Tissue Inhibitor of Metalloproteinases-1 Promotes Liver Metastasis by Induction of Hepatocyte Growth Factor Signaling

Tissue Inhibitor of Metalloproteinases-1 Promotes Liver Metastasis

by Induction of Hepatocyte Growth Factor Signaling

Charlotte Kopitz,1Michael Gerg,

1Obul Reddy Bandapalli,

4Dilek Ister,

1Caroline J. Pennington,

6

Stephanie Hauser,1Christin Flechsig,

1Hans-Willi Krell,

7Dalibor Antolovic,

5Keith Brew,

8

Hideaki Nagase,9Manfred Stangl,

2Claus W. Hann von Weyhern,

3Bjorn L.D.M. Brucher,

2

Karsten Brand,4Lisa M. Coussens,

10Dylan R. Edwards,

6and Achim Kruger

1

1Institut fur Experimentelle Onkologie und Therapieforschung, 2Chirurgische Klinik, and 3Institut fur Allgemeine Pathologie undPathologische Anatomie, Klinikum rechts der Isar der Technischen Universitat Munchen, Munich, Germany; 4Institute of Pathologyand 5Department of Surgery, University Hospital Heidelberg, Heidelberg, Germany; 6School of Biological Sciences, Universityof East Anglia, Norwich, Norfolk, United Kingdom; 7Roche Diagnostics GmbH, Penzberg, Germany; 8Department ofBiomedical Science, Florida Atlantic University, Boca Raton, Florida; 9The Kennedy Institute of RheumatologyDivision, Imperial College School of Medicine, London, United Kingdom; and 10Department of Pathologyand Comprehensive Cancer Center, University of California, San Francisco, California

Abstract

Balanced expression of proteases and their inhibitors is oneprerequisite of tissue homeostasis. Metastatic spread of tumorcells through the organism depends on proteolytic activity andis the death determinant for cancer patients. Paradoxically,increased expression of tissue inhibitor of metalloproteinases-1 (TIMP-1), a natural inhibitor of several endometalloprotei-nases, including matrix metalloproteinases and a disintegrinand metalloproteinase-10 (ADAM-10), in cancer patients isnegatively correlated with their survival, although TIMP-1itself inhibits invasion of some tumor cells. Here, we show thatelevated stromal expression of TIMP-1 promotes liver metas-tasis in two independent tumor models by inducing thehepatocyte growth factor (HGF) signaling pathway andexpression of several metastasis-associated genes, includingHGF and HGF-activating proteases, in the liver. We also foundin an in vitro assay that suppression of ADAM-10 is inprinciple able to prevent shedding of cMet, which may be oneexplanation for the increase of cell-associated HGF receptorcMet in livers with elevated TIMP-1. Similar TIMP-1–associ-ated changes in gene expression were detected in livers ofpatients with metastatic colorectal cancer. The newly identi-fied role of TIMP-1 to create a prometastatic niche may alsoexplain the TIMP-1 paradoxon. [Cancer Res 2007;67(18):8615–23]

Introduction

Metastasis and progressive scattering of tumor cells throughouttissues is the main cause of organ failure and subsequent deathof cancer patients. Proteolytic remodeling of components of theextracellular matrix (ECM) in tissue interstitium is a prerequisite ofthe metastatic process (1). Matrix metalloproteinases (MMP) havelong been thought to be major regulators of the pericellularprotease-dependent steps of tumor progression (2). Alteredexpression profiles of some MMPs have been correlated with poorclinical prognosis for several human tumor types (3). Based on

these facts, the historical view of MMPs and cancer was that theyexerted proinvasive and prometastatic activity by mere ECMremodeling (1). Thus, it was hypothesized that therapeutic broad-spectrum inhibition of MMPs would result in antimetastaticactivity (1). Indeed, under certain experimental circumstances,overexpression of the endogenous broad-spectrum metalloprotei-nase [including a disintegrin and metalloproteinase-10 (ADAM-10);ref. 4] inhibitor tissue inhibitor of metalloproteinases-1 (TIMP-1;ref. 5) can reduce tumor cell invasion (6, 7).In contrast to these observations are other results, which are

actually in agreement with clinical studies showing a correlationbetween elevated TIMP-1 expression and poor prognosis in manyhuman cancer types (5, 8), where elevated levels of TIMP-1 eitherhad no effect on metastasis (9) or led to increased incidence(10, 11) or growth (12) of primary tumors. These effects of TIMP-1on cancer development have been attributed to its knownproproliferative, antiapoptotic (5), or proangiogenic (13) activities,which are necessary but not sufficient for metastasis.Tumor cell invasion and metastasis are the primary determi-

nants of the survival of cancer patients (13). Signaling pathwaysleading to the expression of proteolytic enzymes have beensuggested to be key mediators of invasive growth (14). Hepatocytegrowth factor (HGF/scatter factor) signaling regulates a multitudeof downstream prometastatic effector molecules, such as urokinase-type plasminogen activator (uPA), uPA receptor (uPAR), plasminogenactivator inhibitor-1 (PAI-1), and MMPs (15–17). HGF stimulatesmetastasis of diverse tumor cells (15) and elevated levels ofcirculating HGF (18) or aberrant activity of cMet (14) has beendetected in patients with many cancer types, including non-Hodgkin’s lymphoma (19) and colorectal cancer (14). Althoughelevated TIMP-1 expression in these patients correlates withdecreased survival, a link between TIMP-1 and metastasis-promoting pathways has previously not been described. Here,we address this issue and report that elevated stromal levels ofTIMP-1 induce HGF signaling, leading to enhanced expression ofother metastasis-promoting genes. Consequently, susceptibilityof the liver for metastasis of tumor cell lines of different origin wasshown to be increased, assigning a new role to TIMP-1 as a factorcreating a prometastatic niche.

Materials and Methods

Cell culture and adenoviruses. HEK 293 (20), HT1080lacZ-K15 (21),

L-CI.5s (22), and 293T cells (23) were cultured as described previously.

Note: Supplementary data for this article are available at Cancer Research Online(http://cancerres.aacrjournals.org/).

Requests for reprints: Achim Kruger, Institut fur Experimentelle Onkologie undTherapieforschung, Klinikum rechts der Isar der Technischen Universitat Munchen,Ismaninger Str. 22, D-81675 Munich, Germany. Phone: 49-89-4140-4463; Fax: 49-89-4140-6182; E-mail: [email protected].

I2007 American Association for Cancer Research.doi:10.1158/0008-5472.CAN-07-0232

www.aacrjournals.org 8615 Cancer Res 2007; 67: (18). September 15, 2007

Research Article

Research. on January 25, 2016. © 2007 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

NIH3T3 cells were cultured similarly to HT1080lacZ-K15 (21). Amino acids1 to 126 of the human TIMP-1 cDNA (N-TIMP-1; ref. 24) were cloned into

pSecTag2/Hygro A (Invitrogen). A DNA fragment consisting of Ign-leadersequence and N-TIMP-1 cDNA or full-length human TIMP-1 cDNA

(pGEMhTIMP-1; ref. 25), respectively, was obtained and cloned into the

adenoviral shuttle plasmid pKA1 (26). Recombinant adenoviruses weregenerated, purified, and plaque titered as described previously (26). As

control, Addl70-3, an adenovirus without transgene, was used (27).

Functionality of adenoviral constructs was tested in vitro (data not shown).

Adenoviral infection in vitro was done as described previously (26).Experimental metastasis assays. Pathogen-free, athymic CD1nu/nu mice

(Charles River) were injected i.v. with 2 � 109 plaque-forming units of

Addl70-3, AdTIMP-1, and AdN-TIMP-1, respectively, or vehicle (1% glycerin

in PBS) alone. Three days later, 5,000 lacZ-tagged murine T-cell lymphomacells (L-CI.5s; ref. 22) or 1 � 106 lacZ-tagged human fibrosarcoma cells

(HT1080lacZ-K15; ref. 21) were injected i.v. Alternatively, inoculation of

tumor cells preceded the adenoviral transfer by 2 days (Fig. 1C). Mice were

sacrificed and their livers and lungs were removed at 6 days (L-CI.5s) or 21days (HT1080lacZ-K15) after inoculation. The right lobe of each lung and

three samples of each liver, with a volume off0.5 cm3, were snap frozen in

liquid nitrogen for biochemical analysis, and left lung lobes and remainingliver were stained with 5-bromo-4-chloro-3-indolyl-h-D-galactopyranoside(X-Gal; Roche Diagnostics) as described (22). Blue multicellular foci on

the surface of the livers (L-CI.5s) or lungs (HT1080lacZ-K15) were counted.

Total metastasis burden was assessed by quantitative real-time PCR(RT-PCR) for lacZ . All animal experiments were done in compliance with

the guidelines of the Tierschutzgesetz des Landes Oberbayern.

Immunostaining. Fixed sections of livers of the different groups were

paraffin embedded for immunostaining. Dewaxing, rehydration, blocking,and antigen retrieval of liver sections (4 Am) were done as described (28).

Sections were incubated with primary antibodies [proliferating cell nuclear

antigen (PCNA) antibody, 1:2,000, Novocastra Laboratories Ltd.; cMetantibody, 1:200, Santa Cruz Biotechnology] for 20 h at 4jC and washed in

TBS, 0.1% (v/v) Tween 20. StreptAB Complex/HRP Duett Mouse/Rabbit kit

(Dako Cytomation) and 3,3¶-diaminobenzidine (DAB) substrate (DAKO

Liquid DAB+ Substrate-Chromogen Solution, DAKO Corp.) were used fordetection. Sections were counterstained with Mayer’s hemalaun (Merck).

For detection of cMet activation, unfixed fresh liver samples were

Figure 1. TIMP-1–mediated reduction of formation of macrometastases butinduction of liver-specific secondary spread of tumor cells. A and B, CD1nu/nu

mice received the respective adenovirus 3 d before inoculation of L-CI.5s (A )or HT1080lacZ -K15 cells (B ), respectively. Six or 21 d, respectively, after tumorcell inoculation, mice were sacrificed and livers and lungs were removed.A, top left, number of X-Gal–stained L-CI.5s macrometastases (>0.2 mm) overthe whole liver surface was reduced by overexpression of human TIMP-1or N-TIMP-1 compared with the virus control group. Columns, mean; bars, SE.Addl70-3: 67.60 F 7.73, n = 5; AdTIMP-1: 38.50 F 8.14, n = 6, P = 0.03;AdN-TIMP-1: 32.83 F 8.32, n = 6, P = 0.009. Close-up pictures of the surfacesof representative X-Gal–stained L-CI.5s metastasis-bearing livers (bottom )show increased micrometastasis in livers with elevated levels of TIMP-1or N-TIMP-1. Top right, the total L-CI.5s tumor burden in livers overexpressingthe TIMP-1 variants was significantly increased compared with the viruscontrol as detected by quantitative RT-PCR of lacZ in metastasis-bearing livers.Columns, mean of relative lacZ mRNA amount versus 18S RNA; bars, SE.Addl70-3: 0.22 F 0.04, n = 3; AdTIMP-1: 1.99 F 0.80, n = 4, P = 0.044;AdN-TIMP-1: 1.96 F 0.76, n = 4, P = 0.041. B, significant reduction of HT1080lung metastasis by overexpression of human TIMP-1. Top left, columns,mean of the number of X-Gal–stained metastases over the whole surface of theleft lung lobe; bars, SE. Addl70-3: 38.1 F 16.1, n = 16; AdTIMP-1: 2.5 F 1.4,n = 10; P = 0.018. Significant increase of HT1080 metastasis in livers withelevated levels of TIMP-1. lacZ mRNA and 18S RNA levels were detected intotal RNA of HT1080lacZ -K15 metastasis-bearing livers by quantitative RT-PCR.Top right, columns, mean of relative lacZ mRNA amount versus 18S RNA;bars, SE. Addl70-3: 0.016 F 0.002; AdTIMP-1: 0.040 F 0.007, n = 3, P = 0.04.Bottom, close-up pictures of the surfaces of representative X-Gal–stainedHT1080 metastasis-bearing livers. C, effects of TIMP-1 elevation after tumorcell inoculation on liver metastasis. CD1nu/nu mice received the respectiveadenovirus 2 d after inoculation of L-CI.5s. Six days after tumor cell inoculation,mice were sacrificed and livers were removed and X-Gal stained. The number ofmacrometastases (>0.2 mm) was counted on the whole surface of the liver.Top left, columns, mean; bars, SE. Addl70-3: 45 F 4, n = 4; AdTIMP-1: 44 F 3,n = 4. Total metastasis burden was assessed by quantitative RT-PCR of themarker gene lacZ. Top right, columns, mean of relative lacZ mRNA versus18S RNA; bars, SE. Addl70-3: 0.29 F 0.16, n = 4; AdTIMP-1: 2.25 F 0.62,n = 4. Bottom, X-Gal–stained livers. Representative pictures of each group.Bar , 200 Am. Asterisks, macrometastases; arrows, micrometastases.

Cancer Research

Cancer Res 2007; 67: (18). September 15, 2007 8616 www.aacrjournals.org



embedded in Tissue-Tek OCT compound (Sakura Finetek), shock frozen ondry ice, and stored at �80jC until sectioning. Sections were dried, fixed in

ice-cold acetone, dried at room temperature, rehydrated in TBS, blocked

with 3% goat serum in TBS-Tween 20, and incubated with phosphorylated

Met (Tyr1349) antibody [Cell Signaling Technology; 1:100 in antibody dilutingbuffer (DAKO)] overnight at 4jC. Counterstaining was done with 100 ng/mL4¶,6-diamidino-2-phenylindole (DAPI; AppliChem). Sections were mounted

with Moviol (Merck). Green fluorescent signal was displayed using

stereomicroscope Axiovert 135 (Zeiss) with integrated digital cameradisposing the AxioVision LE 4.2 software (Zeiss Vision GmbH).

HGF ELISA. Livers were perfused with 100 mL PBS before excision,

sample preparation, purification (HiTrap heparin column, Amersham), and

ELISA [rat HGF ELISA (Institute of Immunology Co. Ltd.)] were done

following the manufacturers’ instructions. This kit detects mouse HGF

equally well (manufacturers’ manual). The ELISA was done in duplicates

and results were expressed as picogram HGF per milligram extracted liver.

In situ zymography and MMP-9 activity assay. In situ zymography wasdone and documented as described previously (29). To detect the origin of

gelatinolytic activity, serial sections were overlaid with DQ-gelatin

(Molecular Probes; ref. 29) containing either 10 Amol/L E64 (AppliChem),

10 Amol/L aprotinin (AppliChem), or 400 Amol/L 1,10-phenanthroline(AppliChem). MMP-9 activity was quantified by MMP-9 activity assay

(Amersham) according to the manufacturer’s instructions. The antibody

used in this assay binds human and murine species of MMP-9(manufacturer’s information). MMP-9 activity in 80 Ag total protein was

determined. Assay was done in triplicates.

Patient material. Liver tissue of 14 patients with metastatic colorectal

cancer (7 males and 7 females; mean age, 61.0 F 2.3 years) was obtained

from patients who had signed informed consent for scientific study of their

tumors and surrounding tissues at the Department of Surgery of the

Klinikum rechts der Isar or of the University Hospital Heidelberg. Tissue

samples were harvested under identical, rigorous procurement protocol

and snap frozen in liquid nitrogen immediately after surgical resection.

‘‘Normal’’ liver tissue was harvested from zones 2 cm distant from visible

liver metastases. Three experienced pathologists from two independent

institutions used cryodissected sections for confirmation of histologic

identity. All samples were stored at �80jC until use.RNA isolation, reverse transcription, design of primer and probes,

and quantitative RT-PCR. RNA isolation and reverse transcription were

done as described (28). Design of primers and probes for murine uPA,

murine uPAR, murine PAI-1, murine matriptase, murine MMP-9, murineMMP-2, and murine ADAM-10 was described previously (28, 30, 31). The

lacZ primer and probes were designed and obtained from Applied

Biosystems ( forward primer: 5¶-CAAGCCGTTGCTGATTCGA-3¶; reverse

primer: 5¶-GCTCATCCATGACCTGACCAT-3¶; probe: 5¶-FAM-ACCGTCAC-GAGCATCAT-nFQ-3¶). All other quantitative RT-PCRs were done with

inventoried primers and probes from Applied Biosystems. Quantitative RT-

PCR was done as described previously (28).Inhibition of HGF signaling by a specific kinase inhibitor. Three

days before i.v. tumor cell inoculation, four mice were transduced with

adenoviral vectors. Mice received 100 mg/kg of 2-(2,6-dichloro-4-(2-

hydroxyethoxy)phenyl-4-(3-benzyloxyphenyl)-5-(2-[3-hydroxypropyl-ami-no]pyrimidin-4-yl)-N-H-imidazole (cMet inhibitor), with high affinity for

cMet [dissolved in 6% (v/v) Tween 80 with two equivalents HCl], i.p. 1 h

before and 1 h after tumor cell inoculation. I.p. treatment with the cMet

inhibitor was continued daily (100 mg/kg/d). As control, mice were treatedwith vehicle alone. Three days after tumor cell inoculation, mice were

sacrificed and livers were removed and X-Gal stained as described

previously (22). Inhibition of cMet by this inhibitor was measured by anELISA-type assay: Lysate of human colon adenocarcinoma HT29 cells was

exposed to a microtiter plate with an antihuman HGF receptor antibody

(BAF 358 ATP phosphorylation; R&D Systems). ATP phosphorylation of the

receptor kinase (cMet kinase) was detected in the presence or absence ofthe test compounds using a phosphorylated tyrosine mouse IgG (Santa

Cruz Biotechnology). The IC50 value for cMet inhibition is 2 nmol/L. Kinase

inhibition activity against other kinases was measured as described by

Fabian et al. [IC50 values: cyclin-dependent kinase 2, 290 nmol/L; HER4,

99 nmol/L; c-Jun NH2-terminal kinase (JNK) 1, 12 nmol/L; JNK3, 2 nmol/L;p38a, 10 nmol/L; p38h, 18 nmol/L; other kinases (254 tested), >1,000 nmol/L;data not shown; ref. 32].

Knockdown of ADAM-10 in NIH3T3 cells. Oligodeoxynucleotides

coding for small hairpin RNA interference (shRNAi) directed against murineADAM-10 (small interfering RNA sequence: AACCTACGAATGAAGAGG-

GAC) were designed and obtained from Eurogenetec. Annealed shRNAis

were cloned into pSiren-RetroQ (Clontech). Recombinant retroviruses were

generated as described (23) by transient cotransfection of 293T cells withpHIT60 (23), pHCMV-G (33), and pSiren-RetroQ coding for anti-ADAM-10

shRNAi. Infection of NIH3T3 cells was done according to standard protocols

(23). Transduced cells were selected with 20 Ag/mL puromycin (Clontech).

cMet ELISA. For detection of membrane-bound cMet, NIH3T3 cellstransduced with the different viral constructs were resuspended in 0.1 mol/L

Tris, 10 mmol/L EDTA, and 0.1% Triton X-114 (AppliChem) at pH 7.5,

supplemented with 10 AL Protease Inhibitor Cocktail (Sigma). Subsequently,cells were incubated for 5 min on ice and 5 min at 37jC and centrifuged

for 5 min at 300 � g at 30jC. The lower phase containing the membraneproteins was used for ELISA. For detection of cMet shedding, transduced

NIH3T3 cells were incubated with serum-free medium for 36 h. Total andcMet protein in membrane fractions and supernatants were quantified by

bicinchoninic acid assay and cMet ELISA (Mouse HGF R/c-MET ECD

DuoSet, R&D Systems), respectively, following the manufacturers’ instruc-

tions. Assays were done in triplicates.Statistical analysis. Data were tested for normal distribution and

normally distributed data were tested by Student’s t test, and where the

normality test failed, Mann-Whitney test was applied. P < 0.05 wasconsidered significant.

Results

Reduction of experimental macrometastases but inductionof scattered micrometastases in the liver by TIMP-1. To achieveelevated TIMP-1 expression in the host, we transduced full-lengthcDNAs of TIMP-1 by i.v. inoculation of adenoviral vectors intomice. Efficient adenoviral transduction of liver cells was deter-mined by quantitative RT-PCR (Supplementary Fig. S1). Mainte-nance of transgene overexpression achieved by the adenoviralvectors until the end of all following metastasis assays was verified(Supplementary Fig. S1). To test the effects of elevated stromalTIMP-1 on tumor cell invasion, we used the well-establishedsyngeneic in vivo metastasis assay with lacZ-tagged lymphomacells (L-CI.5s), which form hepatic metastatic foci following i.v.injection. Colonies that we defined as macrometastases in liversreached the cutoff size of 0.2 mm in this assay at day 6 after tumorcell inoculation. At this time point, the combination of successfulextravasation right after tumor cell inoculation and proliferation ofthe extravasated tumor cells for at least six days resulted inmulticellular foci of this size. Detected micrometastases (<0.2 mm)resulted from additionally invaded tumor cells if their numberexceeds the 5,000 inoculated tumor cells.Three days after adenoviral gene transfer, TIMP-1–transduced

and control mice were challenged with L-CI.5s tumor cells. Thepattern of metastasis in TIMP-1–transduced livers was significantlyaltered compared with the control virus–transduced livers:Elevated levels of TIMP-1 correlated with significant reduction ofthe number of macrometastases (by 43%, P = 0.003; Fig. 1A, topleft). However, on the level of micrometastases, no decrease but adrastic increase of scattered infiltrating tumor cells was detected inlivers with elevated TIMP-1 (Fig. 1A). In one photographic frame(1.3 mm � 1.5 mm), already 623 micrometastatic foci at the liversurface were visible (Fig. 1A, bottom), showing that the number ofscattered micrometastases in the entire liver by far exceeded thenumber of inoculated cells (5,000). Quantification of the tumor cell

TIMP-1 Promotes Liver Metastasis via HGF Signaling




tag lacZ revealed an overall increase of total metastasis burden inlivers transduced with TIMP-1 cDNA compared with the control(P = 0.044; Fig. 1A, top right), reflecting induction of micro-metastasis.To determine whether inhibition of macrometastasis but

promotion of micrometastasis was a function of the metal-loproteinase-inhibitory activity of TIMP-1, we investigated theeffect of the N-terminal domain of human TIMP-1 (N-TIMP-1)harboring the metalloproteinase-inhibitory bioactivity on livermetastasis. Indeed, elevated levels of N-TIMP-1 significantlysuppressed macrometastasis by 51% (P = 0.009; Fig. 1A) andpromoted micrometastasis (P = 0.041; Fig. 1A), indicating thatN-TIMP-1 is sufficient to evoke the described effects.We used lacZ-tagged human fibrosarcoma cells (HT1080lacZ-

K15) that typically form large pulmonary metastatic foci and onlyfew single tumor cells as micrometastases, but no macrometa-stases, in livers (26) to exclude that the TIMP-1 effect was cell typedependent. After challenge of CD1nu/nu mice with HT1080lacZ-K15, significant reduction of lung metastasis by 93% was observedin TIMP-1–transduced animals compared with mice with viruscontrol (P = 0.018; Fig. 1B, top left). Again, in livers with elevatedTIMP-1 levels, a significant increase of the metastatic burden wasfound (P = 0.04; Fig. 1B, top right). No induction of HT1080lacZ-K15micrometastasis was detected in lungs (data not shown). In theabove experiments, no differences in metastasis between controlvirus–transduced mice and mice treated with vehicle alone werefound, excluding effects of viral transduction itself (data notshown).

Effects of TIMP-1 elevation on liver metastasis after tumorcell inoculation. To determine whether TIMP-1 affects livermetastasis on two different levels, [i.e., on the level of initialmanifestation and subsequent growth (leading to formation ofmacrometastases) or subsequent infiltration (micrometastases)],we inoculated L-CI.5s cells into CD1nu/nu 2 days before adenoviraltransfer of TIMP-1 cDNA. The number of macrometastatic colonieswas not affected compared with the control (Fig. 1C, top left).However, we still observed induction of micrometastasis comparedwith the control (Fig. 1C, bottom), resulting in a significantlyincreased total metastasis burden in livers with elevated TIMP-1(Fig. 1C, top right). This indicates that micrometastases takeadvantage of the TIMP-1–modulated host microenvironment. Incontrast, macrometastasis formation was only reduced whenTIMP-1 was already elevated in the host at the time point ofextravasation right after tumor cell inoculation (Fig. 1A).

Proliferation activity of TIMP-1–induced scattered livermicrometastases. To assess if the increased presence of invadedL-CI.5s cells in the liver was a function of TIMP-1–inducedproliferation, we determined their proliferative status by examiningPCNA expression. Indeed, infiltrating T-cell lymphoma cells inlivers with elevated TIMP-1 or N-TIMP-1 were proliferatively active(Fig. 2A), indicating a proproliferative effect of the TIMP-1–modulated environment on these cells, accounting for the increaseof total metastasis burden in the liver. A proproliferativeenvironment was further indicated by the presence of proliferatinghepatocytes in these groups (Fig. 2A, arrows). In the control livers,proliferation was restricted to the macrometastases (Fig. 2A, left).

TIMP-1 induces HGF signaling in the liver. To elucidate amolecular mechanism responsible for the TIMP-1–induced pro-proliferative and proinvasive environment in the liver, we tested forexpression of HGF, also known as scatter factor. Elevated levels ofHGF correlated with the TIMP-1–induced scattered metastatic

Figure 2. Induction of hepatocyte proliferation and HGF signaling byelevated levels of TIMP-1. A, PCNA staining of L-CI.5s metastasis-bearingliver sections (4 Am) 9 d after adenoviral gene transfer and 6 d afterL-CI.5s inoculation. Representative sections. Asterisks, PCNA-positivehepatocytes; arrows, PCNA-positive micrometastases. B, left, HGF proteinlevel 9 d after gene transfer and 6 d after L-CI.5s inoculation. HGFprotein in pools of three livers per group was quantified by HGF ELISA(Addl70-3: 30.3 pg/mg; AdTIMP-1: 75.8 pg/mg; AdN-TIMP-1: 80.6 pg/mg).Right, HGF protein in metastasis-free livers 3 d after gene transfer.HGF protein in pools of three livers per group was quantified by ELISA(Addl70-3: 25.2 pg/mg; AdTIMP-1: 41.1 pg/mg; AdN-TIMP-1: 45.7 pg/mg).C, increased cMet expression in AdTIMP-1–transduced livers. Paraffin-embedded sections of L-CI.5s metastasis-bearing livers 9 d afteradenoviral gene transfer and 6 d after tumor cell inoculation wereimmunohistochemically analyzed for cMet localization. Bottom inset,stronger staining for cMet was found within metastases compared with liverparenchyma; top inset, controls without primary antibody for each slide.Black arrows, additionally stained micrometastases. D, TIMP-1–inducedactivation of cMet. Cryosections of metastasis-bearing (top ) andmetastasis-free (bottom ) livers, respectively, were immunohistochemicallyanalyzed for phosphorylated cMet (phospho cMet ). Counterstaining wasdone with DAPI. Asterisks, lumen of blood vessels; arrows, staining ofphosphorylated cMet. Bar , 100 Am.

Cancer Research




phenotype (Fig. 2B, left). In fact, increase of HGF expression wasalready a component of the TIMP-1–altered liver environment as itcould already be detected 3 days after adenoviral gene transfer (i.e.,at the time point of tumor cell challenge; Fig. 2B, right). In addition,cMet, the tyrosine kinase receptor of HGF, was elevated throughoutthe tissue of metastasis-bearing livers transduced by either TIMP-1species (Fig. 2C) and in macrometastases (Fig. 2C, insets).Elevated expression of either TIMP-1 variant led to markedly

increased HGF signaling, as detected by immunohistochemicalstaining of phosphorylated cMet throughout the liver parenchyma(Fig. 2D). In control animals, activated cMet was very low even inmacrometastases (Fig. 2D, top). Importantly, this TIMP-1–mediatedinduction of HGF signaling was already present 3 days after TIMP-1gene transfer (Fig. 2D, bottom), at the time point of tumor cellchallenge, suggesting that a HGF pathway-connected prometastaticniche has been formed.

Causal relationship between HGF signaling and TIMP-1–induced scattering of experimental liver metastasis. In TIMP-1–transduced animals, the scattered metastasis phenotype wasalready apparent 3 days after inoculation of L-CI.5s cells (Fig. 3A,top). To investigate whether HGF signaling was essential forformation of scattered metastasis, we first transduced mice withAdTIMP-1 or Addl70-3 virus, treated the mice with a cMetinhibitor, and challenged with L-CI.5s cells. Kinase inhibitionprevented formation of micrometastasis (Fig. 3A, bottom), indicat-ing that HGF signaling was essential for TIMP-1–induced micro-metastatic spread. The TIMP-1–induced significant increase oftotal metastasis burden compared with the control (P < 0.001;Fig. 3B) was significantly reduced by cMet inhibition (P < 0.001;Fig. 3B). The number of multicellular metastatic foci was signifi-cantly reduced by elevated levels of TIMP-1 compared with thecontrol (P = 0.048; Fig. 3C), indicating that formation of macro-metastasis is not affected by HGF signaling.

Overexpression of TIMP-1 reduces metalloproteinase activ-ity and increases activity of other proteases in the liver. Asproteolytic activity is one prerequisite of metastasis, we nextassessed whether TIMP-1–related changes in metastasis rely onmodulated proteolytic activity. In situ zymography revealed

gelatinolytic activity in macrometastatic foci in Addl70-3–trans-duced livers but not in foci of AdTIMP-1–transduced animals(Fig. 4A, left). As this was inhibitable by the canonical metal-loproteinase inhibitor 1,10-phenanthroline (Fig. 4B, left), we showedthat the achieved TIMP-1 levels indeed inhibited MMP activityin vivo . No MMP-9–derived proteolytic activity was detected inlivers with elevated TIMP-1 (Fig. 4C), indicating the expectedfunctional activity of TIMP-1 to inhibit proMMP-9 activation.The gelatinolytic activity detectable in TIMP-1–elevated liver

parenchyma distant from metastases (Fig. 4A, right) was by andlarge metalloproteinase independent, as it was only slightlyinhibitable by 1,10-phenanthroline (Fig. 4B). Addition of broad-spectrum cysteine protease inhibitor E64 (34) or broad-spectrumserine protease inhibitor aprotinin (35) to the in situ zymographyDQ-overlay revealed the remaining gelatinolytic activity as partlyderived from serine and cysteine proteases (Fig. 4D). Elevated levelsof N-TIMP-1 had the same effects (data not shown). These dataindicate that increased expression of TIMP-1 in the liver led to anincrease of metalloproteinase-independent proteolytic activity.

TIMP-1–induced expression of metastasis-related genes inthe liver. Induction of HGF signaling and nonmetalloproteinaseproteolytic activity in liver parenchyma by TIMP-1 suggested analtered HGF signaling-related gene expression favoring metastasis.Therefore, we used quantitative RT-PCR to characterize theexpression of HGF-inducible genes, HGF-activating proteases, aswell as gelatinolytic proteases. Elevation of TIMP-1 led to increaseof mRNA expression of uPA, uPAR, PAI-1, tissue plasminogenactivator (tPA), matriptase, MMP-9, MMP-2, ADAM-10, cathepsinG, and neutrophil elastase in the tumor-free liver (Fig. 5). Elevatedlevels of N-TIMP-1 had the same effects (data not shown).To test whether these TIMP-1–regulated changes in gene

expression are of more general relevance, we measured TIMP-1mRNA expression in human liver tissue samples distant fromcolorectal metastases. In normal tissue, elevated expression ofTIMP-1 (relative mRNA expression: 98.92 F 19.84, n = 8) wasassociated with increased mRNA expression of uPA, uPAR,matriptase, MMP-9, MMP-2, ADAM-10, cathepsin G, and neutro-phil elastase compared with livers with significantly lower TIMP-1

Figure 3. Reduction of TIMP-1–induced micrometastatic spread by cMet inhibition. A to C, CD1nu/nu mice were transduced with AdTIMP-1 or Addl70-3. Threedays later, mice were treated with a cMet-specific kinase inhibitor [2-(2,6-dichloro-4-(2-hydroxyethoxy)phenyl-4-(3-benzyloxyphenyl)-5-(2-[3-hydroxypropyl-amino]-pyrimidin-4-yl)-N-H-imidazole] or vehicle alone and challenged by inoculation of L-CI.5s cells. Three days after tumor cell inoculation, mice were sacrificed and liverswere removed and X-Gal stained. A, pictures of the surface of representative livers of each treatment group. Bar , 0.1 mm. B, total metastasis burden in the liveras detected by quantitative RT-PCR of lacZ. Columns, mean; bars, SE. Addl70-3/vehicle: 0.059 F 0.021; AdTIMP-1/vehicle: 0.728 F 0.233; Addl70-3/cMet inhibitor:0.040 F 0.016; AdTIMP-1/cMet inhibitor: 0.051 F 0.030; all n = 3. C, number of multicellular metastatic foci was counted on the whole surface of X-Gal–stained livers.Columns, mean; bars, SE. Addl70-3/vehicle: 55 F 10; AdTIMP-1/vehicle: 36 F 2; Addl70-3/cMet inhibitor: 42 F 9; AdTIMP-1/cMet inhibitor: 25 F 6; all n = 5.





expression (relative mRNA expression: 21.22 F 1.89, n = 6, P <0.001; Fig. 5). This difference also correlated with shorter timeinterval between detection of primary tumor and liver metastases[26.83 F 5.81 months (n = 6) versus 13.75 F 2.62 months (n = 8);P = 0.049].

Inhibition of cMet shedding by TIMP-1. Interestingly,although cMet protein was increased in livers with elevatedTIMP-1 (Fig. 2C), mRNA expression of cMet was not up-regulatedin these livers, neither at the time point of tumor cell inoculation(Fig. 5) nor in metastasis-bearing livers (Supplementary Fig. S2).Reduced shedding of cMet by TIMP-1–induced inhibition ofmetalloproteinase activity may account for the accumulation ofcMet protein, but it was thus far unknown that TIMP-1 can inhibitits shedding. As this hypothesis cannot be addressed in vivo , wechose a murine nontumorigenic cell line expressing inhibitablemetalloproteinases as well as cMet, thus providing all prerequisitesfor a proof-of-concept study. Overexpression of either TIMP-1 orN-TIMP-1 in NIH3T3 cells (Fig. 6A) resulted in significantlyaugmented membrane-bound cMet protein in these cells com-pared with the virus control (both P V 0.026; Fig. 6B), although RNAexpression of cMet was not increased (both P > 0.33; Fig. 6C).

TIMP-1 indeed inhibited cMet shedding, as reduced levels ofsoluble cMet were detected in supernatant of cells overexpressingTIMP-1/N-TIMP-1 compared with the control (both P V 0.033;Fig. 6D). The elevated level of cell-associated cMet in cellsoverexpressing TIMP-1/N-TIMP-1 correlated with increased phos-phorylation of cMet (Supplementary Fig. S3).

Activity of ADAM-10 regulates cMet shedding. Knockdown ofADAM-10 mRNA expression in NIH3T3 cells by 88% (Fig. 6A) led toa significant increase of membrane-bound cMet protein comparedwith the control (P = 0.018; Fig. 6B), whereas cMet mRNA expres-sion was not enhanced (P = 0.2; Fig. 6C). ADAM-10 knockdown wasverified on protein level (Supplementary Fig. S4). A significantreduction of shed cMet in the supernatant of cells with reducedADAM-10 expression was detected (P = 0.021; Fig. 6D), indicatingthat depletion of ADAM-10 was sufficient to accumulate cMet onthe cell surface. In addition, suppression of ADAM-10 mRNAexpression significantly increased cMet signaling (SupplementaryFig. S3).Together, these data suggest a molecular pathway regulated by

TIMP-1 that could lead to a microenvironment with highersusceptibility to metastases.

Figure 4. Reduction of gelatinase activity but increaseof non-MMP proteolytic activity by overexpression ofTIMP-1 detected by in situ zymography. A, B, andD, cryosections of L-CI.5s metastasis-bearing livers,transduced by either AdTIMP-1 or Addl70-3, were overlaidwith agarose containing DQ-gelatin (A). Parallel sectionswere overlaid with the same solution supplementedwith 1,10-phenanthroline (B), E64, and aprotinin (D ),respectively. Counterstaining was done with DAPI. Whiteline, boundaries of metastases; asterisks, lumen of bloodvessels; arrows, areas of micrometastases. C, inhibitoryactivity of TIMP-1 in the liver 3 d after adenoviral genetransfer was tested using a MMP-9 activity assay.Columns, mean; bars, SE. Addl70-3: 100F 23.23%, n = 4;AdTIMP-1: 0 F 0%, n = 4.

Cancer Research




Discussion

This study shows that host-derived TIMP-1 can promote liver

metastasis by induction ofHGFsignaling. Until recently, the dominantconcept of natural inhibitors of tumor-associated proteases focused

on their capacity to repress activity of ECM remodeling enzymes,

such as metalloproteinases. This view has undergone revision basedon increased awareness of ‘‘degradome’’ functions of metalloprotei-

nases and the tumor microenvironment and recognition of their

participation during many stages of tumor progression (3).The original hypothesis that TIMP-1 overexpression would yield

a reduction in the number of successfully extravasated tumor cells

into a target organ was based on numerous reports on the

importance of MMPs for metastasis in many tumor models (36). In

agreement with these and our previous studies indicating the

importance of gelatinase inhibition for formation of macrometa-

stases of the lymphoma model (28, 29), elevated host levels of

TIMP-1 prevented macrometastases only when it was already

increased at the time point of tumor cell inoculation. This confirms

the already known effect of metalloproteinase inhibition by TIMP-1

on the level of extravasation.In contrast to this inhibitory action on initial invasion, we reveal

here for the first time an unexpected metastasis-promoting feature

of TIMP-1 (i.e., induction of scattered invasion of tumor cells inliver parenchyma). The bulk of these micrometastases cannotderive from augmented extravasation of the L-CI.5s cells becausethe number of the proliferative active and scattered micro-metastases by far exceeded the number of inoculated tumor cells.Previous studies have reported that, even when metalloproteinasesare required for metastasis in a particular tumor model, TIMP-1does not always suppress metastasis (9, 37), supporting the ideathat TIMP-1 possesses a broad range of biological activities, someof which may be independent of its metalloproteinase-inhibitoryfunction (11, 38–40). Those variations of antitumorigenic efficacy ofTIMP-1 have been suggested to depend on the concentrations ofTIMP-1 (10, 41) or related to the proproliferative and antiapoptoticfeatures of TIMP-1 (11, 12, 38).The new aspect is that we found a proinvasive feature of TIMP-1

when elevated in the liver based on TIMP-1–induced stimulationof HGF signaling and downstream expression of metastasis-promoting genes. The N-terminal domain of TIMP-1, which includesthe metalloproteinase-inhibitory domain, was sufficient to exhibitthis effect, indicating that broad-spectrum inhibition of metal-loproteinases is responsible for this scattering. In agreement withthis notion, treatment with a synthetic broad-spectrum metal-loproteinase inhibitor also leads to liver-specific promotion of

Figure 5. Common TIMP-1–associated alterations in gene expression in mouse and human. Mice were transduced with either control virus (Addl70-3) or AdTIMP-1.Three days later, mice were sacrificed and livers were removed. Normal liver samples of patients with metastatic colorectal cancer were obtained at least 2 cm distantfrom colorectal liver metastases during curative surgical interventions. Total RNA was isolated from murine and human liver samples and gene expression wasassessed by quantitative RT-PCR. Human samples were initially tested for TIMP-1 expression and then divided into two subgroups according to their TIMP-1expression levels (low TIMP-1: 21.22 F 1.89, n = 6; high TIMP-1: 98.92 F 19.84, n = 8; P < 0.001). All samples were tested for uPA, uPAR, PAI-1, tPA, matriptase,PCNA, MMP-2, MMP-9, ADAM-10, cathepsin G (Cath G), neutrophil elastase (NE ), and Met mRNA expression. Columns, mean of gene expression in murine[Addl70-3 (n = 4) versus AdTIMP-1 (n = 3): uPA: 100 F 16.3% versus 193.3 F 20.3%; uPAR: 100 F 15.3% versus 557.9 F 63.2%; PAI-1: 100 F 21.2% versus5,083.0F 3,052.9%; tPA: 100 F 18.4% versus 1,673.8F 465.2%; matriptase: 100 F 4.4% versus 451.5F 33.8%; PCNA: 100F 82.7% versus 554.6F 80.3%; MMP-9:100 F 47.2% versus 494.1 F 34.6%; MMP-2: 100 F 45.9% versus 289.4 F 63.2%; ADAM-10: 100 F 35.7% versus 542.8 F 397.4%; cathepsin G: 100 F 52.6%versus 2,040.0 F 721.1%; neutrophil elastase: 100 F 66.0% versus 1,024.3 F 541.1%; Met: 100 F 9.3% versus 109.6 F 8.9%] and human (low TIMP-1 versus highTIMP-1: uPA: 100 F 16.6% versus 615.5 F 61.2%; uPAR: 100 F 20.7% versus 691.6 F 131.1%; PAI-1: 100 F 15.7% versus 107.1 F 182%; tPA: 100 F 10.8%versus 116.9 F 18.6%; matriptase: 100 F 18.1% versus 213.0 F 24.4%; PCNA: 100 F 24.3% versus 231.7 F 50.7%; MMP-9: 100 F 29.5% versus 2,418.6 F 396.4%;MMP-2: 100 F 30.6% versus 695.3 F 590.9%; ADAM-10: 100 F 11.1% versus 150.6 F 33.9%; cathepsin G: 100 F 31.9% versus 379.5 F 107.8%; neutrophilelastase: 100 F 75.5% versus 1621.6 F 339.2%; Met: 100 F 17.0% versus 111.5 F 27.6%) livers; bars, SE. *, P < 0.05.





metastasis in several xenograft models as well as in the syngeneicT-cell lymphoma model (42, 43). In addition, in the present study,promotion of liver metastasis was found, whereas lung metastasiswas significantly reduced. Either TIMP-1 was not able to induceHGF signaling in the lung or induction of HGF signaling in the lungtissue has different effects on the formation of a premetastatic

niche than in the liver parenchyma. However, several studiesrevealed that inhibition or overexpression of proteases orprotease families evolves different effects on metastasis to liverversus lung (26, 42–44). This leads to the hypothesis thatmetastasis to the liver requires a different repertoire of proteasesthan metastasis to the lung and that the different microenviron-ments present in these organs react differentially on changedproteolysis. The underlying mechanism of the organ specificity ofthe prometastatic effect of TIMP-1 will need to be determined infuture studies.Upon elevation of TIMP-1, we observed augmented HGF as well as

cMet protein in livers. Simultaneous increase of HGF and functionalcell-associated cMet can explain the overall increase of HGF/cMetsignaling. Elevated presence of the active form of HGF, which isrequired for signaling (15), can derive from the observed up-regulation of the HGF-activating proteases uPA and matriptase(45, 46) in the liver. Active matriptase and uPA (also activated bymatriptase; ref. 46) can lead to an additional increase of proteolyticactivity by initiating a protease activation cascade, including theactivation of plasminogen by uPA and of some pro-MMPs by activeplasmin (47). Elevated levels of TIMP-1 in the liver also increasedexpression of cathepsin G, also a gelatinolytic enzyme (48), as well asexpression of neutrophil elastase, which also enhances plasminogenactivation (49). Such proteases are likely the source of the increasednonmetalloproteinase-derived proteolytic activity observed in me-tastasis-bearing liver sections with augmented TIMP-1 levels in situ .Interestingly, the TIMP-1–provoked gene expression signature of theexperimental model resembled in some instances (matriptase, uPA,uPAR, PCNA, MMP-9, and cathepsin G) the signature observed inliver samples of colorectal cancers in patients with elevated levels ofTIMP-1, which is a hint on general molecular mechanisms provokedby elevated stromal TIMP-1. In the future, stable knockdown ofcMet in host and/or tumor cells will further characterize the role ofTIMP-1–induced cMet signaling in liver metastasis.A link between metalloproteinase inhibition and enhancement

of HGF/cMet signaling was revealed by an in vitro biochemicalassay: Elevated levels of TIMP-1/N-TIMP-1 preserved activatablecMet on cell surfaces by reduced cMet shedding. Interestingly,suppression of ADAM-10 also prevented shedding of cMet,indicating that ADAM-10 is involved in the regulation of cMetshedding. Sheddases are already known for their capacity toregulate cell surface–associated tyrosine kinase receptors (50).These data indicate that preservation of cMet is induced by TIMP-1

Figure 6. Preservation of cell-associated cMet by either TIMP-1 overexpressionor ADAM-10 knockdown. NIH3T3 cells were infected either with Addl70-3(NIH/Addl70-3 ), AdTIMP-1 (NIH/AdTIMP-1 ), AdN-TIMP-1 (NIH/AdN-TIMP-1 alladenoviruses with a multiplicity of infection of 100), with a retroviral vector codingfor shRNAi targeting murine ADAM-10 (NIH-shADAM-10 ), or with a retroviralvector encoding scrambled shRNAi (NIH-scr ). A, quantitative RT-PCR ofhuman TIMP-1 (huTIMP-1 ; left ) and murine ADAM-10 (muADAM-10 ; right ),respectively, verified overexpression of TIMP-1/N-TIMP-1 and suppression ofADAM-10. Columns, mean of relative huTIMP-1 and muADAM-10 mRNA,respectively, versus 18S RNA; bars, SE. B, membrane preparations oftransduced cells were tested for the presence of cMet by ELISA. Elevated levelsof either TIMP-1 variant (left ) or reduced levels of ADAM-10 (right ) led toincreased membrane-bound cMet protein compared with the controls. C, murinecMet (mu-cMet ) expression in cells with either increased expression of TIMP-1/N-TIMP-1 (left ) or reduced expression of ADAM-10 (right ), respectively, wasassessed by quantitative RT-PCR. Columns, mean of murine cMet mRNAversus 18S RNA; bars, SE. D, quantification of soluble cMet. FCS-free culturemedium of transduced NIH3T3 cells was tested by cMet ELISA. Columns, meanof soluble cMet in culture medium; bars, SE. ELISAs were done in triplicates.Results of three independent experiments.

Cancer Research




or N-TIMP-1 and may be a consequence of direct inhibition ofADAM-10 and thereby loss of its sheddase activity (4) or indirectlyby inhibition of ADAM-10 activation.Many tumor cell lines respond to elevated levels of HGF (scatter

factor) with increased invasive behavior (15, 16). Indeed, the alteredhost environment fed back to the inoculated tumor cell lines,which expressed functional cMet (Supplementary Fig. S5). Thus,increased scattering of tumor cells in the liver parenchyma seemsto be based on a reaction of the tumor cells to the TIMP-1–provoked change of homeostasis in the host. Identification of thisTIMP-1–provoked prometastatic niche reveals a novel role forTIMP-1 during cancer progression and illuminates the intricacy ofprotease and protease inhibitor networks that regulate the

penultimate step of cancer evolution and are one explanation forthe association between tumor aggressiveness and increased levelsof TIMP-1 in cancer patients.

Acknowledgments

Received 1/18/2007; revised 6/22/2007; accepted 7/11/2007.Grant support: Deutsche Forschungsgemeinschaft grant SFB 469 (A. Kruger),

European Union Framework Programme 6, project LSHC-CT-2003-503297, Cancer-degradome (A. Kruger, K. Brand, and D.R. Edwards), and NIH grant AR40994 (K. Brewand H. Nagase).

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

We thank Dr. Susanne Schaten and Katja Honert (both from Technische UniversitatMunchen) for expert technical support.



References

1. Overall CM, Lopez-Otin C. Strategies for MMPinhibition in cancer: innovations for the post-trial era.Nat Rev Cancer 2002;2:657–72.

2. Overall CM, Kleifeld O. Tumour microenvironment—opinion: validating matrix metalloproteinases as drugtargets and anti-targets for cancer therapy. Nat RevCancer 2006;6:227–39.

3. Egeblad M, Werb Z. New functions for the matrixmetalloproteinases in cancer progression. Nat RevCancer 2002;2:161–74.

4. Amour A, Knight CG, Webster A, et al. The in vitroactivity of ADAM-10 is inhibited by TIMP-1 and TIMP-3.FEBS Lett 2000;473:275–9.

5. Chirco R, Liu XW, Jung KK, Kim HR. Novel functions ofTIMPs in cell signaling. Cancer Metastasis Rev 2006;25:99–113.

6. Elezkurtaj S, Kopitz C, Baker AH, et al. Adenovirus-mediated overexpression of tissue inhibitor of metal-loproteinases-1 in the liver: efficient protection againstT-cell lymphoma and colon carcinoma metastasis.J Gene Med 2004;6:1228–37.

7. Brand K. Cancer gene therapy with tissue inhibitors ofmetalloproteinases (TIMPs). Curr Gene Ther 2002;2:255–71.

8. Hammer JH, Basse L, Svendsen MN, et al. Impact ofelective resection on plasma TIMP-1 levels in patientswith colon cancer. Colorectal Dis 2006;8:168–72.

9. Kruger A, Fata JE, Khokha R. Altered tumor growthand metastasis of a T-cell lymphoma in Timp-1transgenic mice. Blood 1997;90:1993–2000.

10. Goss KJ, Brown PD, Matrisian LM. Differing effects ofendogenous and synthetic inhibitors ofmetalloproteinaseson intestinal tumorigenesis. Int J Cancer 1998;78:629–35.

11. Rhee JS, Diaz R, Korets L, Hodgson JG, Coussens LM.TIMP-1 alters susceptibility to carcinogenesis. CancerRes 2004;64:952–61.

12. Guedez L, McMarlin AJ, Kingma DW, Bennett TA,Stetler-Stevenson M, Stetler-Stevenson WG. Tissueinhibitor of metalloproteinase-1 alters the tumorigenic-ity of Burkitt’s lymphoma via divergent effects on tumorgrowth and angiogenesis. Am J Pathol 2001;158:1207–15.

13. Wurtz SO, Schrohl AS, Sorensen NM, et al. Tissueinhibitor of metalloproteinases-1 in breast cancer.Endocr Relat Cancer 2005;12:215–27.

14. Trusolino L, Comoglio PM. Scatter-factor andsemaphorin receptors: cell signalling for invasivegrowth. Nat Rev Cancer 2002;2:289–300.

15. Jiang WG, Martin TA, Parr C, Davies G, Matsumoto K,Nakamura T. Hepatocyte growth factor, its receptor, andtheir potential value in cancer therapies. Crit Rev OncolHematol 2005;53:35–69.

16. Fujiuchi Y, Nagakawa O, Murakami K, Fuse H,Saiki I. Effect of hepatocyte growth factor on invasionof prostate cancer cell lines. Oncol Rep 2003;10:1001–6.

17. Tacchini L, Matteucci E, De Ponti C, Desiderio MA.Hepatocyte growth factor signaling regulates trans-activation of genes belonging to the plasminogenactivation system via hypoxia inducible factor-1. ExpCell Res 2003;290:391–401.

18. Kuroi K, Toi M. Circulating angiogenesis regulatorsin cancer patients. Int J Biol Markers 2001;16:5–26.

19. Hsiao LT, Lin JT, Yu IT, et al. High serum hepatocytegrowth factor level in patients with non-Hodgkin’slymphoma. Eur J Haematol 2003;70:282–9.

20. Hitt M, Bett AJ, Addison CL, Prevec L, Graham FL.Techniques for human adenovirus vector constructionand characterization. In: Adolph KW, editor. Viral genetechniques. San Diego: Academic Press; 1995. p. 13–30.

21. Schweinitz A, Steinmetzer T, Banke IJ, et al. Design ofnovel and selective inhibitors of urokinase-type plas-minogen activator with improved pharmacokineticproperties for use as antimetastatic agents. J Biol Chem2004;279:33613–22.

22. Kruger A, Schirrmacher V, von Hoegen P. Scatteredmicrometastases visualized at the single-cell level:detection and re-isolation of lacZ-labeled metastasizedlymphoma cells. Int J Cancer 1994;58:275–84.

23. Soneoka Y, Cannon PM, Ramsdale EE, et al. Atransient three-plasmid expression system for theproduction of high titer retroviral vectors. Nucleic AcidsRes 1995;23:628–33.

24. Wei S, Chen Y, Chung L, Nagase H, Brew K. Proteinengineering of the tissue inhibitor of metalloproteinase1 (TIMP-1) inhibitory domain. In search of selectivematrix metalloproteinase inhibitors. J Biol Chem 2003;278:9831–4.

25. Parkes R, Meng QH, Siapati KE, McEwan JR,Hart SL. High efficiency transfection of porcine vascularcells in vitro with a synthetic vector system. J Gene Med2002;4:292–9.

26. Kopitz C, Anton M, Gansbacher B, Kruger A.Reduction of experimental human fibrosarcoma lungmetastasis in mice by adenovirus-mediated cystatin Coverexpression in the host. Cancer Res 2005;65:8608–12.

27. Jones N, Shenk T. Isolation of deletion and substitu-tion mutants of adenovirus type 5. Cell 1978;13:181–8.

28. Arlt M, Kopitz C, Pennington C, et al. Increase ingelatinase-specificity of matrix metalloproteinase inhib-itors correlates with antimetastatic efficacy in a T-celllymphoma model. Cancer Res 2002;62:5543–50.

29. Kruger A, Arlt MJ, Gerg M, et al. Antimetastaticactivity of a novel mechanism-based gelatinase inhibi-tor. Cancer Res 2005;65:3523–6.

30. Banke IJ, Arlt MJ, Pennington C, et al. Increase ofanti-metastatic efficacy by selectivity- but not affinity-optimization of synthetic serine protease inhibitors. BiolChem 2003;384:1515–25.

31. Palosaari H, Pennington CJ, Larmas M, Edwards DR,Tjaderhane L, Salo T. Expression profile of matrixmetalloproteinases (MMPs) and tissue inhibitors ofMMPs in mature human odontoblasts and pulp tissue.Eur J Oral Sci 2003;111:117–27.

32. Fabian MA, Biggs WH III, Treiber DK, et al. A smallmolecule-kinase interaction map for clinical kinaseinhibitors. Nat Biotechnol 2005;23:329–36.

33. Yee JK, Miyanohara A, LaPorte P, Bouic K, Burns JC,Friedmann T. A general method for the generation ofhigh-titer, pantropic retroviral vectors: highly efficientinfection of primary hepatocytes. Proc Natl Acad SciU S A 1994;91:9564–8.

34. Barrett AJ, Kembhavi AA, Brown MA, et al. L-trans-Epoxysuccinyl-leucylamido(4-guanidino)butane (E-64) andits analogues as inhibitors of cysteine proteinases includingcathepsins B, H and L. Biochem J 1982;201:189–98.

35. Fritz H, Wunderer G. Biochemistry and applicationsof aprotinin, the kallikrein inhibitor from bovine organs.Arzneimittelforschung 1983;33:479–94.

36. Coussens LM, Werb Z. Matrix metalloproteinases andthe development of cancer. Chem Biol 1996;3:895–904.

37. Soloway PD, Alexander CM, Werb Z, Jaenisch R.Targeted mutagenesis of Timp-1 reveals that lung tumorinvasion is influenced by Timp-1 genotype of the tumorbut not by that of the host. Oncogene 1996;13:2307–14.

38. Chesler L, Golde DW, Bersch N, Johnson MD.Metalloproteinase inhibition and erythroid potentiationare independent activities of tissue inhibitor of metal-loproteinases-1. Blood 1995;86:4506–15.

39. Porter JF, Shen S, Denhardt DT. Tissue inhibitor ofmetalloproteinase-1 stimulates proliferation of humancancer cells by inhibiting a metalloproteinase. Br JCancer 2004;90:463–70.

40. Kong Y, Poon R, Nadesan P, et al. Matrix metal-loproteinase activity modulates tumor size, cell motility,and cell invasiveness in murine aggressive fibromatosis.Cancer Res 2004;64:5795–803.

41. Yamauchi K, Ogata Y, Nagase H, Shirouzu K. Inhibitionof liver metastasis from orthotopically implanted coloncancer in nude mice by transfection of the TIMP-1 geneinto KM12SM cells. Surg Today 2001;31:791–8.

42. Kruger A, Soeltl R, Sopov I, et al. Hydroxamate-typematrix metalloproteinase inhibitor batimastat promotesliver metastasis. Cancer Res 2001;61:1272–5.

43. Della Porta P, Soeltl R, Krell HW, et al. Combinedtreatment with serine protease inhibitor aprotinin andmatrixmetalloproteinase inhibitor batimastat (BB-94) doesnot prevent invasion of human esophageal and ovariancarcinoma cells in vivo . Anticancer Res 1999;19:3809–16.

44. Tester AM, Waltham M, Oh SJ, et al. Pro-matrixmetalloproteinase-2 transfection increases orthotopicprimary growth and experimental metastasis of MDA-MB-231 human breast cancer cells in nude mice. CancerRes 2004;64:652–8.

45. Mars WM, Zarnegar R, Michalopoulos GK. Activationof hepatocyte growth factor by the plasminogenactivators uPA and tPA. Am J Pathol 1993;143:949–58.

46. Lee SL, Dickson RB, Lin CY. Activation of hepatocytegrowth factor and urokinase/plasminogen activator bymatriptase, an epithelial membrane serine protease.J Biol Chem 2000;275:36720–5.

47. Reuning U, Magdolen V, Wilhelm O, et al. Multifunc-tional potential of the plasminogen activation system intumor invasion and metastasis [review]. Int J Oncol1998;13:893–906.

48. Capodici C, Berg RA. Cathepsin G degrades dena-tured collagen. Inflammation 1989;13:137–45.

49. Machovich R, Owen WG. An elastase-dependentpathway of plasminogen activation. Biochemistry 1989;28:4517–22.

50. Blobel CP. ADAMs: key components in EGFRsignalling and development. Nat Rev Mol Cell Biol2005;6:32–43.



2007;67:8615-8623. Cancer Res Charlotte Kopitz, Michael Gerg, Obul Reddy Bandapalli, et al. SignalingMetastasis by Induction of Hepatocyte Growth Factor Tissue Inhibitor of Metalloproteinases-1 Promotes Liver

Updated version

http://cancerres.aacrjournals.org/content/67/18/8615

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2008/03/13/67.18.8615.DC1.html

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/67/18/8615.full.html#ref-list-1

This article cites 49 articles, 16 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/67/18/8615.full.html#related-urls

This article has been cited by 17 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

[email protected] at

To request permission to re-use all or part of this article, contact the AACR Publications


http://cancerres.aacrjournals.org/content/67/18/8615

http://cancerres.aacrjournals.org/content/suppl/2008/03/13/67.18.8615.DC1.html

http://cancerres.aacrjournals.org/content/67/18/8615.full.html#ref-list-1

http://cancerres.aacrjournals.org/content/67/18/8615.full.html#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

mailto:[email protected]


Documents

Tissue Inhibitor of Metalloproteinases-1 Promotes Liver Metastasis by Induction of Hepatocyte Growth Factor Signaling