Numerous studies have demonstrated that overexpression of Met, the hepatocyte growth factor(HGF) receptor, plays an important role in tumorigenesis. Met activation can either occur through ligand-independent or -dependent mechanisms, both of which are mediated by a series of proteases and modulators. We studied the protein expression of several components of the HGF/Met pathway on a cohort of 330 node-negative breast carcinomas using a tissue microarray annotated with 30-year, disease-specific patient follow-up data. We examined HGF, matriptase (an activator of HGF expressed on mammary epithelial cell surfaces), HAI-I (the cognate inhibitor of matriptase), and the Met receptor itself. Our studies demonstrate tight correlation between the expression of HGF, matriptase, and Met in breast carcinoma. High-level expression of Met, matriptase, and HAI-I were associated with poor patient outcome. Met and HAI-I showed independent prognostic value when compared with traditional breast markers in a multivariate analysis. Intriguingly, antibodies against the intracellular but not the extracellular domain of Met were prognostic, suggesting that overexpression of the cytoplasmic-tail of Met, perhaps through cleavage or truncating mutation, may play an important role in breast cancer progression.

Many studies have demonstrated the importance of the HGF3 pathway in carcinogenesis (1). HGF is produced both by tumor cells as well as by surrounding stromal elements, and can act in either a paracrine or autocrine fashion (2, 3, 4, 5). HGF is secreted as an inactive propeptide, which must be cleaved to become biologically active. One enzyme responsible for this cleavage is matriptase, an epithelial-localized transmembrane serine protease (6, 7, 8). Matriptase is, in turn, regulated by a naturally occurring inhibitor, HAI-1 (9, 10). When cleaved, HGF can bind to its receptor, Met, thereby stimulating multiple downstream pathways, leading to mitogenesis, motogenesis, and morphogenesis (11).

Several studies have analyzed individual components of the HGF pathway for their association with tumor aggression and/or patient survival. Early biochemical studies demonstrated that overall levels of HGF in breast cancers correlated with worse patient outcome (2, 3, 5). Whether HGF production by tumors and/or surrounding stroma is an important prognostic feature is unclear; although tumor cells themselves are a major producer of HGF (4, 5, 12). The use of matriptase and HAI-1 as prognostic markers in breast cancer has not been reported previously. However, recently, one study demonstrated that high matriptase and low HAI-1 levels were associated with advanced-stage ovarian tumors (13). Another report demonstrated that the glycosylation of matriptase stabilized and enhanced its proteolytic activity, and promoted tumor aggression (14).

The expression of Met has been more extensively studied. Met overexpression associates with poor prognosis in a variety of tumors (1). Whether such expression is ligand- (HGF) dependent or independent is unclear; however, the constitutive activation of Met, via several ligand-independent mechanisms, is established. These mechanisms include activating point mutations (15, 16, 17, 18, 19), chromosomal translocations (20, 21), and truncations of the cytoplasmic domain (22, 23). In addition, dysregulation of Met-associated phosphatases may also lead to Met activation (24).

We have now studied several elements of the HGF pathway including HGF, matriptase, HAI-1, and Met in a single cohort of node-negative breast cancer patients with 30-year follow-up, correlating the expression of each element and determining their prognostic value. This study was facilitated by the use of tissue microarrays: arrays of hundreds of patient histological samples on a single glass slide. Our study demonstrates a significant correlation between members of the HGF pathway and shows that several members have independent prognostic value in determining patient outcome.

Cohort Design and Tissue Microarray Construction.

Tissue microarrays were constructed, as described previously, and reviewed recently (25, 26). Three hundred and thirty cases of formalin-fixed, paraffin-embedded, node-negative breast carcinoma were obtained from the archives of the Department of Pathology, Yale University. Cases were taken sequentially, as available, from 1962 to 1980, with a median survival time of 15.6 years. Complete treatment history is not available from this cohort, but the vast majority of patients in this era were not treated with chemotherapy. Representative tumor regions were selected for coring by a pathologist (R. L. C.). Because prior studies have demonstrated that a single core adequately represents the staining pattern of an entire slide, all of the studies were performed using a single sample of each tumor (27, 28). Our previous study has also demonstrated the durability of antigens from archival specimens as old as 70 years (27). In the present study, all of the tumors demonstrated some degree of staining with one or more of the antibodies tested, demonstrating that no cases were antigenically “dead” because of fixation artifacts or tissue age.

Immunohistochemistry.

Briefly, 5-μm tissue microarray slides were deparaffinized with xylene and ethanol. Antigen retrieval was performed using citrate buffer (pH 6.0) pressure-cooking (29). Primary antibodies were incubated overnight at 4°C, with the exception of antibodies to ERs, PRs, and Her2, which were incubated at room temperature for 1 h. Monoclonal anti-matriptase and anti-HAI-1 antibodies were prepared as described previously (9, 30). Commercially acquired antibodies included: polyclonal (goat) anti-HGF antibody (R&D Systems, Minneapolis, MN); monoclonal antibody to the extracellular domain of Met (DO-24; Upstate Biotechnology, Lake Placid, NY); and monoclonal antibody to the intracellular domain of Met (3D4; Zymed, South San Francisco, CA). The specificities of all of the antibodies used were verified using immunoprecipitation and Western blotting. Antibodies to ER, PR, and HER-2/neu were obtained from DAKO (Carpinteria, CA) and used according to the manufacturer’s specifications. Antibodies were either detected using a Vectastain ABC kit (Vector Laboratories, Burlingame, CA) for anti-HAI-1 or the DAKO Envision TM + System (DAKO) for the others. Signal from the HGF antibody was amplified using biotin-tyramide signal amplification followed by a streptavidin-horseradish peroxidase conjugate (TSA kit; Perkin-Elmer Life Sciences, Boston, MA). Staining was visualized using diaminobenzidine and counterstained with acidified hematoxylin. Slides were also stained in the absence of primary antibody to evaluate nonspecific secondary antibody reactions.

Evaluation of Immunostaining.

Immunostaining was scored on a scale of 0 to 3+ (negative/weak/moderate/intense staining). Distinctions between membrane and cytoplasmic staining were impractical given the diffuse staining of the antigens (visualized using the chromogenic substrate, diaminobenzidine). Therefore, scores represent the combined staining intensity of membranous and cytoplasmic staining. Histospots with <10% of their area covered by tumor were excluded from analysis. Scoring was performed by two independent observers (J. Y. K. and M. D-F.), and histocores with discrepant scores were re-examined by both observers to achieve a consensus score. Cases with scores of 2+ or 3+ were designated as “high,” whereas cases with scores of 0 or 1+ were designated as “low.”

Statistical Analysis.

All of the analyses were completed using Statview 5.0.1 (SAS Institute Inc., Cary, NC). Correlations between markers were performed using a χ2 test. Prognostic significance was assessed using both univariate and multivariate Cox proportional hazards models with 30-year survival as an end point. Survival curves were calculated using the Kaplan-Meier method, with significance evaluated using the Manel-Cox log rank test.

We analyzed the expression of several components of the Met pathway including HGF, matriptase, HAI-1, and Met itself (using both intracellular- and extracellular-specific antibodies; Fig. 1). The number of tumors expressing high levels of these antigens ranged from 18 to 46% (Table 1). Although tissue microarrays have been shown to adequately represent tumor antigen expression (27), they most likely do not fully represent the expression of stromal markers. Therefore, we limited our analysis to the expression of markers by tumor cells and not stroma. In the case of HGF, this resulted in our selective study of autocrine (tumor) expression.

To elucidate potential associations between these markers, we performed χ2 analyses, which revealed highly significant associations between the expression of Met, HGF, and matriptase (P < 0.0002; Table 2). HAI-1 expression was independent of these markers, although it trended toward coexpression with matriptase (P = 0.0597). Interestingly, the expression of antibodies to the intracellular and extracellular domains of Met were highly correlated (P < 0.0001; Table 2), but not coincident; 38 cases were scored as entirely Met-extracellular negative (score of 0) and Met-intracellular high (12.2%), whereas only 1 case was judged as entirely Met-intracellular negative and Met-extracellular high (0.3%). This result suggests that relative overexpression of the intracellular domain of Met is far more common than relative overexpression of the extracellular domain.

To determine the predictive power of the Met pathway, we initially performed a univariate analysis of individual Met pathway components and compared them with traditional breast cancer markers (Table 3). Because breast carcinoma can recur and kill patients decades after its initial diagnosis, we studied 30-year disease-related survival. Using univariate analysis, only the cytoplasmic tail of Met, not the extracellular portion, showed prognostic power (P = 0.0029; Table 3). Sixty-one percent of patients overexpressing the cytoplasmic domain of Met died of breast cancer within 30 years compared with 41% with lower levels. High-level matriptase expression was also predictive of poor outcome (51% dead of disease versus 40%; P = 0.0279; Table 3), as were elevated levels of HAI-1 (59% versus 40% survival; P = 0.0110; Table 3). Of the traditional markers of tumor aggression, only tumor size was predictive of outcome (P = 0.0001; Table 3). Kaplan-Meier curves demonstrated that these markers were prognostic over the entire 30-year follow-up period (Fig. 2).

Previous studies have suggested that the ratio of HAI-1 and matriptase may play an important role in promoting tumor aggression (13). Therefore, we compared the survival of patients with tumors expressing high and/or low levels of each. Among tumors expressing high levels of matriptase, HAI-1 coexpression predicted a worse outcome (relative risk = 1.88; 95% CI, 1.05–3.36; P = 0.0335). Comparison of tumors expressing both markers to those expressing neither demonstrated that patients with double-positive tumors have an increase relative risk of 2.43 (95% CI, 1.36–4.34; P = 0.0026). Addition of Met (cytoplasmic domain) to this analysis showed that patients with Met, matriptase, and HAI-1-positive tumors exhibit an increased relative risk of 3.25 (95% CI, 1.24–8.50; P = 0.0165).

We then determined the independent predictive power of the Met pathway. First, we limited our analysis solely to the Met pathway components. Using multivariate analysis, both the cytoplasmic tail of Met and HAI-1 retained independent predictive power (Table 4). When we included these two markers with traditional breast cancer markers, they retained their independence. Tumor size was the only other independent predictor of poor outcome (Table 5).

The relative importance of ligand-dependent and ligand-independent Met activation in carcinogenesis is a matter of continued debate (1). Our studies demonstrate that members of the HGF pathway, namely HGF, Met, and matriptase, are often coexpressed on breast cancers, and that high-level expression of two of these members, Met and matriptase, associates with more aggressive tumors. Such observations would be expected if HGF-mediated Met stimulation played a role in tumorigenesis. Although tumoral HGF levels were not found to be predictive of outcome, the expression of matriptase was prognostic in a univariate analysis, suggesting that as an activator of HGF, it plays a rate-dependent (“gate-keeping”) role in the ligand-dependent stimulation of Met.

In addition to their role in the activation of HGF, matriptase and its cognate inhibitor, HAI-1, also play a role in the plasminogen activator cascade (6, 31, 32). This cascade culminates in the activation of plasmin and the coactivation of matrix metalloproteinases, both of which degrade extracellular matrix components and potentiate tumor cell invasion, extravasation, and metastasis (33). Matriptase promotes this pathway by activating latent uPA, which, in turn, activates plasmin (6, 34). Like matriptase, uPA can also cleave pro-HGF, providing another level of interaction between the Met/HGF pathway and plasmin cascade (35). Given the multiple functions of matriptase, it is not surprising that aggressive breast tumors produce higher levels of this enzyme.

Likewise, HAI-1, as an inhibitor of matriptase, may help modulate both the Met/HGF and plasmin pathways. Interestingly, the expression of HAI-I was independent of the other members of the Met pathway indicating that its expression is regulated differently. Previous reports have suggested that HAI-I is down-regulated in colon carcinoma and high-grade ovarian carcinomas (13, 36). In contrast, our study demonstrates that HAI-1 expression is associated with aggressive breast carcinomas, being an independent predictor of poor outcome (Table 5). Although this may be puzzling in light of the role of HAI-1 in inhibiting HGF-dependent Met activation, the coordinated expression of both matriptase and HAI-1 may be far more important in promoting tumor aggression than the unopposed production of active matriptase in the absence of its inhibitor. Coordinated regulation of another inhibitor of the uPA/plasmin cascade, PAI-1, is crucial for inducing tumor invasion (37). Furthermore, a recent meta-analysis of the uPA/PAI-1 system in breast cancer demonstrated conclusively that overexpression of both uPA and its inhibitor PAI-1 were associated with poor outcome in breast cancer (32). In our study, the importance of matriptase and HAI-1 coexpression is demonstrated in the elevated relative risk of patients with tumor expressing both markers.

Because of lot-to-lot inconsistencies in polyclonal Met-antibodies,4 we analyzed two different monoclonal antibodies, one to the extracellular and one to the intracellular domain. Comparison of Met expression as assessed by these two antibodies showed some interesting results. First, the expression of the intra- and extracellular domains of Met, although highly associated, was not coincident. Second, of cases with mixed expression of intra- and extracellular Met, overexpression of the intracellular domain was far more common, with 12.2% of all of the cases expressing solely the intracellular domain. Third, high levels of the cytoplasmic tail of Met were predictive of poor outcome, whereas expression of the extracellular portion was not. Although this result could be explained by differences in the affinity of the antibodies for Met in formalin-fixed, paraffin-embedded tissue, both antibodies gave strong staining and similar results across a range of titrations (data not shown). A more likely explanation is that the cytoplasmic tail of Met is either cleaved (e.g., after activation) or that mutations in Met lead to an overexpression of the cytoplasmic tail in some tumors. Indeed, recent studies have suggested that the cleavage of the cytoplasmic tail of Met may be important in signal transduction (1, 22, 23). Whether overexpression of the Met intracellular domain relative to the extracellular domain is a ligand-independent or -dependent phenomenon is unclear. Interestingly, expression of the Met extracellular domain correlates with HGF levels but not matriptase levels, whereas the Met intracellular domain correlates with both. This observation would be expected if the binding of matriptase-potentiated HGF to Met induced a subsequent cleavage of the Met intracellular domain.

In summary, we have made use of tissue microarray technology to analyze various components of the Met-signaling pathway. Our studies provide evidence that the expression of the stimulatory members of this pathway (Met, HGF, and matriptase) is tightly correlated. High-level HAI-1 expression is an independent predictor of outcome. Furthermore, studies using antibodies to different domains of the Met receptor suggest that overexpression of the cytoplasmic domain is a strong independent predictor of outcome.

Fig. 1.

Immunohistochemical staining of breast cancer tissue microarrays for four members of the HGF/Met pathway. Each histospot shows a representative positive case (3+) for matriptase (A), HAI-1 (B), HGF (C), and Met (D), at low (×40) and high (×200, insets) magnification. Matriptase, HAI-1, and Met show a membrano-cytoplasmic localization, whereas HGF is predominantly cytoplasmic. The cellular localization of Met using antibodies to the intracellular and extracellular domains was comparable (D, left and right insets, respectively). Representative histospots stained using an antibody to the intracellular domain of Met and scored as 0, 1, 2, or 3+ are shown in E–H, respectively.

Fig. 1.

Immunohistochemical staining of breast cancer tissue microarrays for four members of the HGF/Met pathway. Each histospot shows a representative positive case (3+) for matriptase (A), HAI-1 (B), HGF (C), and Met (D), at low (×40) and high (×200, insets) magnification. Matriptase, HAI-1, and Met show a membrano-cytoplasmic localization, whereas HGF is predominantly cytoplasmic. The cellular localization of Met using antibodies to the intracellular and extracellular domains was comparable (D, left and right insets, respectively). Representative histospots stained using an antibody to the intracellular domain of Met and scored as 0, 1, 2, or 3+ are shown in E–H, respectively.

Close modal
Fig. 2.

Kaplan-Meier analysis of disease-related survival demonstrates that Met (intracellular domain), matriptase, and HAI-1 show long-term prognostic benefit. Statistical significance was assessed using the log rank test.

Fig. 2.

Kaplan-Meier analysis of disease-related survival demonstrates that Met (intracellular domain), matriptase, and HAI-1 show long-term prognostic benefit. Statistical significance was assessed using the log rank test.

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

Supported by grants from the Patrick and Catherine Weldon Donaghue Foundation for Medical Research, The Connecticut Breast Cancer Alliance, and grants from the NIH, including National Institute of Environmental Health Sciences Grant K0-8 ES11571 (to R. L. C.), National Cancer Institute Grant RO-1 GM57604 (to D. L. R.), United States Army Grant DAMD 01-000436, and NIH Breast Cancer Specialized Programs of Research Excellence 2P50CA72460.

3

The abbreviations used are: HGF, hepatocyte growth factor; HAI, hepatocyte growth factor activator inhibitor; ER, estrogen receptor; PR, progesterone receptor; CI, confidence interval; uPA, urokinase-type plasminogen activator; PAI, plasminogen activator inhibitor.

4

Unpublished observations.

Table 1

Marker expressiona

MarkerLow (%)High (%)Total
Met (cytoplasmic) 229 (72) 91 (28) 320 
Met (extracellular) 251 (78) 72 (22) 323 
Matriptase 181 (55) 148 (45) 329 
HGF 176 (54) 147 (46) 323 
HAI-1 260 (82) 56 (18) 316 
ER 212 (69) 95 (31) 307 
PR 203 (68) 97 (32) 300 
HER2/neu 260 (87) 38 (13) 298 
Nuclear grade III 241 (81) 56 (19) 297 
Patient age (>50 y) 225 (68) 105 (31) 330 
Tumor size (>2 cm) 180 (45) 150 (55) 330 
MarkerLow (%)High (%)Total
Met (cytoplasmic) 229 (72) 91 (28) 320 
Met (extracellular) 251 (78) 72 (22) 323 
Matriptase 181 (55) 148 (45) 329 
HGF 176 (54) 147 (46) 323 
HAI-1 260 (82) 56 (18) 316 
ER 212 (69) 95 (31) 307 
PR 203 (68) 97 (32) 300 
HER2/neu 260 (87) 38 (13) 298 
Nuclear grade III 241 (81) 56 (19) 297 
Patient age (>50 y) 225 (68) 105 (31) 330 
Tumor size (>2 cm) 180 (45) 150 (55) 330 
a

The expression of experimental and traditional markers in a cohort of 330 node-negative breast cancer patients. Immunohistochemical stains were score on a four-point scale (0–3+), and divided into low (0–1+) and high (2+–3+) categories.

Table 2

Met pathway associations: χ2 analysisa

Met (cyto)Met (extra)MatriptaseHGF
Met (extra) <0.0001    
Matripase <0.0001 0.2588   
HGF 0.0001 0.0002 <0.0001  
HAI-I 0.3716 0.8320 0.0597 0.2902 
Met (cyto)Met (extra)MatriptaseHGF
Met (extra) <0.0001    
Matripase <0.0001 0.2588   
HGF 0.0001 0.0002 <0.0001  
HAI-I 0.3716 0.8320 0.0597 0.2902 
a

The association among expression of matriptase, HGF, HAI-1, and Met (using antibodies to both the cytoplasmic and extracellular domains) were determined using χ2 analysis. Statistically significant observations are in boldface. All significant associations are direct (i.e. high-expression of one marker correlates with high-expression of the other).

Table 3

Univariate analysisa

Marker—high expressionPRelative risk95% CI
Met (cytoplasmic) 0.0029 1.826 1.228–2.715 
Met (extracellular) 0.9771 1.007 0.639–1.585 
Matriptase 0.0279 1.527 1.047–2.226 
HGF 0.5026 1.140 0.778–1.670 
HAI-1 0.0110 1.808 1.145–2.853 
ER 0.3663 1.214 0.797–1.850 
PR 0.7881 0.945 0.627–1.424 
HER2/neu 0.4535 0.786 0.420–1.473 
Nuclear grade III 0.7993 0.934 0.555–1.574 
Patient age (>50 yr) 0.3240 1.227 0.817–1.843 
Tumor size (>2 cm) 0.0001 2.134 1.448–3.145 
Marker—high expressionPRelative risk95% CI
Met (cytoplasmic) 0.0029 1.826 1.228–2.715 
Met (extracellular) 0.9771 1.007 0.639–1.585 
Matriptase 0.0279 1.527 1.047–2.226 
HGF 0.5026 1.140 0.778–1.670 
HAI-1 0.0110 1.808 1.145–2.853 
ER 0.3663 1.214 0.797–1.850 
PR 0.7881 0.945 0.627–1.424 
HER2/neu 0.4535 0.786 0.420–1.473 
Nuclear grade III 0.7993 0.934 0.555–1.574 
Patient age (>50 yr) 0.3240 1.227 0.817–1.843 
Tumor size (>2 cm) 0.0001 2.134 1.448–3.145 
a

Univariate analysis of 30-year disease-related survival was performed using the Cox-proportional hazards model. Statistically significant observations are in boldface.

Table 4

Multivariate analysis: Met pathwaya

Marker—high expressionPRelative risk95% CI
Met (cytoplasmic) 0.0064 1.862 1.191–2.910 
Met (extracellular) 0.7710 1.079 0.645–1.805 
Matriptase 0.2413 1.290 0.843–1.974 
HGF 0.5189 1.159 0.740–1.815 
HAI-1 0.0291 1.721 1.057–2.803 
Marker—high expressionPRelative risk95% CI
Met (cytoplasmic) 0.0064 1.862 1.191–2.910 
Met (extracellular) 0.7710 1.079 0.645–1.805 
Matriptase 0.2413 1.290 0.843–1.974 
HGF 0.5189 1.159 0.740–1.815 
HAI-1 0.0291 1.721 1.057–2.803 
a

Multivariate analysis of 30-year disease-related survival was performed using the Cox-proportional hazards model using only the experimental markers. Statistically significant observations are in boldface.

Table 5

Multivariate analysis: traditional markersa

Marker—high expressionPRelative risk95% CI
Met (cytoplasmic) 0.0098 1.841 1.159–2.925 
HAI-1 0.0483 1.805 1.004–3.242 
ER 0.8552 1.597 0.568–1.597 
PR 0.9040 0.970 0.590–1.596 
HER2/neu 0.5284 0.783 0.367–1.674 
Nuclear grade III 0.7886 1.084 0.602–1.952 
Patient age (>50 yr) 0.5320 1.177 0.706–1.963 
Tumor size (>2 cm) 0.0007 2.265 1.415–3.627 
Marker—high expressionPRelative risk95% CI
Met (cytoplasmic) 0.0098 1.841 1.159–2.925 
HAI-1 0.0483 1.805 1.004–3.242 
ER 0.8552 1.597 0.568–1.597 
PR 0.9040 0.970 0.590–1.596 
HER2/neu 0.5284 0.783 0.367–1.674 
Nuclear grade III 0.7886 1.084 0.602–1.952 
Patient age (>50 yr) 0.5320 1.177 0.706–1.963 
Tumor size (>2 cm) 0.0007 2.265 1.415–3.627 
a

Multivariate analysis of 30-year disease-related survival was performed using the Cox-proportional hazards model using both the independent markers from Table 4 and traditional histopathologic measures. Statistically significant observations are in boldface.

We thank Thomas D’Aquila and Lori Charette for their help in this effort.

1
Danilkovitch-Miagkova A., Zbar B. Dysregulation of Met receptor tyrosine kinase activity in invasive tumors.
J. Clin. Investig.
,
109
:
863
-867,  
2002
.
2
Yao Y., Jin L., Fuchs A., Joseph A., Hastings H. M., Goldberg I. D., Rosen E. M. Scatter factor protein levels in human breast cancers: clinicopathological and biological correlations..
Am. J. Pathol.
,
149
:
1707
-1717,  
1996
.
3
Jin L., Fuchs A., Schnitt S. J., Yao Y., Joseph A., Lamszus K., Park M., Goldberg I. D., Rosen E. M. Expression of scatter factor and c-met receptor in benign and malignant breast tissue.
Cancer (Phila.)
,
79
:
749
-760,  
1997
.
4
Tsao M. S., Yang Y., Marcus A., Liu N., Mou L. Hepatocyte growth factor is predominantly expressed by the carcinoma cells in non-small-cell lung cancer.
Hum. Pathol.
,
32
:
57
-65,  
2001
.
5
Wang Y., Selden A. C., Morgan N., Stamp G. W., Hodgson H. J. Hepatocyte growth factor/scatter factor expression in human mammary epithelium.
Am. J. Pathol.
,
144
:
675
-682,  
1994
.
6
Lee S. L., Dickson R. B., Lin C. Y. Activation of hepatocyte growth factor and urokinase/plasminogen activator by matriptase, an epithelial membrane serine protease.
J. Biol. Chem.
,
275
:
36720
-36725,  
2000
.
7
Tanimoto H., Underwood L. J., Wang Y., Shigemasa K., Parmley T. H., O’Brien T. J. Ovarian tumor cells express a transmembrane serine protease: a potential candidate for early diagnosis and therapeutic intervention.
Tumor Biol.
,
22
:
104
-114,  
2001
.
8
Oberst M., Anders J., Xie B., Singh B., Ossandon M., Johnson M., Dickson R. B., Lin C. Y. Matriptase and HAI-1 are expressed by normal and malignant epithelial cells in vitro and in vivo.
Am. J. Pathol.
,
158
:
1301
-1311,  
2001
.
9
Lin C. Y., Anders J., Johnson M., Dickson R. B. Purification and characterization of a complex containing matriptase and a Kunitz-type serine protease inhibitor from human milk.
J. Biol. Chem.
,
274
:
18237
-18242,  
1999
.
10
Shimomura T., Denda K., Kitamura A., Kawaguchi T., Kito M., Kondo J., Kagaya S., Qin L., Takata H., Miyazawa K., Kitamura N. Hepatocyte growth factor activator inhibitor, a novel Kunitz-type serine protease inhibitor.
J. Biol. Chem.
,
272
:
6370
-6376,  
1997
.
11
Boccaccio C., Ando M., Tamagnone L., Bardelli A., Michieli P., Battistini C., Comoglio P. M. Induction of epithelial tubules by growth factor HGF depends on the STAT pathway.
Nature (Lond.)
,
391
:
285
-288,  
1998
.
12
Edakuni G., Sasatomi E., Satoh T., Tokunaga O., Miyazaki K. Expression of the hepatocyte growth factor/c-Met pathway is increased at the cancer front in breast carcinoma.
Pathol. Int.
,
51
:
172
-178,  
2001
.
13
Oberst M. D., Johnson M. D., Dickson R. B., Lin C. Y., Singh B., Stewart M., Williams A., al-Nafussi A., Smyth J. F., Gabra H., Sellar G. C. Expression of the serine protease matriptase and its inhibitor HAI-1 in epithelial ovarian cancer: correlation with clinical outcome and tumor clinicopathological parameters..
Clin. Cancer Res.
,
8
:
1101
-1107,  
2002
.
14
Ihara S., Miyoshi E., Ko J. H., Murata K., Nakahara S., Honke K., Dickson R. B., Lin C. Y., Taniguchi N. Prometastatic effect of N-acetylglucosaminyltransferase V is due to modification and stabilization of active matriptase by adding β 1–6 GlcNAc branching.
J. Biol. Chem.
,
277
:
16960
-16967,  
2002
.
15
Park W. S., Dong S. M., Kim S. Y., Na E. Y., Shin M. S., Pi J. H., Kim B. J., Bae J. H., Hong Y. K., Lee K. S., Lee S. H., Yoo N. J., Jang J. J., Pack S., Zhuang Z., Schmidt L., Zbar B., Lee J. Y. Somatic mutations in the kinase domain of the Met/hepatocyte growth factor receptor gene in childhood hepatocellular carcinomas.
Cancer Res.
,
59
:
307
-310,  
1999
.
16
Di Renzo M. F., Olivero M., Martone T., Maffe A., Maggiora P., Stefani A. D., Valente G., Giordano S., Cortesina G., Comoglio P. M. Somatic mutations of the MET oncogene are selected during metastatic spread of human HNSC carcinomas.
Oncogene
,
19
:
1547
-1555,  
2000
.
17
Moon Y. W., Weil R. J., Pack S. D., Park W. S., Pak E., Pham T., Karkera J. D., Kim H. K., Vortmeyer A. O., Fuller B. G., Zhuang Z. Missense mutation of the MET gene detected in human glioma.
Mod. Pathol.
,
13
:
973
-977,  
2000
.
18
Schmidt L., Duh F. M., Chen F., Kishida T., Glenn G., Choyke P., Scherer S. W., Zhuang Z., Lubensky I., Dean M., Allikmets R., Chidambaram A., Bergerheim U. R., Feltis J. T., Casadevall C., Zamarron A., Bernues M., Richard S., Lips C. J., Walther M. M., Tsui L. C., Geil L., Orcutt M. L., Stackhouse T., Zbar B., et al Germline and somatic mutations in the tyrosine kinase domain of the MET proto-oncogene in papillary renal carcinomas.
Nat. Genet.
,
16
:
68
-73,  
1997
.
19
Lee J. H., Han S. U., Cho H., Jennings B., Gerrard B., Dean M., Schmidt L., Zbar B., Vande Woude G. F. A novel germ line juxtamembrane Met mutation in human gastric cancer.
Oncogene
,
19
:
4947
-4953,  
2000
.
20
Cooper C. S., Park M., Blair D. G., Tainsky M. A., Huebner K., Croce C. M., Vande Woude G. F. Molecular cloning of a new transforming gene from a chemically transformed human cell line.
Nature (Lond.)
,
311
:
29
-33,  
1984
.
21
Rodrigues G. A., Park M. Dimerization mediated through a leucine zipper activates the oncogenic potential of the met receptor tyrosine kinase.
Mol. Cell. Biol.
,
13
:
6711
-6722,  
1993
.
22
Zhen Z., Giordano S., Longati P., Medico E., Campiglio M., Comoglio P. M. Structural and functional domains critical for constitutive activation of the HGF-receptor (Met).
Oncogene
,
9
:
1691
-1697,  
1994
.
23
Wallenius V., Hisaoka M., Helou K., Levan G., Mandahl N., Meis-Kindblom J. M., Kindblom L. G., Jansson J. O. Overexpression of the hepatocyte growth factor (HGF) receptor (Met) and presence of a truncated and activated intracellular HGF receptor fragment in locally aggressive/malignant human musculoskeletal tumors.
Am. J. Pathol.
,
156
:
821
-829,  
2000
.
24
Rusciano D., Lorenzoni P., Burger M. M. Constitutive activation of c-Met in liver metastatic B16 melanoma cells depends on both substrate adhesion and cell density and is regulated by a cytosolic tyrosine phosphatase activity.
J. Biol. Chem.
,
271
:
20763
-20769,  
1996
.
25
Rimm D. L., Camp R. L., Charette L. A., Olsen D. A., Provost E. Amplification of tissue by construction of tissue microarrays.
Exp. Mol. Pathol.
,
70
:
255
-264,  
2001
.
26
Kononen J., Bubendorf L., Kallioniemi A., Barlund M., Schraml P., Leighton S., Torhorst J., Mihatsch M. J., Sauter G., Kallioniemi O. P. Tissue microarrays for high-throughput molecular profiling of tumor specimens.
Nat. Med.
,
4
:
844
-847,  
1998
.
27
Camp R. L., Charette L. A., Rimm D. L. Validation of tissue microarray technology in breast carcinoma.
Lab. Investig.
,
80
:
1943
-1949,  
2000
.
28
Torhorst J., Bucher C., Kononen J., Haas P., Zuber M., Kochli O. R., Mross F., Dieterich H., Moch H., Mihatsch M., Kallioniemi O. P., Sauter G. Tissue microarrays for rapid linking of molecular changes to clinical endpoints.
Am. J. Pathol.
,
159
:
2249
-2256,  
2001
.
29
Norton A. J., Jordan S., Yeomans P. Brief, high-temperature heat denaturation (pressure cooking): a simple and effective method of antigen retrieval for routinely processed tissues..
J. Pathol.
,
173
:
371
-379,  
1994
.
30
Lin C. Y., Wang J. K., Torri J., Dou L., Sang Q. A., Dickson R. B. Characterization of a novel, membrane-bound, 80-kDa matrix-degrading protease from human breast cancer cells. Monoclonal antibody production, isolation, and localization.
J. Biol. Chem.
,
272
:
9147
-9152,  
1997
.
31
Oberst M., Lin C. Y., Dickson R. B., Johnson M. Role of Proteases in Breast Cancer.
J. Women Cancer
,
2
:
201
-216,  
2002
.
32
Look M. P., van Putten W. L., Duffy M. J., Harbeck N., Christensen I. J., Thomssen C., Kates R., Spyratos F., Ferno M., Eppenberger-Castori S., Sweep C. G., Ulm K., Peyrat J. P., Martin P. M., Magdelenat H., Brunner N., Duggan C., Lisboa B. W., Bendahl P. O., Quillien V., Daver A., Ricolleau G., Meijer-van Gelder M. E., Manders P., Fiets W. E., Blankenstein M. A., Broet P., Romain S., Daxenbichler G., Windbichler G., Cufer T., Borstnar S., Kueng W., Beex L. V., Klijn J. G., O’Higgins N., Eppenberger U., Janicke F., Schmitt M., Foekens J. A. Pooled analysis of prognostic impact of urokinase-type plasminogen activator and its inhibitor PAI-1 in 8377 breast cancer patients.
J. Natl. Cancer Inst.
,
94
:
116
-128,  
2002
.
33
Andreasen P. A., Kjoller L., Christensen L., Duffy M. J. The urokinase-type plasminogen activator system in cancer metastasis: a review..
Int. J. Cancer
,
72
:
1
-22,  
1997
.
34
Takeuchi T., Harris J. L., Huang W., Yan K. W., Coughlin S. R., Craik C. S. Cellular localization of membrane-type serine protease 1 and identification of protease-activated receptor-2 and single-chain urokinase-type plasminogen activator as substrates.
J. Biol. Chem.
,
275
:
26333
-26342,  
2000
.
35
Mars W. M., Zarnegar R., Michalopoulos G. K. Activation of hepatocyte growth factor by the plasminogen activators uPA and tPA.
Am. J. Pathol.
,
143
:
949
-958,  
1993
.
36
Kataoka H., Hamasuna R., Itoh H., Kitamura N., Koono M. Activation of hepatocyte growth factor/scatter factor in colorectal carcinoma.
Cancer Res.
,
60
:
6148
-6159,  
2000
.
37
Bajou K., Noel A., Gerard R. D., Masson V., Brunner N., Holst-Hansen C., Skobe M., Fusenig N. E., Carmeliet P., Collen D., Foidart J. M. Absence of host plasminogen activator inhibitor 1 prevents cancer invasion and vascularization.
Nat. Med.
,
4
:
923
-928,  
1998
.