Background: Matriptase, a type II transmembrane serine protease is involved in angiogenesis, degradation of extracellular matrix, and in the progression of some epithelial cancers. Here, we establish the clinical significance of matriptase and its inhibitor, hepatocyte growth factor activator inhibitor-1 (HAI-1), during the progression of human prostate cancer (CaP).

Methods: The expression patterns of matriptase and HAI-1 were determined in primary cultures of normal human prostate epithelial (NHPE) cells, human CaP cells LNCaP, DU-145, CWR22Rν1, and PC-3, and in tissue samples of 172 patients with normal prostate, benign prostatic hyperplasia (BPH), prostatic intraepithelial neoplasia (PIN), and adenocarcinoma of different tumor grades.

Results: The protein and mRNA levels of matriptase were significantly higher in all carcinoma cells as compared with NHPE cells. Conversely, all CaP cells exhibited a reduced expression of HAI-1 as compared with NHPE cells. A progressive increase in the protein levels of matriptase was observed with increasing tumor grade in CaP specimens as compared with normal and BPH tissue specimens. Tissue samples of normal prostate exhibited a high constitutive protein level of HAI-1 compared with BPH and low-grade cancer with a progressive loss with increasing tumor grade.

Conclusion: The increased expression of matriptase and loss of HAI-1 may be an important event during the progression of CaP in humans. We suggest that the ratio of these two gene products may serve as a promising biomarker for CaP progression and a potential marker for establishing the efficacy of therapeutic and chemopreventive interventions. (Cancer Epidemiol Biomarkers Prev 2006;15(2):217–27)

Prostate cancer (CaP) is the most common visceral cancer diagnosed in men; it is the second leading cause of cancer-related deaths in males in the U.S. and in the Western world (1). The lack of effective therapies for advanced CaP reflects, to a large extent, the paucity of knowledge about the molecular pathways involved in CaP development. Thus, the identification of new predictive biomarkers will be important for improving clinical management, leading to improved survival of patients with CaP. Such molecular targets, especially those that are indicative of invasiveness of the disease, will also be excellent candidate targets for staging the disease and establishing the effectiveness of therapeutic and chemopreventive intervention of CaP (2, 3).

Among the critical pathologic processes that occur during the progression and metastasis of CaP is the breakdown of extracellular matrix (ECM) and interstitial stroma (4). An array of ECM-degrading proteases have been identified and implicated in cancer invasion and metastasis (5). The serine proteases with conserved residues of histidine, aspartate, and serine constitute a novel family of transmembrane enzymes involved in numerous biological processes, including activation of growth and angiogenic factors, and in the degradation of ECM components (6). The high expression of serine and other proteases in CaP are reported to be correlated with poor survival of the patients (7, 8). Because of a high level of serine protease expression in several human solid tumors, these enzymes have recently been proposed as potential diagnostic, prognostic, and/or therapeutic targets for several human tumors (9).

Matriptase, a type II transmembrane protein also known as tumor-associated differentially expressed gene-15 (TADG-15), is a serine protease and has been detected in various tissues of epithelial origin along with many cell lines in vitro and has been characterized as an activator of tumor cell invasion and metastases (10, 11). Matriptase has been shown to be overexpressed during tumor growth, invasion, and progression of breast, cervical, and ovarian cancer of humans (12-18). Virtually nothing is known about the function of matriptase in the prostate, with the exception of a study that identified a serine protease, MT-SP1, identical to matriptase in the prostatic cell line PC-3 (19). In addition, a recent study has shown that the specific inhibition of matriptase in CaP cells inhibits tumorigenesis in an athymic nude mouse model (20). The hepatocyte growth factor activator inhibitor-1 (HAI-1), a Kunitz-type serine protease inhibitor found predominantly in the columnar epithelium of many tissue types including breast, stomach, lung, kidney, prostate, and uterus, has been shown to modulate the activity of matriptase in several cancer types (10, 12, 21-24). HAI-1 has been shown to suppress the growth and motility of carcinoma cells by inhibiting the matriptase activity (12, 13).

In the current study, we provide evidence for the first time for overexpression of matriptase with concomitant loss of HAI-1 in human CaP cells and tissue specimens and show that this correlates with the clinical stages of human CaP. We propose a role for matriptase and HAI-1 protein in CaP development and suggest their potential use as a biomarker in the clinical management of CaP.

Primary Cell Culture

Prostate tissues were obtained under an approved Institutional Review Board protocol from men (ages 44-66) undergoing cystoprostatectomy for bladder cancer at the University of Wisconsin Hospital and Clinics. Histology confirmed that no bladder or prostate cancer was present in the prostate tissue harvested for our studies. Prostate epithelial cultures were established, as described previously (25). Normal human prostate epithelial (NHPE) cells were maintained in Ham's F-12 media (Invitrogen, Carlsbad, CA) supplemented with 0.25 units/mL regular insulin, 1 μg/mL hydrocortisone, 5 μg/mL human transferrin, 2.7 mg/mL dextrose, 0.1 mmol/L nonessential amino acids, 100 units/mL penicillin, 100 μg/mL streptomycin, 2 mmol/L l-glutamine, 10 ng/mL cholera toxin, 25 μg/mL bovine pituitary extract, and 1% fetal bovine serum.

Cell Lines

PC-3, CWR22Rν1, DU-145, and LNCaP cancer cells were obtained from American Type Culture Collection (Manassas, VA). PC-3, CWR22Rν1, and LNCaP cancer cells were cultured in RPMI 1640 (American Type Culture Collection), whereas DU-145 cells were grown in EMEM. The cells were cultured in a humidified atmosphere of 95% air and 5% CO2 in an incubator at 37°C.

Monoclonal Antibodies

Anti-matriptase (M32) and anti-HAI-1(M19) monoclonal antibodies were developed in the Lombardi Comprehensive Cancer Center, Georgetown University Medical Center, Washington, DC. The development and specificity of monoclonal antibodies has been previously described (11, 26).

Immunohistochemistry

Paraffin-embedded sections of human prostate tissues of 172 patients with normal, benign prostate hyperplasia (BPH), PIN, and adenocarcinoma were obtained from the Department of Pathology, University of Wisconsin Hospital, Madison, WI and Cybrdi Corporation (Gaithersburg, MD). Immunohistochemical staining was done using an automated Benchmark immunostainer (Ventana Medical Systems, Tucson, AZ). After antigen retrieval, thick-paraffin-embedded sections (5 μm) were dewaxed, rehydrated, and endogenous peroxidase activity was blocked. Sections were washed in water and PBS and were blocked in blocking buffer (2% goat serum/5% bovine serum albumin in PBS) for 30 minutes followed by incubation with primary antibody of matriptase-specific monoclonal antibody M32 (IgG1), and HAI-1-specific monoclonal antibody clone M19 (IgG1), at the dilution of 1:50 for 1 hour at room temperature. A negative control was included, in which prostate tissues were incubated with normal mouse IgG1 replacing the anti-HAI-1 or matriptase monoclonal antibody. In addition, we also included positive and negative tissue controls using the breast carcinoma specimens to show the specificity of the antibodies and quality of the immunohistochemical staining in our study. After incubation in the primary antibody, sections were washed twice in PBS to remove unbound antibody, followed by incubation for 2 hours at room temperature with appropriate horseradish peroxidase–conjugated secondary antibody. Immunoreactive complexes were detected using 3,3′-diaminobenzidene (Dako Corp., Carpinteria, CA). Slides were then counterstained in hematoxylin, mounted in crystal mount, and coverslipped in 50:50 xylene/Permount. Sections were visualized on a Zeiss-Axiophot DM HT microscope. Images were captured with an attached camera linked to a computer.

Histopathologic Grading of Prostate Cancer Specimens

The Gleason system and the WHO grading system were used for evaluation of levels of matriptase and HAI-1 in the prostatic tissue by immunoperoxidase staining. Prostate adenocarcinoma was first graded in Gleason patterns 1, 2, 3, 4 and 5. The intensity of immunoperoxidase staining for matriptase and HAI-1 was scored as 0 (negative), 1 (weak), 2 (moderate), and 3 (strong) based on the Gleason patterns. The primary Gleason pattern and the secondary Gleason pattern were added to arrive at a Gleason score, ranging from 2 (1 + 1) to 10 (5 + 5). The Gleason scoring system was then converted to the WHO grading system as grade 1 (well-differentiated prostatic adenocarcinoma) corresponding to Gleason score 5 to 6, and grade 3 (poorly differentiated or undifferentiated adenocarcinoma) corresponding to Gleason score 7 to 10.

Western Blot Analysis

Cell lysates were prepared in cold lysis buffer [0.05 mmol/L Tris-HCl, 0.15 mmol/L NaCl, 1 mol/L EGTA, 1 mol/L EDTA, 20 mmol/L NaF, 100 mmol/L Na3VO4, 0.5% NP40, 1% Triton X-100, 1 mol/L phenylmethylsulfonyl flouride (pH 7.4)] with freshly added protease inhibitor cocktail (Protease Inhibitor Cocktail Set III; Calbiochem, La Jolla, CA). The lysate was collected, cleared by centrifugation, and the supernatant aliquoted and stored at −80°C. The protein content in the lysates was measured by bicinchoninic acid protein assay (Pierce, Rockford, IL), as per the manufacturers' protocol. For Western blot analysis, 40 μg protein was resolved over 12% Tris-glycine polyacrylamide gels (Novex, Carlsbad, CA) under nonreduced conditions, transferred onto nitrocellulose membranes and subsequently incubated in blocking buffer [5% nonfat dry milk/1% Tween 20; in 20 mmol/L TBS (pH 7.6)] for 2 hours. The blots were incubated with the appropriate primary (human reactive matriptase and HAI-1), washed and incubated with appropriate secondary horseradish peroxidase–conjugated antibody (Amersham Biosciences, Piscataway, NJ). The blots were detected with chemiluminescence (ECL kit, Amersham Biosciences) and autoradiography, using XAR-5 film (Eastman Kodak, Rochester, NY). Equal loading of protein was confirmed by stripping the blots and reprobing with β-actin (Sigma, St. Louis, MO). Semiquantitative RT-PCR of matriptase and HAI-1: total RNA was isolated from cell lines using the RNeasy kit (Qiagen, Valencia, CA) and was checked for its purity and concentration. Two micrograms of total RNA was reverse-transcribed using Moloney murine leukemia virus–reverse transcriptase and oligo dT primer from Promega (Madison, WI) for cDNA synthesis. Two microliters of the cDNA were subjected to PCR in PCR buffer [20 mmol/L Tris-HCl (pH 8.4) containing 50 mmol/L KCl], 1.5 mmol/L MgCl2, 0.2 mmol/L deoxynucleotide triphosphate, 2.5 units of Platinum Taq DNA polymerase and 0.2 μmol/L of each primer for matriptase (forward, 5′-TCATCGCCTACTACTGGTCTGAGT-3′; reverse, 5′-GATCTTGTTGCTGTTGCTGGTGAC-3′) and HAI-1 (forward, 5′-AAGACTACTGCCTCGCATCCAACA-3′, reverse, 5′-AGCTCGTCAAAGCCACTCGT-GTAT-3′) obtained from Integrated DNA Technologies, Inc. (Coralville, IA) in a total volume of 25 μL. A constitutively expressed GAPDH gene (forward, 5′-AATCCCATCACCATCTTCCAGGAG-3′; reverse, 5-GCATTGCTGATGATCTTGAGGAGG-CTG-3′) was coamplified to confirm the equal loading of the cDNA. Fifteen-microliter aliquots of PCR products were electrophoresed on 1.5% agarose gel and visualized by ethidium bromide staining under UV transilluminator.

Statistical Analysis

All measures were summarized as means ± SE. Graphical summaries of the distribution of measures were made using box plots. Measures were examined for the appropriateness of a normality assumption by density estimation (data not shown). Associations of categorical variables were evaluated using Fisher's exact test. Sample correlations were estimated using Spearman's rank correlation. All tests were two-sided and conducted at the α = 0.05 significance level. All statistical analyses were done with the S-plus, Professional Version 6.2 software (Insightful Corp., Seattle, WA).

Matriptase and HAI-1 Expression in NHPE and CaP Cells

Matriptase and HAI-1 protein expression in NHPE and CaP cells. As an attempt towards identifying the expression of matriptase and HAI-1 in CaP progression, we first measured protein expression levels by immunoblot analysis in several human prostate carcinoma cell lines, LNCaP, CWR22Rν1, DU-145, and PC-3, and compared them to NHPE cells. Among the four cell lines used, LNCaP and CWR22Rν1 are androgen-sensitive, whereas DU-145 and PC-3 are androgen-independent. The choice of these cells was based on the fact that 80% of patients with CaP present with androgen-dependent disease at the time of diagnosis, which later transforms into more aggressive, androgen-independent disease (27). As shown in Fig. 1A, all CaP cell lines exhibited a higher expression of matriptase protein than in NHPE. When the protein expression of matriptase was compared among the four cancer lines, based on the densitometric analysis of the immunoblots, highly aggressive PC-3 cells exhibited 2.5-fold (P < 0.001) higher expression than in LNCaP, CWR22Rν1, and DU-145 cells. These data suggest a possibility that expression of matriptase protein may be correlated with disease progression and may play a role in aggressiveness of human CaP. Conversely, immunoblot analysis for HAI-1 protein exhibited moderate constitutive expression in NHPE and LNCaP cells, and weak expression in CWR22Rν1, DU-145, and PC-3 cells (Fig. 1B). Densitometric analysis of immunoblots revealed a 50% (P < 0.05) lower HAI-1 protein expression in PC-3 cells compared with that in NHPE cells. When the expression level of HAI-1 in different CaP cell lines was compared, highly aggressive PC-3 cells exhibited weaker expression than DU-145, LNCaP, and CWR22Rν1 cells (Fig. 1B).

Figure 1.

Expression of (A) matriptase and (B) HAI-1 protein in NHPE and prostate cancer cells LNCaP, CWR22Rν1, DU145, and PC-3 by Western blotting. Equal loading of protein was confirmed by stripping the blots and reprobing with β-actin antibody. The histogram indicates the relative density of the bands normalized to β-actin. Columns, mean of relative densities; bars, ± SE; NS, nonsignificant.

Figure 1.

Expression of (A) matriptase and (B) HAI-1 protein in NHPE and prostate cancer cells LNCaP, CWR22Rν1, DU145, and PC-3 by Western blotting. Equal loading of protein was confirmed by stripping the blots and reprobing with β-actin antibody. The histogram indicates the relative density of the bands normalized to β-actin. Columns, mean of relative densities; bars, ± SE; NS, nonsignificant.

Close modal

Matriptase and HAI-1 mRNA expression in NHPE and CaP cells. We next determined the mRNA expression levels for matriptase and HAI-1 by RT-PCR analysis in NHPE, LNCaP, CWR22Rν1, DU-145, and PC-3 cells. The observations of RT-PCR analysis for matriptase and HAI-1 proteins were consistent with the results of immunoblot analysis. As shown in Fig. 2A, all CaP cell lines exhibited higher mRNA expression of matriptase than in NHPE cells. Furthermore, expression of matriptase mRNA showed a pattern similar to the expression of matriptase protein, with highly aggressive PC-3 cells showing higher expression than LNCaP, CWR22Rν1, and DU-145 cells. Next, we examined the levels of HAI-1 mRNA expression in NHPE, PC-3, LNCaP, DU-145, and CWR22Rν1 cells. As evident from the densitometric analysis of bands, all CaP cell lines showed a significantly lower mRNA expression of HAI-1 as compared with that in NHPE cells (Fig. 2B). When expression levels were compared within CaP cell lines, PC-3 cells exhibited significantly lower (50%; P < 0.05) mRNA expression of HAI-1 as compared with LNCaP cells (Fig. 2B).

Figure 2.

Expression of (A) matriptase and (B) HAI-1 mRNA in NHPE and prostate cancer cells LNCaP, CWR22Rν1, DU145, and PC-3 by RT-PCR. Equal loading of RT-PCR product was confirmed by GAPDH as an internal control. The histogram indicates the relative density of bands normalized to GAPDH. Columns, mean of relative densities; bars, ± SE; NS, nonsignificant.

Figure 2.

Expression of (A) matriptase and (B) HAI-1 mRNA in NHPE and prostate cancer cells LNCaP, CWR22Rν1, DU145, and PC-3 by RT-PCR. Equal loading of RT-PCR product was confirmed by GAPDH as an internal control. The histogram indicates the relative density of bands normalized to GAPDH. Columns, mean of relative densities; bars, ± SE; NS, nonsignificant.

Close modal

Immunohistochemical Analysis of Matriptase and HAI-1 Protein in Normal, BPH, PIN, and CaP Specimens

In the next series of experiments, we used immunoperoxidase to determine matriptase and HAI-1 protein expression in specimens of normal, BPH, PIN, and CaP representing all tumor grades. These specimens were obtained as tissue array slides from Cybrdi Corporation. In the first experiment, a total of 88 samples were obtained out of which 48 were evaluated for matriptase expression and 40 samples were evaluated for HAI-1 expression. Breast carcinoma specimens were used as positive controls to show the specificity of the antibodies and the quality of the immunohistochemical staining. The breast carcinoma specimens showed that matriptase expression was present in the ductal and lobular epithelial cells, whereas the stroma was negative for matriptase (Fig. 3). These results are consistent with the results in the literature (12). The staining intensity in tissue specimens were scored on a scale of 0 to 3.

Figure 3.

Immunostaining for matriptase and HAI-1 in representative specimens of positive and negative controls. A negative control in which CaP tissues were incubated with normal mouse IgG replacing the anti-HIA-1 or matriptase monoclonal antibody (top). Positive controls for matriptase (middle) and HAI-1 (bottom) using the breast carcinoma specimens to show the specificity of the antibodies and quality of the immunohistochemical staining. The breast cancer specimens showed that both HAI-1 and matriptase were only present in the ductal and lobular epithelial cells, whereas the stroma was negative for both markers. Arrows, staining for matriptase and HAI-1 staining in the ductal and lobular epithelial cells. Original magnification, ×10; insets, ×40.

Figure 3.

Immunostaining for matriptase and HAI-1 in representative specimens of positive and negative controls. A negative control in which CaP tissues were incubated with normal mouse IgG replacing the anti-HIA-1 or matriptase monoclonal antibody (top). Positive controls for matriptase (middle) and HAI-1 (bottom) using the breast carcinoma specimens to show the specificity of the antibodies and quality of the immunohistochemical staining. The breast cancer specimens showed that both HAI-1 and matriptase were only present in the ductal and lobular epithelial cells, whereas the stroma was negative for both markers. Arrows, staining for matriptase and HAI-1 staining in the ductal and lobular epithelial cells. Original magnification, ×10; insets, ×40.

Close modal

Scoring pattern for matriptase in tissue specimens. Based on the scoring patterns, a significant difference was observed in matriptase protein expression between cancer tissues and normal tissues (Table 1). The staining for matriptase protein was moderate to strong in CaP specimens as compared with normal prostate specimens which exhibited either none or weak staining (Table 1). In a total of 12 specimens of normal (n = 6) and BPH (n = 6), the staining for matriptase was weak in 1 (8%), moderate in 3 (25%) and negative in the remaining 8 (67%) specimens (Table 1). Staining for matriptase protein in PIN specimens (n = 4) showed moderate staining in 1 (25%) specimen and weak staining in the remaining 3 (75%) specimens (Table 1). In BPH and PIN specimens, staining for matriptase was either weak or moderate as compared with normal prostate, suggesting a possible role of matriptase during progression of CaP because BPH and PIN conditions are considered as high-risk factors for CaP development. Staining for matriptase in stromal cells was either occasional or negative in normal as well as cancer specimens.

Table 1.

Expression of matriptase and HAI-1 in normal human prostate, BPH, PIN, and adenocarcinoma specimens

Specimen typeMatriptase staining intensity
HAI-1 staining intensity
TotalNoneWeakModerateStrongPTotalNoneWeakModerateStrongP
Total 48 11 20 11  40 13 14 12  
Normal 0.05* 0.05* 
BPH  16  
PIN 0.05*  
Carcinoma (all grades) 32 16 11 0.0015* 13 0.05* 
Grade 1   
Grade 2 10   
Grade 2 20 0.01*  
Specimen typeMatriptase staining intensity
HAI-1 staining intensity
TotalNoneWeakModerateStrongPTotalNoneWeakModerateStrongP
Total 48 11 20 11  40 13 14 12  
Normal 0.05* 0.05* 
BPH  16  
PIN 0.05*  
Carcinoma (all grades) 32 16 11 0.0015* 13 0.05* 
Grade 1   
Grade 2 10   
Grade 2 20 0.01*  

NOTE: The expression of matriptase and HAI-1 was evaluated as staining of the tissue as none (0), weak (1), moderate (2), and strong (3). Fisher exact test was used to examine the association between staining intensity and tissue type or staining intensity and tumor grade (for tumor specimen only).

*

P < 0.05 was considered significant.

Scoring pattern for HAI-1 in tissue specimens. Based on the scoring patterns, a significant difference in HAI-1 protein expression was also observed between normal prostate and CaP specimens (Table 1). Among the total sample size of 22 (normal, n = 6; BPH, n = 16) specimens, 19 (86%) specimens exhibited weak to moderate staining for HAI-1 protein, whereas 3 (14%) specimens did not exhibit any HAI-1 protein staining (Table 1). The staining for HAI-1 protein in PIN specimens (n = 5) showed weak staining in 2 (40%) and no staining in the remaining 3 (60%) specimens (Table 1). The staining for HAI-1 protein was generally moderate in normal prostate specimens and weak to moderate in BPH specimens (Table 1). However, in sharp contrast, CaP specimens exhibited weak or no staining suggesting a loss of HAI-1 protein during progression of CaP (Table 1).

Scoring pattern of matriptase and HAI-1 in pair-matched CaP specimens. Because we observed a significant difference in staining pattern for matriptase and HAI-1 protein between normal and CaP specimens, we next asked whether the overexpression of matriptase and loss of its inhibitor HAI-1 protein, was a general phenomenon that occurs during the progression of CaP. To achieve this objective, we examined the staining pattern for matriptase and HAI-1 proteins in 42 pair-matched specimens. These samples were obtained as tissue array slide from CaP patients at University Hospitals (University of Wisconsin-Madison). As is evident from the data, the scoring pattern for matriptase protein was moderate to strong in all grades of CaP specimens, being mostly moderate in grades 1 to 2 carcinomas and generally strong in grade 3 carcinomas (Table 2). In sharp contrast, the scoring pattern for HAI-1 staining in pair-matched samples was weak to moderate in all grades of CaP specimens being generally moderate in grades 1 to 2 carcinomas and mostly weak in grade 3 carcinomas (Table 2).

Table 2.

Expression of matriptase and HAI-1 in pair-matched human prostate adenocarcinoma specimens

Specimen TypeMatriptase staining intensity
HAI-1 staining intensity
TotalNoneWeakModerateStrongPTotalNoneWeakModerateStrongP
Carcinoma (all grades) 42 20 19 0.0001* 42 17 15 0.05* 
Grade 1   
Grade 2 14 13 0.001* 14  
Grade 3 24 17 0.001* 24 16  
Specimen TypeMatriptase staining intensity
HAI-1 staining intensity
TotalNoneWeakModerateStrongPTotalNoneWeakModerateStrongP
Carcinoma (all grades) 42 20 19 0.0001* 42 17 15 0.05* 
Grade 1   
Grade 2 14 13 0.001* 14  
Grade 3 24 17 0.001* 24 16  

NOTE: The expression of matriptase and HAI-1 was evaluated as staining of the tissue as none (0), weak (1), moderate (2), and strong (3). Fisher exact test was used to examine the association between staining intensity and tumor grade.

*

P < 0.05 was considered significant.

Average scoring pattern of matriptase in CaP specimens. The box plots of the cumulative data (48 specimens + 42 pair-matched specimens) for matriptase protein expression exhibited a wide interspecimen variation in cancer specimens, compared with the normal tissues and revealed a significant difference in the level of protein expression between normal and CaP tissues (Fig. 4A). The average score for matriptase protein in normal prostate tissues was 0.25 ± 0.02 (mean ± SE; n = 6) and was significantly higher, 2.5 ± 0.2 (mean ± SE; n = 44) in high-grade cancer specimens (grade 3), a 5.5-fold increase (P < 0.001; Fig. 4A). Although the staining for matriptase in BPH specimens was higher than normal, it was significantly lower (0.70 ± 0.05; mean ± SE; n = 6, P < 0.001), compared with CaP tissues.

Figure 4.

Box plots for matriptase (A) and HAI-1 (B) protein based on staining pattern in normal, BPH, PIN, and prostate cancer specimens of tumor grades 1 to 3. The intensity of staining was graded semiquantitatively by assigning a score to each tissue specimen. *, P < 0.05 compared with normal; black bar in gray box, median values.

Figure 4.

Box plots for matriptase (A) and HAI-1 (B) protein based on staining pattern in normal, BPH, PIN, and prostate cancer specimens of tumor grades 1 to 3. The intensity of staining was graded semiquantitatively by assigning a score to each tissue specimen. *, P < 0.05 compared with normal; black bar in gray box, median values.

Close modal

Average scoring pattern for HAI-1 protein in CaP specimens. The box plots (cumulative data of 40 specimens + 42 pair-matched specimens) for HAI-1 protein indicated a wide interindividual variation in cancer specimens compared with normal tissue (Fig. 4B). The average scoring pattern for HAI-1 protein was 2.8 ± 0.15 (mean ± SE; n = 6) in normal and 1.8 ± 0.08 (mean ± SE; n = 16) in BPH specimens, however, the scoring pattern was significantly reduced to 1.0 ± 0.05 (mean ± SE; n = 27) in grade 3 cancer specimens (P < 0.05). These data show that prostate tissue significantly loses the inhibitory protein HAI-1 with the onset of CaP and is almost completely lost at the highly advanced stages of the disease (Fig. 4B).

Staining pattern of matriptase protein.Figure 5A and B shows the immunoreactive matriptase protein in a coarsely granular pattern in the cytoplasm of epithelial cells of normal, BPH, PIN, and of grade 1 (Gleason pattern 2, Gleason scores 5-6) to grade 3 (Gleason pattern 4, Gleason scores 7-10) prostatic adenocarcinoma. The staining for matriptase was highly positive in epithelial cells of CaP specimens (cumulative sample size) and was negative in epithelial cells of normal prostate specimens. A cumulative analysis of all CaP specimens (n = 74) suggested higher levels of matriptase expression, with moderate to strong staining in 66 (90%) specimens, weak staining in 3 (4%) specimens, and no staining in 5 (6%) specimens (Tables 1 and 2). Because PIN has been identified as the most significant risk factor for CaP development and the expression of matriptase protein was found to be increased in both PIN and CaP specimens, a strong link could be suggested between the expression of matriptase protein and the development of human CaP.

Figure 5.

Immunostaining for matriptase in representative specimens of (A) normal, BPH, PIN (B) CaP specimens of tumor grades 1 to 3, and (C) in specimens with adjacent normal and cancerous regions. CaP specimens were assigned tumor grades on the basis of Gleason pattern and Gleason score as described in Materials and Methods. Immunoreactive matriptase protein was observed in a coarsely granular pattern in cell cytoplasms of epithelial cells of BPH and of grade 1, grade 2 and grade 3 prostatic adenocarcinoma. There was minimal staining of occasional stromal cells. Matriptase expression was weak in normal and moderate to strong in BPH, PIN and advanced CaP specimens. , staining for matriptase in cancer regions; , staining for matriptase in normal regions. Original magnification, ×10; insets, ×40.

Figure 5.

Immunostaining for matriptase in representative specimens of (A) normal, BPH, PIN (B) CaP specimens of tumor grades 1 to 3, and (C) in specimens with adjacent normal and cancerous regions. CaP specimens were assigned tumor grades on the basis of Gleason pattern and Gleason score as described in Materials and Methods. Immunoreactive matriptase protein was observed in a coarsely granular pattern in cell cytoplasms of epithelial cells of BPH and of grade 1, grade 2 and grade 3 prostatic adenocarcinoma. There was minimal staining of occasional stromal cells. Matriptase expression was weak in normal and moderate to strong in BPH, PIN and advanced CaP specimens. , staining for matriptase in cancer regions; , staining for matriptase in normal regions. Original magnification, ×10; insets, ×40.

Close modal

The staining pattern of matriptase protein was further compared in grade 1, grade 2, and grade 3 CaP specimens. The mean matriptase expression was 1.75 ± 0.15 (mean ± SE; n = 6) in grade 1 specimens, 2.0 ± 0.20 (mean ± SE; n = 24) in grade 2 specimens, and 2.7 ± 0.15 (mean ± SE; n = 44) in grade 3 specimens indicating a progressive increase of matriptase as the disease progresses from lower to higher stages (Fig. 5B). In specimens belonging to grade 1 and grade 2 CaP (n = 30), 27 specimens exhibited moderate to strong staining for matriptase protein whereas no staining was observed in 3 specimens (Tables 1 and 2). Similarly, in grade 3 CaP specimens (n = 44), 39 (89%) exhibited moderate to strong staining for matriptase protein, 3 (7%) exhibited weak staining, and the remaining 2 (4%) specimens showed no staining (Tables 1 and 2). These data show a progressive increase of matriptase protein expression corresponding with increasing tumor grade in human CaP.

Next, we evaluated the staining pattern of matriptase protein in the tissues which had normal regions and cancer regions adjacent to each other in the same specimen. In these tissues, the scoring pattern for matriptase in the normal region was 0, whereas it was significantly higher (2-3) in cancer regions (Fig. 5C). In these tissues, matriptase protein expression was undetectable in normal regions adjacent to cancer regions, whereas its expression was significantly high in all grades of cancer of all specimens (Fig. 5C).

Staining pattern of HAI-1 protein. Normal glandular epithelial cells were generally negative for HAI-1 protein (with some occasional epithelial staining); however, intense cytoplasmic staining for HAI-1 was observed in stromal cells in these specimens (Fig. 6A). The normal stromal cells which were found to be positive for HAI-1 staining are most likely to be smooth muscle cells. Furthermore, staining for HAI-1 was found to be positive in the stromal cell cytoplasm of BPH specimens and less staining was observed in PIN specimens (Fig. 6A). Among grade 1 specimens (Gleason pattern 2, Gleason scores 2-4), some specimens exhibited a minimal apical staining for HAI-1 in stromal cells, however, all specimens belonging to grade 2 (Gleason pattern 3, Gleason scores 5-6) and grade 3 (Gleason pattern 4, Gleason scores 7-10) category, exhibited a very low cytoplasmic staining in stromal cells.

Figure 6.

Immunostaining for HAI-1 protein in representative specimens of normal (A), BPH, PIN, CaP of Gleason patterns 2 to 4 (B), and in specimens with adjacent normal and cancerous regions (C). CaP specimens were assigned tumor grades on the basis of Gleason pattern and Gleason score as described in Materials and Methods. Staining for HAI-1 was positive in the cytoplasm of the stromal cells of normal and BPH but was progressively lost in CaP specimens. , cancerous region; , normal regions. HAI-1 protein expression was lost in both cancer as well as adjacent normal regions in CaP specimens of Gleason pattern 4. Original magnification, ×10; insets, ×40.

Figure 6.

Immunostaining for HAI-1 protein in representative specimens of normal (A), BPH, PIN, CaP of Gleason patterns 2 to 4 (B), and in specimens with adjacent normal and cancerous regions (C). CaP specimens were assigned tumor grades on the basis of Gleason pattern and Gleason score as described in Materials and Methods. Staining for HAI-1 was positive in the cytoplasm of the stromal cells of normal and BPH but was progressively lost in CaP specimens. , cancerous region; , normal regions. HAI-1 protein expression was lost in both cancer as well as adjacent normal regions in CaP specimens of Gleason pattern 4. Original magnification, ×10; insets, ×40.

Close modal

Figure 6A and B shows the HAI-1 protein staining in normal, BPH, PIN, and CaP specimens. The staining for HAI-1 protein in CaP tissues (cumulative sample size; Tables 1 and 2) was strong in 2 (3%), moderate in 21 (26%), weak in 29 (35%), and negative in 30 (36%) specimens (Tables 1 and 2). The staining pattern of HAI-1 protein was further compared in grade 1, grade 2, and grade 3 CaP specimens (Fig. 6B). Contrary to matriptase expression, a progressive loss of HAI-1 protein was observed in CaP specimens with increasing tumor grade (Table 1). Analysis of the cumulative samples (Tables 1 and 2) suggested that in CaP specimens belonging to grade 1 (n = 6) and grade 2 (n = 22), the staining for HAI-1 protein was strong in 2 (4%), moderate in 9 (16%), weak in 11 (20%), and the remaining 6 (11%) specimens did not exhibit staining. In grade 3 CaP specimens (n = 27) moderate staining was observed in 1 (2%), weak in 8 (14%) and the remaining 18 (33%) specimens did not exhibit staining (Tables 1 and 2).

Next, we evaluated the staining pattern of HAI-1 protein in CaP tissues with cancerous as well as adjacent normal regions (Fig. 6C). Interestingly, HAI-1 protein expression was lost in both cancer as well as adjacent normal regions in CaP specimens of Gleason pattern 4 (Fig. 6C).

Statistical Analysis

We next examined the association between matriptase and HAI-1 protein expression in BPH and CaP tissues. The Spearman correlation between matriptase and HAI-1 was found to be −0.46 (P < 0.001) based on total specimens (Tables 1 and 2). This analysis indicates an inverse correlation between matriptase and HAI-1 protein expression during the progression of human CaP.

One important consideration in the intervention and prevention of CaP is the development of surrogate end point biomarker(s) that can be correlated with the staging of disease along the course of tumor development. The expression levels of matriptase and HAI-1 genes, which have been proposed to be involved in tumor development and metastases for some cancers, were examined in normal and prostate carcinoma cell lines along with tissue specimens of normal, BPH, PIN, and malignant prostate tissues. The strong expression of matriptase in prostate carcinoma cell lines and in malignant tissues of the human prostate, as compared with a very low constitutive level in normal prostate cells and tissues, indicate that this gene product may be involved in the progression of human prostate adenocarcinoma. In contrast, HAI-1 protein expression was significantly lower in highly aggressive prostate carcinoma PC-3 cells, as compared with LNCaP and CWR22Rν1 cells. In addition, the expression of matriptase inhibitor HAI-1 protein was highly diminished or absent in the high-grade malignant prostate tissues, as compared with normal prostate tissues, and a progressive loss of HAI-1 expression was observed during the progression of disease from low-grade to high-grade CaP. The important finding from our study is that the expression of matriptase protein increases significantly with a concomitant loss of its inhibitor HAI-1 with progressive stages of CaP development in humans. This observation suggests that the matriptase/HAI-1 complex could be employed as a predictive biomarker in human CaP development.

Metastatic spread of CaP to distant sites is regarded as the major cause of CaP-related deaths in humans (4). Tumor cells acquire this increased invasive potential by a complex pathway, which include decreased cell substrate attachment and cell-cell adhesion, as well as increased cell motility (4). To successfully complete the complex invasion and metastatic process, degradation of the ECM including the basement membrane and interstitial stroma, is required for tumor cells to migrate through anatomic barriers and to invade distant tissues (4, 28). In this process, the proteolysis of ECM is thought to be one of the most critical steps. High levels of proteolytic activities have been implicated in neoplastic progression (11, 29). Serine proteases have been implicated in the degradation of ECM in various types of cancers and have been proposed as potential diagnostic and therapeutic targets for human tumors (5, 6, 30). It is known that two members of the serine protease family, prostate-specific antigen and human kallikrein 2, serve as important prostate carcinoma biomarkers (6, 31).

Matriptase is a trypsin-like type II membrane serine protease that was first isolated from human breast milk (10, 32). It has a multidomain structure, containing a putative amino-terminal transmembrane region, a sperm protein, enterokinase, an agrin (SEA) domain, two complement subcomponents (C1r/C1s), urchin embryonic growth factor, and bone morphogenic protein (CUB) domains, four low-density lipoprotein receptor class A repeats, and a carboxyl-terminal serine protease domain (11, 13). The human gene for matriptase is located within chromosome 11q24-25 and is likely to be the human homologue of the mouse serine protease epithin, based on its high degree of sequence identity and chromosomal location (12, 33).

Matriptase has a widespread expression in normal epithelial tissues including the skin, thymic stroma, gastrointestinal tract, kidney, lung, seminal vesicle, and mammary gland (19). High levels of matriptase have been reported in a variety of carcinomas such as ovarian, cervical, breast, and gastric cancer (12, 21). It has been shown to be a potential activator of key molecules associated with tumor invasion and metastasis (16). For example, when activated by matriptase, urokinase plasminogen activator and the protease-activated receptor-2, leads to the degradation of the ECM and activation of other protease systems involved in the spread of cancer cells (16, 34). Matriptase has also been reported to convert hepatocyte growth factor/scatter factor to its active form, resulting in the scattering of tumor cells (35). Therefore, matriptase has been suggested to play an important role in the growth and/or invasion of human cancers via its ability to activate pro-urokinase plasminogen activator and pro-hepatocyte growth factor. In the current study, moderate to strong expression of matriptase was observed in cancer tissue, which increased with tumor grade.

The activity of matriptase has been reported to be modulated by HAI-1, a Kunitz-type inhibitor (21). HAI-1 has been shown to colocalize with matriptase and prevent the undesired proteolysis in breast cancer cells (11, 32). It has been shown that HAI-1 protein expression in human primary colorectal carcinomas is decreased significantly in cells within colon carcinomas relative to adjacent normal mucosa or adenomas (35). Recently, the loss of HAI-1 expression has been implicated in the progression of renal cell and colorectal carcinoma in human patients (22, 36). Earlier studies have implicated HAI-1 protein in the pathogenesis of human hepatocellular carcinoma (22). It has been reported that the invasive nature of several cancer cell lines of breast, prostate, colon, bladder, liver, lung and pancreas is correlated to the degree of HAI-1 protein expression (10, 12, 21, 22). It is important to emphasize here that in the current study, diminution of HAI-1 expression was observed in CaP specimens of all grades. Although HAI-1 protein and mRNA expression was observed in NHPE cells, its expression was observed to be mostly strong in stromal cells than in epithelial cells in CaP specimens. Additional studies with a large sample size of various tumor grades are required to establish a significant inverse correlation between matriptase and HAI-1 protein expression during CaP progression. Similarly, further studies are warranted to validate the role of ratio of matriptase to HAI-1 as a predictive potential predictive indicator for human CaP development.

In conclusion, this study indicates that the enhancement of matriptase and diminution of HAI-1 expression during the progressive stages of CaP may be associated with increasing biological aggressiveness of the disease. Thus, the combined analysis of matriptase and HAI-1 in tissue biopsies may serve as a prognostic marker in cases of prostate adenocarcinoma and may ultimately lead to monitoring therapeutic response during CaP treatment protocols and for establishing efficacy of CaP chemopreventive regimens.

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
Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005.
CA Cancer J Clin
2005
;
55
:
10
–30.
2
Saleem M, Adhami VM, Siddiqui IA, Mukhtar H. Tea beverage in chemoprevention of prostate cancer: a mini-review.
Nutr Cancer
2003
;
47
:
13
–23.
3
Klein EA, Thompson IM. Update on chemoprevention of prostate cancer.
Curr Opin Urol
2004
;
14
:
143
–9.
4
Stewart DA, Cooper CR, Sikes RA. Changes in extracellular matrix (ECM) and ECM-associated proteins in the metastatic progression of prostate cancer.
Reprod Biol Endocrinol
2004
;
2
:
2
.
5
Hooper JD, Clements JA, Quigley JP, Antalis TM. Type II transmembrane serine proteases. Insights into an emerging class of cell surface proteolytic enzymes.
J Biol Chem
2001
;
276
:
857
–60.
6
Yousef GM, Diamandis EP. An overview of the kallikrein gene families in humans and other species: emerging candidate tumour markers.
Clin Biochem
2003
;
36
:
443
–52.
7
Tangen CM, Faulkner JR, Crawford ED, et al. Ten-year survival in patients with metastatic prostate cancer.
Clin Prostate Cancer
2003
;
2
:
41
–5.
8
Oefelein MG, Agarwal PK, Resnick MI. Survival of patients with hormone refractory prostate cancer in the prostate specific antigen era.
J Urol
2004
;
171
:
1525
–8.
9
Sawasaki T, Shigemasa K, Gu L, Beard JB, O'Brien TJ. The transmembrane protease serine (TMPRSS3/TADG-12) D variant: a potential candidate for diagnosis and therapeutic intervention in ovarian cancer.
Tumour Biol
2004
;
25
:
141
–8.
10
Lin CY, Anders J, Johnson M, Dickson RB. Purification and characterization of a complex containing matriptase and a Kunitz-type serine protease inhibitor from human milk.
J Biol Chem
1999
;
274
:
18237
–42.
11
Lin CY, Anders J, Johnson M, Sang QA, Dickson RB. Molecular cloning of cDNA for matriptase, a matrix-degrading serine protease with trypsin-like activity.
J Biol Chem
1999
;
274
:
18231
–6.
12
Oberst M, Anders J, Xie B, et al. Matriptase and HAI-1 are expressed by normal and malignant epithelial cells in vitro and in vivo.
Am J Pathol
2001
;
158
:
1301
–11.
13
Suzuki M, Kobayashi H, Kanayama N, et al. Inhibition of tumor invasion by genomic down-regulation of matriptase through suppression of activation of receptor-bound pro-urokinase.
J Biol Chem
2004
;
279
:
14899
–908.
14
Kang JY, Dolled-Filhart M, Ocal IT, et al. Tissue microarray analysis of hepatocyte growth factor/Met pathway components reveals a role for Met, matriptase, and hepatocyte growth factor activator inhibitor 1 in the progression of node-negative breast cancer.
Cancer Res
2003
;
63
:
1101
–15.
15
Benaud CM, Oberst M, Dickson RB, Lin CY. Deregulated activation of matriptase in breast cancer cells.
Clin Exp Metastasis
2002
;
19
:
639
–49.
16
Santin AD, Cane' S, Bellone S, et al. The novel serine protease tumor-associated differentially expressed gene-15 (matriptase/MT-SP1) is highly overexpressed in cervical carcinoma.
Cancer
2003
;
98
:
1898
–904.
17
Johnson MD, Oberst MD, Lin CY, Dickson RB. Possible role of matriptase in the diagnosis of ovarian cancer.
Expert Rev Mol Diagn
2003
;
3
:
331
–8.
18
Santin AD, Zhan F, Bellone S, et al. Gene expression profiles in primary ovarian serous papillary tumors and normal ovarian epithelium: identification of candidate molecular markers for ovarian cancer diagnosis and therapy.
Int J Cancer
2004
;
112
:
14
–25.
19
Takeuchi T, Shuman MA, Craik CS. Reverse biochemistry: use of macromolecular protease inhibitors to dissect complex biological processes and identify a membrane-type serine protease in epithelial cancer and normal tissue.
Proc Natl Acad Sci U S A
1999
;
96
:
11054
–61.
20
Galkin AV, Mullen L, Fox WD, et al. CVS-3983, a selective matriptase inhibitor, suppresses the growth of androgen independent prostate tumor xenografts.
Prostate
2004
;
61
:
228
–35.
21
Oberst MD, Johnson MD, Dickson RB, et al. 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
2002
;
8
:
1101
–7.
22
Nagaike K, Kohama K, Uchiyama S, et al. Paradoxically enhanced immunoreactivity of hepatocyte growth factor activator inhibitor type 1 (HAI-1) in cancer cells at the invasion front.
Cancer Sci
2004
;
95
:
728
–35.
23
Kataoka H, Suganuma T, Shimomura T, et al. Distribution of hepatocyte growth factor activator inhibitor type 1 (HAI-1) in human tissues. Cellular surface localization of HAI-1 in simple columnar epithelium and its modulated expression in injured and regenerative tissues.
J Histochem Cytochem
1999
;
47
:
673
–82.
24
Itoh H, Kataoka H. Roles of hepatocyte growth factor activator (HGFA) and its inhibitor HAI-1 in the regeneration of injured gastrointestinal mucosa.
J Gastroenterol
2002
;
37
:
15
–21.
25
Schwarze SR, Fu VX, Jarrard DF. Cdc37 enhances proliferation and is necessary for normal human prostate epithelial cell survival.
Cancer Res
2003
;
63
:
4614
–9.
26
Hung RJ, Hsu IaW, Dreiling JL, et al. Assembly of adherens junctions is required for sphingosine 1-phosphate-induced matriptase accumulation and activation at mammary epithelial cell-cell contacts.
Am J Physiol Cell Physiol
2004
;
286
:
C1159
–69.
27
Chen CD, Welsbie DS, Tran C, et al. Molecular determinants of resistance to antiandrogen therapy.
Nat Med
2004
;
10
:
33
–9.
28
Liotta LA, Clair T. Cancer. Checkpoint for invasion.
Nature
2000
;
405
:
287
–8.
29
Netzel-Arnett S, Hooper JD, Szabo R, et al. Membrane anchored serine proteases: a rapidly expanding group of cell surface proteolytic enzymes with potential roles in cancer.
Cancer Metastasis Rev
2003
;
22
:
237
–58.
30
Duffy MJ. Proteases as prognostic markers in cancer.
Clin Cancer Res
1996
;
2
:
613
–8.
31
Balk SP, Ko YJ, Bubley GJ. Biology of prostate-specific antigen.
J Clin Oncol
2003
;
21
:
383
–91.
32
Lin CY, Wang JK, Torri J, Dou L, Sang QA, Dickson RB. 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
1997
;
272
:
9147
–52.
33
Kim MG, Chen C, Lyu MS, et al. Cloning and chromosomal mapping of a gene isolated from thymic stromal cells encoding a new mouse type II membrane serine protease, epithin, containing four LDL receptor modules and two CUB domains.
Immunogenetics
1999
;
49
:
420
–8.
34
Lee SL, Dickson RB, Lin CY. Activation of hepatocyte growth factor and urokinase/plasminogen activator by matriptase, an epithelial membrane serine protease.
J Biol Chem
2000
;
275
:
36720
–5.
35
Kataoka H, Hamasuna R, Itoh H, Kitamura N, Koono M. Activation of hepatocyte growth factor/scatter factor in colorectal carcinoma.
Cancer Res
2000
;
60
:
6148
–59.
36
Yamauchi M, Kataoka H, Itoh H, Seguchi T, Hasui Y, Osada Y. Hepatocyte growth factor activator inhibitor types 1 and 2 are expressed by tubular epithelium in kidney and down-regulated in renal cell carcinoma.
J Urol
2004
;
171
:
890
–6.