Oxidative Stress and Prostate Cancer Progression Are ...Hoang-Lan Nguyen1,2,3, Stanley Zucker2,3, Kevin Zarrabi1, Pournima Kadam2, Cathleen Schmidt3, and Jian Cao1,2 Abstract Oxidative

Angiogenesis, Metastasis, and the Cellular Microenvironment

Oxidative Stress and Prostate Cancer Progression AreElicited by Membrane-Type 1 Matrix Metalloproteinase

Hoang-Lan Nguyen1,2,3, Stanley Zucker2,3, Kevin Zarrabi1, Pournima Kadam2,Cathleen Schmidt3, and Jian Cao1,2

AbstractOxidative stress caused by high levels of reactive oxygen species (ROS) has been correlated with prostate cancer

aggressiveness. Expression of membrane-type 1 matrix metalloproteinase (MT1-MMP), which has beenimplicated in cancer invasion and metastasis, is associated with advanced prostate cancer. We show here thatMT1-MMP plays a key role in eliciting oxidative stress in prostate cancer cells. Stable MT1-MMP expression inless invasive LNCaP prostate cancer cells with low endogenous MT1-MMP increased activity of ROS, whereasMT1-MMP knockdown in DU145 cells with high endogenous MT1-MMP decreased activity of ROS.Expression of MT1-MMP increased oxidative DNA damage in LNCaP and in DU145 cells, indicating thatMT1-MMP–mediated induction of ROS caused oxidative stress. MT1-MMP expression promoted a moreaggressive phenotype in LNCaP cells that was dependent on elaboration of ROS. Blocking ROS activity using theROS scavenger N-acetylcysteine abrogated MT1-MMP–mediated increase in cell migration and invasion. MT1-MMP–expressing LNCaP cells displayed an enhanced ability to grow in soft agar that required increased ROS.Using cells expressing MT1-MMP mutant cDNAs, we showed that ROS activation entails cell surface MT1-MMP proteolytic activity. Induction of ROS in prostate cancer cells expressing MT1-MMP required adhesion toextracellular matrix proteins and was impeded by anti-b1 integrin antibodies. These results highlight a novelmechanism of malignant progression in prostate cancer cells that involves b1 integrin–mediated adhesion, inconcert withMT1-MMP proteolytic activity, to elicit oxidative stress and induction of a more invasive phenotype.Mol Cancer Res; 9(10); 1305–18. �2011 AACR.

Introduction

Prostate cancer, a disease which currently accounts formore than 27,000 deaths per year in the United States (1), ismost often a disease effectively treated by surgery or radi-ation. However, in many cases, prostate cancer progressesfrom an indolent to a more aggressive state that is refractoryto conventional therapy. The factors underlying prostatecancer transition to the more advanced state remain poorlyunderstood and present a significant challenge for improv-ing treatment and survival of prostate cancer patients.Oxidative stress, a state in which cellular reactive oxygen

species (ROS) level exceeds the ability to detoxify the ROSor to repair the ROS-mediated damage, has been shown tosignificantly increase the risk of developing prostate cancer

(2–5). Oxidative stress in prostate cancer cells has beenshown to be necessary for their invasive phenotype, withmore aggressive prostate cancer cells displaying greateroxidative stress than less aggressive cells (6). While thereis mounting evidence that oxidative stress plays an impor-tant role in development and progression of prostate cancer,there is incomplete understanding of the source and role ofoxidative stress in prostate cancer.A sizable body of literature has accumulated implicating

membrane-type 1 matrix metalloproteinase (MT1-MMP),a member of a family of 24 zinc-dependent endopeptidasesthat mediates extracellular matrix (ECM) degradation andremodeling (7), in invasion and metastasis of many dif-ferent cancer types including prostate cancer (8, 9). Theability of MT1-MMP to enhance cell migration is believedto be one of the key factors fostering cancer invasion andmetastasis (10). As a member of the MMP family, MT1-MMP is able to degrade ECM components, includingcollagens, laminins, fibronectin, and vitronectin, whichcan clear a path to facilitate cell migration and invasion(10). In prostate cancer cell lines, increased MT1-MMP hasbeen associated with the transition from androgen-depen-dent to -independent growth (11, 12). Higher MT1-MMPmRNA levels were reported in human precancerous pros-tatic intraepithelial neoplasia and in prostate cancer tissuethan in benign epithelial tissue (13). Despite a body of

Authors' Affiliations: Divisions of 1Cancer Prevention and 2Hematology-Oncology, Department of Medicine, Stony Brook University, Stony Brook;and 3Research Service, VA Medical Center, Northport, New York

Note: Supplementary data for this article are available at Molecular CancerResearch Online (http://mcr.aacrjournals.org/).

Corresponding Author: Jian Cao, Department of Medicine, Stony BrookUniversity, Life Sciences Bldg. Room 004, Stony Brook, NY 11794. Phone:631-632-1815; Fax: 631-632-1820; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-11-0033

�2011 American Association for Cancer Research.

MolecularCancer

Research

www.aacrjournals.org 1305

on January 8, 2021. © 2011 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst August 17, 2011; DOI: 10.1158/1541-7786.MCR-11-0033

http://mcr.aacrjournals.org/

evidence implicating MT1-MMP in prostate cancer inva-sion and in promoting prostate cancer aggressiveness, theunderlying mechanism by which this occurs is poorlyunderstood.A link between a matrix metalloproteinase and oxidative

stress had been previously described, in which MMP-3expression in breast cancer cells caused oxidative stressvia upregulation of Rac1b (14), prompting us to askwhether there is also an association between MT1-MMPand enhanced ROS production. We report herein thatMT1-MMP expression triggers oxidative stress in prostatecancer cells. The presence of MT1-MMP produced a moreaggressive phenotype in prostate cancer cells, as illustratedby enhanced cell migration and invasion and anchorage-independent cell growth, all of which required increasedROS production. We found that MT1-MMP–mediatedelaboration of ROS requires MT1-MMP proteolytic activ-ity and its localization at the cell surface. Because degrada-tion and remodeling of the ECM by MT1-MMP play anintegral role in promoting cell migration and invasion, weasked whether the interaction of MT1-MMP-expressingcells with the ECM can elicit oxidative stress. By deter-mining ROS production on different substrates, we notedthat MT1-MMP induction of ROS was influenced by b1integrin–mediated adhesion to specific ECM substratesincluding collagen, laminin, and fibronectin. Our resultsallow us to propose a novel pathway of prostate cancerprogression that entail integrin-mediated cell adhesion tothe ECM, together with MT1-MMP proteolytic activity,which collectively generates oxidative stress and enhancedprostate cancer aggressiveness.

Materials and Methods

Cell cultureLNCaP cells stably transfected with green fluorescent

protein (GFP) or with MT1-MMP-GFP were describedpreviously (15). DU145 cells were from American TypeCulture Collection. MT1-MMP small hairpin RNA(shRNA) constructs and retrovirus preparation were previ-ously described (15). DU145 cells infected with retrovirus-encoding GFP or MT1-MMP shRNAs were cultured for2 weeks with 4 mg/mL puromycin.

Detection and quantitative determination of ROSCells were stained with 25 mmol/L dihydrorhodamine 6G

(DHR), 1 mmol/L PF-H2TMRos, 5 mmol/L dihydroethi-dium (DHE), or 25 mmol/L 5-(and-6)-carboxy-20,70-dichlorodihydrofluoresceindiacetate (CM-H2DCFDA) andviewed with 510 to 560 nm excitation and 575 to 590 nmemission for DHR and PF-H2TMRos and 450 to 490 nmexcitation and 505 to 520 nm emission for CM-H2DCFDA.All dyes were purchased from Invitrogen Corp.To quantify ROS by fluorescence images of adherent

cells, cell images were captured with a Nikon Digital Sightcamera attached to a Nikon TE2000S microscope (NikonInstruments, Inc.). Mean intracellular fluorescence intensityof 10 field images, containing more than 200 cells, were

automatically calculated for each group using NIS ElementsBR 3.0 imaging software.To quantify ROS by flow cytometry, adherent cells were

stained with either DHE or CM-H2DCFDA for 45 min-utes in PBS at 37�C in the dark. DHE is oxidized bysuperoxide in live cells to ethidium (16). Cells were liftedwith trypsin-EDTA, washed, and fluorescence data wereacquired within 60 minutes using the BD FACSCalibur,with 488 nm laser and 585 nm (for CM-H2DCFDA) and585 nm (for DHE) bandpass filters.

Western blottingCells were lysed using SDS sample buffer and protein

concentrations were determined using the bicinchoninicacid method (Thermo Fisher Scientific). Equal amounts ofprotein were electrophoresed through a NuPAGE 4% to12% Bis–Tris acrylamide and transferred according tomanufacturer's instructions (Invitrogen Corp.). Westernblotting was conducted as previously described (15).

8-HydroxydeoxyguanosineGenomic DNA was extracted using the DNEasy Blood

and Tissue Kit (Qiagen Inc.) according to manufacturer'sinstructions. Ten micrograms of each DNA sample wasdenatured with heat, digested with S1 nuclease, and depho-sphorylated with shrimp alkaline phosphatase. 8-Hydroxy-deoxyguanosine (8-OHdG) measurement was conductedby ELISA (Cell Biolabs, Inc.) according to manufacturer'sinstructions.Immunohistochemistry staining of 8-OHdG was con-

ducted on rehydrated 5-mm paraffin sections from mousexenograft tumors, using goat anti-8-OHdG (Abd Serotech)diluted in Tris-buffered saline with 1% normal rabbitserum. Secondary antibody staining was conducted usingthe Vector ABC system for anti-goat primary antibodiesusing biotin-labeled rabbit anti-goat horseradish peroxi-dase–conjugated streptavidin. Detection was conductedwith 3,30-diaminobenzidine reagents. Secondary antibodyand detection reagents were from Vector Laboratories, Inc.and were used according to manufacturer's instructions.

Cell migration and cell scatteringCell migration was conducted using the BD Falcon

FluoroBlok 96 System (BD Biosciences). Cells labeled with5 nmol/L of DiIC18 for 30 minutes were seeded in 4replicates wells with 27,000 cells in serum-free, phenolred–free RPMI-1640 medium. Migrated cells throughthe opaque polyethylene terephthalate (PET) membraneafter 8 hours were detected relative to control wells lackingcells using the SpectraMax M2 Microplate Reader (Molec-ular Devices, Inc.) at excitation/emission wavelengths of549/565 nm.To assay scattering, cells were seeded at 20,000 cells

per mL in 0.2% rat tail Type I collagen (BD Biosciences)in RPMI-1640 medium at pH 7.5 and analyzed as previ-ously described (15). Cells were imaged using Zeiss LSM510 META NLO Two-Photon Laser Scanning ConfocalMicroscope System coupled to a Zeiss Inverted Axiovert

Nguyen et al.

Mol Cancer Res; 9(10) October 2011 Molecular Cancer Research1306




200 M Microscope using fluorescein isothiocyanate filtersets, and color-coded depth was rendered on Zeiss LSMImage Browser version 4.2.0.121 software.

Xenograft prostate tumor growthA total of 1.5 � 106 LNCaP/MT1-MMP and LNCaP/

GFP cells were suspended in 50 mL of PBS, mixed with50 mL Matrigel (BD Biosciences), and injected subcutane-ously in the right flank of 6-week-old athymic (nu/nu) malemice (Taconic Farms). Animals were weighed, tumor di-mension was measured with calipers, and volume wasdetermined by the formula: length � width2/2. Animalswere euthanized when the tumor diameter reached 1.5 cm,if the tumor was ulcerated, or if the animal lost 10% or moreof its body weight. At euthanasia, tumor tissue was collectedin 4% paraformaldehyde and embedded in paraffin.

Soft agar clonogenic assayCells were cultured in 96-well plates at 500 cells per well

in 50 mL of 0.35% agar over a base of 50 mL of 0.5% agar incomplete medium. Live colonies in the soft agar werecounted in 3 replicate wells 14 days after seeding. Relativenumber of viable cells in each well was monitored by addingalamarBlue (Invitrogen Corp.) solution to cells per manu-facturer instructions for 24 hours before measuring fluo-rescence at 570-nm excitation and 585-nm emission.

MT1-MMP mutants and gelatin zymographyCOS-1 cells were transiently transfected with MT1-

MMP mutant constructs including MT1D535 (17) andMT1E240!A (17). The MTDPEX construct was preparedas previously described (18). In brief, an N-terminal MT1-MMP fragment using the MT1-MMP forward primer(50-CACGAATTCCGGACCATGTCTCCCGCCCCAAGA-30) and reverse primer (50-AGCCGCGCTCACCGCCCCACAGATGTTGGGCCCATA-30) and a C-termi-nal fragment using the forward primer (50-GGGGCG-GTGAGCGCGGCTGCCGTG-30) and reverse MT1-MMP primer (50-ACCCTGGATGGCGTAGAAGCT-GCTGCTTTG-30) were generated from the MT1-MMPcDNA template. The products were used as template togenerate MT1-MMP with a deleted PEX domain usingMT1-MMP forward and reverse primers. Eighteen hoursafter transfection, medium was replaced with fresh serum-free DMEM containing recombinant pro-MMP-2 (19).Conditioned media from the transfected cells were analyzedby gelatin zymography as previously described (15).

Fluorescence-labeled gelatin degradation assayAcid-washed cover glasses were coated with a solution of

Texas Red–labeled gelatin, 1 mg/mL in 70% glycerol for 1hour at room temperature in the dark. Coated cover glasseswere washed once with 70% ethanol and air dried andequilibrated with PBS before use. Transfected COS-1 cellswere seeded at 50,000/cm2, cultured for 18 hours over-night, washed twice with PBS, fixed in 4% paraformalde-hyde for 10 minutes, and mounted onto Fluoromount Gmounting medium (Southern Biotech). Cover glasses were

viewed and photographed at excitation/emission wave-lengths of 549/565 nm.

Cell adhesionCells were prelabeled with 5 mmol/L Calcein AM (Invi-

trogen Corp.), counted on the Cellometer Auto T4 (Nex-celom Bioscience) cell counter, and seeded at 1.5� 105 cellsper cm2 in triplicate wells of 96-well Millipore MillicoatECM-coated plates [collagen I, collagen IV, laminin, fibro-nectin, vitronectin, and bovine serum albumin (BSA)] withor without anti-integrin antibodies in serum-free media for1 to 2 hours. ECM-coated plates were purchased fromMillipore. Prelabeled cells seeded with 2 mg/mLmonoclonalanti-b1 integrin (clone 4B4; Beckman-Coulter) or mono-clonal anti-a6 integrin (Mab 13444-20; Millipore) antibo-dies were preincubated with the antibodies for 30 minutes atroom temperature before seeding. Fluorescence was deter-mined at excitation/emission wavelengths of 487/520 nmbefore and after washing twice with serum-free media. Ad-hesion was expressed for each respective well, as the percent-age of fluorescence after washing relative to before washing.

Aconitase assayAll reagents and chemicals were purchased from Cayman

Chemical Company and the assay was conducted accordingto manufacturer's instructions. Briefly, cells cultured in 10-cm diameter plates were washed with cold PBS, scraped,transferred to Eppendorf tubes, and pelleted at 800 � g for10 minutes at 4�C. The cell pellet was resuspended in 50mmol/L Tris, pH ¼ 7.4, containing sodium citrate, son-icated to homogenize, and spun at 20,000 � g for 15minutes at 4�C. Total protein level was determined usingthe bicinchoninic acid assay (Thermo Fisher Scientific) onthe supernatant. Fifty micrograms of each supernatantsample, with or without oxalomalate inhibitor, was mixedwith NADPþ, isocitric dehydrogenase, and sodium citrateand incubated at 37�C for 15 minutes. A340 was deter-mined every 30 seconds for a total of 15 minutes and theslope of A340 versus time in minutes was determined. Theslope of each sample was subtracted from the slope of blanksamples, and aconitase activity for each cell sample wasdetermined by the following formula:

Slope Sampleð Þ½ � � Slope Sample� Inhibitorð Þ½ �0:0313 ðmmol=LÞ�1 � Total volume

Volume sample

where 0.313 (�mol/L)-1 is the extinction coefficient ofthe NADPH adjusted for the path length of samples in a96-well plate.

Results

MT1-MMP induces oxidative stress in prostate cancercellsBoth oxidative stress (2, 3, 6) and MT1-MMP expression

(12, 13, 20, 21) have been reported to be important forprostate cancer pathogenesis and aggressiveness. To testwhether there is a link between MT1-MMP expressionand ROS activity in prostate cancer, we assayed ROS

MT1-MMP Induction of ROS in Prostate Cancer Invasion

www.aacrjournals.org Mol Cancer Res; 9(10) October 2011 1307




activity in MT1-MMP–expressing prostate cancer cells,using a number of different dyes designed to detectROS. These include DHE, DHR, CM-H2DCFDA, andPF-H2TMRos (Fig. 1).Using androgen-dependent LNCaP cells which express

an undetectable level of MT1-MMP, we measured ROS

levels in LNCaP cells stably transfected with MT1-MMP-GFP cDNA or with control GFP cDNA (15). By addingDHE to adherent cells, followed by flow cytometry, wefound that LNCaP/MT1-MMP-GFP cells had significantlygreater DHE fluorescence than LNCaP/GFP or untrans-fected LNCaP cells (Fig. 1A, left). We were unable to use

Nguyen et al.





CM-H2DCFDA for flow cytometry in these cells becausethe GFP excitation/emission properties in these cells over-lapped with those of CM-H2DCFDA and interfered withflow cytometric measurements. To confirm results obtainedfrom DHE staining, we used DHR, which, like CM-H2DCFDA, is sensitive to oxidation by hydrogen peroxideand PF-H2TMRos, an indicator of intracellular redox po-tential, for ROS staining in cells. In accordance with resultsfrom flow cytometry, we noted increased intracellular fluo-rescence intensity of both DHR and PF-H2TMRos byfluorescence microscopy, in MT1-MMP-GFP–expressingLNCaP cells relative to LNCaP/GFP (Fig. 1A, right). Theseobservations were consistent with a link between over-expression of MT1-MMP and elevated ROS.To further confirm the role of MT1-MMP in ROS induc-

tion in prostate cancer, we used DU145 cells, an invasive,androgen-independent prostate cancer cell line, which pro-duces high levels of endogenous MT1-MMP. Two indepen-dent DU145 PCa cell lines each infected with differentretrovirus constructs encoding MT1-MMP shRNA (MT1shRNA1 and MT1 shRNA2) resulted in downregulatedMT1-MMP expression by real-time reverse transcriptasePCR (Fig. 1B, left) and byWestern blotting (data not shown)compared with DU145 cells expressing an irrelevant GFPshRNA (DU145/GFP shRNA).We assessedROSproductionin these cell lines using PF-H2TMRos and CM-H2DCFDA.Intracellular CM-H2DCFDA fluorescence intensity, as deter-mined by flow cytometry, showed significantly decreasedmean fluorescence intensity in both DU145/MT1 shRNA1and DU145/MT1 shRNA2 cells than in uninfected DU145cells or DU145/GFP shRNA (Fig. 1B, middle). Decreasedredox potential, as indicated by PF-H2TMRos intensity, wasalso observed in DU145/MT1 shRNA than in DU145/GFPshRNA cells (Fig. 1B, right). We further confirmed impor-tance of MMP proteolytic activity in ROS generation bycomparing ROS levels of DU145/GFP and DU145/MT1shRNA with the respective cell lines treated with 1 mmol/Lof the broad-spectrum MMP inhibitor BB3103 (Fig. 1C).

Decreased ROS in DU145/GFP cells upon BB3103 treat-ment provides further evidence that activity of MMPs, inparticular, that of MT1-MMP can induce ROS.One of the consequences of increased ROS and oxidative

stress is oxidative DNA damage, measured by determiningthe level of 8-OHdG, a byproduct of DNA damage. Thus,8-OHdG is a commonly used marker of oxidative stress.Moreover, oxidative damage to DNA is thought to havesignificant implications for prostate cancer tumorigenesis bycontributing to increased mutation rates and genomicinstability (22, 23). Using an ELISA to measure 8-OHdG,we found that LNCaP/MT1-MMP-GFP cells displayedsignificantly increased 8-OHdG levels than LNCaP/GFPand untransfected LNCaP cells (Fig. 1D, left). Conversely,DU145/MT1 shRNA1 cells had significantly decreasedlevel of 8-OHdG relative to control DU145/GFP shRNA(Fig. 1D, left). To determine whether MT1-MMP expres-sion can cause prostate cancer oxidative stress in vivo,LNCaP/GFP and LNCaP/MT1-MMP-GFP cells wereinjected into nude mice. As expected from previous studies(24), mean tumor volume was significantly greater within25 days in mice injected withMT1-MMP-GFP–transfectedLNCaP cells than that of GFP-transfected LNCaP cells(data not shown). Immunohistochemical staining of 8-OHdG in tumor tissue sections from these mice showedgreater staining in LNCaP/MT1-MMP-GFP cancer cellsthan in LNCaP/GFP cells (Fig. 1D, right), consistent withfindings by 8-OHdG ELISA in the cell lines (Fig. 1D, left).To confirm the existence of oxidative stress, we measured

the aconitase activity of LNCaP/GFP and LNCaP/MT1-MMP-GFP. Aconitase is an iron–sulfur protein that cata-lyzes the conversion of citrate to isocitrate (25). Exposure ofaconitase to pro-oxidants has been shown to inhibit itsactivity (26); thus, loss of aconitase activity is a sensitiveindicator of oxidative damage, and aconitase suppression is asensitive endogenous marker of ROS. We found thatLNCaP/MT1-MMP-GFP had more than 3-fold loweraconitase activity than control LNCaP/GFP cells

Figure 1. Expression of MT1-MMP in prostate cancer cells elicits oxidative stress. A, left, untransfected LNCaP, GFP-transfected control (LNCaP/GFP),and MT1-MMP-GFP–transfected LNCaP cells (LNCaP/MT1-MMP-GFP) were stained with the DHE for 1 hour, washed, and fluorescence intensity of10,000cells per groupwas analyzedby flowcytometry at excitation/emissionwavelengthsof 490/585nm.Data represent the geometricmeanof 3 experiments,each result of which was normalized to LNCaP/GFP � SEM. Statistical significance was determined with a 2-tailed Student's t test. *, P < 0.05. A, right,LNCaP/GFP and LNCaP/MT1-MMP-GFP cells were stained with PF-H2TMRos (top) to assess intracellular redox potential and DHR (bottom) to stain for ROS.Bar represents 50 mm. B, left, real-time reverse transcriptase PCR analysis of MT1-MMP levels from total RNA extracted fromDU145/GFP shRNA, DU145/MT1shRNA1, and DU145/MT1 shRNA2. Results shown are the mean of 2 replicates � SEM, normalized to the human hypoxanthine phosphoribosyltransferase1 (HPRT1) gene. Middle, wild-type DU145, DU145/GFP shRNA, DU145/MT1 shRNA1, and DU145/MT1 shRNA2 cells were stained with 10 mmol/LCM-H2DCFDA, and fluorescence intensity of 10,000 cells per group was analyzed by flow cytometry at excitation/emission wavelengths of 490/520 nm. Datarepresent the mean of 3 experiments, each result of which was normalized to DU145/GFP shRNA � SEM (black bar). **, P < 0.01; ***, P < 0.001. Right,representative results of PF-H2TMRos staining inDU145, andDU145/MT1 shRNA.Bar represents 50mm.C,DU145/GFP shRNAandDU145/MT1 shRNA1cellswere treated for 1 hour with 1 mmol/L BB3103 before staining with 10 mmol/L CM-H2DCFDA and fluorescence intensity determined (as described in B). Datashown represent the mean of 3 experiments normalized to DU145/GFP shRNA � SEM (black bar). *, P < 0.05. D, left, 8-OHdG content from prostate cancercell genomicDNAwasassayedbyELISA. FourmicrogramofDNA fromeachsamplewas loaded into eachof duplicatewells; results shownare themean�SEM.8-OHdG level in DU145/MT1 shRNA was below the detection limit of the 8-OHdG standard curve used and was considered not detectable (N.D.).Statistically significant differences between each groupwere determined using a 2-tailed Student's t test. *,P < 0.05 for the respective groups. Right, nudemicewere implanted with tumors derived from LNCaP/GFP or from LNCaP/MT1-MMP-GFP. Tumors were dissected, fixed in formalin, and embedded in paraffin.Five-micrometer paraffin sections from these tumors were stained with a goat anti-8-OHdG antibody, horseradish peroxidase–conjugated anti-goat IgG,anddetectedusing3,30-diaminobenzidine substrate (shownasdark staining). Images in the topwere viewedandphotographedat 200� and those in thebottomat 400 �magnification. Bars in top and bottom right represent 100 and 50 mm, respectively. E, aconitase activity was measured from 50 mg of protein lysatesprepared from LNCaP/GFP and LNCaP/MT1-MMP-GFP cells. Aconitase activity shown was measured in untreated lysates (�) or in lysates treated withthe aconitase competitive inhibitor, oxalomalate (OMA). Mean aconitase activity was determined from triplicate samples for each cell line and shown � SEM.Levels of aconitase that were below the detection limits of the fluorescence reader was labeled not detectable. ***, P < 0.001. wt, wild-type.






(Fig. 1E). Inhibition with the aconitase competitive inhib-itor oxalomalate inhibited all aconitase activity down tonondetectable levels in both LNCaP/GFP and LNCaP/MT1-MMP-GFP, confirming the specificity of the assay foraconitase activity.Taken together, we have shown for the first time that

MT1-MMP, expressed in transfected cells or producedendogenously, induces ROS activity and causes oxidativestress in prostate cancer.

MT1-MMP promotes a more invasive, aggressivephenotype via an ROS-dependent mechanismTo determine whether ROS influences the ability of

MT1-MMP to promote invasion, we cultured LNCaP/GFP and LNCaP/MT1-MMP-GFP cells in 0.2% type Icollagen with or without 1 mmol/L of the ROS scavengerN-acetylcysteine (NAC). We had verified independentlythat 1 mmol/L NAC is not cytotoxic by alamarBluefluorescence detection of viable cells (data not shown).Presence of GFP in our LNCaP cells allowed us to useconfocal laser scanning microscopy and Zeiss LSM ImageBrowser version 4.2.0.121 software to render relative depthusing color codes, with cells colored from red to bluerepresenting closest to farthest, respectively. Thus, cellsdisplaying 3-dimensional scattering would be expected todisplay a greater spectral range, indicating variety of depth.LNCaP cells are minimally invasive and are unable tomigrate and scatter in a 3-dimensional collagen matrix(15), and as expected, LNCaP/GFP displayed very littlecolor variations in the same field, indicating lack of3-dimensional scattering. As we anticipated, expression offull-length MT1-MMP in LNCaP cells stimulated cellscattering/invasion at 4 days, in a 3-dimensional collagenmatrix relative to control LNCaP/GFP cells, as shown inFigure 2A. Both differential photomicrograph images frominterference contrast and confocal laser scanning microsco-py with color-coded depth show 3-dimensional scattering ofLNCaP/MT1-MMP-GFP cells, with cells in the same fielddisplaying depth coloration ranging from red to blue.Treatment of LNCaP/MT1-MMP-GFP with NAC wasfound to inhibit MT1-MMP–mediated cell scattering,resulting in all cells in the depth-colored field displayinglittle color variation. These results support the notion thatMT1-MMP–induced cell scattering/invasion is dependenton ROS activity.Because ROS is required for MT1-MMP–mediated cell

invasion,we askedwhetherMT1-MMP–inducedROSactiv-ity plays a role in cellmigration.To this end, we tested relativecell migration in a modified Boyden chamber in the presenceof 1 mmol/L NAC. In agreement with previous reports(10, 24, 27), MT1-MMP promoted LNCaP cell migrationas compared with controls (Fig. 2B). Interestingly, additionof NAC reduced MT1-MMP–enhanced cell migration tothe level ofGFP-expressing LNCaP cells (Fig. 2B), suggestingthat MT1-MMP–mediated prostate cancer cell migrationand invasion occurs through elevated ROS activity. Theseresults are consistent with previously published observations(28) that repeated exposure of epithelial cells to sublethal

doses of hydrogen peroxide, over as short a period as 2 days,caused an increase in invasive behavior.Previous studies had suggested that oxidative stress was

correlated with an aggressive phenotype in prostate cancercells (6). Accordingly, we asked whether increased oxida-tive stress mediated by MT1-MMP can lead to a moreaggressive cancer phenotype. To address this question, weused an in vitro soft agar assay. We found that LNCaP/MT1-MMP-GFP had significantly enhanced ability toproliferate in soft agar, as assayed by alamarBlue fluores-cence in the first 7 days (Fig. 2C, left). The alamarBluedye was added to a set of cells immediately after seeding tomonitor the initial number of viable cells. Viable cells werefirst detectable by alamarBlue on the second day after cellswere seeded in soft agar and approximately 24 hours afterthe dye was added. Although all cells were counted anddiluted to the same seeding density, actual cell count onthe second day after seeding revealed that there were moreviable LNCaP/GFP cells than LNCaP/MT1-MMP-GFP.Nevertheless, by the seventh day after cell seeding, pro-liferation rate of LNCaP/MT1-MMP-GFP cells was sig-nificantly greater than of control, LNCaP/GFP cells(Fig. 2C, left). All cells cultured in the presence of asublethal dose of 1 mmol/L NAC displayed profoundinhibition of growth in soft agar, even by the second dayafter cell seeding (Fig. 2C, left). We were able to count cellcolonies by 14 days after cell seeding and found thatLNCaP/MT1-MMP-GFP had enhanced ability to formcolonies in soft agar compared with LNCaP/GFP cells(Fig. 2C, middle and right). Consistent with results fromdetermining alamarBlue fluorescence, addition of a non-toxic concentration of NAC inhibited colony formation insoft agar for both LNCaP/GFP and LNCaP/MT1-MMP-GFP (Fig. 2C, middle and right).These results collectively support the notion that

increased ROS production, triggered by prostate cancercell expression of MT1-MMP, can lead to a more invasivephenotype and to enhanced malignancy.

Induction of ROS requires MT1-MMP proteolyticactivity and membrane anchorageTo shed light on the mechanism by which MT1-MMP

elicits oxidative stress in prostate cancer cells, we beganby asking which functional domains of MT1-MMP areimportant in inducing ROS. We had found that full-lengthMT1-MMP can induce ROS in COS-1 African greenmonkey kidney epithelial cells, and that COS-1 cells canbe transfected more efficiently than LNCaP cells. We thuschose to compare ROS levels of COS-1 cells transfectedwith full-length MT1-MMP with those of COS-1 cellstransfected with mutant MT1-MMP constructs to assessthe roles of different domains of MT1-MMP in ROSinduction. Accordingly, we transiently transfected COS-1cells with a control empty vector, a vector expressingfull-length MT1-MMP, or deletion mutant constructsthat included a deleted PEX domain (MTDPEX), a non-functional catalytic domain mutant with glutamine toalanine substitution at position 240 (MT1E240!A;

Nguyen et al.





ref. 15), and a tethering-terminal domain mutant thatremoves both the cytoplasmic and transmembrane domainsand thus converts the MT1-MMP molecule into a soluble,secreted form (MT1D535), as described schematically in

Figure 3A (left). Transfection efficiency was observed tobe 40% to 60%, based on estimates from control transfec-tions of GFP-expressing vector (data not shown). Westernblot analysis of equal amounts of protein from each

Figure 2. MT1-MMP expression promotes a more invasive, aggressive phenotype in prostate cancer cells. A, LNCaP, LNCaP/GFP, and LNCaP/MT-MMP-GFP cells were cultured in 0.2% collagen with or without 1 mmol/L NAC. On the left, cells shown were photographed under differential interferencecontrast (DIC) microscopy 4 days after initial seeding. On the right, different views of the same cell samples were photographed under confocalmicroscopy and shown represented with color-coded depth. Blue to red spectrum represents cells ranging from closest to farthest away from viewer. Barrepresents 50 mm. B, relative migration of LNCaP/GFP and LNCaP/MT-MMP-GFP cells treated with 1 mmol/L NAC (light gray) or vehicle (dark gray),through BD Falcon HTS FluoroBlok 96 inserts, 7 hours after cell seeding into the inserts. Results are expressed as mean of 4 replicate wells � SEM.Statistically significant differences determined using a 2-tailed Student's t test. *, P < 0.05. C, LNCaP/GFP and LNCaP/MT1-MMP-GFP cells were cultured in 3replicate wells of 96-well plates, with 1 mmol/L NAC (light gray) or vehicle (dark gray) in soft agar for 14 days. Top, relative number of viable cells as determinedby alamarBlue staining at days 2 and 7 are shown. The alamarBlue data points below the detection limit of the fluorimeter were designated nondetectable(N.D.). Middle, the number of viable colonies formed at 14 days, as shown graphically. Bottom, representative micrograph of colonies formed at 14 days.Bars for all graphs represent the mean of 3 replicate wells � SEM. *, P < 0.05 by a 2-tailed Student's t test. Bar in micrograph represents 500 mm.RFU, relative fluorescence units.






transfected cell sample showed that the expression level ofMT1-MMP wild-type and deletion mutants were similar(Fig. 3A, right).Full-length MT1-MMP was able to proteolytically

activate pro-MMP-2 to its fully active form, as shown bygelatin zymography, whereas MT1-MMP deletion mutantswere unable to convert pro-MMP-2 to active MMP-2(Fig. 3B). COS-1 cells transfected with full-length MT-MMP were also able to directly degrade Texas Red–labeledgelatin substrate, as shown by a Texas Red–free cleared areasurrounding some MT1-MMP–transfected cells (Fig. 3C).Significant Texas Red–labeled gelatin degradation wasalso observed by COS-1cells transfected with MTDPEX(Fig. 3C), suggesting that the PEX domain of MT1-MMPis not required for direct degradation of ECM substrates.However, cells transfected with either MT1/E240!A orMT1D535 were unable to effect degradation of Texas Red–labeled gelatin to a visually appreciable extent (Fig. 3C),suggesting a necessity for the catalytic domain and for cell

membrane localization in direct MT1-MMP proteolysis ofECM substrates.Transfected cells were stained with DHE, and mean

intracellular fluorescence was determined by flow cytome-try. Results from 3 independent experiments were normal-ized relative to full-length MT1-MMP–transfected cells,and the mean ROS levels are displayed graphically(Fig. 3D). Of the mutants tested, the MT1/E240!Acatalytic domain mutant displayed approximately 67%reduction in ROS relative to full-length MT1-MMP–trans-fected cells, using DHE fluorescence of vector-transfectedcells as baseline. The MT1D535 membrane tetheringdomain deletion mutants displayed a more modest, yetconsistent approximately 27% reduction in ROS, relative tofull-length MT1-MMP–transfected cells (Fig. 3D). Theseresults suggest that MT1-MMP catalytic function and itsability to be localized to the cell membrane are importantproperties for induction of ROS. The MTDPEX transfec-tants did not display decreased ROS, indicating that the

A B

C D

Figure 3. Induction of ROS requires MT1-MMP proteolytic activity and membrane anchorage. COS-1 cells were transiently transfected with expressionvectors expressing full-length MT1-MMP or MT1-MMP mutants. A, MT1-MMP constructs are shown schematically (left). Relative expression of eachtransfectant was monitored by Western blotting using anti-MT1-MMP antibody to the hinge region to stain MT1-MMP and anti-a-tubulin antibodyto stain tubulin loading control (right). B, ability of each MT1-MMP construct to activate pro-MMP-2 was determined by assessing the activity of fully activeMMP-2 in the conditioned media of each COS-1–transfected sample via gelatin zymography. Latent, intermediate, and fully active MMP-2 are shownhighlighted by arrows. C, transfected cells were seeded onto cover glasses coated with Texas Red–labeled gelatin in serum-free medium and incubatedovernight. Cover glasses were fixed with 4% paraformaldehyde and mounted onto glass slides. The capability of each MT1-MMP construct to effectdegradation of Texas Red–labeled gelatin, shown as black areas, is shown highlighted by arrows. Because transfection efficiency of COS-1 cells isapproximately 50% (data not shown), only 50% of the cells in each panel would be expected to be expressing MT1-MMP constructs. Bar represents 50 mm.D, transfected cells were assayed for ROS by labeling with 5 mmol/L DHE for 2 hours and determining the mean fluorescence intensity of each sample byflow cytometry. Each data point in the graph represents the mean of 3 independent experiments, each normalized to full-length MT1-MMP–transfectedcontrols,� SEM. *, P < 0.05 as determined from 2-tailed Student's t test in comparing DHE levels with full-length MT1-MMP transfectants using fluorescenceof vector-transfected cells as baseline. TM, transmembrane.

Nguyen et al.





PEX domain of MT1-MMP is not vital for MT1-MMPinduction of ROS.

MT1-MMP induction of ROS involves b1integrin–mediated adhesion to ECM componentsBecause induction of ROS requires MT1-MMP pro-

teolytic activity localized to the cell surface, and MT1-MMP degrades and remodels the ECM (10), we askedwhether MT1-MMP can induce ROS by mediating cellinteractions with the ECM. To address this hypothesis,LNCaP/GFP and LNCaP/MT1-MMP-GFP cells werecultured in serum-free medium on tissue culture platescoated with selected ECM components. Cell attachment totissue culture plastic has been shown to depend directly onthe ability of the tissue culture plastic material to selectivelyadsorb ECM proteins such as fibronectin and vitronectin(29–31). The cells, themselves, produce and secrete ECMproteins that become adsorbed onto the tissue cultureplastic and influence cell attachment. We chose to controlthe interactions of our experimental cells with the sub-stratum by culturing cells in serum-free media on tissueculture plastic plates coated with collagen I, collagen IV,fibronectin, laminin, vitronectin, or BSA control.The cells displayed morphologic variations on different

ECM substrates with very little cell attachment andspreading on collagen types I and IV and, as expected,on BSA (Fig. 4A, left). Measurement of adhesion 2 hoursafter cell seeding confirmed our visual observation of lowadhesion to collagen I, collagen IV, vitronectin, and BSAand high levels of adhesion in cells cultured on fibronectinand laminin (Fig. 4A, right). We noted, furthermore, thatLNCaP/MT1-MMP-GFP cells displayed greater cell ad-hesion to all ECM substrates tested and were markedlygreater in the cases of fibronectin and laminin (Fig. 4A,right). These results suggest a role for MT1-MMP inaltering adhesion of cells on ECM substrates.As we were studying the interaction between the cells and

the ECM, we chose to quantify intracellular ROS ofadherent cells in situ, by photographing the cells at constantexposure times and measuring intracellular PF-H2TMRosfluorescence intensity via image analysis software. We hadobserved that intracellular ROS levels quantified by thismethod were highly reproducible and closely matchedresults obtained using flow cytometry.LNCaP/MT1-MMP-GFP displayed significantly in-

creased ROS relative to LNCaP/GFP cells when culturedon laminin or fibronectin substrates, as shown by quanti-tative determination of PF-H2TMRos fluorescence(Fig. 4B, left) and in fluorescent micrographs (Fig. 4B,right). In contrast, no significant differences in ROS levelbetween LNCaP/MT1-MMP-GFP and LNCaP/GFP cellswere noted when cells were cultured on collagen I, collagenIV, vitronectin, or BSA (data not shown). These dataindicated that ROS induction by MT1-MMP occurredonly on substrates to which the cells adhered well, inparticular, laminin and fibronectin. Because LNCaP/MT1-MMP-GFP cells displayed markedly greater adhesionto laminin and fibronectin than control LNCaP/GFP cells,

this suggested that cell adhesion plays an important role inMT1-MMP–mediated induction of ROS.As the functional interactions of cells with the ECM are

known to involve integrins (32), we asked whether disrup-tion of cell adhesion using anti-integrin antibodies can alsoprevent induction of ROS in MT1-MMP–expressing cells.Seeding of LNCaP/MT1-MMP-GFP cells in the presenceof anti-b1 integrin antibody (2 mg/mL) lowered theiradhesion to all ECM substrates tested including laminin,fibronectin, and vitronectin (Fig. 4C, left). In contrast,culture of these cells in the presence of an a6 integrinantibody (2 mg/mL) had no significant effect on celladhesion to any ECM substrates tested (Fig. 4C, left).Inhibition of adhesion by anti-b1 integrin antibody alsoresulted in loss of LNCaP/MT1-MMP-GFP ability toinduce ROS when cultured on laminin, as shown usingDHE staining (Fig. 4C, right). These results suggested thatadhesion of LNCaP/MT1-MMP-GFP to the ECM wasmediated specifically by b1 integrin, and that this adhesionplays a key role in MT1-MMP induction of ROS.To confirm the role of MT1-MMP and cell adhesion on

ROS induction, we also assessed adhesion of DU145/GFPshRNA and DU145/MT1 shRNA1 in serum-free media oncollagen I, collagen IV, fibronectin, laminin, vitronectin, andBSA. We found that, unlike LNCaP cells, both controlDU145/GFP shRNA and DU145/MT1 shRNA adheredwell to collagen I, collagen IV, as well as to fibronectin andlaminin (Fig. 4D, left). Adhesion was, as for LNCaP cells,inhibited with anti-b1 integrin antibodies and not by anti-a6integrin antibodies (Fig. 4D, middle). Anti-b1 integrinantibodies also inhibited ROS production in DU145/GFPcells (Fig. 4D, right) and in DU145/MT1 shRNA (notshown), as evidenced bymoderately decreased PF-H2TMRosintracellular fluorescence intensity (Fig. 4D, right). Theseresults further lend support for the role of b1 integrin–mediated cell adhesion in ROS induction by MT1-MMP.

Induction of ROS does not require MT1-MMPactivation of pro-MMP-2Among the many proteolytic functions of MT1-MMP is

its ability to convert the zymogen pro-MMP-2 to itsactive form. Because induction of ROS by MT1-MMPrequires active MT1-MMP, we asked whether MT1-MMPactivation of pro-MMP-2 is involved in the pathway. Toaddress this question, we cultured LNCaP/GFP andLNCaP/MT1-MMP-GFP cells in serum-free media withor without recombinant active MMP-2 for up to 24 hours(Fig. 5A). As shown in Figure 5A and B, active MMP-2 didnot increase the level of ROS in LNCaP/GFP cells. Theseresults indicate that MMP-2 activation by MT1-MMP isnot required for its pro-ROS activation function.

Discussion

An association between oxidative stress and prostatecancer is well established, implicating oxidative stress inboth the pathogenesis and progression of prostate cancer(4–6, 33). However, the mechanism by which oxidative






***

**

**

**

***

*** *** ********

**

****

**

*

**

D

C

B

A

Nguyen et al.





stress is provoked and the role it plays in prostate cancerdevelopment and etiology are poorly understood. The"oxidative aging" theory proposes that accumulated oxida-tive damage associated with aging increases an individual'srisk of developing cancer (33, 34). This is supported byfindings of increasing oxidative mitochondrial damage andmutations with age (35, 36).In this report, we show a mechanism by which expression

of MT1-MMP, a molecule associated with cancer cellinvasion and metastasis (8) in prostate cancer cells triggersan oxidative stress pathway leading to aggressive malignancy.In cells without appreciable levels of MT1-MMP, as

exemplified by LNCaP, ectopic expression of MT1-MMPtriggered induction of ROS. In DU145 cells, an androgen-independent cell line that possesses a more malignantphenotype than LNCaP cells (6), MT1-MMP knockdownresulted in approximately 50% decrease in ROS. CellularROS levels were quantified by staining adherent cells withDHE or CM-H2DCFDA, followed by cell lifting withtrypsin-EDTA and flow cytometric analysis. This methodprovides a reliable measurement of ROS in adherent cells, asthe ROS dyes oxidized by elevated ROS are retained in thecells (37, 38) even after the cells are placed in suspension.Because ROS determinations using redox-sensitive dyes canbe argued as having nonspecific effects (39), we confirmedour findings using aconitase as an endogenous oxidativesubstrate. Because aconitase activity is inhibited by oxida-tion, our results showing more than 3-fold decreasedaconitase activity in LNCaP/MT1-MMP-GFP cellsthan in control LNCaP/GFP cells provide evidence thatMT1-MMP expression leads to oxidative damage to anendogenous enzyme. Oxidative stress can lead to progressiveaccumulation of oxidativeDNAdamage, which is thought tocontribute to increased genomic instability and greatermalignancy (40). In accordance with this idea, previousstudies have shown that exposure of mammary epithelialcells to low levels of hydrogen peroxide over a prolongedperiod resulted in conversion to amoremalignant phenotype(28). Radisky and colleagues (14) showed that MMP-3 wasable to induce genomic instability in breast cancer cellsthrough a ROS-mediated pathway. Of interest, results froma number of different studies have showed that expression ofMT1-MMP in epithelial cells caused chromosomal insta-bility and aneuploidy (41) and leads to malignant transfor-mation (41, 42).In the current study, we have shown that MT1-MMP

expression in prostate cancer cells caused these cellsto acquire a more aggressive phenotype, as displayed byincreased migration, invasion, and anchorage-independentgrowth and that these qualities were dependent on enhancedROS production.We noted that inhibition of ROS with theantioxidant NAC had an antiproliferative effect on bothLNCaP/MT1-MMP and control LNCaP/GFP in soft agar.This is consistent with the well-documented observationsthat low levels of ROS have a growth-promoting effect on

Figure 4. MT1-MMP induction of ROS is influenced by integrin-mediated adhesion to ECM components. A, Control LNCaP or LNCaP/GFP and LNCaP/MT1-MMP-GFP cells were cultured for �24 hours on various ECM substrates, including collagen I (COL I), collagen IV (COL IV), laminin (Ln), fibronectin(Fn), and vitronectin (Vn), and control bovine serum albumin (BSA) in serum-free media. Differences in cell spreading between LNCaP and LNCaP/MT1-MMP-GFP (A, left) and adhesion (A, right) on different ECM substrates are shown. #, P < 0.01 between LNCaP/GFP and LNCaP/MT1-MMP-GFP on eachrespective ECM; **, P < 0.01; ***, P < 0.001 relative to BSA control. B, ROS was increased in LNCaP/MT1-MMP-GFP only in cells cultured on lamininand fibronectin, as shown graphically (B, left), and as displayed in a representative image (B, right). Intracellular ROS levels were determined on adherentcells by quantifying the mean intracellular fluorescence intensity per cell area of images of H2-PFTMRos-stained cells. Bar represents 100 mm. *, P < 0.05;**, P < 0.01. C, left, cells were seeded, in the presence of 2 mg/mL of antibody to b1 integrin or to a6 integrin, onto 96-well plates coated with fibronectin,laminin, or vitronectin. Adhesion was assayed for each group relative to untreated cells, as shown graphically; only anti-b1 integrin was able to disruptcell adhesion. **, P < 0.01. Right, incubation of LNCaP/MT1-MMP-GFP cells in the presence of b1 integrin also decreased ROS. A representative ROSexperimental result is shown using DHE staining. Bar represents 100 mm. D, left, adhesion of DU145/GFP shRNA and DU145/MT1 shRNA1 to collagen I,collagen IV, fibronectin, laminin, vitronectin, and BSA. Middle, adhesion of DU145/GFP shRNA and DU145/MT1 shRNA1 to collagen I, collagen IV, fibronectin,and laminin in the presence and absence of 2 μmol/L anti-b1 or anti-a6 integrins. Right, representative images of DU145/GFP shRNA cells cultured on collagenIV and on fibronectin cultured with or without anti-b1 integrin antibodies. Cells stained with PFH2TmRos, shown on the right side, are also shownphotographed under DIC, on the left. *, P < 0.05, **, P < 0.01, ***, P < 0.001.

Figure 5. Induction of ROS does not require MT1-MMP activation ofpro-MMP-2. LNCaP/GFP and LNCaP/MT1-MMP-GFP cells wereincubated with media containing recombinant pro-MMP-2. A, gelzymography of conditionedmedia from LNCaP/GFP (left lane) and LNCaP/MT1-MMP-GFP (right lane) incubated overnight with recombinant pro-MMP-2. Conditioned media from LNCaP/MT1-MMP-GFPdisplayed fully activated MMP-2. B, LNCaP/GFP and LNCaP/MT1-MMP-GFP cells were cultured with conditioned media from cells (treated asdescribed in A) containing activated MMP-2 (þMMP-2). Images weretaken 24 hours after seeding, and ROS levels, in RFU of PF-H2TMRos areshown graphically (B) and in representative images. C, representativeimages of PF-H2TMRos–stained cells treated (as described in B).






cultured mammalian cells (43–46). A modest level of ROS(�3 mmol/L) in cells, rather than promoting oxidative stress,is thought to play an integral role in the signaling pathwayregulating cell proliferation (45). Thus, the antiproliferativeeffect of ROS inhibition in cells that are not in a state ofoxidative stress is consistent with the notion that a low levelof ROS is important for regulating cell proliferation How-ever, treating cells with higher levels of ROS (�120 mmol/L)have been reported to profoundly alter the cellular geneexpression profile (45). Exposure of the mouse mammaryepithelial cells NMuMG to similarly high concentrations ofROS caused their transition to a more invasive phenotype(28). While we do not know the intracellular ROS concen-tration of our experimental prostate cancer cells, our resultsindicate that MT1-MMP expression in these cells promoteoxidative stress, as evidenced by increased 8-OHdG levels.The changes brought on by oxidative stress, including alteredexpression of numerous genes, as previously reported (45),can facilitate prostate cancer cells to acquire a more invasivephenotype and to enhance anchorage-independent growthin soft agar.Induction of ROS required MT1-MMP proteolytic ac-

tivity, as evidenced by a loss of ROS induction in MT1/E240-A–transfected COS-1 cells relative to full-lengthMT1-MMP–transfected cells. Loss of the transmembranedomain, rendering MT1-MMP soluble, rather than mem-brane-bound, also consistently resulted in decreased ROSinduction, suggesting the importance of membrane local-ization. It is not directly clear from our results why neitherthe MT1/E240!A nor the MT1D535 mutant constructsinhibited ROS to the level of vector control. We have foundthat ROS levels of LNCaP/MT1/E240!A-GFP stablecell lines were also not decreased to the level of controlLNCaP/GFP cells (data not shown), which supports thenotion that catalytic activity by MT1-MMP, alone, cannotactivate ROS to the level of full-length MT1-MMP. Theobservation that both the MT1/E240!A mutant and theMT1D535 partially decreased ROS level relative to vectorcontrol leads us to hypothesize that both catalytic activityand membrane localization by MT1-MMP are required forfull activation of ROS.Activation of pro-MMP-2 does not appear to play a

significant role in MT1-MMP–mediated ROS production,as evidenced by lack of ROS response in LNCaP cells toaddition of active MMP-2. These results suggest that ROSinduction occurs via direct MT1-MMP proteolytic activityat the cell surface, rather than indirectly via pro-MMP-2activation. Local targets in close proximity to the cellsurface, such as components of ECM or associated withthe ECM, or molecules located on the cell surface wouldbe the most likely targets of MT1-MMP proteolyticactivity along the pathway leading to oxidative stress.Because MT1-MMP interacts with and can proteolyticallyremodel the ECM, we chose to investigate whetherMT1-MMP induction can be influenced by the presenceof different ECM substrates.We found that induction of ROS by MT1-MMP

requires adhesion to ECM substrates such as collagen,

laminin, and fibronectin and this adhesion can be regulatedby b1 integrin. LNCaP/MT1-MMP-GFP cells displayedboth greater adhesion and generated significantly greaterROS than control LNCaP/GFP cells when cultured oneither laminin or fibronectin. ROS levels in LNCaP/MT1-MMP and in DU145/GFP shRNA control cells weredecreased by the presence of anti-b1 integrin antibodies,which also abrogated adhesion in these cells. These resultssuggest that one of the ways MT1-MMP promotes oxida-tive stress is via modulating cell adhesion to the ECM.These results raise the question of how cell adhesion to

the ECM plays a role in MT1-MMP–mediated generationof oxidative stress. Our laboratory had previously deter-mined that MT1-MMP promigratory influence is depen-dent on signaling by the small GTPase Rac1 (17). ActivatedGTP-Rac1 is a regulator of the membrane-associatedNADPH oxidase (Nox) enzymes that reduce molecularoxygen to superoxide, which then go on to generate avariety of different ROS species (47). Furthermore, Nox1-mediated generation of ROS has been shown to be regu-lated by Rac1 (48). We have recently found that treatingMT1-MMP-GFP cells with the Nox inhibitor dipheny-liodonium eliminated much of the observed ROS increase(Nguyen and colleagues, unpublished data). Rac actsreciprocally with Rho in actin assembly and focal adhesionstability. Rho promotes the formation of stress fibers andfocal adhesions whereas Rac1 can suppress Rho activity,destabilizing focal adhesions and support focal complexessuch as lamellipodia (49). It is interesting to note thatMT1-MMP has been reported to destabilize focal adhesioncomplexes (50), suggesting that it may play an importantrole in the same or parallel pathway as that in which Rac1resides.While data presented here do not fully elucidate thepathway by which MT1-MMP produces oxidative stress,the evidence that cell adhesion to the ECM accompaniedbyMT1-MMP proteolytic activity is involved, that ROS isgenerated by the Nox system, and that increased ROSinfluences cell migration, collectively suggest that theRac1-Nox pathway is involved. Additional research alongthis line of thought will need to be pursued to betterunderstand the role of MT1-MMP in this pathway.Data presented here suggest that MT1-MMP expres-

sion modulates adhesion of cancer cells to the ECM.These results appear consistent with previous data thatshowed prolonged exposure of cultured cells to hydrogenperoxide induced upregulation of select integrins (28).We showed here that both, adhesion to the ECM andMT1-MMP proteolytic activity, are important for MT1-MMP–mediated ROS induction. It is uncertain fromthese results whether MT1-MMP proteolytic activityregulates cell adhesion or if the proteolytic activity isrequired subsequent to cell adhesion in ROS generation.We also found that inhibition of ROS with NAC

abrogated ability of both LNCaP/GFP and LNCaP/MT1-MMP-GFP to form colonies in soft agar. Thesefindings are in agreement with previous reports indicatingthat the transformed state of cancer cells is regulated byROS (51, 52). However, they also raise the question of

Nguyen et al.





the role of integrin-mediated adhesion in ROS inductionin the setting of anchorage-independent growth, wherethere is presumably little or no adhesion involved. Wehave observed that LNCaP/MT1-MMP-GFP displayedincreased ROS than LNCaP/GFP, rapidly after additionto tissue culture plates, before significant adhesion wasapparent (Supplementary Fig. S1), suggesting that whileadhesion to the ECM can play an important role ineliciting ROS, MT1-MMP can also induce ROS inde-pendently of adhesion to the ECM.Our results nevertheless show that MT1-MMP expres-

sion elicits oxidative stress in prostate cancer cells and thatthis oxidative stress plays an important role in cell migrationand invasion, promotes oxidative DNA damage, and con-tributes to increased malignancy. These results suggest anovel pathway by which MT1-MMP, an important com-ponent of the cancer cell migration and invasion machinery,can contribute to prostate cancer malignancy by triggeringoxidative stress. Additional studies will be needed to furtherdissect the signaling pathway(s) involved.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interests were disclosed.

Acknowledgments

The authors thank Dr. Holly Colognato, SUNY Stony Brook Department ofPharmacological Sciences, for valuable scientific insight and discussions. In addition,the authors would like to thankDr. Ghassan Samara, SUNY Stony BrookDepartmentof Surgery, for editing and proofreading assistance in the preparation of themanuscript.

Grant Support

This work was supported by NIH 5RO1CA11355301A1 and a grant fromthe Walk-for-Beauty Foundation (J. Cao); an NIH supplemental grantRO1CA11355301S1 and a grant from the Walk for Beauty Foundation (H.-L.Nguyen); and a VA Merit Review grant and a grant from the Carol Baldwin BreastCancer Foundation (S. Zucker).

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.

Received January 19, 2011; revised July 7, 2011; accepted August 8, 2011;published OnlineFirst August 17, 2011.

References1. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Thun MJ. Cancer statistics,

2009. CA Cancer J Clin 2009;59:225–49.2. Chomyn A, Attardi G. MtDNA mutations in aging and apoptosis.

Biochem Biophys Res Commun 2003;304:519–29.3. Dakubo GD, Parr RL, Costello LC, Franklin RB, Thayer RE. Altered

metabolism and mitochondrial genome in prostate cancer. J ClinPathol 2006;59:10–6.

4. Wu JD, Lin DW, Page ST, Lundgren AD, True LD, Plymate SR.Oxidative DNA damage in the prostate may predisposemen to a higher risk of prostate cancer. Transl Oncol 2009;2:39–45.

5. Frohlich DA, McCabe MT, Arnold RS, Day ML. The role of Nrf2 inincreased reactive oxygen species and DNA damage in prostatetumorigenesis. Oncogene 2008;27:4353–62.

6. Kumar B, Koul S, Khandrika L, Meacham RB, Koul HK. Oxidativestress is inherent in prostate cancer cells and is required for aggres-sive phenotype. Cancer Res 2008;68:1777–85.

7. Sternlicht MD, Werb Z. How matrix metalloproteinases regulate cellbehavior. Annu Rev Cell Dev Biol 2001;17:463–516.

8. Poincloux R, Lizarraga F, Chavrier P. Matrix invasion by tumour cells: afocus on MT1-MMP trafficking to invadopodia. J Cell Sci 2009;122:3015–24.

9. Sato H, Takino T. Coordinate action of membrane-type matrix metal-loproteinase-1 (MT1-MMP) and MMP-2 enhances pericellular prote-olysis and invasion. Cancer Sci 2010;101:843–7.

10. Itoh Y. MT1-MMP: a key regulator of cell migration in tissue. IUBMBLife 2006;58:589–96.

11. Nagakawa O, Murakami K, Yamaura T, Fujiuchi Y, Murata J, Fuse H,et al. Expression of membrane-type 1 matrix metalloproteinase (MT1-MMP) on prostate cancer cell lines. Cancer Lett 2000;155:173–9.

12. Jennbacken K, Gustavsson H, Welen K, Vallbo C, Damber JE. Pros-tate cancer progression into androgen independency is associatedwith alterations in cell adhesion and invasivity. Prostate 2006;66:1631–40.

13. Cardillo MR, Di Silverio F, Gentile V. Quantitative immunohistochem-ical and in situ hybridization analysis of metalloproteinases in prostatecancer. Anticancer Res 2006;26:973–82.

14. Radisky DC, Levy DD, Littlepage LE, Liu H, Nelson CM, Fata JE, et al.Rac1b and reactive oxygen species mediate MMP-3-induced EMTand genomic instability. Nature 2005;436:123–7.

15. Cao J, Chiarelli C, Richman O, Zarrabi K, Kozarekar P, Zucker S.Membrane type 1 matrix metalloproteinase induces epithelial-to-

mesenchymal transition in prostate cancer. J Biol Chem 2008;283:6232–40.

16. Zhao H, Kalivendi S, Zhang H, Joseph J, Nithipatikom K, Vasquez-Vivar J, et al. Superoxide reacts with hydroethidine but forms afluorescent product that is distinctly different from ethidium: potentialimplications in intracellular fluorescence detection of superoxide. FreeRadic Biol Med 2003;34:1359–68.

17. Cao J, Kozarekar P, Pavlaki M, Chiarelli C, Bahou WF, Zucker S.Distinct roles for the catalytic and hemopexin domains of membranetype 1-matrix metalloproteinase in substrate degradation and cellmigration. J Biol Chem 2004;279:14129–39.

18. Cao J, Drews M, Lee HM, Conner C, Bahou WF, Zucker S. Thepropeptide domain of membrane type 1 matrix metalloproteinase isrequired for binding of tissue inhibitor of metalloproteinases and foractivation of pro-gelatinase A. J Biol Chem 1998;273:34745–52.

19. Cao J, Rehemtulla A, Pavlaki M, Kozarekar P, Chiarelli C. Furin directlycleaves proMMP-2 in the trans-Golgi network resulting in a nonfunc-tioning proteinase. J Biol Chem 2005;280:10974–80.

20. Upadhyay J, Shekarriz B, Nemeth JA, Dong Z, Cummings GD, Frid-man R, et al. Membrane type 1-matrix metalloproteinase (MT1-MMP)and MMP-2 immunolocalization in human prostate: change in cellularlocalization associated with high-grade prostatic intraepithelial neo-plasia. Clin Cancer Res 1999;5:4105–10.

21. Trudel D, Fradet Y, Meyer F, Harel F, Tetu B. Membrane-type-1 matrixmetalloproteinase, matrix metalloproteinase 2, and tissue inhibitor ofmatrix proteinase 2 in prostate cancer: identification of patients withpoor prognosis by immunohistochemistry. Hum Pathol 2008;39:731–9.

22. Miyake H, Hara I, Gleave ME, Eto H. Protection of androgen-depen-dent human prostate cancer cells from oxidative stress-induced DNAdamage by overexpression of clusterin and its modulation by andro-gen. Prostate 2004;61:318–23.

23. Tam NN, Nyska A, Maronpot RR, Kissling G, Lomnitski L, Suttie A,et al. Differential attenuation of oxidative/nitrosative injuries in earlyprostatic neoplastic lesions in TRAMP mice by dietary antioxidants.Prostate 2006;66:57–69.

24. Cao J, Chiarelli C, Kozarekar P, Adler HL. Membrane type 1-matrixmetalloproteinase promotes human prostate cancer invasion andmetastasis. Thromb Haemost 2005;93:770–8.

25. Villafranca JJ, Mildvan AS. The mechanism of aconitase action. I.Preparation, physical properties of the enzyme, and activation by iron(II). J Biol Chem 1971;246:772–9.






26. Gardner PR. Aconitase: sensitive target and measure of superoxide.Methods Enzymol 2002;349:9–23.

27. Bair EL, Chen ML, McDaniel K, Sekiguchi K, Cress AE, Nagle RB,et al. Membrane type 1 matrix metalloprotease cleaves laminin-10and promotes prostate cancer cell migration. Neoplasia 2005;7:380–9.

28. Mori K, ShibanumaM, Nose K. Invasive potential induced under long-term oxidative stress in mammary epithelial cells. Cancer Res 2004;64:7464–72.

29. Steele JG, Johnson G, Norris WD, Underwood PA. Adhesion andgrowth of cultured human endothelial cells on perfluorosulphonate:role of vitronectin and fibronectin in cell attachment. Biomaterials1991;12:531–9.

30. Steele JG, Johnson G, Underwood PA. Role of serum vitronectin andfibronectin in adhesion of fibroblasts following seeding onto tissueculture polystyrene. J Biomed Mater Res 1992;26:861–84.

31. Wilson CJ, Clegg RE, Leavesley DI, Pearcy MJ. Mediation of bioma-terial-cell interactions by adsorbed proteins: a review. Tissue Eng2005;11:1–18.

32. Hynes RO. Integrins: bidirectional, allosteric signaling machines. Cell2002;110:673–87.

33. Minelli A, Bellezza I, Conte C, Culig Z. Oxidative stress-relatedaging: a role for prostate cancer? Biochim Biophys Acta 2009;1795:83–91.

34. Muller FL, Lustgarten MS, Jang Y, Richardson A, Van Remmen H.Trends in oxidative aging theories. Free Radic Biol Med 2007;43:477–503.

35. Bokov A, Chaudhuri A, Richardson A. The role of oxidative damageand stress in aging. Mech Ageing Dev 2004;125:811–26.

36. Van Remmen H, Richardson A. Oxidative damage to mitochondriaand aging. Exp Gerontol 2001;36:957–68.

37. Bucana C, Saiki I, Nayar R. Uptake and accumulation of the vital dyehydroethidine in neoplastic cells. J Histochem Cytochem 1986;34:1109–15.

38. Kirkland RA, Saavedra GM, Franklin JL. Rapid activation of antioxi-dant defenses by nerve growth factor suppresses reactive oxygenspecies during neuronal apoptosis: evidence for a role in cytochromec redistribution. J Neurosci 2007;27:11315–26.

39. Wardman P. Fluorescent and luminescent probes for measurement ofoxidative and nitrosative species in cells and tissues: progress,pitfalls, and prospects. Free Radic Biol Med 2007;43:995–1022.

40. Kregel KC, Zhang HJ. An integrated view of oxidative stress in aging:basic mechanisms, functional effects, and pathological considera-tions. Am J Physiol Regul Integr Comp Physiol 2007;292:R18–36.

41. Golubkov VS, Chekanov AV, Savinov AY, Rozanov DV, GolubkovaNV, Strongin AY. Membrane type-1 matrix metalloproteinase confersaneuploidy and tumorigenicity on mammary epithelial cells. CancerRes 2006;66:10460–5.

42. Ha HY, Moon HB, Nam MS, Lee JW, Ryoo ZY, Lee TH, et al. Over-expression of membrane-type matrix metalloproteinase-1 geneinduces mammary gland abnormalities and adenocarcinoma in trans-genic mice. Cancer Res 2001;61:984–90.

43. Burdon RH, Alliangana D, Gill V. Hydrogen peroxide and the prolif-eration of BHK-21 cells. Free Radic Res 1995;23:471–86.

44. Burdon RH. Superoxide and hydrogen peroxide in relation to mam-malian cell proliferation. Free Radic Biol Med 1995;18:775–94.

45. Davies KJ. The broad spectrum of responses to oxidants in prolifer-ating cells: a new paradigm for oxidative stress. IUBMB Life 1999;48:41–7.

46. Galli A, Svegliati-Baroni G, Ceni E, Milani S, Ridolfi F, Salzano R, et al.Oxidative stress stimulates proliferation and invasiveness of hepaticstellate cells via a MMP2-mediated mechanism. Hepatology 2005;41:1074–84.

47. Sumimoto H. Structure, regulation and evolution of Nox-familyNADPH oxidases that produce reactive oxygen species. FEBS J2008;275:3249–77.

48. Cheng G, Diebold BA, Hughes Y, Lambeth JD. Nox1-dependentreactive oxygen generation is regulated by Rac1. J Biol Chem2006;281:17718–26.

49. Sander EE, ten Klooster JP, van Delft S, van der Kammen RA, CollardJG. Rac downregulates Rho activity: reciprocal balance between bothGTPases determines cellular morphology and migratory behavior. JCell Biol 1999;147:1009–22.

50. Takino T, Watanabe Y, Matsui M, Miyamori H, Kudo T, Seiki M, et al.Membrane-type 1matrix metalloproteinase modulates focal adhesionstability and cell migration. Exp Cell Res 2006;312:1381–9.

51. Jurgensmeier JM, Panse J, Schafer R, Bauer G. Reactive oxygenspecies as mediators of the transformed phenotype. Int J Cancer1997;70:587–9.

52. Urbano A, Lakshmanan U, Choo PH, Kwan JC, Ng PY, Guo K, et al.AIF suppresses chemical stress-induced apoptosis and maintainsthe transformed state of tumor cells. EMBO J 2005;24:2815–26.

Nguyen et al.





2011;9:1305-1318. Published OnlineFirst August 17, 2011.Mol Cancer Res Hoang-Lan Nguyen, Stanley Zucker, Kevin Zarrabi, et al. Membrane-Type 1 Matrix MetalloproteinaseOxidative Stress and Prostate Cancer Progression Are Elicited by

Updated version

10.1158/1541-7786.MCR-11-0033doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2011/08/16/1541-7786.MCR-11-0033.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/9/10/1305.full#ref-list-1

This article cites 52 articles, 16 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/9/10/1305.full#related-urls

This article has been cited by 3 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]

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

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mcr.aacrjournals.org/content/9/10/1305To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-11-0033

http://mcr.aacrjournals.org/content/suppl/2011/08/16/1541-7786.MCR-11-0033.DC1

http://mcr.aacrjournals.org/content/9/10/1305.full#ref-list-1

http://mcr.aacrjournals.org/content/9/10/1305.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/9/10/1305


Oxidative Stress and Prostate Cancer Progression Are ...Hoang-Lan Nguyen1,2,3, Stanley Zucker2,3, Kevin Zarrabi1, Pournima Kadam2, Cathleen Schmidt3, and Jian Cao1,2 Abstract Oxidative

Documents