identify a possible MMP-induced shift in molecular weight, Western blots were performed with glycosylated and deglycosylated samples (deglycosylation was performed after MMP treatment). Because the transmembrane and cytoplasmic segment of EMMPRIN exhibit a molecular weight of only 7 kDa (24 and 39 amino acids, respectively),10 such a shift in molecular weight is not clearly visible with glycosylated EMMPRIN, which appears as broad band in Western blots because of different degrees of glycosylation. Deglycosylated EMMPRIN without MMP pretreatment exhibited a molecular weight of 27 kDa, corresponding to the reported molecular weight of the native protein.10 After incubation with MMP-1 or MMP-2, an additional band with a molecular weight of 20 kDa was detectable. EMMPRIN isolation from supernatant of HCA-SMC incubated with oxLDL yielded a hydrophilic (73%) and a hydrophobic (27%) EMMPRIN-containing fraction in anion exchange chromatography. Deglycosylated EMMPRIN from the hydrophilic fraction of HCA-SMC supernatant exhibited a molecular weight of 20 kDa, corresponding to the size of the additional band of MMP-treated deglycosylated EMMPRIN from cell lysate. Glycosylated and deglycosylated EMMPRIN from the hydrophilic fraction of HCA-SMC supernatant as well as the 20-kDa form of MMP-treated deglycosylated EMMPRIN from cell lysate were not detectable with an EMMPRIN antibody against the intracellular domain.
These findings suggest that MMP-1 and MMP-2 promote shedding of soluble EMMPRIN from the cell surface. Fitting with our results, very recently published data also have suggested MMP-dependent proteolytic cleavage of soluble EMMPRIN in human breast cancer cells.22 Another recently published study has demonstrated release of full-length EMMPRIN via budding of microvesicles in a lung carcinoma line.21 This finding might appear contradictory to the presented results; however, we also have observed that supernatant of HCA-SMC incubated without oxLDL yielded mainly a hydrophobic EMMPRIN-containing fraction in anion exchange chromatography (data not shown). Thus, the presently available data suggest that under variable conditions different mechanisms might be involved in EMMPRIN release. In addition, it must be considered that results from in vitro studies do not necessarily reflect the in vivo situation.
As mentioned above, oxLDL-induced upregulation of MMP production in endothelial cells and monocyte-derived macrophages has been described in several studies, and it has been suggested that the oxLDL-induced MMP expression involves extracellular signal regulated kinase activation.24 In this study, we have demonstrated predominant MMP-2 activity in HCA-SMC. MMP-1 and MMP-2 release was significantly increased by oxLDL and hydrogen peroxide, and the oxLDL-induced MMP-1 and MMP-2 release was partly antagonized by antioxidants. In addition, we have demonstrated that soluble EMMPRIN, isolated from macrophage supernatants, significantly enhanced MMP-1 and MMP-2 release by HCA-SMC, whereas recombinant human EMMPRIN slightly reduced MMP-1 release and did not significantly alter MMP-2 release. In contrast to isolated EMMPRIN, recombinant human EMMPRIN mainly contained a lower molecular weight form of EMMPRIN of 30 kDa (data not shown), corresponding to unglycosylated or only slightly glycosylated EMMPRIN. Similar observations have been made in previous studies, where bacterially produced recombinant EMMPRIN with a molecular weight of 29 kDa was inactive in stimulating MMP production by human fibroblasts25, and deglycosylated EMMPRIN abolished the stimulatory effect of glycosylated EMMPRIN on MMP-1 production in dermal fibroblasts.12 These studies have also demonstrated that the degree and pattern of MMP stimulation by EMMPRIN varies in different fibroblast preparations25 and that EMMPRIN may be a counterreceptor for itself and might induce MMP production by homophilic interaction.12 The signaling pathways by which EMMPRIN regulates MMP production still remain unclear; however, recent studies have suggested involvement of mitogen-activated protein kinase p38 and phospholipase A2/5-lipoxygenase activation, respectively.23,26
To our knowledge, this is the first study that demonstrates EMMPRIN expression in HCA-SMC. Our data provide evidence for a pathophysiological role of EMMPRIN in the progression of coronary atherosclerosis, and on the basis of our findings and recently published results from tumor cells we propose the following mechanism: oxLDL and oxidative stress stimulate the release of soluble EMMPRIN as well as MMP-1 and MMP-2 synthesis in HCA-SMC. EMMPRIN release results from enhanced shedding, which, at least in part, seems to be MMP-dependent. In turn, soluble EMMPRIN might enhance MMP-1 and MMP-2 synthesis by autocrine or paracrine mechanisms. Thus oxLDL might induce a circular cascade of increased MMP activity, enhanced MMP-dependent shedding of soluble EMMPRIN, and EMMPRIN-induced upregulation of MMP production. This cascade might accelerate extracellular matrix degradation in atherosclerotic plaques and thereby promote plaque growth and plaque destabilization.
Acknowledgments
We thank Martina de Groot, Martina Adam-J?ger, and Gisela Sailer for expert technical assistance.
This work was supported by a grant from Deutsche Forschungsgemeinschaft (SFB 451, Teilprojekt B3 to C.H. and M.G.B.)
References
Rajavashisth TB, Liao JK, Galis ZS, Tripathi S, Laufs U, Tripathi J, Chai NN, Xu XP, Jovinge S, Shah PK, Libby P. Inflammatory cytokines and oxidized low density lipoproteins increase endothelial cell expression of membrane type 1-matrix metalloproteinase. J Biol Chem. 1999; 274: 11924–11929.
Rajavashisth TB, Xu XP, Jovinge S, Meisel S, Xu XO, Chai NN, Fishbein MC, Kaul S, Cercek B, Sharifi B, Shah PK. Membrane type 1 matrix metalloproteinase expression in human atherosclerotic plaques: evidence for activation by proinflammatory mediators. Circulation. 1999; 99: 3103–3109.
Xu XP, Meisel SR, Ong JM, Kaul S, Cercek B, Rajavashisth TB, Sharifi B, Shah PK. Oxidized low-density lipoprotein regulates matrix metalloproteinase-9 and its tissue inhibitor in human monocyte-derived macrophages. Circulation. 1999; 99: 993–998.
Matsumoto S, Kobayashi T, Katoh M, Saito S, Ikeda Y, Kobori M, Masuho Y, Watanabe T. Expression and localization of matrix metalloproteinase-12 in the aorta of cholesterol-fed rabbits: relationship to lesion development. Am J Pathol. 1998; 153: 109–119.
Cho A, Reidy MA. Matrix metalloproteinase-9 is necessary for the regulation of smooth muscle cell replication and migration after arterial injury. Circ Res. 2002; 91: 845–851.
Mason DP, Kenagy RD, Hasenstab D, Bowen-Pope DF, Seifert RA, Coats S, Hawkins SM, Clowes AW. Matrix metalloproteinase-9 overexpression enhances vascular smooth muscle cell migration and alters remodeling in the injured rat carotid artery. Circ Res. 1999; 85: 1179–1185.
Shah PK, Falk E, Badimon JJ, Fernandez-Ortiz A, Mailhac A, Villareal-Levy G, Fallon JT, Regnstrom J, Fuster V. Human monocyte-derived macrophages induce collagen breakdown in fibrous caps of atherosclerotic plaques. Potential role of matrix-degrading metalloproteinases and implications for plaque rupture. Circulation. 1995; 92: 1565–1569.
Choi EY, Kim D, Hong BK, Kwon HM, Song YG, Byun KH, Park HY, Whang KC, Kim HS. Upregulation of extracellular matrix metalloproteinase inducer (EMMPRIN) and gelatinases in human atherosclerosis infected with Chlamydia pneumoniae: the potential role of Chlamydia pneumoniae infection in the progression of atherosclerosis. Exp Mol Med. 2002; 34: 391–400.
Major TC, Liang L, Lu X, Rosebury W, Bocan TM. Extracellular matrix metalloproteinase inducer (EMMPRIN) is induced upon monocyte differentiation and is expressed in human atheroma. Arterioscler Thromb Vasc Biol. 2002; 22: 1200–1207.
Toole BP. Emmprin (CD147), a cell surface regulator of matrix metalloproteinase production and function. Curr Top Dev Biol. 2003; 54: 371–389.
Li R, Huang L, Guo H, Toole BP. Basigin (murine EMMPRIN) stimulates matrix metalloproteinase production by fibroblasts. J Cell Physiol. 2001; 186: 371–379.
Sun J, Hemler ME. Regulation of MMP-1 and MMP-2 production through CD147/extracellular matrix metalloproteinase inducer interactions. Cancer Res. 2001; 61: 2276–2281.
Cominacini L, Garbin U, Davoli A, Micciolo R, Bosello O, Gaviraghi G, Scuro LA, Pastorino AM. A simple test for predisposition to LDL oxidation based on the fluorescence development during copper-catalyzed oxidative modification. J Lipid Res. 1991; 32: 349–358.
Labarca C, Paigen K. A simple, rapid and sensitive DNA assay procedure. Anal Biochem. 1980; 102: 344–352.
Biswas C, Zhang Y, DeCastro R, Guo H, Nakamura T, Kataoka H, Nabeshima K. The human tumor cell-derived collagenase stimulatory factor (renamed EMMPRIN) is a member of the immunoglobulin superfamily. Cancer Res. 1995; 55: 434–439.
Miyauchi T, Masuzawa Y, Muramatsu T. The basigin group of the immunoglobulin superfamily: complete conservation of a segment in and around transmembrane domains of human and mouse basigin and chicken HT7 antigen. J Biochem. 1991; 110: 770–774.
Allen RW, Trach KA, Hoch JA. Identification of the 37-kDa protein displaying a variable interaction with the erythroid cell membrane as glyceraldehyde-3-phosphate dehydrogenase. J Biol Chem. 1987; 262: 649–653.
Caudroy S, Polette M, Nawrocki-Raby B, Cao J, Toole BP, Zucker S, Birembaut P. EMMPRIN-mediated MMP regulation in tumor and endothelial cells. Clin Exp Metastasis. 2002; 19: 697–702.
Guo H, Majmudar G, Jensen TC, Biswas C, Toole BP, Gordon MK. Characterization of the gene for human EMMPRIN, a tumor cell surface inducer of matrix metalloproteinases. Gene. 1998; 220: 99–108.
Ellis SM, Nabeshima K, Biswas C. Monoclonal antibody preparation and purification of a tumor cell collagenase-stimulatory factor. Cancer Res. 1989; 49: 3385–3391.
Sidhu SS, Mengistab AT, Tauscher AN, LaVail J, Basbaum C. The microvesicle as a vehicle for EMMPRIN in tumor-stromal interactions. Oncogene. 2004; 23: 956–963.
Tang Y, Kesavan P, Nakada MT, Yan L. Tumor-stroma interaction: positive feedback regulation of extracellular matrix metalloproteinase inducer (EMMPRIN) expression and matrix metalloproteinase-dependent generation of soluble EMMPRIN. Mol Cancer Res. 2004; 2: 73–80.
Taylor PM, Woodfield RJ, Hodgkin MN, Pettitt TR, Martin A, Kerr DJ, Wakelam MJO. Breast cancer cell-derived EMMPRIN stimulates fibroblast MMP2 release through a phospholipase A2 and 5-lipoxygenase catalyzed pathway. Oncogene. 2002; 21: 5765–5772.
Song L, Xu M, Lopes-Virella MF, Huang Y. Quercetin inhibits matrix metalloproteinase-1 expression in human vascular endothelial cells through extracellular signal-regulated kinase. Arch Biochem Biophys. 2001; 391: 72–78.
Guo H, Zucker S, Gordon MK, Toole BP, Biswas C. Stimulation of matrix metalloproteinase production by recombinant extracellular matrix metalloproteinase inducer from transfected Chinese hamster ovary cells. J Biol Chem. 1997; 272: 24–27.
Lim M, Martinez T, Jablons D, Cameron R, Guo H, Toole B, Li JD, Basbaum C. Tumor-derived EMMPRIN (extracellular matrix metalloproteinase inducer) stimulates collagenase transcription through MAPK p38. FEBS Lett. 1998; 441: 88–92.(Cornelia Haug; Christina )