SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References
  9. Supporting Information

Increased expression of heparin-binding EGF-like growth factor (HB-EGF) and membrane-type matrix metalloproteinase-1 (MT1-MMP) is frequently associated with various types of malignant tumor. HB EGF-like growth factor has been reported to promote the malignant progression of ovarian carcinoma. Based on this finding, inhibition of HB-EGF activity with CRM197 is now under phase I clinical evaluation. On the other hand, MT1-MMP expressed in ovarian carcinoma cells is thought to promote invasion and growth of tumor cells by degrading the extracellular matrix. However, we recently demonstrated that co-expression of MT1-MMP and HB-EGF in gastric carcinoma cells leads to cleavage of HB-EGF within its N-terminal heparin-binding region, converting it into a potent heparin-independent growth factor. In this study, we evaluated the importance of regulation of HB-EGF by MT1-MMP in clinical samples of ovarian carcinoma. We detected co-expression of HB-EGF and MT1-MMP in clear cell ovarian carcinoma tissues, particularly at the invasion front and in tumor cells that had disseminated into the ascites, whereas HB-EGF alone was expressed in non-invasive borderline ovarian tumor tissue. Furthermore, a soluble HB-EGF fragment that corresponds to that processed by MT1-MMP was detected in malignant ascites obtained from patients with metastatic ovarian carcinoma. Ovarian carcinoma cells that express MT1-MMP and HB-EGF exhibited enhanced cell growth in a 3D-collagen matrix and anchorage-independent growth in suspension. These results indicate that MT1-MMP co-expressed with HB-EGF in ovarian carcinoma cells potentiates the activity of HB-EGF to promote invasive tumor growth and spreading in vivo. (Cancer Sci 2011; 102: 111–116)

Ovarian carcinoma has the greatest mortality among gynecological cancers, with a rate of cure that has not improved over the last 30 years.(1) The growth of many types of tumor cells is dependent on signals mediated by ErbB family receptors.(2) Indeed, inhibition of signaling by these receptors is a promising therapeutic approach to cancer, as has been shown with drugs such as gefitinib, lapatinib and erlotinib.(3) Members of the EGF family of growth factors bind to ErbB receptors and one or more such ligands have been exploited by tumor cells to support their growth.(4)

Heparin-binding EGF-like growth factor (HB-EGF) is the predominant growth factor responsible for supporting the growth of human ovarian carcinomas and it plays a key role in the acquisition of malignant tumor phenotypes such as rapid growth, spreading into the abdominal cavity and resistance to chemotherapy.(5) CRM197 is a mutant diphtheria toxin fragment that binds HB-EGF and prevents the latter from binding ErbB receptors. It is currently being evaluated in a phase I clinical trial to treat ovarian carcinoma patients.(5)

Similarly to other members of the EGF family, HB-EGF is synthesized as a transmembrane protein (proHB-EGF).(6) Ectodomain shedding of proHB-EGF is mediated by members of the ADAM (A Disintegrin And Metalloprotease) family of proteinases(7) and proHB-EGF is thereby converted into a soluble form. Therefore, soluble HB-EGF acts as a growth factor that stimulates ErbB receptors.(6) In addition, recent studies revealed that the remaining membrane-bound portion of the protein following ectodomain shedding transmits signals to the nucleus via its cytoplasmic domain.(8) ProHB-EGF is also cleaved by MMP7 expressed in tumor cells at a site similar to that cleaved by ADAM or at a site in the N-terminal portion of the EGF-like domain.(9,10)

Most tumor cells express multiple MMP that regulate the tumor microenvironment via the processing of proteins in the extracellular space. We have been interested in membrane-type matrix metalloproteinase-1 (MT1-MMP), which is a potent pro-invasive and growth promoting membrane protease often expressed in invasive tumor cells. Important substrates for MT1-MMP include collagen and many other ECM proteins.(11) MT1-MMP cleaves a variety of membrane proteins and modulates their biological activities.(11) We recently identified an interesting link between MT1-MMP and HB-EGF using various cell lines, including human gastric carcinoma cells.(12) Co-expression of MT1-MMP and HB-EGF led to cleavage of a site within the N-terminal portion of proHB-EGF that is different from the site cleaved by MMP7, thereby generating a membrane-bound fragment of HB-EGF smaller than that produced by MMP7.(12) Following this proteolytic processing, the ectodomain of HB-EGF can be shed by ADAM and released from the cells. Elimination of an N-terminal peptide by MT1-MMP abrogated the ability of HB-EGF to bind heparin, which generally acts as a co-factor for the mitogenic activity of HB-EGF in vitro.(13) Gastric carcinoma cells that express both HB-EGF and MT1-MMP grew invasively in a collagen matrix in vitro and their growth was dependent on both HB-EGF and MT1-MMP.(12) Forced expression of MT1-MMP and HB-EGF in tumor cell lines had a synergistic effect on their growth in mice.(12)

The aim of the present study was to confirm the significance of our previous studies in vitro and in mouse models in clinical specimens and to evaluate the possible involvement of MT1-MMP-dependent regulation of HB-EGF in ovarian carcinomas using clinical specimens and cell lines. We observed co-expression of HB-EGF and MT1-MMP in clinical tissue samples and in tumor cells disseminated into the ascites. A soluble HB-EGF fragment corresponding to that predicted to be produced by processing by MT1-MMP can be detected in ascites from patients. The processing of HB-EGF by MT1-MMP and the biological impact of this processing were confirmed using ovarian carcinoma cell lines. Detection of the processed HB-EGF fragment in ascites may serve as a biomarker for involvement of MT1-MMP in the activation of HB-EGF to promote malignant progression of ovarian carcinomas.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References
  9. Supporting Information

Cells and cell culture.  Human ovarian carcinoma OVCAR-8 cells were a gift from Dr Vito. Quaranta (Vanderbilt University, Nashville, TN, USA) and SKOV-3 cells were purchased from the American Type Culture Collection. All cells were cultured in RPMI-1640 (Sigma, St Louis, MI, USA) supplemented with 10 mM HEPES, 1.2 mg/mL of NaHCO3, 2 mM glutamate and 10% fetal calf serum (FCS) (HyClone, Logan, UT, USA). These cells were cultured in a humidified atmosphere of 5% CO2/95% air.

Anchorage-independent growth.  Mock and HB-EGF transfectant SKOV-3 cells were suspended in RPMI-1640 complete growth medium containing 1% FCS, seeded onto a 96-well plate coated with non-adhesive polymer (100 μL each, 1 × 105 cells/well) (PrimeSurface, Sumitomo Bakelite, Tokyo Japan), and incubated for 3 days in an incubator.(12) The morphology of multicellular spheroids in suspension culture was observed by phase-contrast microscopy (Olympus, Tokyo, Japan). The spheroid cells were collected, suspended in 0.1% (w/v) trypsin/0.02% (w/v) EDTA/PBS solution for 3 min at 37°C, and treated with 0.1% trypan blue solution. The number of live cells was counted with a hemocytometer by phase-contrast microscopy.

Detection of sHB-EGF in ovarian cancer ascites.  Ascites from ovarian cancer patients was collected at Fukuoka University Hospital. We obtained informed consent from all patients according to protocol approved by the internal review board for clinical study (05-05-3, April 16, 2008). Ascites from randomly selected patients was centrifuged to remove debris. The supernatant was incubated with CRM197-conjugated Sepharose beads for 2 h at 4°C to purify HB-EGF. The CRM197-Sepharose was washed three times with buffer (50 mM Tris–HCl, pH 7.4, 150 mM NaCl, 1% Triton X-100, 5 mM o-phenanthrolin) and the bound materials were eluted with 2× SDS-PAGE sample buffer. The eluate was subjected to western blot analysis using anti-HB-EGF polyclonal antibody (pAb).

Immunohistochemistry.  Tumor tissue sections of human ovarian clear cell adenocarcinoma and borderline ovarian tumor tissues had been surgically obtained at Fukuoka University Hospital (Fukuoka, Japan). We obtained informed consent from all patients according to protocol approved by the internal review board for clinical study (05-05-3, April 16, 2008). The tissue was snap frozen and embedded in OCT compound. The thickness of each section was 8 μm. Samples were treated with either monoclonal antibody (mAb) against MT1-MMP (10 μg/mL) or HB-EGF (5 μg/mL) overnight at 4°C. Other experimental methods were described previously.(14)

Antibodies.  Anti-MT1-MMP mAb (1D8) was a gift from Dai-ich Fine chemical (Takaoka, Japan), and anti-HB-EGF mAb was produced in our laboratory; anti-FLAG (M2) mAb was purchased from Sigma; HB-EGF pAb were from R&D systems (Minneapolis, MN, USA) and Santa Cruz Biotechnology (Santa Cruz, CA, USA).

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References
  9. Supporting Information

Co-expression of HB-EGF and MT1-MMP in invasive human ovarian carcinoma.  Published reports have examined the expression of HB-EGF and MT1-MMP separately in ovarian carcinoma tissues and cells.(15,16) However, it is not known whether these proteins are co-expressed in the same tumor cells. We performed immunohistochemical staining of serial sections of surgically dissected tumor samples and representative results are presented in Figure 1. Expression of HB-EGF was detected in samples of both invasive clear cell adenocarcinoma (Fig. 1A and S1) and non-invasive borderline ovarian tumor tissue (Fig. 1B). The HB-EGF protein was detected in nearly all tumor cells in the sampled tissues. In contrast, MT1-MMP expression was only detected in confined areas of the invasive carcinoma tissue that corresponded to the invasion front (Fig. 1A and S1). Expression of MT1-MMP was not detected in non-invasive ovarian carcinoma, although such tumor cells expressed HB-EGF (Fig. 1B).

image

Figure 1.  Co-localization of HB-EGF and MT1-MMP in invasive ovarian carcinoma tissue. Serial sections of clear cell carcinoma (A) and non-invasive borderline ovarian tumor tissues (B) were examined by immunostaining with anti-HB-EGF mAb (a and d) or anti-MT1-MMP mAb (b and e). As a negative control, IgG was used for staining (c and f). The staining patterns were observed under a microscope (×100 or ×400) and representative fields are presented (a, b and c). The regions within the squares are enlarged (d, e and f, ×400 magnification). Arrows indicate typical positive signals. T and N indicate cancer and interstitial normal tissues, respectively.

Download figure to PowerPoint

Malignant clear cell ovarian carcinoma cells frequently disseminate into the abdominal cavity and induces accumulation of ascites. Cells within the ascites obtained from patients were subjected to immunohistochemical staining (Fig. 2). A significant number of the cells were positive for both MT1-MMP and HB-EGF expression. Thus, it appears that HB-EGF is co-expressed with MT1-MMP in invasive carcinoma cells obtained from patients with ovarian clear cell carcinoma.

image

Figure 2.  Co-expression of HB-EGF and MT1-MMP in ovarian carcinoma ascites. Invasive ovarian tumor cells that had metastasized to the ascites were obtained from a patient and were subjected to immunohistochemical analysis of the expression of HB-EGF (A) and MT1-MMP (B) using anti-MT1-MMP and anti-HB-EGF mAb. As a negative control, mouse IgG was used for staining (C). The staining patterns were observed under a microscope (×400). Ascites tumor cells formed spherical aggregates as indicated by arrows.

Download figure to PowerPoint

Detection of HB-EGF fragments in ascites obtained from ovarian carcinoma patients.  We previously reported that co-expression of MT1-MMP and HB-EGF in cultured cells gave rise to cleavage of the latter, as schematically illustrated in Figure 3A.(12)Membrane-bound forms of N-terminally processed (mN) HB-EGF fragments are designated the mN1- to mN4-fragments (Fig. 3A). The propeptide of HB-EGF is usually cleaved by furin or other pro-protein convertases to generate the mN1-fragment. The proteases that generate the mN2-fragment have not been identified. Cleavage by MT1-MMP generates mN3-HB-EGF and the membrane-bound portion of the protein remaining after ectodomain shedding by ADAM corresponds to mN4-HB-EGF. Ectodomain shedding of the mN1-, mN2- and mN3-fragments generates the soluble sN1-, sN2- and sN3-fragments, respectively (Fig. 3A).

image

Figure 3.  Detection of HB-EGF in ascites of ovarian carcinoma. (A) Domain structure and processing sites of HB-EGF. Furin, MT1-MMP and ADAM (A Disintegrin And Metalloprotease) cleave the membrane bound form of HB-EGF at the sites indicated by the arrows. mN1–mN4 represent the membrane anchored forms and the soluble forms of HB-EGF (sN1–sN3) are generated by ADAM activity. The protease(s) responsible for generation of mN2 is (are) not known.(24) (B) The HB-EGF fragments in the ascites harvested from 10 ovarian cancer patients were concentrated by affinity beads conjugated to recombinant CRM197, and were then subjected to western blot analysis using anti-HB-EGF goat IgG (upper) or control goat IgG as a negative control (lower). Arrows indicate the location of the sN1-, sN2- and sN3-fragments at the right (control). Recombinant sHB-EGF were used as a positive control.(12)

Download figure to PowerPoint

However, the findings referred to above were generated using cells cultured in vitro following ectopic overexpression of HB-EGF. The level of expression of endogenous HB-EGF is usually very low and is not sufficient to allow detection by western blot analysis. Detection of the expression of both HB-EGF and MT1-MMP in disseminated ovarian carcinoma cells by immunostaining suggested that soluble HB-EGF fragments might have accumulated in the ascites. If this were the case, it should be possible to detect them by western blot analysis after concentrating the HB-EGF fragments from the ascites. We used CRM197-conjugated Sepharose beads to concentrate soluble HB-EGF fragments in the ascites collected from 10 patients and subjected the concentrate to western blot analysis (Fig. 3B, upper panel). A fragment with a size corresponding to that of sN1 was detected in all samples, although its level varied. Three of the 10 samples contained relatively high levels of this sN1-like fragment and a fragment of size similar to that of sN3 was detected in these samples (Fig. 3B, upper panel). No such discrete fragments were detected in the absence of the anti-HB-EGF antibody (Fig. 3B, lower panel).

MT1-MMP-dependent processing of HB-EGF in a human ovarian carcinoma cell line.  To confirm the results obtained using the clinical samples, we next studied two ovarian carcinoma cells (SKOV-3 and OVCAR-8). Expression of MT1-MMP and HB-EGF was analyzed by RT-PCR analysis (Fig. 4A). OVCAR-8 cells expressed high levels of both HB-EGF and MT1-MMP mRNA, whereas SKOV-3 cells expressed the MT1-MMP mRNA at a more moderate level and only a negligible level of the HB-EGF mRNA. Therefore, we used SKOV-3 cells for a gain-of-function analysis following stable, forced expression of either HB-EGF or a mutant HB-EGF (ucHB) containing alanine substitutions at the cleavage site, which cannot therefore be cleaved by MT1-MMP (Fig. 4B). We also co-expressed TIMP-2 and HB-EGF in these cells in order to evaluate the effect of inhibition of the activity of endogenous MT1-MMP. TIMP-2 is a natural inhibitor of MT1-MMP, which inhibits MT1-MMP-dependent HB-EGF processing. Cell lysates prepared from the stable transfectants were analyzed by western blot analysis for the presence of HB-EGF fragments. We detected each of the mN1- through mN4-fragments in lysates of cells transfected with wild-type HB-EGF (Fig. 4C), whereas the mN3-fragment was missing from lyste prepared from cells transfected with ucHB-EGF (Fig. 4B). Co-expression of TIMP-2 with HB-EGF also suppressed generation of the mN3-fragment (Fig. 4C). These results suggest that endogenous MT1-MMP can cleave HB-EGF in SKOV-3 cells at the previously reported site.

image

Figure 4.  MT1-MMP-dependent HB-EGF processing in ovarian carcinoma cells. (A) Expression of HB-EGF and MT1-MMP mRNA in SKOV-3 and OVCAR-8 cells was analyzed by RT-PCR using specific primers as described previously.(12) The GAPDH mRNA was used as an internal control and cDNA encoding MT1-MMP and HB-EGF were used as positive controls. (B) Expression of MT1-MMP and HB-EGF mRNA in mock-transfected SKOV-3 cells (mock) and in SKOV-3 transfectants expressing either HB-EGF, uncleavable mutant HB-EGF (ucHB), or co-expressing HB-EGF and TIMP-2 (HB/TIMP-2). (C) Detection of membrane bound HB-EGF (mHB-EGF) in cell lysates by western blot using anti-HB-EGF pAb (Santa Cruz).

Download figure to PowerPoint

MT1-MMP promotes HB-EGF-dependent proliferation of ovarian carcinoma cells in collagen gel and in suspension.  Using the stable transfectants of SKOV-3 cells prepared above, we next analyzed the effect of MT1-MMP on HB-EGF-dependent cell growth in 3D-collagen matrix (Fig. 5A). Expression of HB-EGF enhanced the growth of the cells markedly compared with the parental cells (Fig. 5A). Cells stably expressing ucHB also exhibited an enhanced rate of growth that was nevertheless much less than that of cells expressing wild type HB-EGF (Fig. 5A). We have already reported that the mN1-fragment derived from ectodomain shedding of ucHB exhibits a growth stimulatory activity comparable with that of the corresponding wild-type mN1-fragment.(12) Cells co-expressing TIMP-2 and HB-EGF exhibited a rate of growth similar to that of cells expressing ucHB (Fig. 5A). These results suggest that MT1-MMP potentiates the growth stimulatory activity of HB-EGF via specific cleavage of the latter in SKOV-3 cells, similar to the findings we reported previously in gastric carcinoma cells.(12)

image

Figure 5.  Processing of HB-EGF and MT1-MMP enhances ovarian carcinoma cell proliferation. Proliferation of SKOV-3 cells expressing the indicated HB-EGF constructs or expressing HB-EGF and TIMP-2 (Fig. 4C) was analyzed in collagen gels. The cells (1 × 105) were seeded and incubated in collagen gels for a week (A). To measure growth activity of SKOV-3 cells in collagen gels, the cells were extracted from the collagen gels. The number of extracted cells was counted with a hemocytometer. (B) These cells (1 × 105) were incubated in culture media in plates coated with a non-adhesive polymer for analysis of suspension culture proliferation for 5 days (B). To measure growth activity of the suspended cells, the aggregated cells were isolated with trypsin in PBS for 10 min at 37°C. The number of extracted cells was counted with a hemocytometer. Each bar represents the mean ± SD from triplicate experiments. The relative increase in cell number is indicated. The Students’t test was used for statistical analysis.

Download figure to PowerPoint

Dissemination of ovarian carcinoma cells into the abdominal cavity is an important step in the metastatic spread of the tumor. However, the free-floating nature of cells in the ascites is not favorable for proliferation and instead induces an apoptotic process called anoikis. Therefore, we examined whether co-expression of HB-EGF and MT1-MMP could support cell growth in suspension. SKOV-3 cells were seeded into multi-well chambers coated with a non-adhesive hydrophilic polymer. Mock cells formed a loosely aggregated spheroid, although they did not proliferate (Fig. 5B). Expression of HB-EGF in the cells promoted their proliferation (Fig. 5B) and led to the formation of larger spheroids (not shown). However, expression of ucHB that could not be cleaved by MT1-MMP significantly attenuated this effect on cell proliferation (Fig. 5B). A similar attenuation of the growth stimulatory activity was observed in cells co-expressing HB-EGF and TIMP-2. These results suggest that MT1-MMP co-operates with HB-EGF through processing of the latter to support anchorage-independent growth of SKOV-3 cells in suspension.

Growth of OVCAR-8 ovarian carcinoma cells in 3D-collagen gel.  OVCAR-8 cells expressed both endogenous MT1-MMP and HB-EGF (Fig. 4A) and these cells grew well in collagen gels. To confirm that growth of the cells required HB-EGF and MT1-MMP, expression of the corresponding mRNA was inhibited using siRNA specific for each gene. Two different siRNA sequences were used to knockdown expression of each gene (Fig. 6A). The parental OVCAR-8 cells exhibited efficient proliferation in collagen gel when analyzed 2 or 5 days after inoculation into the gel (Fig. 6B). However, knockdown of either HB-EGF or MT1-MMP using these siRNA suppressed proliferation of the cells significantly (Fig. 6B). Because expression of the genes was only transiently knocked down by the siRNA, the effect of the silencing diminished over several days.

image

Figure 6.  Effect of HB-EGF and MT1-MMP on growth of OVCAR-8 cells in collagen gels. (A) Expression of HB-EGF and MT1-MMP mRNA in ovarian tumor OVCAR-8 cells was knocked down using two different siRNA sequences specific for each gene (#1 and #2). Expression of each gene was analyzed by RT-PCR using specific primers.(12) GAPDH mRNA was used as an internal control and cDNA encoding MT1-MMP and HB-EGF were used as positive controls. (B) Growth of OVCAR-8 cells treated with siRNA targeting MT1-MMP or HB-EGF mRNA. Treated and control OVCAR-8 cells (2.5 × 103/well) were cultured in collagen gels for 2 or 5 days. Each bar represents the mean ± SD from triplicate experiments. The relative increase in cell number is indicated. The Students’t test was used for statistical analysis. *P < 0.002.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References
  9. Supporting Information

We have recently reported that HB-EGF is a substrate of MT1-MMP and that removal of the N-terminal fragment of HB-EGF by MT1-MMP converts the former into a hyperactive growth factor that does not require heparin as a co-factor. Many human tumors have been reported to express HB-EGF and current evidence points to an important role for HB-EGF in supporting malignant progression of ovarian carcinomas.(5) We have thus become interested in the possible involvement of MT1-MMP in the HB-EGF-dependent growth of ovarian carcinoma cells. We confirmed that invasive clear cell ovarian carcinoma cells expressed both HB-EGF and MT1-MMP by immunohistochemical analysis of serial tissue sections (Fig. 1A and S1). In contrast to the uniform expression of HB-EGF in almost every tumor cell, MT1-MMP expression was detected focally within a confined area at the invasion front. Non-invasive borderline ovarian tumor tissue expressed HB-EGF but not MT1-MMP (Fig. 1B). Malignant tumor cells that had disseminated into the ascites expressed both HB-EGF and MT1-MMP (Fig. 2). However, these immunohistochemical findings do not constitute sufficient proof of co-expression of both proteins in the same tumor cells. To further study this point and to study the functional analysis of the processing of HB-EGF further, we made use of ovarian carcinoma cell lines. OVCAR-8 cells expressed both HB-EGF and MT1-MMP, whereas SKOV-3 cells expressed only MT1-MMP. Therefore, we have confirmed our previous findings in gastric carcinoma cells.(12) The results suggest that MT1-MMP potentiates the growth stimulatory activity of HB-EGF in ovarian carcinoma.

It is usually very difficult to detect endogenous HB-EGF fragments in tumor samples due to their low level of expression. Accumulation of ascites and the disseminated tumor cells within it allowed us to concentrate the endogenous HB-EGF fragments so that they could be detected by western blot analysis. We detected a HB-EGF fragment corresponding to sN1 in each of the 10 samples we tested. Three of the samples exhibited a greater level of the sN1 fragment compared with the others and also expressed the sN3-fragment produced following processing of HB-EGF by MT1-MMP. We cannot rule out the possibility that the sN3-fragment is present in the other samples at a level below our ability to detect it. Unfortunately there was insufficient sample to determine the N-terminal amino acid sequence of the fragment.

Dissemination of ovarian carcinoma cells into the abdominal cavity is an important step for metastatic spread of the tumor. Such tumor cells in the ascites must be resistant to anoikis until finding a secondary lodging site for the following metastatic growth.(17) However, little information is available about the fate of tumor cells in the ascites. We found that HB-EGF promoted anchorage-independent growth of SKOV-3 cells in cooperation with MT1-MMP (Fig. 5B). This result suggests that disseminated ovarian carcinoma cells can grow in ascites if the cells express both HB-EGF and MT1-MMP.

MT1-MMP is a potent promoter of invasion and proliferation of tumor cells in vitro and in vivo.(18) Tumor cells expressing MT1-MMP formed rapidly growing tumors in mice, and such tumor formation could be inhibited by downregulating MT1-MMP expression or by inhibiting MT1-MMP activity in tumor cells. The mechanism by which MT1-MMP promotes tumor cell growth in collagen matrix or in tissue has been proposed to be through degradation of collagen.(19) However, our results suggest that MT1-MMP might stimulate tumor growth by activating HB-EGF. Therefore, MT1-MMP is a promising therapeutic target to retard tumor growth and invasion by suppressing either ECM degradation or the activation of HB-EGF. However, clinical development of MMP inhibitors has not been successful.(20) Nevertheless, Sabeh et al.(21) have suggested that the serum concentrations of these inhibitors were not sufficient to inhibit MT1-MMP in most cases. Although a phase I clinical trial is evaluating CRM197 for treatment of advanced ovarian cancer patients, combination therapy together with MT1-MMP inhibitors might provide a better therapeutic outcome.(22,23)

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References
  9. Supporting Information

This work was partly supported by a grant-in-aid for Scientific Research on Priority Areas “Integrative Research Toward the Conquest of Cancer” to N.K., M.S. and E.M. and by the Global COE Program “Center of Education and Research for the Advanced Genome-Based Medicine – For personalized medicine and the control of worldwide infectious diseases”, MEXT, Japan, to M.S.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References
  9. Supporting Information

Figure S1. Detection of HB-EGF and MT1-MMP in clear cell carcinoma.

Data S1. Methods.

FilenameFormatSizeDescription
CAS_1748_sm_FigureS1.eps9769KSupporting info item
CAS_1748_sm_DataS1.doc44KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.