Radiol Oncol 2000; 34(1): 41-7. Cathepsin D and plasminogen activator inhibitor type 1 in normal, benign and malignant ovarian tissues: a preliminary report Marina Sprem1, Damir Babic--z, Mariija Abramic--3, Dusko Milicic--1, Ivan Vrhovec4, Janez Skrk4, and Maja Osmak5 department of Obstetrics and Gynaecology; 2Department of Gynaecological and Perinatal Pathology, University Hospital and School of Medicine, Zagreb, Croatia; 3Department of Organic Chemistry and Biochemistry, Ruder Boskovic Institute, Zagreb, Croatia; 4Institute of Oncology, Ljubljana, Slovenia; 5Department of Molecular Genetics, Ruder Boskovic Institute, Zagreb, Croatia Background. The aim of the present study was to determine the concentration of cathepsin D (CathD) and plasminogen activator inhibitor type 1 (PAI-1) in normal ovarian tissues, benign and malignant ovarian tumor tissues, and to asses relationship between Cath D and PAI-1 content, and some clinical and patho-histological parameters. Materials and methods. Cath D contents and PAI-1 concentrations were determined (using immunora-diometric ELSA-Cath D assay and commercial IMUDINDR ELISA immunoassay, respectively) in 35 samples: 10 normal ovarii, 10 benign, 10 primary malignant and 5 metastatic ovarian tumors. Results. The concentrations of Cath D were significantly higher in malignant (32.89±14.26 pmol/mg protein ) and metastatic (31.42±1024 pmol/mg protein), than in normal (13.68±4.03 pmol/mg protein) and benign (17.89±13.13 pmol/mg protein) ovarian tissues. There was no statistical differences in the concentrations of PAI-1 between normal, benign, malignant and metastatic tumor specimens. The concentrations of Cath D as well as PAI-1 did not correlate to the age of patients, menopausal status, parity, GOG risk group, clinical stage or pathohistological grading. Conclusion. Concentrations of Cath D (but not PAI-1) were significantly increased in malignant and metastatic ovarian tumor tissues when compared to normal and benign ovarian tumor samples; they were independent from pathohistological and clinical parameters. Key words: ovarian neoplasms; cathepsin D, plasminogen activator inhibitor type 1 Correspondence to: Maja Osmak, Ph.D., Department of Molecular Genetics, Ruder Boskovic Institute, Bijenicka cesta 54, 10000 Zagreb, Croatia. Phone: +385-1-4561-145; Fax: +385-1-4561-177; E-mail: osmak@rudjer.irb.hr Introduction Among gynaecological malignancies, ovarian cancer has been the major cause of death because they are usually presented in advanced clinical stage, with extensive intraab- 42 Sprem Metal. / Cath D and PA1-1 in ovarian cancer dominal metastasis and the unavoidable development of resistance to chemotherapy.1'2 Invasion and metastases are multi-step processes that require a complex cascade of interrelated events including extracellular matrix degradation, migration, proliferation and induction of neovascularisation.3-5 These processes are complicated and only partially understood. Proteases associated with the invasive capacity of tumor cells are metallo-proteinases (collagenases, gelatinases, stro-melysin), cysteine proteases (cathepsin B, H, L), aspartyl proteases (cathepsin D) and serine proteases (urokinase type plasminogen activator and its activated product, plasmin). Cathepsin D (Cath D) is a lysosomal acidic protease. It might be involved in tumor growth, invasion and metastasis by different mechanisms: a) by inactivating a secreted growth inhibitor, b) by releasing growth and angiogenesis factors from the extracellular matrix, c) by providing amino acids following phagocytosis of extracellular matrix, d) by degrading extracellular matrix and basal membranes, and e) by activating latent precursors from other proteinases involved in the invasive step of metastatic process.5'7 Thus, Cath D may facilitate the spread of neoplastic cells and promote tumor invasiveness and metastatic potential through different mechanisms by acting at different levels of the metastatic process. Positive correlation between Cath D level and aggressiveness of breast tumor, as well as shorter relapse free interval and overall survival of breast cancer patients 5,8,9, have stimulated the investigation on tumors of other origin. So, elevated levels of Cath D have been found in other tumors like melanoma, head and neck carcinoma, genital carcinoma.6'10'12 In the multi-step process of invasion, plas-minogen/plasmin system plays an important role. Urokinase plasminogen activator (uPA) is a highly specific serine protease that converts plasminogen in plasmin.4,13 Plasmin degrades fibronectin, laminin and other non- collagenous proteins of the extracellular matrix, and is able to activate latent collage-nases. That leads to further degradation of extracellular matrix and invasion and metastasis. UPA activity is controlled by specific plasminogen activator inhibitors (PAI), which include plasminogen activator inhibitor type 1 (PAI-1), plasminogen activator inhibitor type 2 (PAI-2) and protease nexin.14,15 PAI-1, a serine protease inhibitor (serpin), is the major inhibitor of PA in the plasma. High levels of PAI-1 in tumor extracts from breast, lung and gastric cancerous tissues appear to be highly significant and independent predictors for shorter overall survival.14 The importance of proteases in the regulation of ovarian carcinogenesis, invasion and metastases may further promote its clinical application in the detection of early stage of this disease, prognostic assessment of patients with ovarian cancer as well as in the development of new cancer specific treatment modalities. Limited information is available on the prognostic value of cathepsin D and plasminogen activator inhibitor type 1 regarding gynaecogolical malignancies. The aim of the present study was to determine the concentration of Cath D and PAI-1 in normal, benign and malignant ovarian tissues and to asses the relationship between Cath D and PAI-1 content and some clinical and patho-histological parameters. Materials and methods Tissue samples The tissue samples were obtained from fresh specimens removed during surgical procedure at the Department of Obstetrics and Gynaecology of Medical School, University of Zagreb. The Ethics of Research Committee at the Medical School of the University of Zagreb approved the protocol. Thirty-five tissue samples were analyzed: 10 normal ovarian tissues, 10 benign ovarian Radiol Oncol 2000; 34(1): 41-7. Sprem Met al. / Cath D mid PAH in ovarian cancer 43 tumor specimens, 10 primary ovarian carcinoma and 5 metastatic ovarian carcinoma samples. After surgery, the samples were frozen in liquid nitrogen. For biochemical analysis, samples were they were prepared as previously described.16 Briefly, each specimen was minced, homogenized in lysis buffer, and centrifuged for 45 min at 15000 g. Supernatant (tumor tissue cytosol) was used for biochemical studies. Cathepsin D determination Cath D content was measured in the cytosol of homogenized samples using a commercially available solid-phase two site immunora-diometric assay ( ELSA-CATH-D, CIS Bio International, Gif-sur-Ivette, France). It detects precursors and mature form of Cath D. The values were normalized according the total protein assessed by Bradford's method17, and expressed as pmol/mg protein. Plasminogen activator inhibitor determination The concentration of PAI-1 inhibitor was determined by immunoassay using a commercially available ELISA kit (IMUBINDR, American Diagnostica Inc., Greenwhich, USA). It detects latent and active forms of PAI-1 and complexes. The concentrations of PAI-1 were expressed in ng/mg protein. Statistics The statistical significance of difference between the concentrations of cathepsin D and PAI-1 in the particular groups was tested by Kruskal-Wallis one way ANOVA test. Their relationship to pathohistological and clinical factors were calculated by Spearman's rank correlation test. Data are reported as mean ± SD. Differences were considered significant at p<05. Results The pathohistological characteristics of ovarian tumors that were entered into present study are given in Table l. Their classification and histopathological grade, as well as clinical staging, was defined according to FIGO. Among the patients with benign ovarian tumors, those with cystadenofibroma sero-sum prevailed, while among the patients with primary malignant tumors, those with cys- Table l. Pathohistological characteristics of ovarian tumor specimens Benign Primary malignant Metastatic carcinoma teratoma dermoides cystadenocarcinoma serosum breast cancer 2 cysticum 1 ovarii 5 carcinoma ventriculi cystadenofibroma serosum cystadenocarcinoma mucinosum (Krukenberg) 1 ovarii 8 ovarii 2 colorectal carcinoma cysta serosa 1 cystadenocarcinoma endometrial carcinoma 1 anaplasticum ovarii 1 adenocarcinoma endometrioides ovarii 1 cystadenocarcinoma mixtum ovarii 1 Radiol Oncol 2000; 34(1): 41-7. 44 Sprem Metal. / Cath D and PA1-1 in ovarian cancer Table l. Pathohistological characteristics of ovarian tumor specimens Total 10 Clinical stage I A 1 IB 1 IC 4 III 2 IV 2 Tumor grade GI (well differentiated) 4 GII (moderately differentiated) 3 GIII (poorly differentiated) 3 Hystological type mucinosus 2 serosus 5 endometrioid 1 mixed 1 anaplastic 1 Age ( mean, range) <50 4 >50 6 Parity O 3 >l 7 GOG( Gynecology Oncology group) low risk (stage IA and B, grade 1 or 2) 2 high risk (stage IC or II or grade 3) 8 tadenocarcinoma serosum were prevalent. (Table 1). Primary ovarian carcinomas were further classified as shown in Table 2. More than half of these malignant tumors belong, according to GOG, to high risk group. The cytosol supernatant obtained after homogenization of ovarian tissue samples was analysed for the content of cathepsin D and plasminogen activator inhibitor type 1. Figure 1 shows the concentrations of cathepsin D in normal, benign and malignant ovarian specimens. Cath D levels were significant- Radio/ Oncol 2000; 34(1): 41-7. ly higher in primary malignant (32.89 ± 14.26 pmol/ mg protein) and metastatic (31.42 ± 10.24 pmol/ mg protein) malignant ovarian carcinomas than in normal (13.68 ± 4.03 pmol/ mg protein) or benign (17.89 ± 13.13 pmol/ mg protein) tissue samples (p=0.003 for normal versus primary malignant tissues, p=0.035 for normal versus metastatic malignant tissues, p=0.028 for benign versus primary malignant tissues, p=0.029 for benign versus metastatic malignant tissues). There was no difference between the levels of Cath D in primary and metastatic ovarian tumor tissues (p=0.995), or between normal and benign tumor tissues (p=0.838). Figure 2 presents the PAI-1 concentrations in normal, benign and malignant ovarian specimens. Although the mean concentration of PAI-1 was lower in normal (2.96 ± 1.49 ng/mg protein) than in malignant tumor tissues (8.80 ± 10.56 ng/mg protein), this difference was not statistically significant (p=0.170). Also, there was no statistically significant difference between the levels of PAI-1 in primary and metastatic ovarian carcinomas (p=0.864), or between normal and benign tumor tissues (p=0.948). The concentration of cathepsin D as well as PAI-1 did not seem to correlate to the age of patients, their menopausal status, parity, or GOG risk group. Neither cathepsin D nor PAI-1 concentrations correlated with clinical staging or pathohistological grading. Discussion The major cause of death among gynaecological malignancies is ovarian cancer: over 70% of women are diagnosed with incurable advanced-stage disease, defined by widespread intraperitoneal metastases.1'2 Tumor invasion is a complex, multistep sequence of events based on a cascade of coordinated cellular processes. The combined action of severa! proteolytic enzymes is involved in tissue Sprem Metal. / Calh D mid PAI-1 in ovarian cancer 45 / m g P 60 50 - 40 30 20 10 N = 10 10 Malignant Metastatic Benign 10 Normal Figure 1. Concentrations of cathepsin D (mean ± SD) in primary and metastatic ovarian carcinomas, benign ovarian tumors and normal ovarian tissues. The number of samples per group are indicated. o o degradation and remodeling in both, normal and tumor tissue.3-5 The collected evidence have proved that proteases and their inhibitors (including Cath D and PAI-1) may serve as prognostic factors in breast cancer, to predict the outcome of the disease.5,8,9 Several tumor-associated proteases are important factors also in solid tumors of other origins, such as lung, gastric, colon, genital, head and neck, bladder, and kidney cancers.4-6,10-12,18,19 The aim of the present study was to determine and compare the levels of cathepsin D and plasminogen activator inhibitor type 1 in normal, benign, primary malignant and metastatic malignant tumor ovarian tissues. As mentioned in the Introduction, Cath D is a proteolytic enzyme that may be involved in invasion and metastasis through different mechanisms.5-7 In the present study, we have determined significantly higher concentrations of Cath D in primary and metastatic ovarian carcinomas than in normal or benign tumor samples (Figure 1), which was is in agreement with literature data.12,20 It is important to point out that Cath D levels were similar in both, normal and benign tumor tissue samples, suggesting that malignant transformation of ovarian tissue was accompanied with the increased level of this protease. The concentration of Cath D was independent from the pathohistological parameters, such as hysto-logical type or tumor grading, as well as clinical markers, such as clinical stage, age, parity and menopausal status. Radiol Oncol 2000; 34(1): 41-7. 46 Sprem Metal. / Cath D and PAI-1 in ovarian cancer g / m g P 10 Malignant Metastatic 10 Benign 10 Normal n Figure 2. Concentrations of urokinase plasminogen activator inhibitor type 1 (mean ± SD) in primary and metastatic ovarian carcinomas, benign ovarian tumors and normal ovarian tissues. The number of samples per group are indicated. Urokinase plasminogen activator has important role in the tissue degradation and in the invasiveness of tumor cells.4,13 In spite of the fact that uPA activity is controlled by a specific inhibitor, such as PAI-1, the increased levels of both, uPA and PAI-1 were found in different tumors: breast cancer20, gastric cancer14 and bladder cancer. 21 Furthermore, increased levels of both, uPA and PAI-1 were associated with poor prognosis, and increased relapse rate and shorter survival. The data in the literature concerning ovarian tumors and PAI-1 levels are contradictory. Some groups have found increased levels of PAI-1 in ovarian cancer22, and even suggested that uPA and PAI-1 may predict the survival of patients with advanced ovarian cancer.23 Radiol Oncol 2000; 34(1): 41-7. Others did not determine any increased levels of uPA and PAI-1 in ovarian tumors.20 Our results concerning PAI-1 are closer to the data of Ruppert et al.20 We did not find a significant increase of PAI-1 in malignant ovarian tumor tissues. However, the observed difference in absolute values for mean PAI-1 concentrations between normal and malignant ovarian tissues suggests that in order to draw important conclusions more samples should be examined. In addition, comparing the levels of Cath D and PAI-1 in normal, benign and malignant tumor tissues, we have not found any correlation between these parameters. In conclusion, Cath D overexpression in malignant tumor ovarian tissue may be relat- Sprem Met til. / Cath D and PAI-1 in ovarian cancer 47 ed to intraabdominal dissemination of ovarian tumor cells. In order to determine more precisely the diagnostic and prognostic values of Cath D and PAI-1 in ovarian cancer patients, we have started a clinical study with a higher number of patients with ovarian cancer and a longer follow up. References 1. Pickel H. Prognostic factors in ovarian cancer. CME J Gynecol Oncol 1999; 4: 8-12. 2. Friedlander ML. Prognostic factors in ovarian cancer. Seiiiiii Oncol 1998; 3: 305-14. 3. Liotta LA, Steeg PS, Stetler-Stevenson WG. Cancer metastasis and angiogenesis: an imbalance of positive and negative regulation. Cell 1991; 64: 327-36. 4. Schmitt M, Janicke F, Graeff H. Tumor-associated proteases. Fibrinolysis 1992; 6, Suppl 4: 3-26. 5. Sloane BF, Moin K, Lach TT. Lysosomal enzymes and their endogenous inhibitors in neoplasia. In: Pretlow TG, Pretlow TH (eds) Biochemical and molecular aspects of selected cancers. New York: Acad Press; 1993. Vol 2, p 411-66. 6. Leto G, Gebbia N, Rausa L, Tumminello FM. Cathepsin D in malignant progression of neoplastic diseases (review). Anticancer Res 1992; 12: 23540. 7. Rochefort H, Liaudet E, Garcia M. Alterations and role of human cathepsin D in cancer metastasis. Enzyme Protein 1996; 49: 106-16. 8. Rochfort H. The prognostic value of cathepsin D in breast cancer. A long road to the clinic. Eur J Cancer 1996, 32A: 7-8. 9. Westly BR, May FEB. Cathepsin D and breast cancer. Eur J Cancer 1996; 32A: 15-24. 10. Zeillinger R, Eder S, Schneeberger CH, Ullrich R, Speiser P, Swoboda H. Cathepsin D and PAI-1 expression in human head and neck cancer. Anticancer Res 1996; 16: 449-54. 11. Osmak M, Babic D, Abramic M, Vrhovec I, Milicic D, Skrk J. Cathepsin D content in malignant tumours of corpus uteri. Eur J Cancer 1997; 33: 699700. 12. Scambia G, Benedetti P, Ferradina G, BAttaglia F, Biaocchi G, Mancuso S. Cathepsin D assay in ovar- ian cancer: correlation with pathological features and receptors for oestrogen, progesterone and epidermal growth factor. Br J Cancer 1991; 64: 182-4. 13. Aquirre Ghiso J, Alonso DF, Farias EF, Gomez DE, BA1 de Kier Joffe E. Deregulation of the signaling pathways controlling urokinase production. Its relationship with the invasive phenotype. Eur J Biochem 1999; 263: 295-304. 14. Pappot H, Gardsvoll H, Romer J, Pedersen AN, Hansen JG, Pyke C, Brunner N. Plasminogen activator inhibitor type 1 in cancer: therapeutic and prognostic implications. Biol Chem Hoppe -Seyler 1995; 376: 259-67. 15. Rijken DC. Plasminogen activators and plasmino-gen activator inhibitors: biochemical aspects. Bailliere Clin Haematol 1995; 8: 291-312. 16. Simaga S, Babic D, Osmak M, Ilic-Forko J, Vitale Lj, Milicic D, Abramic A. Dipeptidyl peptodase III in malignant and non-malignant gynaecological tissue. Eur J Cancer 1998; 34: 399-405. 17. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein, utilizing the principle of protein-dye binding. Anal Biochem 1976; 72: 248-54. 18. Nekarda H, Sievert JR, Schmitt M, Ulm K. Tumour associated proteolytic factors uPA and PAI-1 and survival in totally resected gastric cancer. Lancet 1994; 343: 117-9. 19. Volm M, Mattern J, Koomagi R. Relationship of urokinase and urokinase receptor in non-small cell lung cancer to proliferation, angiogenesis, metastasis and patient survival. Oncology Reports 1999; 6: 611-5. 20. Ruppert C, Ehrenforth S, Scharer I, Halberstadt E. Protease levels in breast, ovary, and other gynaecological tumor tissues-prognostic importance in breast cancer. CancerDetect & Prev 1997; 21: 452-9. 21. Hasui Y, Maratsuka K, Nishi S, Kitada S, Osada Y, Simiyosha A. The content of urokinase-type plas-minogen activator and tumor recurrence in superficial bladder cancer. J Urol 1994; 151: 16-20. 22. Casslen B, Bossmar T, Lecander I, Astedt B. Plasminogen activators and plasminogen ctivator inhibitors in blood and tumor fluids of patients with ovarian cancer. Eur J Cancer 1994; 30:1302-9. 23. Kuhn W, Pache L, Schmalfeldt B, Dettmar P, Schmitt M, Janicke F, Graeff H. Urokinase (uPA) and PAI-1 predict survival in advanced ovarian cancer patients (FIGO III} after radical surgery and platinum-based chemotherapy. Gynecol Oncol 1994; 55: 401-9. Radiol Oncol 2000; 34(1): 41-7.