275

Molecular and Cellular Biochemistry 254: 275–287, 2003.© 2003 Kluwer Academic Publishers. Printed in the Netherlands.

Identification, purification and partialcharacterization of tissue inhibitor of matrixmetalloproteinase-2 in bovine pulmonary arterysmooth muscle

Malay Mandal, Amritlal Mandal, Sudip Das, Tapati Chakraborti andSajal ChakrabortiDepartment of Biochemistry and Biophysics, University of Kalyani, Kalyani, West Bengal, India

Received 23 January 2003; accepted 14 May 2003

Abstract

Bovine pulmonary artery smooth muscle possesses the tissue inhibitor of matrix metalloproteinase-2 (TIMP-2) as revealed byWestern immunoblot study of its cytosol fraction with bovine polyclonal TIMP-2 antibody. This potent polypeptide inhibitorof matrix metalloproteinases (MMPs) was purified to homogeneity from cytosol fraction of bovine pulmonary artery smoothmuscle. This inhibitor was purified by ammonium sulfate precipitation followed by gelatin sepharose and lentil lectin sepharoseaffinity chromatography and continuous elution electrophoresis by Prep Cell Model 491 (Bio-Rad, USA). SDS-PAGE revealedthat the inhibitor has an apparent molecular mass of 21 kDa and was confirmed as TIMP-2 by (i) Western immunoblot assayusing bovine polyclonal TIMP-2 antibody; and also by (ii) amino terminal amino acid sequence analysis of the purified inhibi-tor is found to be identical with TIMP-2 obtained from other sources. The purified 21 kDa inhibitor was found to be activeagainst matrix metalloproteinase-2 (MMP-2, 72 kDa gelatinase) and matrix metalloproteinase-9 (MMP-9, 92 kDa gelatinase),the ambient MMPs in the pulmonary artery smooth muscle. The inhibitor was also found to be sensitive to the activated 72kDa gelatinase-TIMP-2 complex and also active human interstitial collagenase. By contrast, it was found to be insensitive tothe serine proteases: trypsin and plasmin. The inhibitor was heat and acid resistant and it had the sensitivity to trypsin degra-dation and reduction-alkylation. Treatment of the inhibitor with hydrogen peroxide, superoxide generating system (hypoxan-thine plus xanthine oxidase) and peroxynitrite inactivated the inhibitor. (Mol Cell Biochem 254: 275–287, 2003)

Key words: Tissue inhibitor of matrix metalloproteinase-2, matrix metalloproteinase-2, matrix metalloproteinase-9, collagenase,gelatinase, interstitial collagenase, pulmonary artery smooth muscle, cytosol, oxidant, antioxidant

Abbreviations: TIMP-2 – tissue inhibitor of matrix metalloproteinase-2; MMP-2 – matrix metallopoteinase-2; MMP-9 – martrixmetalloproteinase-9; ECM – extracellular matrix; SMC – smooth muscle cell; HBPS – Hank’s buffered physiological saline;APMA – aminophenylmercuric acetate; ONOO– – peroxynitrite; H

2O

2 – hydrogen peroxide, O

2.– – superoxide radical

Introduction

The interactions of cells with extracellular matrix (ECM) arecritical for the normal development and function of organ-

isms. Modulation of cell-matrix interactions occurs throughthe action of unique proteolytic systems responsible forhydrolysis of a variety of ECM components. The matrixmetalloproteinases (MMPs) are an important family of en-

Address for offprints: S. Chakraborti, Department of Biochemistry and Biophysics, University of Kalyani, Kalyani 741235, West Bengal, India(E-mail: s_chakraborti@hotmail.com)

276

zymes, produced by connective-tissue cells, which can de-grade virtually all the components of ECM. By regulating theintegrity and composition of the ECM structure, MMPs playa pivotal role in the control of signals elicited by matrixmolecules, which regulate growth and development of cells.The turnover and remodeling of ECM must be highly regu-lated since uncontrolled proteolysis contributes to abnormaldevelopment and to the generation of many pathologicalconditions characterized by either excessive degradation ora lack of degradation of ECM components [1–5]. Manypathological conditions have been shown to be associatedwith changes in collagen metabolism. The unrestrained ac-tivity of these MMPs may result in extensive tissue damageand these enzymes have been implicated in a variety of dis-ease processes including the lung diseases, for example,pulmonary hypertension [6–10]. A number of different con-trol mechanisms exist to regulate the extracellular activity ofthese potent enzymes. These include the control of the syn-thesis and the secretion of the enzymes. Maintenance of ex-tracellular matrix architecture is achieved by the balance ofthe action of MMPs with simultaneously elaborated coun-ter regulatory inhibitors, tissue inhibitor of matrix metallo-proteinases (TIMPs) [11, 12].

The inhibitors can be categorized into three types [13]. Onetype includes low molecular weight cataionic proteins (Mr= 11 kDa) partially purified from cartilage and other tissues[14]. A second type consists of α2-macroglobulin, whichaccounts for the major anticollagenase activity of plasma.This glycoprotein inhibits all the matrix metalloproteinasesknown to date [15, 16]. And, the third type consists of bothglycosylated and non-glycosylated proteins of molecularmasses between 20 and 30 kDa known as tissue inhibitor ofmetalloproteinases (TIMPs) [17]. TIMP-2 is a non-glyco-sylated protein with a molecular mass of 21 kDa, in mostcases isolated as a complex with the proenzyme of gelatinase,i.e. latent Mr 72 kDa gelatinase from different culture media[18–20]. It has been found that TIMP-2 is able to form acomplex with Mr 72 kDa gelatinase. Upon activation withAPMA the complex shows gelatinolytic activity, which canbe inhibited by the addition TIMP-2 [20–22]. The regulationof the matrix degradation is very complex, and TIMP-2 playsan important role in controlling the activity of matrix metallo-proteinases in vivo.

DeClerck et al. [18] purified an inhibitor from the culturemedium of bovine aortic endothelial cells and showed that itshared many properties with TIMP-1, it was not glycosylatedand having an apparent molecular mass of 21 kDa. Stetler-Stevenson et al. [19] described a similar inhibitor secreted byhuman melanoma cells, which they named TIMP-2. Aminoterminal amino acid sequences of TIMP-2 have been ob-tained from several sources [19, 23, 24]. A high degree ofhomology exists between TIMP-2s obtained from differentsources.

Our present study indicates that bovine pulmonary arterysmooth muscle possesses a potent inhibitor of matrix metallo-protease, identified as TIMP-2. In this present communica-tion, we described the purification and partial characterizationof TIMP-2 from the cytosol fraction of bovine pulmonaryartery smooth muscle tissue. Additionally, we also determinedthe effect of the oxidants: H

2O

2, O

2.– generating system and

ONOO–, on the inhibitory activity of the purified TIMP-2.

Materials and methods

Materials

Gelatin Sepharose 4B and lentil lectin sepharose were pur-chased from Pharmacia, Upsala, Sweden. [14C]-gelatin and[3H] acetic anhydride were obtained from New England Nu-clear (Wilmington, DE, USA). 125I was obtained from BhabaAtomic Research Centre, Trombay, Mumbai, India. Poly-clonal antibody of lactate dehydrogenase (LDH) (host goat)was obtained from Fitzgerald Industries International, Inc.(Concord, MA, USA) and the rabbit polyclonal antibody toHistone H3 was the product of Abcam Ltd. (Cambridge,UK). Human interstitial collagenase was purchased fromChemicon International Inc. (Temecula, CA, USA). Matrixmetalloproteinase-2, matrix metalloproteinase-9, tissue in-hibitor of metallo-proteinase-2 and their polyclonal anti-bodies were obtained from Chemicon International Inc.(Temecula, CA, USA). All other chemicals used were of ana-lytical grade and are the products of Sigma Chemical Co. (St.Louis, MO, USA).

Methods

All steps in the purification procedure were performed at 4°Cunless otherwise mentioned.

Isolation of bovine pulmonary artery smooth muscle tissue

Bovine pulmonary artery collected from slaughterhouse waswashed several times with Hank’s buffered physiologicalsaline (HBPS) (pH 7.4) and kept at 4°C. The washed pulmo-nary artery was used for further processing within two hoursafter collection. The intimal and the serosal layer (outer layer)were removed and the tunica media i.e. the smooth muscletissue was collected, characterized histologically (Fig. 1), andused for the present studies.

Preparation of smooth muscle tissue cytosol fraction

The smooth muscle tissue was homogenized with a cyclomixerin ice-cold medium containing 0.25 M sucrose and 10 mM

277

Tris HCl buffer, pH 7.4. The homogenate was centrifuged at3000 g for 30 min at 4°C to separate nucleus and cell debris.The supernatant was centrifuged at 20,000 g for 30 min at4°C to sediment mitochondria and lysosomes. The resultingsupernatant was then centrifuged at 100,000 g for 1 h at 4°C.The pellet was discarded, and the supernatant portion wassaved and used as the cytosol fraction [25].

Purity of the cytosolic fraction

The purity of the studied cytosolic fraction was ascertainedby western immunoblot analysis [26] with lactate dehydro-genase as the cytosolic marker [27–29] and the histone H3as nuclear marker [30, 31]. Briefly, the bovine pulmonaryartery smooth muscle cytosol fraction (~ 50 µg protein), waselectrophoresed under reducing condition by SDS-PAGE andtransferred to nitrocellulose membranes. Non-specific pro-tein binding sites on the nitrocellulose papers were blockedby incubating the papers with 3% BSA in tris buffered sa-line (TBS: 10 mM Tris-HCl, pH 7.4 containing 0.9% NaCl)for 40 min with constant shaking at 40°C. These were thenincubated, respectively, with the polyclonal antibody of lac-tate dehydrogenase and the polyclonal antibody of histoneH3 for 2 h at room temperature with constant shaking. Theblots were then washed three times (20 min each) with TBSand incubated separately for 2 h with horseradish peroxidaseconjugated secondary antibodies. After that the blots werewashed three times with TBS (20 min each) followed by de-velopment with 4-chloro-1-naphthol.

Identification of TIMP-2, MMP-2 and MMP-9 inpulmonary artery smooth muscle cytosol fraction

TIMP-2, MMP-2 and MMP-9 were identified in bovine pul-monary artery smooth muscle cytosol fraction by Westernimmunoblot study as described by Towbin et al. [26] withsome modifications. Briefly, the bovine pulmonary arterysmooth muscle tissue cytosol fraction (~ 50 µg protein), pureTIMP-2, pure MMP-2 and pure MMP-9 were electrophoresedin 10% SDS-PAGE. The resolved proteins were then trans-ferred electrophoretically (1 h, 100 V) to nitrocellulose pa-pers (0.2 µm pore size). Non-specific protein binding siteson the nitrocellulose papers were blocked by incubating thepaper with 3% BSA in tris buffered saline (TBS: 10 mM Tris-HCl, pH 7.4 containing 0.09% NaCl) for 40 min with constantshaking at 40°C. These were then incubated with polyclonalrabbit anti (bovine TIMP-2, MMP-2 and MMP-9) IgGs for2 h at room temperature with constant shaking. Then the blotswere washed three times (20 min each) with TBS and incu-bated separately for 2 h with horseradish peroxidase conju-gated goat anti (rabbit IgG) antibodies. The blots were thenwashed three times with TBS (20 min each), and then theblots were developed with 4-chloro-1-naphthol.

Ammonium sulfate precipitation

Solid ammonium sulfate was added to make 70% ammoniumsulfate saturation of the cytosol fraction in constant stirringcondition. After sufficient time, the suspension was centri-fuged at 15,000 g for 1 h. The precipitate appearing at 0–70%saturation was dissolved in gelatin sepharose column bufferA (50 mM Tris-HCl, pH 7.4 containing 0.2 M NaCl, 5 mMCaCl

2, 0.05% Brij 35 and 0.02% NaN

3) and dialyzed thor-

oughly for 24 h against the same buffer with two changes.

Gelatin sepharose chromatography

The clear dialyzed material was applied to a gelatin sepharosecolumn (2 × 15 cm) equilibrated in buffer A at a flow rateof 0.2 ml/min. The flow through material from the gelatinsepharose column was collected and recirculated three timesand the unretained protein was collected. Then the columnwas washed with five column volume of buffer A, followedby buffer A containing 1 M NaCl at a flow rate of 0.7 ml perminute until the absorbance of the eluent returned to back-ground. The proteins bound to the column was then elutedwith buffer A containing 1 M NaCl and 7.5% dimethyl sul-phoxide (DMSO) at a flow rate of 0.5 ml/min and the fractionswere collected as 5 ml per tube. Each fraction was analyzed

Fig. 1. Transverse section through the tunica media of bovine pulmonaryartery showing the characteristic smooth muscle cells (magnification ×200).(Hematoxylin-Eosin preparation). The arrows demonstrate the smooth mus-cle cells.

278

both by gelatin zymography and 12% SDS-PAGE. The frac-tions containing the progelatinase-TIMP-2 complex waspooled together and concentrated by ultrafiltration (Amicon,YM-10 membrane) and thoroughly dialyzed against bufferB (HEPES buffer, pH 7.5).

A portion of the gelatin-sepharose eluate was blotted witha mixture of the MMP-2 and TIMP-2 antibodies in order todetermine the presence of MMP2/TIMP2 complex by follow-ing the procedure as described by Towbin et al. [26].

Lentil lectin sepharose chromatography

To eliminate the traces of 92 kDa gelatinase and other con-taminants present in the gelatin sepharose affinity eluate, thedialyzed material obtained after gelatin sepharose chroma-tography was loaded on a lentil lectin sepharose column (1.6× 10 cm) pre-equilibrated with the dialyzing buffer B. Theflow through material from the column contains the 72 kDagelatinase-TIMP-2 complex free from the 92 kDa gelatinasecontamination. This material was then concentrated by Amiconultrafiltration cell equipped with YM10 membrane. Then theconcentrated material was dialyzed thoroughly against 25 mMtris-HCl buffer, pH 7.5 containing 200 mM NaCl.

Continuous elution electrophoresis

The dialyzed material was treated with 0.2 % SDS for 30 minat 20°C; it was then loaded on a Prep Cell Model 491 (BioRad,USA) for continuous elution electrophoresis. A 10 cm sepa-rating gel of 10% polyacrylamide and 2.5 cm stacking gel of4% polyacrylamide were cast in the 28 mm-diameter gel tubeof the instrument. Electrophoresis was performed at 40 mA(max 300 V) under cooling conditions, with an anode buffer(25 mM Tris, 192 mM Glycine, pH 8.3) at 60 ml per min forabout 9 h. During the run, all proteins were electrophoresedvertically through a cylindrical sieving gel. As individualbands migrated off the bottom of the gel, they passed directlyinto an elution chamber consisting of a thin frit. A dialysismembrane (Mr cut off 6 kDa), directly underneath the elu-tion frit, trapped proteins within the chamber. A peristalticpump pulled the proteins up through the elution tube with acathode buffer (25 mM Tris, 192 mM Glycine, 0.02% Brij35, pH 8.3) at 30 ml/h on to a fraction collector. Fractions of2.5 ml were collected and analyzed by SDS-PAGE. Fractionscontaining the inhibitory 21 kDa protein were pooled andconcentrated by ultrafiltration (Amicon, YM 10 membrane)and stored under liquid nitrogen.

Inhibitor assay

Collagenolytic activity was determined with collagen (ratType I) that was reductively acetylated with [3H]acetic an-

hydride to a specific activity of 5.6 × 106 cpm/mg as previ-ously described [32]. Samples of radiolabelled collagen weremixed with 1 mg/ml of unlabelled collagen and dialyzed into25 mM Tris, 1 mM CaCl

2, 0.02% NaN

3, pH 7.5. Proteinases

were activated for 2–4 h by incubation at 37°C with 1 mMAPMA from a 100 mM stock made with 100 mM NaOH [33].Preincubation of the collagenase and purified bovine 21 kDainhibitor (TIMP-2) for 30 min at 37°C was followed by 2hincubation with collagen substrate at 37°C. Assays were ter-minated by centrifugation for 10 min and was counted [34].For assay of gelatinolytic activity, the radiolabelled gelatinsubstrate was prepared by diluting 20 µl (1.2 µCi) of [14C]-labelled gelatin with 480 µl of 1 mg/ml cold gelatin. Thesubstrate mixture was then heated at 55°C for 25 min andallowed to cool slowly to room temperature. Gelatinases wereactivated for 2 h by incubation at 37°C with 1 mM APMA.Gelatinase activity was measured by preincubating the pureactivated gelatinases with purified bovine TIMP-2 for 30 minat 37°C followed by 2 h incubation with the labelled gelatinsubstrate. The reaction was then stopped by the addition of0.25 M EGTA. Undigested gelatin was precipitated by theaddition of 20% TCA. After chilling on ice for 10 min, sam-ples were centrifuged for 10 min and the supernatant wascounted [35]. One unit is the amount of enzyme that degrades1 µg collagen/gelatin per min at 37°C. One unit of inhibitoryactivity was defined as the amount of inhibitor that inhibits2 units of collagenase by 50%.

Anti-trypsin and anti-plasmin activity was tested using[125I]-fibrin as substrate [36].

SDS polyacrylamide gel electrophoresis

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis(SDS-PAGE) was performed according to the procedure ofLaemmli [37] using a minigel system apparatus (BioRad,USA). A 12% separation gel was used with a 2.5% stackinggel. Electrophoresis was performed at room temperature at16 mA/plate. Protein containing bands were stained with sil-ver [38].

Zymography

Polyacrylamide gels (10 %; 0.75 mm thick) were cast con-taining 0.1% gelatin. Gelatin solution was made up as a 2%stock solution in distilled water and dissolved by heating.Pure 92 kDa gelatinase (MMP-9), gelatinase-TIMP-2 com-plex and pure 72 kDa gelatinase (MMP-2) were incubatedwith 1 mM aminophenylmercury acetate (APMA) for 2 h at37°C and then these were applied to the gel in standardSDS loading buffer containing 0.1% SDS but lacking 2-mercaptoethanol; it was not boiled before loading. The gel

279

was run at 200 V for 1 h at 4°C in a BioRad Mini protean IIapparatus and then soaked in 200 ml of 2.5% triton X-100 indistilled water in a shaker for 1 h with one change after 30min at 25°C. Next, the gel was soaked in reaction buffer (100mM Tris/HCl, 10 mM CaCl

2, pH 8.0) for 12 h at 37°C and

then stained with Coomassie Brilliant Blue, and this was fol-lowed by washing with distilled water for 1 min. The clearzone against the dark Coomassie background indicates pro-tease activity. The marker lane was separated from the gel andthen destained with destaining solution containing methanol:acetic acid:water (4:1:5 by volume) [39].

The purified 21 kDa inhibitor was added to the Triton X-100 soaked gels for 1 h, then incubated in reaction buffer con-taining this inhibitor for 12 h at 37°C. This was then stainedwith Coomassie Brilliant blue followed by washing with dis-tilled water. The inhibition of the clear zone against the darkCoomassie background indicates the activity of TIMP-2.

Synthesis of peroxynitrite (ONOO–)

ONOO– was synthesized from 0.2 M NaNO2 and 1 M H

2O

2

in 0.5 M HCl, and stored in 1.5 M KOH at –70°C as previ-ously described [40]. Excess H

2O

2 in the ONOO– solution was

removed by MnO2. The concentration of ONOO– was deter-

mined by its absorbance at 302 nm in 1.5 M KOH.

Effects of oxidants on purified TIMP-2

Purified TIMP-2 (1 µg/ml) was treated with H2O

2 (1 mM), O

2.–

generating system {HPX (2 mM) +XO (0.03 unit/ml)} andONOO– (300 µM) for 15 min at 37°C. The doses of the agentsare same as have been used previous investigators [41–43].

Protein sequencing

Sample of inhibitor protein was electrophoresed, transferredonto a PVDF membrane [44] and amino acid sequence wasdetermined.

Protein determination

Protein concentrations were estimated using BCA proteinassay kit (Pierce, USA) using bovine serum albumin as thestandard.

Statistical analysis

Data were analyzed by unpaired t-test and the analysis ofvariance followed by the least significant difference [45] for

comparisons within the groups. Statistical significance wasassumed when the p < 0.001.

Results

Histological study of bovine pulmonary artery smooth mus-cle indicates that the tissue has spindle shaped cells whichconfirms that our studied tissue is enriched with typical smoothmuscle cells (Fig. 1). Immunoblot study with polyclonal anti-body of lactate dehydrogenase, a cytosolic marker, confirmedthat our studied fraction is the cytosol fraction (Fig. 2A).Additionally, immunoblot study of the cytosol fraction withpolyclonal histone H3 antibody, a well known nuclear marker,did not elicit any immunoreactive band indicating that ourstudied cytosol fraction is free from contaminating nucleus(Fig. 2B). Western immunoblot study confirmed that thecytosolic fraction of bovine pulmonary artery smooth mus-cle possesses TIMP-2, MMP-2 and MMP-9 (Fig. 3). For thepurification of TIMP-2, we used cytosol fraction of bovinepulmonary artery smooth muscle, which contains 1452 mgof protein. Proteins in the cytosolic fraction were first con-centrated by 0–70% ammonium sulfate precipitation tech-nique. Then the 72 kDa gelatinase-TIMP-2 complex wasisolated by gelatin-sepharose affinity chromatography (Fig.4A). SDS-PAGE of the fractions obtained from the gelatinsepharose chromatography was shown in Fig. 4B. Immunoblotstudy of the gelatin sepharose eluate with a mixture of twodifferent polyclonal antibodies to the MMP-2 (rabbit anti(bovine MMP-2) IgG) and the TIMP-2 (rabbit anti (bovine

Fig. 2. (A) Western immunoblot of cytosolic fraction of bovine pulmonarysmooth muscle with polyclonal antibody of lactate dehydrogenase (obtainedfrom Fitzgerald Industries International, Inc., Concord, MA, USA). Lane 1,cytosolic fraction; lane 2, molecular mass standards. (B) Western immunoblotof cytosolic fraction of bovine pulmonary smooth muscle with polyclonalantibody of histone H3 (obtained from Abcam Ltd., Cambridge, UK). Lane1, cytosolic fraction; lane 2, molecular mass standards.

280

TIMP-2) IgG) revealed two immunoreactive bands at theirrespective positions in the blot (Fig. 4C). As free TIMP-2 cannot bind to gelatin-sepharose, the above immunoreactivebands suggest that a complex of the 72 kDa gelatinase andTIMP-2 may present in the system as a complex. The gelatinsepharose eluate was then subjected to lentil lectin sepharosechromatography to eliminate contaminating proteins of highmolecular weight, which bind to the column. The lentil lec-tin sepharose flow though containing the 72 kDa gelatinase-TIMP-2 complex was collected. An attempt to isolate the 21kDa inhibitory protein from its complexed form by sephacrylS-200 gel filtration chromatography was performed (data notshown) but the effort has not been successful. We, therefore,proceeded to purify TIMP-2 from its complexed form (72kDa gelatinase-TIMP-2 complex) by continuous elution elec-trophoresis on a Prep Cell Model 491 (Bio-Rad, USA). Thematerial containing 72 kDa gelatinase-TIMP-2 complex wasdialysed against the buffer (25 mM Tris-HCl, pH 7.5 contain-ing 200 mM NaCl). The dialyzed material was then treatedwith 0.2% SDS for 30 min at 22°C and loaded on the PrepCell for continuous elution electrophoresis to isolate the 21kDa inhibitor (TIMP-2) from the complex. The elution pro-file of the continuous elution and the SDS-PAGE of the elutedfractions were shown in Figs 5A and 5B, respectively.

At each stage of the purification procedure, samples weretaken and assayed for inhibitory activity against the humaninterstitial collagenase and for the determination of protein

content. The results were shown in Table 1. The purified in-hibitor had a final specific activity of 925 units/mg. Only0.5% of the original inhibitory activity was recovered by thepurification procedure. The original estimates of inhibitoryactivity apparently appear to include the activities of otherambient protease inhibitors whose purification was not shownhere. Further, the loss of inhibitory activity as observed dur-ing purification procedure could be due to the fact that theinhibitor rapidly loses its activity in dilute solutions.

The purity at each step of the purification was verified bySDS-PAGE (Fig. 6). This purified inhibitor with a molecularweight 21 kDa was clearly identified without any other pro-teins being detected in the silver stained 12% SDS-PAGE (Fig.6). Western immunoblot study confirmed that the purified 21kDa inhibitory protein is the TIMP-2 (Fig. 7). A 10% SDS-PAGE of the purified 21 kDa inhibitor was electrophoreticallytransferred onto a PVDF membrane. The 21 kDa inhibitor bandwas cut out from this membrane and subsequently used fordetermining its amino acid sequence data. This amino-termi-nal sequence of the first 22 amino acids of the purified 21 kDainhibitor, bovine TIMP-2 revealed 95% identity to those res-idues of rat and human TIMP-2 [19, 23, 24] (Table 2).

Specificity and stability

The purified inhibitor blocked the activities of human inter-stitial collagenase, activated 72 kDa gelatinase as well as theactivated 92 kDa gelatinase (Table 3, Figs 8A and 8B). It wasfound that the purified TIMP-2 is most effective to inhibit theactivity of 72 kDa gelatinase than that of the 92 kDa gelatinaseand collagenase. The purified bovine TIMP-2 also inhibitsthe APMA activated 72 kDa-TIMP-2 complex (Table 3, Fig.8). The inhibitor was found to be inactive against the serineprotease trypsin and plasmin (Table 3).

The purified 21 kDa inhibitor (TIMP-2) was tested for theirsensitivity to heat, trypsin, acid and reduction-alkylation. Thetreatment conditions and the inhibitory activity of the puri-fied TIMP-2 were shown in Table 4. The inhibitor is relativelyheat resistant up to 30 min at 100°C. This inhibitor has beenfound to be acid resistant up to pH 2. Treatment of the inhibi-tor with 10-fold excess of trypsin resulted in a complete lossof activity. Reduction of the inhibitor with 2-mercaptoethanoland alkylation with iodoacetamide destroyed the activity,indicating that intact sulfhydryl groups are required for ac-tivity. Finally APMA failed to reverse the inhibitory activityof the inhibitor, an observation which indicated that the in-hibitor forms a tight complex with the enzyme (Table 4).

Oxidative inactivation of TIMP-2

The purified TIMP-2 was shown to be inactivated by theoxidants, H

2O

2 (1 mM), O

2.– generating system {HPX (2 mM)

Fig. 3. (A) Identification of tissue inhibitor of metalloproteinase-2 (TIMP-2) in bovine pulmonary artery smooth muscle cytosol by Western immunoblot.Lane 1, control TIMP-2 (pure TIMP-2 purchased from Chemicon Interna-tional Inc., Temecula, CA, USA); lane 2, bovine pulmonary vascular arterysmooth muscle cytosol. (B) Identification of 72 kDa gelatinase (MMP-2, gelatinase A) in bovine pulmonary artery smooth muscle cytosol byimmunoblot assay. Lane 1, control MMP-2 (pure MMP-2 purchased fromChemicon International Inc., Temecula, CA, USA); lane 2, bovine pulmo-nary vascular artery smooth muscle cytosol. (C) Immunoblot assay for theidentification of 92 kDa gelatinase (MMP-9, gelatinase B) in cytosol ofbovine pulmonary artery smooth muscle. Lane 1, control MMP-9 (PureMMP-9 purchased from Chemicon International Inc., Temecula, CA, USA);lane 2, bovine pulmonary vascular artery smooth muscle cytosol.

281

Fig. 4. (A) Gelatin-Sepharose chromatography; the 0–70% ammonium sulfate precipitated protein suspension was dialysed thoroughly against 50 mM Tris-HCl, pH 7.4 containing 0.5 M NaCl, 5 mM CaCl

2, 0.05% Brij 35 and 0.02% NaN

3 and loaded on gelatin sepharose affinity column. The bound progelatinase-

TIMP-2 complex was eluted from the column with 7.5% DMSO. Fractions (5 ml each) were monitored for protein (A280

–). The fractions containing the 72kDa gelatinase-TIMP-2 complex exhibiting the inhibitory activity was also shown (–•–). (B), 12% SDS-PAGE of the fractions obtained during gelatin sepharosechromatography. The numbers under each lane denotes the fraction number and M indicates the lane for molecular mass standards. (C) Immunoblot study ofthe gelatin sepharose eluate with a mixture of two different polyclonal antibodies to the MMP-2 (rabbit anti (bovine MMP-2) IgG) and the TIMP-2 (rabbitanti (bovine TIMP-2) IgG). Lane 1, control MMP-2 (Pure bovine MMP-2 obtained from Chemicon International Inc., Temecula, CA, USA); lane 2, controlTIMP-2 (Pure bovine TIMP-2 obtained from Chemicon International Inc., Temecula, CA, USA); lane 3, gelatin sepharose affinity eluate; lane 4, molecularmass standards.

A

B

C

282

+XO (0.03 unit/ml)}, and ONOO– (300 µM) (Fig. 9). Pre-treatment with the antioxidant, vitamin E (1 mM) preventedthe effect of the oxidants on TIMP-2 activity (Fig. 9).

Discussion

We have demonstrated that the cytosolic fraction of bo-vine pulmonary artery smooth muscle possesses the tissueinhibitor of matrix metalloproteinases-2 (TIMP-2), matrixmetalloproteinase-2 (MMP-2, 72 kDa gelatinase) and ma-trix metalloproteinase-9 (MMP-9, 92 kDa gelatinase). Thepurification procedure described in this present communica-tion resulted in the isolation of TIMP-2 from the cytosolicfraction of smooth muscle to homogeneity. Although TIMP-2 has been purified from different sources such as fibroblasts,aortic endothelial cells, human melanoma cells and rheuma-toid synovial fluid [18, 46–48] yet no report has, however,been available for the purification of TIMP-2 from pulmo-nary artery smooth muscle tissue. This prompted us to un-dertake the present research.

In the present study, a large amount of bovine pulmonaryartery smooth muscle tissue was required to purify TIMP-2.

A

B

Fig. 5. (A) Elution profile of continuous elution electrophoresis by PrepCell Model 491 (BioRad, USA). (–), indicates absorbance at 280 nm; (–•–), indicates % inhibitory activity. (B) 12% SDS-PAGE of the fractions ob-tained during continuous elution electrophoresis by Prep Cell Model 491(BioRad, USA). The numbers under each lane denotes the fraction numberand M indicates the lane for molecular mass standards.

Fig. 6. 12% SDS-PAGE of purification profile of the 21 kDa inhibitor frombovine pulmonary artery smooth muscle tissue. Lane 1, bovine pulmonaryvascular artery smooth muscle cytosolic fraction; lane 2, suspension of 0–70% ammonium sulfate precipitaton; lane 3, Gelatin sepharose chromatog-raphy eluate; lane 4, lentil lectin sepharose flow through; lane 5, Prep CellModel 491 (BioRad, USA) eluate; lane 6, molecular mass standards.

Table 1. Purification of 21 kDa bovine pulmonary artery smooth muscle tissue inhibitor of metalloproteases

Stage Protein Inhibitor Specific activity Yield Purification factor(mg) (units) (unit/mg) (%)

Cytosol 1452 6924 4.77 100 10–70% (NH

4)

2SO

4 precipitation 474 3377 7.12 49 1.5

Gelatin-Sepharose 1.08 408 377.78 6 79Continuous elution electrophoresis 0.04 37 925 0.5 194

(by Prep Cell Model 491)

283

Table 3. Inhibitory activity of the purified TIMP-2 on collagenase (human interstitial collagenase), gelatinases (MMP-2 and MMP-9) and the serine proteases,trypsin and plasmin

Enzyme Substrate Purified TIMP-2 added Inhibition (%)

Human interstitial collagenase Radiolabelled collagen 5 nM 0(3.5 µg/ml) ([3H]collagen) 10 nM 8

15 nM 7920 nM 9825 nM 98

APMA-activated 72 kDa Radiolabelled gelatin 0.5 nM 20gelatinase-TIMP-2 complex ([14C]gelatin) 1 nM 39(9 mg/ml) 10 nM 85

25 nM 89

APMA-activated 72 kDa Radiolabelled gelatin 0.1 nM 29gelatinase (2.5 µg/ml) ([14C]gelatin) 1 nM 75

5 nM 9510 nM 99

APMA-activated 92 kDa Radiolabelled gelatin 0.1 nM 2 gelatinase (2.5 µg/ml) ([14C]gelatin) 1 nM 37

10 nM 9320 nM 97

Trypsin0.1 µg 125I-fibrin 1.0 µg 01.0 µg 1.0 µg 0

Plasmin0.1 µg 125I-fibrin 1.0 µg1.0 µg 1.0 µg 0

Results are mean of 4 sets of experiments. Samples of pure collagenase and gelatinases were activated by 2 h incubation with 1 mM APMA at 37°C followedby incubation with purified TIMP-2 for 30 min at 37°C. After that these were incubated with labelled collagen and gelatin, respectively for 2 h. The reactionwas stopped by the addition of TCA and the acid soluble peptides were counted. For the serine proteases, purified TIMP-2 was incubated with trypsin andplasmin at 22°C for 15 min and then the mixture was added to the indicated substrate and incubated for 4 h at 37°C.

Table 2. NH2-terminal sequences of tissue inhibitor of matrix metallo-

proteinases-2 (TIMP-2)

TIMP-2 purified from C-S-C-S-P-V-H-P-Q-Q-A-F-C-N-A-D-I-V-I-R-A-K

bovine pulmonary artery

smooth muscle tissue

cytosol fraction

Rat TIMP-2a C-S-C-S-P-V-H-P-Q-Q-A-F-C-N-A-D-V-V-I-R-A-K

Human TIMP-2b,c C-S-C-S-P-V-H-P-Q-Q-A-F-C-N-A-D-V-V-I-R-A-K

aRef. [23] in the text; bRefs [19, 24] in the text.

Fig. 7. Confirmation of the purified 21 kDa inhibitory protein isolatedfrom bovine pulmonary vascular artery smooth muscle cytosolic fractionas TIMP-2 by Western immunoblot. Lane 1, pure TIMP-2 (purchased fromChemicon International Inc., Temecula, CA, USA); lane 2, purified 21 kDainhibitory protein.

A proportion of the inhibitor that forms complex with the72 kDa gelatinase was considered in our study. Only 40 µgof protein was finally recovered. The amount of the inhibi-tor present in normal cells and tissues is very low. In contrastwith the other investigators, who used conditioned culturemedium of various types of cells and tissues to purify TIMP-2 [18, 19, 23, 24, 46–48], we used cytosolic fraction of bo-

284

fied inhibitor with polyclonal bovine TIMP-2 antibody re-vealed the presence of an immunoreactive band having anapparent molecular mass of 21 kDa (Fig. 7), which suggeststhat the purified inhibitor is TIMP-2. Secondly, the aminoterminal sequence analysis of the first 22 amino acids of thepurified 21 kDa inhibitor (Table 2) shows its resemblancewith TIMP-2 obtained from other sources [19, 23, 24]. Thirdly,in agreement with previous studies that 72 kDa progelatinaseis secreted as a complex with TIMP-2 in a variety of systems[19–21], our present study also indicates that TIMP-2 remainsas a complex with the 72 kDa gelatinase (Fig. 4C). Fourthly,the 72 kDa progelatinase-TIMP-2 complex elicits gelatinaseactivity when activated with the organomercurial, APMA(Fig. 8) which is in agreement with the previous study ofKolkenbrock et al. [21]. Additionally, the activity of the APMAtreated complex has been found to be inhibited by the puri-fied TIMP-2 (Table 3, Fig. 8). Fifthly, the purified 21 kDainhibitor inhibits APMA activated 72 kDa gelatinase (MMP-2) and also the APMA-activated 92 kDa gelatinase (MMP-9) as assessed by 14C-labelled gelatin degradation and alsoby gelatin zymogram assays (Table 3, Fig. 8). It is evidentfrom Table 3 that the purified inhibitor preferentially inhib-its the activated MMP-2 to that of MMP-9, which is a wellknown characteristics of TIMP-2 [49]. Sixthly, like the TIMP-

B

A Table 4. Stability of bovine pulmonary artery smooth muscle TIMP-2

Conditions 14C-collagen degraded Inhibition(cpm ± S.E.) (%)

No treatment 0* 100

Heata

37°C, 30 min 29 ± 3* 9950°C, 30 min 0* 10080°C, 30 min 0* 100100°C, 30 min 156 ± 22* 95

Trypsinb

1 µg 498 ± 49* 8410 µg 3021 ± 83 3

Acid (pH 2) 280 ± 41* 91Reduction-Alkylationc 3098 ± 86 0.5APMA (1 mM)d 0* 100No inhibitor 3114 ± 94 –

(0.5 µg of TIMP-2 was incubated with 1 mM of APMA activated collagenase(human interstitial collagenase) (3.5 µg/ml) in a total volume of 1 ml for30 min at 37°C followed by addition of labelled collagen and incubated for2 h at 37°C. The reaction was stopped by the addition of TCA and the acidsoluble peptides were counted by a scintillation counter). Results are mean± S.E. (n = 4). *p < 0.001 compared with ‘no inhibitor’ value. aTIMP-2 wasincubated at the indicated temperatures for 30 min. Loss of inhibitory ac-tivity was calculated in comparison with the untreated sample. bSampleswere incubated for 1 h at 37°C. The reaction was then blocked with the addi-tion of 5-fold excess of soybean trypsin inhibitor. cSamples were reduced bythe addition of 2-mercaptoethanol (20 mM) for 16 h at 4°C and alkylationwith iodoacetamide (20 mM) at 25°C for 1 h. dAPMA (1 mM) was added tothe collagenase-inhibitor mixture and incubated for 1 h at 37°C.

Fig. 8. (A) Zymogram (10%) showing the inhibitory activity of the puri-fied TIMP-2 from cytosol fraction bovine pulmonary vascular smoothmuscle. Lane 1, APMA-activated 92 kDa gelatinase; lane 2, APMA-acti-vated 92 kDa gelatinase incubated with purified TIMP-2; lane 3, APMAactivated 72-kDa gelatinase-TIMP-2 complex; lane 4, APMA activated72-kDa gelatinase-TIMP-2 complex incubated with purified TIMP-2; lane5, APMA activated 72 kDa gelatinase; lane 6, APMA-activated 72 kDagelatinase incubated with purified TIMP-2; lane 7, molecular weight stand-ard. (B) Histographic representation showing the inhibitory activity of thepurified TIMP-2 of APMA activated MMP-9 (92 kDa gelatinase), MMP-2/TIMP-2 complex and MMP-2 (72 kDa gelatinase) in gelatin zymogram.Areas of lysis were measured by laser scanning densitometry. (a) Percentinhibition by the purified TIMP-2 on APMA activated MMP-9. Gelatin lysisby the activated MMP-9 is considered as control (zero %); (b) percent in-hibition by the purified TIMP-2 on APMA activated MMP-2/TIMP-2 com-plex. Gelatin lysis by the activated complex is considered as control (zero%); (c) percent inhibition by the purified TIMP-2 on APMA activated MMP-2. Gelatin lysis by the activated MMP-2 is considered as control (zero %).

vine pulmonary artery smooth muscle. To our knowledge, thisis the first description of purification of TIMP-2 from cytosolicfraction of bovine pulmonary artery smooth muscle tissue.

Several lines of evidence suggest that the purified 21 kDainhibitor is the TIMP-2. First, immunoblot study of the puri-

285

2 (as revealed from other sources [20]), the 21 kDa inhibitoris resistant to heat and acid treatment and also sensitive toreduction followed by alkylation. Seventhly, the purified 21kDa inhibitor does not inhibit the serine proteases, trypsin andplasmin, which is a typical characteristics of TIMP-2 [20, 50].These seven pieces of evidence strongly support that the 21kDa purified inhibitor is TIMP-2.

A significant observation was that this inhibitor was ac-tive against the collagenase and the gelatinases (both 72 kDagelatinase and 92 kDa gelatinase), the metalloproteinaseswith specific degradative activity for gelatin (constituent ofbasement membranes). These enzymes were considered toplay a role in proliferation and migration of smooth musclecells (SMCs) into the intima [51–54]. Treatment of TIMP-2inhibitor has been shown to cause blockage of SMC migra-tion in injured arteries [52, 54]. Previous studies have alsodemonstrated that this inhibitor possesses biological impor-tance in protecting vascular basement membrane from deg-radation by invasive tumor cells [6, 7, 55, 56]. TIMP-2 may

also play a major role in the regulation of these enzymes andmay serve to suppress tumor cell invasion [6, 7, 47]. Theobservations that both free native and recombinant 72 kDagelatinases [21, 22, 57] are significantly more active than theactivated complex suggests that degradation of ECM in vivomay be more efficiently accomplished by generating a 72kDa gelatinase free of TIMP-2. Although the physiologicalmechanisms are still unknown, this could be achieved byinactivating TIMP-2 from the complex, by an enhanced pro-duction of free enzyme and/or by a down-regulation of TIMP-2 expression [58]. The measurement of the relative levels ofTIMP-2 and the 72 kDa gelatinase and the identification ofactivated species in both normal and pathological situationswill help to define the role of the 72 kDa gelatinase in a moreprecise manner. Thus, MMP inhibition by TIMP-2 may pro-vide a novel strategy to ameliorate several disease processes,for example, pulmonary hypertension, and further investiga-tion is required in this direction.

Oxidative stress has been implicated in many disease proc-esses including reperfusion injury. In ischemic reperfusion,significant amounts of oxidants such as H

2O

2, superoxide

(O2

.–) and peroxynitrite are produced [59–61]. Proteolyticprocesses play important roles in experimentally induced andphysiologically occurring changes in cells and tissues [62].The strongest evidence for the participation of proteasescomes from the studies with protease inhibitors [63]. Oxidantshave been demonstrated to produce pulmonary hypertensionin isolated lung [64]. In isolated lung, protease inhibitors havebeen shown to prevent pulmonary hypertension caused byoxidants [65, 66]. Oxidants, such as ONOO- and HOCl havebeen shown to inactivate the tissue inhibitors of metallo-protease-1 [43, 67]. Our present study suggests that oxidantsinactivate the inhibitory activity of TIMP-2. Thus, oxidativeinactivation of TIMP-2, as observed in our present study, couldperturb the protease-antiprotease (MMP-2 and TIMP-2) bal-ance leading to an increase in the activity of MMP-2 whichsubsequently plays a significant role in different lung diseases,for example pulmonary hypertension [10].

Acknowledgement

Financial assistance from the Indian Council of Medical Re-search (ICMR), New Delhi, Life Science Research Board(LSRB), Ministry of Defense, Govt. of India and the Depart-ment of Biotechnology (DBT), Government of India are grate-fully acknowledged.

References

1. Das S, Chakraborti T, Mandal M, Mandal A, Chakraborti S: Role ofmembrane-associated Ca2+ dependent matrix metalloprotease-2 in the

Fig. 9. Effects of oxidants, H2O

2 (1 mM) O

2.– generating system (2 mM HPX

+ 0.03 unit/ml XO), and ONOO– (300 µM), on the inhibitory activity of thepurified TIMP-2. Pure gelatinase A (MMP-2, 2.5 µg/ml) and gelatinase B(MMP-9, 2.5 µg/ml) were activated for 2 h incubation with 1 mM APMAat 37°C followed by incubation with TIMP-2 (1 µg/ml), or TIMP-2 (1 µg/ml) treated with oxidants in presence or absence of the antioxidant, vita-min E (1 mM). After that 14C-labelled gelatin was added and incubated for2 hs at 37°C. Reactions were stopped by addition of TCA and the acid solu-ble peptides were counted by scintillation counter. Results are mean ± S.E.(n = 4). *p < 0.001 compared with ‘no inhibitor’ value.

286

oxidant activation of Ca2+ ATPase by tertiary butylhydroperoxide. MolCell Biochem 237: 85–93, 2002

2. Mandal M, Chakraborti T, Das S, Mandal A, Chakraborti S: Matrixmetalloprotease-2 mediated activation of Ca2+ ATPase by superoxideradical (O

2.–) in plasma membrane of bovine pulmonary vascular

smooth muscle. Indian J Biochem Biophys 39:390–396, 20023. Chakraborti T, Das S, Mandal M, Mandal A, Chakraborti S: Role of

Ca2+ dependent matrix metalloprotease-2 in stimulating Ca2+ ATPaseactivity under peroxynitrite treatment in pulmonary vascular smoothmuscle plasma membrane. IUBMB Life 53: 167–173, 2002

4. Murphy G, Cawston TE, Galloway WA, Barnes MJ, Bunning RA,Mercer E, Reynolds JJ, Burgeson RE: Metalloproteinases from rabbitbone culture medium degrade types IV and V collagens, laminin andfibronectin. Biochem J 199: 807–811, 1981

5. Nagase H, Woessner JF Jr: Matrix metalloproteinases. J Biol Chem 274:21491–21504, 1999

6. Giannelli G, Antonaci S: Gelatinases and their inhibitors in tumor me-tastasis: From biological research to medical applications. HistolHistopathol. 17: 339–345, 2002

7. Johansson N, Ahonen M, Kahari VM: Matrix metalloproteinases intumor invasion. Cell Mol Life Sci 57: 5–15, 2000

8. Chang C, Werb Z: The many faces of metalloproteases: Cell growth,invasion, angiogenesis and metastasis. Trends Cell Biol 11: S37–S43,2001

9. Corbel M, Boichot E, Lagente V: Role of gelatinases MMP-2 andMMP-9 in tissue remodeling following acute lung injury. Braz J MedBiol Res 33: 749–754, 2000

10. Ohbayashi H: Matrix metalloproteinases in lung diseases. Curr ProtPep Sci 3: 409–421, 2002

11. Woessner JF Jr: Matrix metalloproteinases and their inhibitors in con-nective tissue remodeling. FASEB J 5: 2145–2154, 1991

12. Dayer JM, de Rochemonteix B, Burrus B, Demczuk S, Dinarello CA:Human recombinant interleukin 1 stimulates collagenase and prostag-landin E2 production by human synovial cells. J Clin Invest 77: 645–648, 1986

13. Wooley DE: In: R.E. Piez, A.H. Reddi (eds). Extracellular Matrix Bio-chemistry. Elsevier, New York, 1984, pp 119–159

14. Eisenstein R, Sorgente N, Soble LW, Miller A, Kuettner KE: The re-sistance of certain tissues to invasion: Penetrability of explanted tis-sues by vascularized mesenchyme. Am J Pathol 73: 765–774, 1973

15. Travis J, Salvesen GS: Human plasma proteinase inhibitors. Annu RevBiochem 52: 655–709, 1983

16. Birkedal-Hansen H, Moore WGI, Bodden MK, Windsor LJ, Birkedal-Hansen B, DeCarlo A, Engler JA: Matrix metalloproteinases: A review.Crit Rev Oral Biol Med 4: 197–250, 1993

17. Woessner JF Jr: Matrix metalloproteinase inhibition. From the Jurassicto the third millennium. Ann NY Acad Sci USA 878: 388–403, 1999

18. De Clerck YA, Yean TD, Ratzkin BJ, Lu HS, Langley KE: Purifica-tion and characterization of two related but distinct metalloproteinaseinhibitors secreted by bovine aortic endothelial cells. J Biol Chem 264:17445–17453, 1989

19. Stetler-Stevenson WG, Krutzsch HC, Liotta LA: Tissue inhibitor ofmetalloproteinase (TIMP-2). A new member of the metalloproteinaseinhibitor family. J Biol Chem 264: 17374–17378, 1989

20. Murphy G, Willenbrock F: Tissue inhibitors of matrix metallo-endopeptidases. Meth Enzymol 248: 496–510, 1995

21. Kolkenbrock H, Orgel D, Hecker-Kia A, Niack W, Ulbrich N: Thecomplex between a tissue inhibitor of metalloproteinases (TIMP-2) and72-kDa progelatinase is a metalloproteinase inhibitor. Eur J Biochem198: 775–781, 1991

22. Howard EW, Bullen EC, Banda MJ: Regulation of the autoactivation ofhuman 72-kDa progelatinase by tissue inhibitor of metalloproteinases-2. J Biol Chem 266: 13064–13069, 1991

23. Roswit WT, McCourt DW, Partridge NC, Jeffrey JJ: Purification andsequence analysis of two rat tissue inhibitors of metalloproteinases.Arch Biochem Biophys 292: 402–410, 1992

24. Goldberg GI, Marmer BL, Grant GA, Eisen AZ, Wilhelm S, He CS:Human 72-kilodalton type IV collagenase forms a complex with a tis-sue inhibitor of metalloproteases designated TIMP-2. Proc Natl AcadSci USA 86: 8207–8211, 1989

25. Kamp HH, Wirtz KW: Phosphatidylcholine exchange protein from beefliver. Meth Enzymol 32: 140–146, 1974

26. Towbin H, Staehelin T, Gordon J: Electrophoretic transfer of proteinsfrom polyacrylamide gels to nitrocellulose sheets: Procedure and someapplications. Proc Natl Acad Sci 76: 4350–4354, 1979

27. Baer HP, Vriend RA. Cytosolic enzyme leakage from isolated smoothmuscle preparations. Can J Physiol Pharmacol 63: 164–165, 1985

28. Bausback HH, Ward PE: Vascular, post proline cleaving enzyme: Met-abolism of vasoactive peptides. Adv Exp Med Biol 198: 397–404, 1986

29. Tenu JP, Etore F, Le Doan T: A simple method for the study of cytosoliccontent of oligonucleotides in cells. CR Acad Sci III 320: 477–486, 1997

30. Reiss K, Meggs LG, Li P, Olivetti G, Capasso JM, Anversa P: Up-regulation of IGF1, IGF1-receptor, and late growth related genes inventricular myocytes acutely after infarction in rats. J Cell Physiol 158:160–168, 1994

31. Juan G, Traganos F, Darzynkiewicz Z: Histone H3 phosphorylation inhuman monocytes and during HL-60 cell differentiation. Exp Cell Res246: 212–20, 1999

32. Mookhtiar KA, Mallya SK, Van Wart HE: Properties of radiolabeledtype I, II, and III collagens related to their use as substrates in colla-genase assays. Anal Biochem 158: 322–33, 1986

33. Stetler-Stevenson WG, Krutzsch HC, Wacher MP, Margulies IM, LiottaLA: The activation of human type IV collagenase proenzyme. Sequenceidentification of the major conversion product following organomer-curial activation. J Biol Chem 264: 1353–1356, 1989

34. Cawston TE, Barrett AJ: A rapid and reproducible assay for collagenaseusing [1-14C] acetylated collagen. Anal Biochem 99: 340–345, 1979

35. Brown PD, Kleiner DE, Unsworth EJ, Stetler-Stevenson WG: Cellu-lar activation of the 72 kDa type IV procollagenase TIMP-2 complex.Kidney Int 43: 163–170, 1993

36. Laug WE: Secretion of plasminogen activators by cultured bovineendothelial cells: Partial purification, characterization and evidence formultiple forms. Thromb Haemost 45: 219–224, 1981

37. Laemmli UK: Cleavage of structural proteins during the assembly ofthe head of bacteriophage T4. Nature (London) 227: 680–685, 1970

38. Wray W, Boulikas T, Wray VP, Hancock R: Silver staining of proteinsin polyacylamide gels. Anal Biochem 118: 197–203, 1981

39. Billings PC, Habres JM, Liao DC, Tuttle SW: Human fibroblasts con-tain a proteolytic activity which is inhibited by the Bowman-Birk pro-tease inhibitor. Cancer Res 51: 5539–5543, 1991

40. Radi P, Beckam JS, Bush KM, Freeman BA: Peroxynitrite oxidationof sulfhhydryl. The cytotoxic potential of superoxide and nitric oxide.J Biol Chem 266: 4244–4250, 1991

41. Chakraborti S, Chakraborti T: Down regulation of protein kinase Cattenuates the oxidant H

2O

2-mediated activation of phospholipase A

2

in pulmonary smooth muscle cells. Cell Signalling 7: 75–83, 199542. Chakraborti T, Ghosh SK, Michael JR, Chakraborti S: Role of an apro-

tinin sensitive protease in the stimulation of Ca2+ ATPase by superoxideradical (O

2.–) in microsomes of pulmonary smooth muscle. Biochem J

317: 885–890, 199643. Frears ER, Zhang Zhi, Blake DR, O’Connell JP, Winyard PG: Inacti-

vation of tissue inhibitor of metalloproteinase-1 by peroxynitrite. FEBSLett 381: 21–24, 1996

44. Matsudaira P. Sequence from picomole quantities of proteins electro-blotted onto polyvinylidene difluoride membranes. J Biol Chem 262:10035–10038, 1987

287

45. David WW: In: Biostatistics: A Foundation for Analysis in HealthScience. Wiley, New York, 1978, p 219

46. Stefansson S, Aimes RT, Seward NB, Alexander DS, Quigley JP: Na-tive TIMP-free 70 kDa progelatinase (MMP-2) secreted at elevatedlevels by RSV transformed fibroblasts. J Cell Physiol 161: 419–428,1994

47. Stetler-Stevenson W, Krutzsch HC, Liotta LA: TIMP-2: Identificationand characterization of a new member of the metalloproteinase inhibi-tor family. Matrix 1(suppl): 299–306, 1992

48. Osthues A, Knauper V, Oberhoff R, Reinke H, Tschesche H: Isolationand characterization of tissue inhibitors of metalloproteinases (TIMP-1 and TIMP-2) from human rheumatoid synovial fluid. FEBS Lett 296:16–20, 1992

49. Olson MW, Gervasi DC, Mobashery S, Fridman F: Kinetic analysisof the binding of human matrix metalloproteinase-2 and -9 to tissueinhibitor of metalloproteinase (TIMP)-1 and TIMP-2. J Bil Chem 272:29975–29983, 1997

50. DeClerck YA, Yean TD, Lee Y, Tomich JM, Langley KE: Characteriza-tion of the functional domain of tissue inhibitor of metalloproteinases-2 (TIMP-2). Biochem J 289: 65–69, 1993

51. Ben-Yosef Y, Lahat N, Shapiro S, Bitterman H, Miller A: Regulationof endothelial matrix metalloproteinase-2 by hypoxia/reoxygenation.Circ Res 90: 784–791, 2002

52. Rauch BH, Bretschneider E, Braun M, Schror K: Factor Xa releasesmatrix metalloproteinase-2 (MMP-2) from human vascular smoothmuscle cells and stimulates the conversion of pro-MMP-2 to MMP-2.Role of MMP-2 in factor Xa-induced DNA synthesis and matrix inva-sion. Circ Res 90: 1122–1127, 2002

53. Chou FP, Chu SC, Cheng MC, Yang SF, Cheung WN, Chiou HL, HsiehYS: Effect of hemodialysis on the plasma level of type IV collagenasesand their inhibitors. Clin Biochem 35: 383–388, 2002

54. Cho A, Reidy MA: Matrix metalloproteinase-9 is necessary for theregulation of smooth muscle cell replication and migration after arte-rial injury. Circ Res 91: 845–851, 2002

55. Thorgeirsson UP, Liotta LA, Kalebic T, Margulies IM, Thomas K, Rios-Candelore M, Russo RG: Effect of natural protease inhibitors and achemoattractant on tumor cell invasion in vitro. J Natl Cancer Inst 69:1049–1054, 1982

56. Mignatti P, Robbins E, Rifkin DB: Tumor invasion through the human

amniotic membrane: Requirement for a proteinase cascade. Cell 47:487–498, 1986

57. Okada Y, Morodomi T, Enghild JJ, Suzuki K, Yasui A, Nakanishi I,Salvesen G, Nagase H: Matrix metalloproteinase 2 from human rheu-matoid synovial fibroblasts. Purification and activation of the pre-cursor and enzymic properties. Eur J Biochem 194: 721–730, 1990

58. Fridman R, Bird RE, Hoyhtya M, Oelkuct M, Komarek D, Liang C-M, Berman ML, Liotta L, Stetler-Stevenson WG, Fuerst TR: Expres-sion of human recombinant 72 kDa gelatinase and tissue inhibitor ofmetalloproteinase-2 (TIMP-2): Characterization of complex and freeenzyme. Biochem J 289: 411–416, 1993

59. Beckman JS, Beckman TW, Chen J, Marshall PM, Freeman BA: Ap-parent hydroxyl radical production by peroxynitrite: Implications forendothelial injury from nitric oxide and superoxide. Proc Natl AcadSci 87: 1620–1624, 1990

60. Zweier JL, Wang P, Samouilov A, Kuppuswamy P: Enzyme-independ-ent formation of nitric oxide in biological tissues. Nature Med 1: 804–189, 1995

61. Huie RE, Padjama S: The reaction of NO with superoxide. Free RadicRes Commun 18: 195–199, 1993

62. Bond JS, Butler PE: Intracellular proteases. Annu Rev Biochem 56:333–364, 1987

63. Matheson NR, Wong PS, Schuyler M, Travis J: Interaction of humanα1-proteinase inhibitor with neutrophil meyloperoxidase. Biochemistry20: 331–336, 1981

64. Burghuber OC, Strife RJ, Zirrolli J, Henson PM, Henson JE, MathiasMM, Reeves JT, Murphy RC, Voelkel NF: Leukotriene inhibitors at-tenuate rat lung injury induced by hydrogen peroxide. Am Rev RespirDis 131: 778–785, 1985

65. Seeger W, Wolf H, Graubert E, Moser U, Neuhof H, Roka L: 1983Influence of aprotinin and gabexate mesilate on arachidonic acid re-lease by the Ca2+-ionophore A 23187 in the lung. Adv Exp Med Biol156: 553–567, 1983

66. White RP: Pharmacodynamic effects of tosyl-arginine methylester(TAME) on isolated human arteries. J Gen Physiol 19: 387–392, 1988

67. Shabani F, McNeil J, Tippett L: The oxidative inactivation of tissue in-hibitor of metalloproteinase-1 (TIMP-1) by hypochlorous acid (HOCl)is suppressed by anti-rheumatic drugs. Free Radic Res 28 :115–23,1998

288