Original Paper

Pathobiology 2004;71:308–313DOI: 10.1159/000081726

Deregulation of Collagen Metabolismin Human Stomach Cancer

Tomasz Guszczyn Krzysztof Sobolewski

Department of Medical Biochemistry, Medical Academy of Białystok, Białystok, Poland

Received: February 16, 2004Accepted: July 13, 2004

Dr. Tomasz GuszczynDepartment of Medical BiochemistryMedical Academy of Białystok, ul. Mickiewicza 2PL–15-089 Białystok (Poland)Tel. +48 85 748 55 78, Fax +48 85 748 54 16, E-Mail tombial@mp.pl

ABCFax + 41 61 306 12 34E-Mail karger@karger.chwww.karger.com

© 2004 S. Karger AG, Basel1015–2008/04/0716–0308$21.00/0

Accessible online at:www.karger.com/pat

Key WordsCollagen W Stomach cancer W Prolidase W Gelatinases W

ß1 integrin

AbstractObjective: The defects of collagen metabolism are re-sponsible for the disorganization of extracellular matrixin stomach cancer. Collagen through interaction withintegrin receptors regulates the cellular growth, differen-tiation, gene expression, prolidase and gelatinase activi-ty and plays an important role in tumorigenesis and inva-siveness. Although extracellular metalloproteinases ini-tiate the breakdown of collagen in tissues, the final stepof its degradation is mediated by prolidase. Therefore,we decided to compare the degradation of collagen incontrol tissues to gastric cancer tissues. Methods: Weinvestigated the collagen content (hydroxyproline as-say), expressions of ß1 integrin, prolidase and gelati-nases A and B (Western immunoblot) as well activities ofprolidase (colorimetric assay) and gelatinases (zymogra-phy) in stomach cancer tissue (n = 10). The results werecompared with corresponding data obtained for controltissues (n = 10). Results: No differences in the collagencontent were found between the studied tissue samples.However, an increase in free proline pool, enhancedgelatinase expression and elevated gelatinolytic activitywere found in the tumor tissue. These phenomena wereaccompanied by a significant elevation in prolidase ac-

tivity and an increase in ß1 integrin expression in stom-ach cancer, compared to control tissue. Conclusion: Thedata presented suggest an enhancement of collagenturnover in stomach cancer. It may be suggested that theincreased degradation of collagen by gelatinase in can-cer tissue is balanced by an increased biosynthesis ofthis protein.

Copyright © 2004 S. Karger AG, Basel

Introduction

Collagen of various types is present in gastric tissue [1,2]. Most part of the gastric collagen is produced by fibro-blasts, but it can also be synthesized by a variety of other,more differentiated cells [3, 4]. Neoplastic transformationis accompanied by some alterations in biosynthesis ofextracellular matrix (ECM) proteins, mainly fibronectinand type I collagen [5, 6].

It is well known that scirrhous is one of the most inva-sive forms of this gastric carcinoma. Some specific fea-tures of collagen metabolism were reported in gastrictumors. An increase in prolyl-4-hydroxylase expression,an enzyme that plays a crucial role in collagen biosynthe-sis, was observed in gastric carcinoma cells. Furthermore,an increase in prolyl-4-hydroxylase expression was foundin fibroblasts located at the periphery of the tumor [7, 8].

The number of fibroblastic cells in scirrhous stroma isdistinctly higher in comparison to the normal gastric wall.

Collagen Metabolism in Stomach Cancer Pathobiology 2004;71:308–313 309

These cells are stimulated by the growth factors producedby cancer cells to produce the ECM components [9]. Asimilar mechanism seems be responsible for the enhancedproduction of urokinase-type plasminogen activator byfibroblasts stimulated by growth factors released from thetumor cells [10]. The increase in the activity of proteolyticenzymes activated by urokinase-type plasminogen activa-tor may enhance the degradation of stroma componentsand promote the invasion of the gastric wall by the tumorcells [11].

Collagen is required for the maintenance of the archi-tecture and integrity of tissues. It plays this role by inter-action with cell surface adhesion receptors, mainly withintegrins. The interaction between integrins and collagen,may affect the cellular gene expression, as well as the cellgrowth and differentiation [12, 13]. Such an interaction isthought to play an important role in tumorigenicity andinvasiveness [14, 15]. For example; collagen receptor,composed of ·2ß1 integrin subunits is involved in the for-mation of tumor metastasis [16]. The mechanism of thisprocess probably depends on ß1 integrin-mediated signaltransduction that regulates cell migration, [17]. For thisreason, any change in the quantity and distribution of col-lagen and integrins may alter the function of adheringcells.

Metastatic tumor cells produce enhanced amounts ofproteolytic enzymes that enable them to penetrate base-ment membranes and ECM [18]. The tumor progressionmight, therefore, critically depend on the breakdown ofcollagen and other ECM proteins. Although extracellularcollagenases initiate the breakdown of collagen both innormal and in tumor tissues [10], the final step of collagendegradation is mediated by prolidase (E.C.3.4.13.9). Prol-idase is a cytosolic enzyme that catalyzes the hydrolysis ofX-Pro imidodipeptides, where X represents any aminoacid [19]. This enzyme is involved in the recycling of pro-line for collagen synthesis and cell growth. The efficiencyof recycling of proline was found to be about 90% [20, 21].It is thought that prolidase is a step-limiting factor in theregulation of collagen biosynthesis.

It is not known if prolidase plays a role in the pathobio-chemistry of tumor tissues. Some observations suggestthat the activity of prolidase produced by normal fibro-blasts in culture is under the control of ß1 integrin – colla-gen interaction [22]. Therefore, it was decided to comparethe collagen content and the expressions and activities ofprolidase, gelatinases A (matrix metalloproteinase-2;MMP-2) and B (MMP-9) as well as ß1 integrin expressionin normal stomach tissue and in stomach cancer.

Materials and Methods

The protocol of study described in this paper was accepted by theCommittee for Ethics and Supervision on Human and AnimalResearch of the Medical Academy of Białystok.

Tissue MaterialThe samples of human stomach carcinoma (located in the pyloric

area) and the samples of control gastric wall (taken from the samearea) were obtained from the Department of General Surgery, theMedical Academy of Bialystok, Poland. In each case histopatholog-ical examination was performed. The histological typing of thetumors was performed according to WHO classification. Ten cases ofintestinal type of adenocarcinoma (taken from the margins of thetumors) were selected for biochemical studies. Ten control samplesof the gastric wall were taken from human stomachs resected becauseof noncarcinomatous diseases. In each case the full mural specimenof the gastric wall was used for biochemical studies.

Samples of tissues were placed in ice-cold 0.05 mol/l Tris-HCl,pH 7.6 buffer and extensively perfused with the same buffer. Thetissues were stored at –70°C, until assay. The tissue homogenates(20% w/v) were prepared in 0.05 mol/l Tris-HCl, pH 7.6 with the useof knife homogenizer (Polytron) and subsequently were sonicated at0°C. Homogenates were centrifuged at 16,000 ! g for 30 min at4°C. Supernatant (tissue extract) was used for protein determina-tion, enzyme activity assays and Western blot analysis.

Determination of Prolidase ActivityThe activity of prolidase was determined according to the method

of Myara et al. [23] based on the measurement of proline by Chi-nard‘s reagent [24]. Activation of prolidase required preincubationwith manganese; 100 Ìl of tissue supernatant incubated with 100 Ìlof 0.05 mol/l Tris-HCl, pH 7.8 containing 20 mmol/l MnCl2 for 1 h at37°C. After preincubation, the prolidase reaction was initiated byadding 100 Ìl of the preincubated mixture to 100 Ìl of 94 mmol/lglycyl-proline (Gly-Pro) for a final concentration of 47 mmol/l Gly-Pro. After an additional incubation for 1 h at 37 °C, the reaction wasterminated with 1 ml of 0.45 mol/l trichloroacetic acid. In paralleltubes the reaction was terminated at time ‘zero’ (without incubation).The released proline was determined by the addition of 0.5 ml of thetrichloroacetic acid supernatant to 2 ml of a 1:1 mixture of glacialacetic acid: Chinard’s reagent (25 g ninhydrin dissolved at 70°C in600 ml of glacial acetic acid and 400 ml of 6 mol/l orthophosphoricacid) and incubated for 10 min at 90°C. The amount of releasedproline was determined colorimetrically by absorbance at 515 nmand calculated by using proline standards. Protein concentration wasmeasured by the method of Lowry. Enzyme activity was reported asnanomoles of proline released during minute per milligram of super-natant protein.

SDS-PAGESlab SDS/PAGE was used, according to the method of Laemmli

[25]. Samples of tissue supernatants (20 Ìg of protein) were electro-phoresed. The following Bio-Rad’s unstained high-molecular-weightstandards were used: myosin (200 kDa), galactosidase (116.2 kDa),phosphorylase b (97.4 kDa), bovine serum albumin (66.2 kDa), oval-bumin (45 kDa).

310 Pathobiology 2004;71:308–313 Guszczyn/Sobolewski

Fig. 1. Hydroxyproline content in normalgastric (1) and gastric cancer tissues (2).

Fig. 2. Prolidase activity (A) and proline content (B) in normal gas-tric (1) and gastric cancer tissues (2). * p ! 0.001.

ZymographyGelatinolytic activity was determined according to the method of

Unemori and Werb [26]. Tissue extract was mixed with Laemmlisample buffer [25] containing 2.5% SDS (without reducing agent).Equal amounts (about 20 Ìg) of protein were electrophoresed underno reducing conditions on 10% polyacrylamide gels impregnatedwith 1 mg/ml gelatin. After electrophoresis, the gels were incubatedin 2% Triton X-100 for 30 min at 37°C to remove SDS and incu-bated for 18–24 h at 37°C in substrate buffer (50 mM Tris-HCl buff-er, pH 8, containing 5 mM CaCl2). After staining with CoomassieBrilliant Blue R-250, gelatin-degrading enzymes present in tissueextract were identified as clear zones in a blue background.

Western Blot AnalysisAfter SDS-PAGE, the gels were allowed to equilibrate for 5 min in

25 mmol/l Tris, 0.2 mol/l glycine in 20% (v/v) methanol. The proteinwas transferred to 0.2-Ìm pore-sized nitrocellulose at 100 mA for 1hour by using a LKB 2117 Multiphor II electrophoresis unit. Thenitrocellulose was incubated with polyclonal antibody against humanprolidase (was the gift from Prof. J. Pałka – Department of MedicalChemistry, Medical Academy of Białystok) at concentration 1: 3,000;anti-integrin ß1, anti-MMP-2 or -9 monoclonal antibodies (ICN Bio-medicals) at concentration 1:1,000 in 5% dried milk in TBS-T (20mmol/l Tris-HCl buffer, pH 7.4, containing 150 mmol/l NaCl and0.05% Tween 20) for 1 h. In order to analyze the proteins, alkalinephosphatase-conjugated second antibodies against rabbit or mouse FcIgG were added at concentration 1:5,000 or 1:7,500 in TBS-T, respec-tively. Incubation was continued for 30 min by shaking slowly. Thenin both cases nitrocellulose was washed with TBS-T (5 times for 5 min)and submitted to Sigma-Fast BCIP/NBT reagent.

Determination of Protein, Proline and HydroxyprolineProtein was determined according to the method of Lowry et al.

[27]. Hydroxyproline was determined by the method of Woessner JF

Jr [28]. Proline was measured by Chinard’s reagent as described inthe methodology for prolidase activity assay [24].

Statistical AnalysisIn all experiments the mean values for 10 assays B standard

deviations (S.D.) were calculated. The results were submitted to sta-tistical analysis with the use of Student’s t test, accepting p ! 0.05 assignificant.

Results

The collagen content was determined in studied tissuesby the measurement of hydroxyproline. As shown in fig-ure 1 no differences in the total hydroxyproline contentwere found in stomach cancer, compared to the controltissue.

The results presented on figure 2A show a significantincrease in prolidase activity in stomach cancer comparedto the normal tissue. Since the prolidase specificity isdirected against dipeptides containing C-terminal pro-line, the free proline was determined in the extracts of tis-sue homogenates using Chinard’s reagent. As can be seenfrom figure 2B the stomach cancer tissue contains signifi-cantly more proline than the control tissue.

Moreover, the increase in prolidase activity in stomachcancer tissue was accompanied by an increase in thisenzyme expression as shown by Western immunoblotanalysis (fig. 3A). Since prolidase activity is stimulated

Collagen Metabolism in Stomach Cancer Pathobiology 2004;71:308–313 311

Fig. 3. Western blot analysis for prolidase (A) and ß1 integrin (B) ofnormal gastric (1) and gastric cancer tissues (2). ß-actin expressionserves as a control for protein loading. The same amount of extractprotein (20 Ìg) was run in each lane. Western immunoblot analysiswas done on pooled homogenate extracts from 10 control and 10gastric cancer tissues.

Fig. 4. Western blot analysis for MMP-2 (A) and MMP-9 (B) of nor-mal gastric (1) and gastric cancer tissues (2). ß-actin expression servesas a control for protein loading. The same amount of extract protein(20 Ìg) was run in each lane. Western immunoblot analysis was doneon pooled homogenate extracts from 10 control and 10 gastric cancertissues.

Fig. 5. Gelatinolytic activity of normal gas-tric (1) and gastric cancer tissues (2). Thesame amount of extract protein (20 Ìg) wasrun in each lane. Western immunoblot anal-ysis was done on pooled homogenate ex-tracts from 10 control and 10 gastric cancertissues.

through ß1 integrin [22] the expression of this integrin wasmeasured. The results presented on figure 3B show a dis-tinct increase in the ß1 integrin expression in stomachcancer, compared to the control tissue.

Western immunoblot analysis presented on figure 4demonstrates the presence gelatinises A (MMP-2) and B(MMP-9) proteins both in control gastric wall and intumor tissue. An increase in the expression of MMP-2(fig. 4A) and MMP-9 (fig. 4B) was found in extracts ofstomach cancer tissues, compared to extracts of controltissues.

Disturbances in the collagen metabolism are usuallyaccompanied by the deregulation of tissue collagenaseactivity. Therefore, the evaluation of gelatinolytic activityby zymography was performed. As can be seen from fig-ure 5, the control stomach tissue contained gelatinase A(MMP-2) – represented by a 72-kDa protein and gelati-nase B (MMP-9) – represented by a 92-kDa protein inzymogen forms. Only trace amounts of active enzymeswere detected. The tumor tissue contained both zymogenand active forms of these metalloproteinases. The pres-ence of an 82-kDa gelatinase – an active form of the latent92-kDa gelatinase and 62-kDa gelatinase – an active formof the latent 72-kDa gelatinase were found in the extractof cancer tissue. It is of interest that both the amounts of

312 Pathobiology 2004;71:308–313 Guszczyn/Sobolewski

zymogens and active forms of these metalloproteinaseswere distinctly higher in the tumor tissue in comparisonto the control tissue.

Discussion

The impairment of tissue collagen metabolism duringneoplastic transformation is a well-known phenomenon[29]. The data presented here support the concept that thederegulation of the balance between collagen biosynthesisand degradation is an underlying mechanism for collagenmetabolism disturbances in cancer tissue [8]. It has beenshown to be so for at least lung adenocarcinoma [30] andpancreatic cancer [31]. The present report suggests thatthe phenomenon may apply also to stomach cancer. In allcases of cancers studied up to date, aberrations in collagenmetabolism were accompanied by significant differencesin prolidase activity. In fact, the enzyme activity reflectsthe intensity of collagen turnover [32]. The increase inprolidase activity parallels the increased rates of collagendegradation and biosynthesis [20]. The data presented inthis report show a significant increase in the prolidaseactivity in stomach cancer.

Prolidase mediates the final step of collagen degrada-tion, cleaving X-Pro or X-Hyp imidodipeptides, where X-represents any amino acid [20]. Collagen is the mostabundant source of such imidodipeptides. The increase inprolidase activity may enhance collagen biosynthesis byincreasing the proline pool in the cells, thus providing amechanism which enhances collagen deposition. A simi-lar phenomenon was observed in cultured breast cancercells (MCF-7), where the low prolidase expression and thelow activity of this enzyme were correlated with a de-crease in extracellular collagen content [22]. Similar rela-tionship was observed during fibrosis processes, where anincrease in prolidase activity was accompanied by anincrease in tissue collagen deposition [33]. Moreover, thelink between prolidase activity and collagen synthesis hasbeen found in several experimental conditions [22, 34–37].

Although the mechanism and endpoints by which prol-idase is regulated remain largely unknown, it has beenpostulated that at least in normal cells the prolidase is reg-ulated by the interaction of type I collagen with ß1 integrinreceptor [22]. We have found in this study an increase inß1 integrin expression in stomach cancer. Although no dif-ferences in collagen (ligand of ß1 integrin) content be-tween control and stomach cancer tissues were found, theincrease in the ß1 integrin receptor expression in cancer

tissue may augment a signal that enhances prolidase activ-ity. This phenomenon may explain the mechanism forincrease of prolidase activity in stomach cancer tissue. Anincrease in the prolidase activity increases proline poolthat enhances collagen synthesis [30].

It is known that most tumor cells produce enhancedamounts of proteolytic enzymes [38] that increase intra-cellular as well extracellular protein degradation and en-able them to penetrate the basement membrane and ECM[18]. The ECM composed of collagens, proteoglycans andglycoproteins is a major barrier against the invasion ofneoplastic cells. Therefore, a tumor progression criticallydepends on the breakdown of collagen and other ECMproteins [39, 40]. The MMPs, especially MMP-2 and -9,are strongly engaged in tumor invasion and metastasis.MMP-2 and -9 are the main proteolytic enzymes degrad-ing basement membranes. They stimulate angiogenesisand tumor growth [18, 39].

Tissue collagenase is the enzyme, which initiates colla-gen breakdown. This enzyme is known to be involved inlung cancer invasion [10]. Once native extracellular colla-gen is cleaved by tissue collagenase into two macromolec-ular fragments, the fragments themselves become suscep-tible to further degradation by a wide variety of extracel-lular and intracellular proteases [41]. An increased ex-pression of tissue collagenases during neoplastic transfor-mation was proved by numerous studies [42]. However,their expression did not correlate with the collagenolyticactivity. The activation of collagenase precursors occursvia limited cleavage of zymogens by other proteases [42].Zymography is a useful technique to detect the activity ofgelatinases. In our studies we have found that both thecontrol gastric wall and gastric cancer tissue containzymogen forms of gelatinases A (MMP-2) and B (MMP-9). The tumor tissue demonstrated both increased expres-sion of MMP-2 and -9 and an elevated activity of thesemetalloproteinases, compared to control tissue.

The data presented suggest an enhancement of colla-gen turnover in stomach cancer. It may be suggested thatthe increased degradation of collagen by gelatinase in can-cer tissue is balanced by an increased biosynthesis of thisprotein.

Collagen Metabolism in Stomach Cancer Pathobiology 2004;71:308–313 313

References

1 Gillessen A, Voss B, Rauterberg J, DomschkeW: Distribution of collagen types I, III, and IVin peptic ulcer and normal gastric mucosa inman. Scand J Gastroenterol 1993;28:688–689.

2 Kondou S, Kanayama K, Morise K: Immuno-histochemical study of collagen in gastric can-cer tissue. Nihon Rinsho Meneki Gakkai Kai-shi 1995;18:168–177.

3 Bornstein P, Sage H: Regulation of collagengene expression. Prog Nucleic Acid Res MolBiol 1989;37:67–106.

4 Kurita A, Kikuchi S, Kakita A: Immunohisto-chemical study of collagen synthesis in an in-vitro model of scirrhous carcinoma of thestomach. Hepatogastroenterology 2002;49:1235–1238.

5 David L, Nesland JM, Holm R, Sobrinho-Simoes M: Expression of laminin, collagen IV,fibronectin, and type IV collagenase in gastriccarcinoma. An immunohistochemical study of87 patients. Cancer 1994;73:518–527.

6 Slack JL, Parker MI, Robinson VR, BornsteinP: Regulation of collagen I gene expression byras. Mol Cell Biol 1992;12:4714–4723.

7 Matsui H, Kubochi K, Shimada A, Hasumi A:Desmoplastic response in scirrhous gastric car-cinoma: Imbalance between collagen synthesisand degradation. Anticancer Res 2000;20:4733–4738.

8 Matsui H, Kubochi K, Okazaki I, Yoshino K,Ishibiki K, Kitajima M: Collagen biosynthesisin gastric cancer: Immunohistochemical analy-sis of propyl 4-hydroxylase. J Surg Oncol 1999;70:239–246.

9 Tahara E: Growth factors and oncogenes inhuman gastrointestinal carcinomas. J CancerRes Clin Oncol 1990;116:121–131.

10 Bolon I, Gouyer V, Devouassoux M, Vanden-bunder B, Wernert N, Moro D, Brambilla C,Brambilla E: Expression of c-ets-1, collagenase1, and urokinase-type plasminogen activatorgenes in lung carcinomas. Am J Pathol 1995;147:1298–1310.

11 Naito Y, Kino I: Pathogenesis and progressionof scirrhous carcinoma. Gan To Kagaku Ryoho1994;21:2364–2370.

12 Matsuoka T, Yashiro M, Nishimura S, InoueT, Fujihara T, Sawada T, Kato Y, Seki S, Hira-kawa YS, Chung K: Increased expression ofalpha2beta1-integrin in the peritoneal dissemi-nation of human gastric carcinoma. Int J MolMed 2000;5:21–25.

13 Carey DJ: Control of growth and differentia-tion of vascular cells by extracellular matrix.Ann Rev Physiol 1991;53:161–177.

14 Ruoslahti E: Control of cell motility and tumorinvasion by extracellular matrix interaction. BrJ Cancer 1992;66:239–242.

15 Albelda SM, Buck CA: Integrins and other celladhesion molecules. FASEB J 1990;4:2868–2880.

16 Bosco M, Chan C, Matsuura N, Takada Y, Zet-ter BR, Hemler ME: In vitro and in vivo conse-quences of VLA-2 expression on rhabdomyo-sarcoma cells. Science 1991;251:1600–1602.

17 Juliano RL: The role of ß1 integrins in tumor.Cancer Biol 1993;4:277–283.

18 Stamenkovic I: Extracellular matrix remodel-ling: The role of matrix metalloproteinases. JPathol 2003;200:448–464.

19 Myara I, Charpentier C, Lemonnier A: Proli-dase and prolidase deficiency. Life Sci 1984;34:1985–1998.

20 Yaron A, Naider F: Proline-dependent struc-tural and biological properties of peptides andproteins. Crit Rev Biochem Mol Biol 1993;28:31–81.

21 Emmerson KS, Phang JM: Hydrolysis of pro-line dipeptides completely fulfills the prolinerequirement in a proline – Auxotropic ChineseHamster Ovary cell line. J Nutr 1993;123:909–914.

22 Pałka JA, Phang JM: Prolidase activity infibroblasts is regulated by interaction of extra-cellular matrix with cell surface integrin recep-tors. J Cell Biochem 1997;67:166–175.

23 Myara I, Charpentier C, Lemonnier A: Opti-mal conditions for prolidase assay by prolinecolorimetric determination: Application to im-idodipeptiduria. Clin Chim Acta 1982;125:193–205.

24 Chinard FP: Photometric estimation of prolineand ornithine. J Biol Chem 1952;199:91–95.

25 Laemmli UK: Cleavage of structural proteinsduring the assembly of the head of bacterio-phage T4. Nature (London) 1970;227:680–685.

26 Unemori EN, Werb Z: Reorganization of poly-merized actin: A possible trigger for inductionof procollagenase in fibroblasts cultured in andon collagen gels. J Cell Biol 1986;103:1021–1031.

27 Lowry OH, Rosenbrough NI, Far AL, RandallIR: Protein measurement with the Folin phenolreagent. J Biol Chem 1951;193:265–275.

28 Woessner JF Jr: Determination of hydroxypro-line in connective tissue; in Hall DA (ed): TheMethodology of Connective Tissue Research.Oxford, Joynson–Bruvvers, 1976, pp 227–233.

29 Bankowski E, Rzeczycki W, Nowak H, JodczykK: Decrease of collagen biosynthesis ability ofrat kidney fibroblasts transformed with SV-40virus. Mol Cell Biochem 1978;20:77–83.

30 Karna E, Surazynski A, Pałka JA: Collagenmetabolism disturbances are accompanied byan increase in prolidase activity in lung carci-noma planoepitheliale. Int J Exp Path 2000;81:341–347.

31 Pałka JA, Surazynski A, Karna E, Orłowski K,Puchalski Z, Pruszynski K, Laszkiewicz J,Dzienis H: Prolidase activity dysregulation inchronic pancreatitis and pancreatic cancer. He-patogastroenterology 2002;49:1699–1703.

32 Chamson A, Voigtlander V, Myara I, Frey J:Collagen biosynthetic anomalies in prolidasedeficiency: Effect of glycyl-L-proline on thedegradation of newly synthesized collagen. ClinPhysiol Biochem 1989;7:128–136.

33 Myara I, Miech G, Fabre M, Mangeot M,Lemonnier A: Changes in prolinase and proli-dase activity during CCl4 administration in-ducing liver cytolysis and fibrosis in rat. Br JExp Path 1987;68:7–13.

34 Galewska Z, Romanowicz L, Bankowski E, Ja-worski S: Preeclampsia-associated decrease ofpotential collagenolytic and gelatinolytic activ-ities in the wall of the umbilical cord vein. Int JBiochem Cell Biol 2002;34:24–32.

35 Miltyk W, Karna E, Pałka JA: Inhibition ofprolidase activity by non-steroid antiinflam-matory drugs in cultured human skin fibro-blasts. Pol J Pharmacol 1996;48:609–613.

36 Pałka JA, Miltyk W, Karna E, Wołczynski S:Modulation of prolidase activity during in vi-tro aging of human skin fibroblasts the role ofextracellular matrix collagen. Tokai J Exp ClinMed 1996;21:207–213.

37 Pałka JA, Karna E, Miltyk W: Fibroblast che-motaxis and prolidase activity modulation byinsulin-like growth factor II and mannose 6-phosphate. Mol Cell Biochem 1997;168:177–183.

38 Yoon SO, Park SJ, Yun CH, Chung AS: Rolesof matrix metalloproteinases in tumor metasta-sis and angiogenesis. J Biochem Mol Biol 2003;36:128–137.

39 Kleiner DE, Stetler-Stevenson WG: Matrixmetalloproteinases and metastasis. CancerChemother Pharmac 1999;43:S42–S51.

40 Liotta LA, Rao CN: Role of the extracellularmatrix in cancer. Ann NY Acad Sci 1985;460:333–344.

41 Sunada H, Nagai Y: Mechanism of collagendegradation by collagenase: A transition pro-cess of the collagen molecule from collagenasesubstrate to gelatinase substrate. Biomed Res1983;4:61–69.

42 Nakajima M, Chop AM: Tumor invasion andextracellular matrix degradative enzymes: Reg-ulation of activity by organ factors. Semin Can-cer Biol 1991;2:115–127.