Human bladder cancer cell lines [J82, 5637, HT-1376 (cat. no. CRL-1472), UM-UC-3], melanoma cells (M010119), prostate cancer cells (LNCaP), colon cancer cells (HCT116) and lung cancer cells (A549) were procured from the American Type Culture Collection. 253J cells were kindly provided by Dr Wun-Jae Kim (Chungbuk National University, Korea). J82, 5637, HT-1376, UM-UC-3, 253J, M010119 and LNCaP cells were cultured in RPMI 1640 medium (Welgene, Inc.) supplemented with 5% heat-inactivated fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin/streptomycin (Gibco; Thermo Fisher Scientific, Inc.). HCT116 and A549 cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Welgene, Inc.) supplemented with 5% heat-inactivated fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin/streptomycin (Gibco; Thermo Fisher Scientific, Inc.). Cultivation was at 37°C in an atmosphere containing 5% CO₂, with media changes every 2-3 days. To establish cell lines with CD46 overexpression, the lentiviral vector pBlasti-eGFP-CD46 kindly provided by Dr Silvio Hemmi (Institute of Molecular Life Sciences, University of Zurich, Zurich, Switzerland) was co-transfected with helper in the 2nd generation transfection system into 293T cells, as detailed in a previous study (29). The 293T cells were cultured in DMEM supplemented with 10% FBS at a temperature of 37°C and 5% CO₂. Following a 48 h incubation period at 37°C; the supernatant of the 293T cells were collected and the lentivirus was concentrated by subjecting to centrifugation at 2,100 x g for 5 min at 4°C and filtering the supernatant through a 0.45 µm filter. The harvested lentiviral vectors were then introduced into the target cells for transduction. Subsequently, cell lines were plated into a 6-well plate and the cells were cultured until they reached 80% confluence. To generate stable cell lines, cells were transduced with these lentiviral vectors in a milieu containing polybrene at a concentration of 8 µg/ml. Then, lentivirus was added and co-cultured with the cells at 37°C for 24 h (Multiplicity of infection, 30). After that, the medium was replaced, and then culture continued in a 5% CO₂ and 37°C incubator for another 48 h. Following a 72 h incubation period, cells underwent a selection process using 10 µg/ml blasticidin (MilliporeSigma), over a course of 3 days.

For the cancer cells, protein extraction was performed using Pro-Prep (cat. no. 17081; Intron Biotechnology, Inc.). The quantification of proteins was conducted using the BCA Protein Assay kit (cat. no. 23227; Thermo Fisher Scientific, Inc.). Proteins (15 µg) were subjected to separation on a 7.5% SDS-polyacrylamide gel, using the Bio-Rad electroporation system (Bio-Rad Laboratories, Inc.), followed by transfer to PVDF membranes (cat. no. IPVH00010; MilliporeSigma). After the membranes were blocked with 3% BSA (BioShop) in tris-buffered saline containing 0.1% Tween-20 (TBST) at room temperature for 1 h and then incubated at 4°C overnight with primary antibodies. The primary antibodies used included: CD46 (1:2,000; cat. no. ab108307; Abcam); MMP9 (1:1,000; cat. no. 7-11C; Santa Cruz Biotechnology, Inc.), phosphorylated (p-)AKT (1:1,000; cat. no. 05-669; MilliporeSigma), p38α (1:1,000; cat. no. sc-535; Santa Cruz Biotechnology, Inc.), JNK1/3 (1:1,000; cat. no. sc-474; Santa Cruz Biotechnology, Inc.), p-JNK (1:1,000; cat. no. sc-6254 Santa Cruz Biotechnology, Inc.), ERK1 (1:1,000; cat. no. sc-271269 Santa Cruz Biotechnology, Inc.), c-Fos (1:1,000; cat. no. sc-166940; Santa Cruz Biotechnology, Inc.), p-c-Fos (1:1,000; cat. no. sc-81485; Santa Cruz Biotechnology, Inc.), c-Jun (1:1,000; cat. no. sc-1694; Santa Cruz Biotechnology, Inc.), p-c-Jun (1:1,000; cat. no. sc-822; Santa Cruz Biotechnology, Inc.); p-p38 (1:2,000; cat. no. 05-1059; MilliporeSigma), AKT (1:2,000; cat. no. 9272; Cell Signaling Technology, Inc.); MMP2 (1:2,000; cat. no. NB200-114; Novus Biologicals; Bio-Techne); p-ERK (1:2,000; cat. no. 05-797R; MilliporeSigma); and β-actin (1:2,000; cat. no. 26628-22-8; MilliporeSigma). After washing three times with TBST, the membranes were incubated at room temperature for 1 h with the secondary antibody (Anti-mouse IgG HRP-linked antibody; 1:5,000; cat. no. 7076S; Anti-rabbit IgG HRP-linked antibody; 1:10,000; cat. no. 7074; Cell Signaling Technology, Inc.). Following secondary antibody incubation, the membranes were washed four times again and then completely covered with a 1:1 mixture of HRP substrate peroxide solution/HRP substrate Luminol reagent (cat. no. WSKLS0500; MilliporeSigma). Band visualization was conducted using the Fusion Solo system (Vilber Lourmat Deutschland GmbH) and analyzed with ImageJ software (version 1.53m; National Institutes of Health, USA).

Total RNA was extracted from cells using TRIzol^® reagent (Thermo Fisher Scientific, Inc.) according to the manual instructions. cDNA was generated with M-MLV reverse transcriptase (cat. no. M170B; Promega Corporation). mRNA expression was determined by qPCR on a CFX96 real-time PCR System (Bio-Rad Laboratories, Inc.) using TOPreal SYBR Green qPCR premix kit (cat. no. RT500M; Enzynomics), using GAPDH as the internal loading control. The primer sequences for PCR were as follows: human CD46 (Forward: 5′-CCACGACCATTTGAAGCTAT-3′, Reverse: 5′-TCCAGGTGCTGGATCACAC-3′), human MMP-9 (Forward: 5′-CATCGTCATCCAGTTTGG-3′, Reverse: 5′-GATGGATTGGCCTTGGAA-3′) and human GAPDH (Forward: 5′-GAAGGTGAAGGTCGGAGTC-3′, Reverse: 5′-GAAGATGGTGATGGGATTTC-3′). PCR cycling conditions for all samples included: 10 min at 95°C for enzyme activation, followed by 40 cycles consisting of melting (95°C; 15 sec) and annealing/extension (72°C; 30 sec) phases. The data were analyzed as previously described (33). Briefly, using the comparative CT method, the Ct values obtained were converted to picogram quantities based on each gene's standard curve from target cDNA. The quantity of CD46 and MMP9 were normalized to GAPDH and adjusted by subtracting values from no reverse transcriptase controls, averaging the result for each triplicate (34).

Two variants of CD46 siRNA (CCAAAACCCUACUAUGAGA, GGAUACUUCUAUAUACCUCUU) and negative control siRNA (UGCAGGUUUAUAGUCCACAUU) were procured from Bioneer Corporation. 5637 cells underwent transfection with these siRNAs using Lipofectamine^® RNAiMAX transfection reagent (cat. no. 13778-075; Invitrogen; Thermo Fisher Scientific, Inc.), adhering to the manufacturer's protocol. Cells were cultured in six-well plates until they reached 60% confluence. To prepare the transfection mix, 7.5 µl Lipofectamine^® RNAiMAX was diluted in 150 µl Opti-MEM Medium (Gibco; Thermo Fisher Scientific, Inc.) and mixed. DNA siRNA (25 pmol) was diluted in 150 µl Opti-MEM Medium and mixed. The diluted DNA was then combined with the previously diluted Lipofectamine^® RNAiMAX, at a 1:1 ratio, to form a DNA-lipid complex. This complex was allowed to incubate for 15 min at room temperature to a stable DNA-lipid complex. Then, the complex was administered to the cultured cells. Following transfection, the cells were incubated at 37°C. At 6 h post-transfection, the cells were chanced and incubated at 37°C with media containing 5% FBS. At 48 h post-transfection, the cells were lysed for western blot analysis as described earlier.

5637, J82 and 253J cells were cultured in six-well plates until they reached 90% confluence in 2 ml of growth medium. The cells were then washed and incubated in RPMI containing 0.5% FBS for 16 h. A sterile 200 µl pipette tip was used to scrape the cell monolayer. Subsequently, the cells were gently washed thrice with PBS and FBS-free medium was added for starvation culture at 37°C. Images were captured at 0, 12 and 24 h post-scraping using an inverted microscope (Olympus IX51; Olympus Corporation). The open wound area percentages were measured and calculated with ImageJ software (version 1.53m; National Institutes of Health).

The migration devices (cat. no. 3422; Corning, Inc.) and invasion devices (cat. no. 354483; Corning, Inc.) were subjected to room temperature equilibration for 10 min, then the two devices were incubated for 2 h in a 37°C incubator, and 300-µl serum-free RPMI was added in the upper chamber and 600 µl in the lower chamber. Cells at 80-90% confluence were washed and incubated in RPMI with 0.1% FBS for 16 h prior to their application to the chambers. Subsequently, 5637 and J82 cells (5×10⁴) in 200 µl of serum-free medium were seeded into the upper chamber and the lower chambers were filled with 600 µl of RPMI media containing 10% fetal bovine serum. After a 24-h incubation, cells on the bottom membrane of the chamber were gently wiped with a cotton swab and migrating cells were stained with Diff-Quik stain solution (cat. no. 38721; Sysmex Corporation) at room temperature. After drying, images were captured under a light microscope (Olympus IX51; Olympus Corporation) and cells were counted in four fields per chamber.

Cell proliferation was assessed using the WST-1 assay (EZ-Cytox Cell Viability Assay kit; ITSBio, Inc.), following the manufacturer's instructions. Briefly, cells were seeded in 96-well plates with 100 µl of RPMI media containing 10% FBS and cultured at 37°C. Then, at the indicated time points (0, 24, 48 and 72 h), 10 µl of the kit solution was added to the cells, followed by a 30-min incubation at 37°C. Cell viability was measured at 450 nm using an Epoch Microplate Spectrophotometer (BioTek Instruments, Inc.).

Cells at 80-90% confluence were washed and incubated in RPMI without FBS for 24 h in 60 mm plates. The samples were then electrophoresed on a 7.5% SDS-polyacrylamide gel containing 1 mg/ml gelatin. Post-electrophoresis, the gel was washed twice for 2 h at room temperature using a washing buffer [50 mM Tris-HCl (pH 7.5), 5 mM CaCl₂, 1 µM ZnCl₂ and 2.5% Triton X-100] and then briefly rinsed with H₂O. Subsequently, the gel was incubated with 1X of Zymogram developing buffer 10X, pH 7.45 (cat. no. 42620000-2; Bioworld Technology, Inc.) at 37°C for 24 h. After incubation, the gels were stained with Coomassie Brilliant Blue R-250 Staining Solution (cat. no. 1610436; Bio-Rad Laboratories, Inc.) and destained with Coomassie Brilliant Blue R-250 Destaining Solution (cat. no. 1610438; Bio-Rad Laboratories, Inc.).

Transcriptional activity was assessed using a luciferase assay system. The pGL4.17[luc 2/Neo] vector, containing the MMP9 promoter region (−924/+13), known as the MMP9-luc promoter, was used. The MMP9-luc vector was kindly provided by Dr Young-Han Lee (Konkuk University, Korea). The AP-1-luc vector was obtained from Stratagene (Agilent Sumitomo Dainippon Pharma Co., Ltd.). CD46-overexpressing and control cells were seeded in a 24-well plate at a density of 1×10⁵ cells per well and cultured for 24 h. The cells were transfected with 0.1 µg of the reporter vector and 2 ng of Renilla using Lipofectamine 2000^® (Invitrogen; Thermo Fisher Scientific, Inc.) following the manufacturer's instructions. At 6 h post-transfection, the cells were washed and cultured with media containing 5% FBS at 37°C for 18 h. The addition of SB202190 or LY294002 was at 24 h post-transfection. 24 h following addition of inhibitor, luciferase and Renilla activities were measured. Light signals were detected using the Dual-Luciferase Reporter Assay System (Promega Corporation) and a Glomax Navigator instrument (Promega Corporation), in accordance with the manufacturer's instructions.

Cell media or mouse sera were collected under sterile conditions, immediately frozen and stored at −20°C until use. MMP9 concentration was determined using the human MMP9 Quantikine ELISA Kit (cat. no. DMP900; R&D Systems), following the manufacturer's instructions.

A total of 28 female BALB/c nude mice (6-weeks-old; 17-22 g; Orient Bio, Inc.) were housed under specific pathogen-free conditions in the animal facility at the Hwasun Biomedical Convergence Center, with a 12-h light/dark cycle (light from 7:00 am to 7:00 pm) and controlled temperature maintained at 24±2 °C for optimal growth, while maintaining a relative humidity range of 50±10%. The mice were allowed ad libitum access to water and food pellets (PicoLab 5053). All procedures performed in studies were approved by the Animal Use and Care Committees at Chonnam National University Medical School (approval no. H2022-70). Mice were used to establish lung metastasis models through tail vein inoculation with 1.5×10⁶ UM-UC-3 and 5637 cells. For each cell line, mice were randomly assigned to the experimental group (overexpressing CD46) and the control group (7 mice per group). Overall survival and physiological signs of mice in the groups were monitored daily (for a period of 6 weeks). If the animals reached any of the following humane endpoints, they were sacrificed: Animal weight loss >20%; severely decreased mobility/activity, moribund behavior or wasting syndrome. Notably, none of the mice suffered from those signs of illness or mortality during the experimental process. At the end of the experiment, the mice were deeply anesthetized with 5% isoflurane (Ifran; Hana Pharm, Co., Ltd.) to minimize any potential pain or distress during the procedure, followed by cervical dislocation to sacrifice the mice. Their blood was collected via the left ventricle using a 23-25 gauge needle. The lung metastasis tumors' fluorescence was detected using the FOBI Fluorescence In Vivo Imaging System (Cellgentek, Co., Ltd.). Additionally, lung and liver tissues were fixed with 10% buffered formalin at 4°C for 48 h. Then tissues were washed three times with PBS 1X and dehydrated with increasing alcohol concentration from 30 to 100% within 4 h and xylene within 1 h, embedded in paraffin and sliced into 5-µm serial sections and subsequently analyzed through H&E staining at room temperature for 1 min 40 and 15 sec and immunohistochemistry.

Tumor paraffin (5 µm) sections from the mice were deparaffinized and rehydrated in graded alcohols and distilled water. Then they initially treated with 3% H₂O₂ in 60% methanol for 10 min and then incubated with Immuno-Block reagent (BioSB, Santa Barbara, CA) for 30 min at room temperature. These sections were further incubated overnight at 4°C with primary antibodies against MMP9 (1:200; cat. no. HPA001238; Atlas Antibodies AB). For the secondary antibody, anti-rabbit immunoglobulin IgG (1:500; cat. no. 30014; Vector Laboratories, Inc.) was used and incubated for 1 h at room temperature. Immunoreactivity was then visualized using an enhanced DAB kit (Abcam). The results were observed using a light microscope.

GraphPad Prism software (Dotmatics) was employed for all statistical analyses. Comparison of two data sets was done with two-tailed unpaired Student's t-test in case of normal distributed data, or with U-Mann Whitney test in case of non-parametric data, considering P-values <0.05 as statistically significant. Multiple groups were analyzed by one-way ANOVA followed by Tukey's post hoc test or two-way ANOVA followed by Bonferroni's post hoc test. The results are presented as mean ± standard deviation.

The observations indicated that ~50% of bladder cancers exhibit an overexpression of CD46 compared with normal urothelia (29). CD46 typically acts as a protector against complement-mediated cytotoxicity in cells. Previous studies reported that CD46 also shields cancer cells from the indirect cytotoxic effects of cetuximab (an EGFR inhibitor), by modulating AKT and ERK phosphorylation in bladder cancer cells (16). The direct role of intracellular CD46 in cancer cells, however, remains poorly understood. Recently, we found that overexpression and suppression of CD46 in bladder cancer cells regulated several genes associated with migration and invasion, as identified by DNA microarray analysis (30). These genes include matrix Gla protein (MGP) and keratin 13 (KRT13) (30). Therefore, the present study explored whether CD46 affected the expression and activity of MMPs, which are pivotal in cancer cell invasion and metastasis across various types of cancer. Focusing on MMP2 and MMP9, known to play significant roles in bladder cancer metastasis (35), the present study examined if forced CD46 expression could alter their expression levels. Previously described stable transfectants overexpressing CD46 in bladder cancer cells (5637, J82, UM-UC-3, 253J and HT1376) and melanoma cells (M010119) were used (29,36). Additionally, LNCaP prostate cancer cells, HCT116 colon cancer cells and A549 lung cancer cells were transfected with pBlasti-CD46 vectors, using pBlasti as control (vehicle group). Western blotting analysis revealed that CD46 overexpression stimulated MMP9 expression in 5637, J82 and UM-UC-3 cells, but not evidently in 253J cells (Fig. 1A). In HT1376 cells, where CD46 previously showed a stimulatory effect on migration (30), MMP9 expression remained unchanged. By contrast, most non-bladder cancer cells (LNCaP, HCT116 and A549) did not exhibit enhanced MMP9 upregulation with CD46, except for M010119 melanoma cells, which showed a marked increase in MMP9 expression. In all tested cells, CD46 stimulation of MMP2 expression was not observed, with the exception of M010119. Given the consistent upregulation of MMP9 mediated by CD46, the present study focused on three bladder cancer cell lines that exhibited increased MMP9 expression (5637, J82 and UM-UC-3). To determine the effect of CD46 overexpression on cancer cell proliferation, a cell proliferation assay was conducted over three days. The results indicated that CD46 overexpression did not significantly affect cell proliferation in these three bladder cancer cells (Fig. 1B; P>0.05 by Mann-Whitney U test), thus underscoring the CD46-induced MMP9 overexpression. These findings suggested that CD46 selectively enhances the expression of MMP9, a key factor in cancer metastasis, in bladder cancer cells.

To elucidate the mechanism behind CD46's role in elevating MMP9 expression, the present study examined several key signaling pathways that govern MMP9 regulation. There is mounting evidence indicating the involvement of both MAPK and PI3K/AKT cascades in MMP9 regulation (27,37,38). The primary constituents of the MAPK family include ERK, p38 and JNK, while PI3K/AKT is crucial for cell survival and growth (39). Investigating the potential link between CD46 and MMP9, the present study assessed alterations in the MAPK and PI3K/AKT pathways in bladder cancer cells exhibiting CD46-induced MMP9 expression. As depicted in Fig. 2A, CD46 stimulation not only enhanced MMP9 expression but also increased phosphorylation levels of p38 MAPK and AKT proteins in all three cell lines examined. The right panel of Fig. 2A presents a quantitation graph corresponding to the left panel. No significant changes were observed in JNK and ERK phosphorylation across the cell lines, with the exception of p-ERK in UM-UC-3 cells (Student's t-test; P<0.0001). To minimize potential nonspecific effects of overexpression via viral vectors, we introduced CD46 siRNA (siCD46) into 5637 cells through transient transfection. CD46 suppression effectively reduced MMP9 expression and decreased phosphorylation of p38 and AKT (Fig. 2B; Student's t-test; P<0.0001). Jun and Fos family proteins form a dimeric complex of activator protein-1 (AP-1), influenced by both MAPK and PI3K/AKT pathways (37,38,40,41). Fig. 2A illustrated that CD46 overexpression also led to increased phosphorylation of c-Jun, but not c-Fos, in all three cell lines tested (Student's t-test; P<0.0001). These results indicated that CD46 enhanced the activation of p38 MAPK and PI3K/AKT in bladder cancer cells, which appears to be associated with increased MMP9 expression, suggesting a potential role of CD46-mediated activation in bladder cancer metastasis through the p38 MAPK and PI3K/AKT pathways.

To examine the influence of CD46 on MMP9 transcription, the present study conducted RT-qPCR in 5637 cells. The results showed that CD46 overexpression significantly increased MMP9 mRNA levels (Fig. 3A; P<0.01). Additionally, reporter gene transcription assays with MMP9 or AP-1 promoters revealed CD46-induced upregulation of both MMP-9 and AP-1 promoter activities in all three cell types examined (Fig. 3B; P<0.01). Subsequently, the present study explored if CD46-mediated MMP9 overexpression involved p38 MAPK and PI3K/AKT pathway activation. UM-UC-3 cells were treated with specific inhibitors: p38 inhibitor SB202190 (20 µM) and AKT inhibitor LY294002 (10 µM) for 24 h. The results, depicted in Fig. 3C, demonstrated that both inhibitors significantly reduced CD46-induced MMP9 promoter activity (P<0.0001) and AP-1 activity (P<0.01). Moreover, these inhibitors also decreased CD46-stimulated MMP-9 protein overexpression in 5637 and J82 cells, as confirmed by western blot analysis (Fig. 3D; P<0.0001). Collectively, these findings suggested that CD46 enhanced MAPK and AKT signaling, leading to AP-1-mediated transcription and, consequently, increased MMP9 expression in bladder cancer cells.

To explore the functional impact of CD46-stimulated MMP9 expression, the bioactivity of MMP9 secreted by 5637 and J82 cells was analyzed using gelatin zymography. This revealed that CD46 markedly elevated the gelatinase activity of MMP9 in cells overexpressing CD46 (Fig. 4A). It was also observed that CD46 is actively secreted into the cell culture media in CD46-overexpressed cancer cells (bottom panel of Fig. 4A). MMP9 is known to actively cleave membrane CD46 to increase soluble CD46 (42,43). Additionally, MMP9 levels secreted into the media were quantified by ELISA, finding significantly elevated MMP9 levels in the media of both cell types (Fig. 4B; P<0.01 for both 5637 and J82). The influence of CD46 on cell migration and wound healing was further investigated using wound-healing assays. Here, the width of the healed gap was measured after a specified incubation period. As depicted in Fig. 4C, CD46 overexpression significantly accelerated wound repair in 5637 and J82 cells at each time point (P<0.001). In 253J cells in which CD46 did not upregulate MMP9 expression, CD46 did not affect the wound repair capability of cells (Fig. S1). Furthermore, to confirm the role of CD46 in promoting bladder cancer cell migration and invasion, Transwell migration and invasion assays were conducted. Post-incubation, filters were stained and examined under a microscope at randomly chosen areas to count cells. CD46 overexpression substantially increased both migration and invasion in 5637 and J82 cells (Fig. 4D; P<0.01). Correspondingly, suppression of CD46 by siRNA caused 5637 cells to slow wound repair and decreased migration and invasion (Fig. S2). These findings suggested that the upregulation of MMP9 expression induced by CD46 was linked with enhanced bioactivity and secretion of MMP9 in bladder cancer cells.

To explore the effect of CD46 overexpression in a live experimental metastasis model, 5637 cells were injected into the tail vein of nude mice, with seven mice in each group. Over a period of six weeks following the injection, the weight of the mice was monitored. The results indicated that CD46 overexpression did not significantly alter the weight of the mice groups (Fig. S3A; P>0.05; Mann-Whitney U test). At the end of the six-week period, the mice were sacrificed and examined for metastatic tumor nodules in the lung and liver. Gross and fluorescent examinations of the lung surfaces were conducted (Fig. S3B and C). The lungs of the mice injected with 5637 cells displayed multiple positive fluorescence foci, as shown in a representative image (Fig. 5A). The lung tissues were fixed and embedded in paraffin to identify metastatic tumor nodules within the lung parenchyma. Serial lung tissue sections, taken every 1-2 mm up to 5 sections, were stained with H&E and the tumor nodules counted. The number of lung metastases was significantly higher in the CD46-5637 cells group, with 6 out of 7 mice showing more than one micrometastatic nodule, in contrast to almost none in the control group (Fig. 5B; P=0.04). The average tumor area in lung metastatic nodules of the CD46-5637-injected mice was 11.99±0.19 mm², compared with >1 mm² in control cell-injected mice (data not shown). The entire microscopic examination of the lung tissues is shown in Fig. S3D. Fig. 5C (a-f) shows a representative lung tissue with a tumor nodule. Immunohistochemical staining revealed that the metastasized tumor cells did not significantly overexpress MMP9 (Fig. 5C, g-h). No tumor nodules were detected in the liver of any mice. Additionally, serum analysis during sample collection showed that mice injected with CD46-overexpressing cells had higher blood levels of MMP9 compared with those injected with control cells, as determined by ELISA (Fig. 5D). These results demonstrated a significant increase in lung metastasis in mice due to CD46 overexpression. While overexpression of MMP9 in these metastatic tumor nodules was not definitively positive, the higher MMP9 levels in the blood of mice injected with CD46-overexpressing cells suggested a potential association between CD46, MMP9 expression and tumor metastasis.