Free Radical Biology & Medicine 50 (2011) 206–215

Contents lists available at ScienceDirect

Free Radical Biology & Medicine

j ourna l homepage: www.e lsev ie r.com/ locate / f reeradb iomed

Original Contribution

Inhibition of nuclear factor κB regresses cardiac hypertrophy by modulating theexpression of extracellular matrix and adhesion molecules

Sandeep Kumar a,b,c, Rachid Seqqat a,b,c, Sravanthi Chigurupati a,b,c, Rajesh Kumar a,b,c, Kenneth M. Baker a,b,c,David Young d, Subha Sen d, Sudhiranjan Gupta a,b,c,⁎a Division of Molecular Cardiology, Department of Medicine, College of Medicine, Texas A&M Health Science Center, College Station, TX 77840, USAb Scott & White, Temple, TX 76508, USAc Central Texas Veterans Health Care System, Temple, TX 76504, USAd Department of Molecular Cardiology, Cleveland Clinic, Cleveland, OH 44195, USA

⁎ Corresponding author. Division of Molecular CardioCollege of Medicine, Texas A&M Health Science Center, C

E-mail address: sgupta@medicine.tamhsc.edu (S. Gu

0891-5849/$ – see front matter © 2010 Published by Edoi:10.1016/j.freeradbiomed.2010.10.711

a b s t r a c t

a r t i c l e i n f o

Article history:Received 3 August 2010Revised 19 October 2010Accepted 27 October 2010Available online 1 November 2010

Keywords:Cardiac fibrosisExtracellular matrix proteinNF-κBPyrrolidine dithiocarbamateFree radicals

Myocardial remodeling denotes a chronic pathological condition of dysfunctional myocardium that occurs incardiac hypertrophy (CH) and heart failure (HF). Reactive oxygen species (ROS) are major initiators ofexcessive collagen and fibronectin deposition in cardiac fibrosis. Increased production of ROS and nuclearfactor κB (NF-κB) activation provide a strong link between oxidative stress and extracellular matrix (ECM)remodeling in cardiac hypertrophy. The protective inhibitory actions of pyrrolidine dithiocarbamate (PDTC), apharmacological inhibitor of NF-κB and a potent antioxidant, make this a good agent to evaluate the role ofinhibition of NF-κB and prevention of excessive ECM deposition inmaladaptive cardiac remodeling during HF.In this report, we used a transgenic mouse model (Myo-Tg) that has cardiac-specific overexpression ofmyotrophin. This overexpression of myotrophin in the Myo-Tg model directs ECM deposition and increasedNF-κB activity, which result in CH and ultimately HF. Using the Myo-Tg model, our data showed upregulationof profibrotic genes (including collagen types I and III, connective tissue growth factor, and fibronectin) inMyo-Tg mice, compared to wild-type mice, during the progression of CH. Pharmacological inhibition of NF-κBby PDTC in the Myo-Tg mice resulted in a significant reduction in cardiac mass, NF-κB activity, and profibroticgene expression and improved cardiac function. To the best of our knowledge, this is the first report of ECMregulation by inhibition of NF-κB activation by PDTC. The study highlights the importance of the NF-κBsignaling pathway and therapeutic benefits of PDTC treatment in cardiac remodeling.

logy, Department of Medicine,ollege Station, TX 77840, USA.pta).

lsevier Inc.

© 2010 Published by Elsevier Inc.

Cardiac remodeling is a complex geometric alteration that involvesultrastructural and microscopic changes in the muscle fibers of theheart. This process occurs in various stages of cardiac hypertrophythat ultimately transitions to heart failure. It is well known thatcollagen type I (Col I) and collagen type III (Col III) interconnect thecellular components of the heart, maintaining structural andfunctional integrity. In normal heart, collagen turnover is regulatedby matrix metalloproteinases (MMPs) and their endogenous inhibi-tors, tissue inhibitors of metalloproteinases (TIMPs) [1]. A fine balancebetween MMPs and TIMPs maintains the integrity of the extracellularmatrix (ECM). An imbalance between MMPs and TIMPs alters thecollagen turnover and leads to collagen deposition in left ventricle[1–3]. Cardiac fibrosis occurs after the abnormal accumulation ofcollagen within the ventricle, which ultimately leads to cardiacdysfunction. Collagen deposition is a fundamental step in cardiacfibrosis, providing the matrix for cardiac tissue remodeling, leading to

abnormal accumulation of collagen within the ventricle. There areseveral posttranslational steps in collagen synthesis that are sensitiveto reactive oxygen species (ROS), especially the activity of theenzymes prolyl hydroxylase (PHD), lysyl hydroxylase, and lysyloxidase (LOX). PHD is required to convert proline residues tohydroxyproline, which allows the procollagen peptide chains toassume the triple-helix configuration [4]. A recent report suggeststhat ROS influence cofactor concentrations required for the function ofPHD [6]. Once the procollagen has assumed the triple-helix confor-mation and been excreted, the individual collagen fibers are arrangedinto linear fibrils via cross-linking of lysyl hydroxylase and finallythrough cross-linking between large fibrils by LOX [5]. Theseextracellular cross-linkages are ultimately responsible for cardiacfibrosis. Furthermore, it has been reported that an increased level ofROS triggers the excessive accumulation of ECM, including collagenand fibronectin that lead to the progression of cardiac fibrosis.

Therefore, inhibition of myocardial collagen deposition andfibrosis may result in improvement of cardiac function and attenuatecardiac hypertrophy. It has been reported recently that inhibition ofPHD prevents left ventricular remodeling in the thoracic aortic-

207S. Kumar et al. / Free Radical Biology & Medicine 50 (2011) 206–215

banded rat model [2]. Also, using explanted human hearts, it has beenshown recently that upregulation of LOX plays an important role inhuman dilated cardiomyopathy [7]. Correspondingly, a number ofstudies have shown that inhibition or knockout of MMPs suppressesventricular remodeling, cardiac dysfunction, and progression of heartfailure [1,8–11]. However, there is a lack of information about thesystemic changes in the expression of collagen, MMPs and TIMPs,PHD, and LOX during progression of cardiac hypertrophy orcompensated hypertrophy to decompensation and heart failure. Thisstudy demonstrates a relationship between NF-κB inhibition by theantioxidant pyrrolidine dithiocarbamate (PDTC) and its effects on theECM and ECM regulatory enzymes LOX and PHD during cardiachypertrophy (CH).

In humans, it is not possible to study the molecular changesoccurring in the heart during CH; therefore, a myotrophin-over-expressing transgenic mouse model (Myo-Tg) has been suggested forstudying various stages of CH in vivo [12]. This experimental mousemodel has heart-specific myotrophin overexpression through the useof an α-myosin heavy chain promoter. The overexpression ofmyotrophin in the myocardium of the Myo-Tg mouse initiateshypertrophy that gradually progresses from chronic hypertrophy toheart failure (HF) over a span of 36 weeks. Myo-Tg mice exhibit leftventricular (LV) hypertrophy, multiple focal fibrosis, myocytenecrosis, pleural effusion, and compromised cardiac function, whichclosely mimic human heart failure [12]. Therefore, this model wasused to study the molecular events during the progression ofhypertrophy and its transition to HF.

NF-κB, a ubiquitous inducible transcription factor, activates anumber of genes, including inflammatory cytokines [13–15]. A keyrole for NF-κB is becoming apparent in the pathophysiology ofischemia–reperfusion injury, ischemia preconditioning, and unstableangina [16–21]. Furthermore, NF-κB regulates MMPs and thus playsan important role in the pathogenesis of cardiac remodeling [22].However, the most direct mechanism, its regulation in cardiaccollagen remodeling, has not been considered to date.

Our previous findings showed that NF-κB activation is implicatedin the development of CH in spontaneously hypertensive rats, as wellas in the failing human hearts [23,24]. In a separate study, wedemonstrated that NF-κB expression increased during the initiation,progression, and culmination of cardiac hypertrophy at 4, 16, and36 weeks, respectively, in the Myo-Tg model compared to its wild-type counterpart [25]. Furthermore, to establish a cause-and-effectrelationship between NF-κB activation and remodeling of cardiacmass, we showed that knocking down the NF-κB p65 gene usinglentivirus-mediated siRNA delivery led to attenuation of cardiac masswith improved cardiac functions. As ECM has a major role in cardiacremodeling, we decided to examine the status of ECM protein changesduring progression of cardiac hypertrophy in the Myo-Tg model. Thisinteresting correlation between the progression of cardiac hypertro-phy associated with NF-κB activation led us to investigate further therole of NF-κB in the ECM remodeling; especially during cardiacregression when NF-κB activity is inhibited.

A previous report suggested that inhibition of NF-κB could be agood therapeutic target for treating CH and HF [25]. It has been shownthat PDTC, an antioxidant and inhibitor of NF-κB, was able toattenuate CH in other experimental models [24,26]; but no informa-tion was available on the regulation of ECM proteins by inhibition ofNF-κB activity.

ROS are the major initiators of myocardial damage during cardiachypertrophy. The role of ROS in cardiac fibrosis, during theprogression of cardiac hypertrophy, remains poorly understood.Therefore, a protective inhibitory action of antioxidant therapywould be considered useful for reversing the process of cardiachypertrophy via inhibition of NF-κB and ECM proteins.

In this study, we tested our hypothesis that inhibition of NF-κB byan antioxidant, PDTC, will suppress the expression of ECM proteins

during the progression of CH in Myo-Tg mouse models. This studywas, therefore, aimed at inhibiting NF-κB activation by PDTC and itsrole in modulation of ECM proteins during the progression of cardiachypertrophy to heart failure.

In the first part of this report, we elucidate the changes in ECM geneexpression including collagen, MMP, TIMP, LOX, and PHD in the initiationand progression of CH and HF that were responsible for maladaptivecardiac remodeling in theMyo-Tgmodel to establish the cause-and-effectrelationship, and in the secondpart,we compare theECMgeneexpressionin cardiac reverse remodeling in the same experimental model in whichPDTC was used to inhibit the NF-κB activity.

Materials and methods

Generation of transgenic mice overexpressing myotrophin in the heart

The generation of the Myo-Tg mouse model has been describedpreviously [12]. As described previously, the onset of hypertrophy inthe Myo-Tg mice is at 4 weeks, followed by progression ofhypertrophy until 16 weeks and transition from hypertrophy to HFby 36 weeks. The progression of CH at 4, 16, and 36 weeks wasdetermined based on the cardiac mass changes and cardiac functiondata as described previously [25]. The NF-κB activity also increasedwith the progression of cardiac hypertrophy at 4, 16, and 36 weeks inthe Myo-Tg model compared to wild type (WT) as shown previously[25]. Considering this, all the experiments in this study were carriedout at 4, 16, and 36 weeks. Age- and sex-matchedWTmice were usedfor comparison with the Myo-Tg mice.

Electrophoretic mobility-shift assay (EMSA)

An EMSA was performed using a double-stranded NF-κB bindingsite oligonucleotide as a probe, as described previously [27].

Western blot analysis

Hearts of WT, Myo-Tg, and PDTC–Myo-Tg mice were excised andwashed with cold phosphate-buffered saline (PBS) to remove blood.Western blot analysis was performed as described previously [23].Membranes were probed with primary antibodies for IκBα (CellSignaling Technology; at 1:1000 dilution), Col I (Rockland Immuno-chemical; at 1:5000), Col III (EMD Biosciences; at 1:1000), andconnective tissue growth factor (CTGF; ABCAM; at 1:5000); washedthree times in Tris-Buffered SalineTween-20 (TBS-T); and thendetectedusing a horseradish peroxidase-conjugated secondary antibody andenhanced chemiluminescence (NEN). For quantification of LOX, mousehearts were homogenized in cold T-PER buffer (Pierce) and LOX levelwas determined using its specific antibody as described previously [7].The antibody detects the mature enzyme at the 32 kDa position.

For quantification of MMP-2 and MMP-9, WT and Myo-Tg mousehearts were homogenized in cold cacodylic extraction buffer andWestern blot was performed with anti-MMP-2 and anti-MMP-9monoclonal antibodies (1:5000; Millipore) as described previously[7]. The images were digitized and band densities were analyzed byImagePro software. For normalization, we used glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibody as a probe.

RNA extraction and Northern blot analysis

Total RNAwas isolated from the heart of WT andMyo-Tg mice andPDTC–Myo-Tg according to the protocol of Chomczynski and Sacchi[28]. RNA was resuspended in DEPC water, quantitated by opticaldensity at 260 nm. Transcript levels of Col I and Col III weredetermined by Northern blot analysis. Northern blotting andhybridization were performed as described previously [27]. cDNAprobes were used for IκBα and p65 mRNA expression studies.

208 S. Kumar et al. / Free Radical Biology & Medicine 50 (2011) 206–215

RT-PCR

For quantification of LOX and PHD, RT-PCR was performed. Thetotal RNA was treated with DNase I to remove genomic DNAcontamination. The first-strand cDNA reaction was performed withMMLV reverse transcriptase (Invitrogen) and oligo(dT) primers from1 μg of total RNA. PCR amplification was performed with the sameamount of the first-strand cDNA using Taq DNA polymerase(Invitrogen). The RT-PCR for LOX was performed using the followingprimers: forward primer, 5′-ATATAGGGGCGGATGTCAGAG-3′, andreverse primer, 5′-CGAATGTCACAGCGTACAAC-3′. The PCR protocolfor LOX consisted of 30 cycles of 94 °C for 30 s, 55 °C for 1 min, and72 °C for 30 s, followed by an extension at 72 °C for 5 min. GAPDHwasused as a loading control for the amount of input RNA.

Extracellular matrix and adhesion molecules RT2 PCR array

For obtaining the holistic profile of the gene expression changes inthe ECM and adhesion molecules affected by the PDTC treatment inMyo-Tg animals, the mouse-specific ECM and adhesionmolecules RT2

PCR array was performed on one representative animal from eachgroup. Briefly, RNAwas extracted and purified from the heart tissue ofWT, Myo-Tg, and PDTC-treated Myo-Tg mice using the RNeasy MiniKit and RNeasy Minelute Cleanup kit (Qiagen) according to themanufacturer's protocol. RNA was quantified and integrity wasassessed spectrophotometrically. The 84 genes assessed by RT-PCRusing the mouse extracellular matrix and adhesion molecules array(RT2 Profiler PCR Array PAMM-013; SuperArray Biosciences, Freder-ick, MD, USA) are listed in Supplementary Table 1. Details of theprocedure have been previously reported [29]. Data were importedinto an Excel database and analyzed using the comparative cyclethreshold method with normalization of the raw data to housekeep-ing genes. Results are presented as fold changes versus the values inthe WT control. Gene expression of Myo-Tg mice with a fold changegreater than two compared with WT mice was obtained.

Validation of selected genes identified from the ECM and adhesionmolecule array

The gene expression data from the extracellular matrix andadhesion molecules RT2 PCR array were validated for the CTGF, ColI, Col IIIa, Tgf-β1, Tgf-β2, Tgf-β3, and fibronectin genes using the iQSYBR Green Supermix (Bio-Rad) and gene-specific primers designedfor Q-PCR by real-time PCR at 95 °C for 10 min followed by 40 cycles of95 °C for 15 s, 60 °C for 30 s, and 72 °C for 30 s.

Preparation of cardiac tissue extract for zymography

Samples for zymography were extracted using extraction buffer(10 mM Tris–Cl, pH 7.5, 150 mM NaCl, 20 mM CaCl2, 1 μM ZnSO4,0.01% (v/v) Triton X-100, 1.5 mM NaN3, 0.5% PMSF). Myocardialmatrix metalloproteinase activity in the gel was measured asdescribed by Tyagi et al. [30]. A gelatinase zymography standard(human MMP-2 and MMP-9; AnaSpec, Inc., Fremont, CA, USA) wasused to detect the correct band.

Treatments of mice with PDTC

All mice were housed under identical conditions and had freeaccess to food and water ad libitum. The studies were conducted withthe approval of the Institutional Animal Care and Use Committee atthe Texas A&M Health Science Center. PDTC treatment began at12 weeks of age in bothWT andMyo-Tgmice. Separate groups of age-matched WT mice receiving standard chow and water served as timecontrols. Treatments involved administration of the NF-κB inhibitorPDTC (100 mg/kg/day) in the drinking water with 3% sucrose for a

period of 12 weeks. The dose for PDTC was standardized using threedifferent doses of PDTC (50, 100, and 150 mg/kg/day) and 100 mg/kg/day showed the best inhibition without having any detrimentaleffects to the animal. The above set of experiments consisted of fiveanimals in each group. At the end of the treatment period, the micewere sacrificed by asphyxiation. The hearts were removed and theblood was squeezed out, then the hearts were washed in cold PBS andweighed on an analytical balance.

Determination of cardiac function data collection and analysis

Murine transthoracic echocardiography was conducted in anes-thetizedmice as previously described using the Visualsonic Vevo 2100echocardiograph machine using a 35-MHz probe [25]. Briefly, micewere anesthetized with 3.5% isoflurane initially and then 1.5% duringthe entire procedure to keep the heart rate between 450 and 500beats per minute. The heart was imaged in the two-dimensionalparasternal short-axis view and an M-mode echocardiogram of themidventricle was recorded at the level of papillary muscles. Heartrate, posterior wall thickness, and end-diastolic and end-systolicinternal dimensions of the left ventricle were measured from theM-mode image. Fractional shortening and ejection fraction werederived using three cycles of LV trace in M-mode.

Statistical analysis

All experiments were performed at least three times for eachdetermination. Data are expressed as means±standard error andwere analyzed using one-way analysis of variance and secondaryanalysis for significance with Tukey–Kramer posttests using Prism 4.0GraphPad software (GraphPad, San Diego, CA, USA). A p value lessthan 0.05 was considered statistically significant.

Results

Expression profiling of LOX, PHD, collagen, MMP, and TIMP duringprogression of cardiac hypertrophy and its transition to HF in Myo-Tg mice

In vivo myotrophin overexpression triggers Col I and Col III duringdisease progression

Collagens, the major components of fibrotic tissue, are the mainsubstrate for MMPs. The net balance of collagen deposition is afunction of both synthesis and degradation. To assess the contributionof increased collagen expression, the levels of mRNA expression ofboth Col I and Col III in Myo-Tg andWTmice were investigated duringprogression of cardiac hypertrophy by Northern blotting. Collagentype I mRNA levels increased in 4-, 16-, and 36-week Myo-Tg animalsby 1.19- (pb0.01), 1.4- (pb0.01), and 1.9-fold (pb0.01), respectively.Likewise, Col III mRNA levels in 4-, 16-, and 36-week-oldMyo-Tgmicealso increased, by 1.43- (pb0.01), 1.95- (pb0.01), and 2.6-fold(pb0.01), respectively, compared with WT mice (Figs. 1A and B).The housekeeping gene 18S RNA was used as an internal loadingcontrol (Fig. 1C). The quantification of the mRNA expression fromNorthern blotting is graphically depicted in Figs. 1D and E.

Myo-Tg mice show enhancement of translational levels of MMP andTIMP during progression of hypertrophy

To evaluate the functional activity of MMP-2 and MMP-9 duringthe progression of cardiac hypertrophy and its transition to HF,zymography was performed by using gelatin-incorporated polyacryl-amide gels. Our data show that there was increased activity of MMP-2and MMP-9 starting at 16 weeks and more pronounced activity at36 weeks of age in Myo-Tg mice (Fig. 2A). To further evaluate thetranslational regulation of MMP and TIMP during the progression ofcardiac hypertrophy and its transition to HF, Western blotting wasperformed using their corresponding antibodies. Our data showed a

Fig. 1. (A) mRNA expression of Col I. (B) mRNA expression of Col III. (C) mRNA expression of 18S (loading control) in WT vs Myo-Tg mice at 4, 16, and 36 weeks. (D and E)Normalized image intensity quantification (fold change) of the Col I and Col III expression. Data represent means±SEM from three experiments. *pb0.05 compared to the respectiveage-matched WT.

209S. Kumar et al. / Free Radical Biology & Medicine 50 (2011) 206–215

marked enhancement of MMP-2, viz., 1.13-, 1.54- (pb0.01), and1.71-fold (pb0.01) for 4, 16, and 36 weeks, respectively. Likewise forMMP-9 the increase was 1.16-, 1.57- (pb0.01), and 1.75-fold(pb0.01) for 4, 16, and 36 weeks, respectively (Figs. 2B, C, and D).The expression levels of TIMP-1 remained unchanged in the 4-weekduration; however, they increased significantly (pb0.01) at 16(1.71-fold) and 36 weeks (3.54-fold) (Figs. 2E and F). Similarly, thelevels of TIMP-2 showed a significant increase in the expression at 4,16, and 36 weeks with fold changes of 1.17, 1.66 (pb0.01), and 1.72(pb0.01), respectively (Figs. 2E and G). The levels of TIMP-3 startedincreasing at 16 weeks of age and weremore pronounced at 36 weeksof age but did not show statistically significant change compared withthe age-matched WT group (Figs. 2E and H). Interestingly, TIMP-4was significantly downregulated (pb0.05) during the HF stage (0.56-fold) at 36 weeks of age (Figs. 2E and I). In all experiments, WT of thesame age group was considered a comparative control and nosignificant changes were observed.

Myo-Tg mice show acceleration of transcription of LOX and PHD duringprogression of hypertrophy

To check the two essential enzymes regulating the cross-linkingand biosynthesis, we determined the levels of LOX and PHD duringthe progression of cardiac hypertrophy in Myo-Tg mice. Our datashowed significant upregulation of both LOX (2.1-, 1.4-, and 1.7-foldfor 4, 16, and 36 weeks, respectively) and PHD (3.3- and 3.6-fold for16 and 36 weeks, respectively) genes during transition of hypertro-phy to HF stage (Figs. 3A and B). Additionally, the translational levelsof LOX during the similar process were also determined. The datarevealed significant upregulation of LOX starting in Myo-Tg heartscompared to WT counterparts, viz. 1.3- (pb0.05), 2- (pb0.05), and2.4-fold (pb0.05) at 4, 16, and 36 weeks, respectively (Figs. 3C and D).Additionally, we examined the expression of CTGF protein during thetransition of hypertrophy to HF and it showed a significant increase at16 (1.6-fold) and 36 weeks (1.8-fold) (Figs. 3C and E). In allexperiments, WT of the same age group was considered a compar-ative control and no significant changes were observed.

Establishment of a cause-and-effect relationship between NF-κB activityand cardiac remodeling

PDTC treatment regresses cardiac mass and improves cardiac function inMyo-Tg mice

With regard to the regression of hypertrophy in PDTC-treatedmice, we determined their heart weight to body weight ratio. As

shown in Fig. 4A, PDTC-treated Myo-Tg mice show a significantattenuation of heart weight to body weight ratio in comparison tountreated Myo-Tg mice (9.6±0.05 vs 6.8±0.33, pb0.001).

Echocardiographic data from 12-week-old PDTC-treated Myo-Tgmice showed an improvement in cardiac function compared tountreated Myo-Tg mice (Fig. 4B). In three of four untreated animals amassively enlarged left atrial (LA) appendagewasnoted,whereas itwasconsidered normal in three of three PDTC-treated animals. LA area wasmarginally decreased in PDTC-treated animals. Compared to theuntreated group, PDTC-treated mice showed a trend toward improve-ment of both ejection fraction (34.0±5.3% vs 50.5±15.2%, pb0.09) andfractional shortening (27.82±1.16 vs 32.57±1.9, pb0.06). Similarly,Doppler myocardial tissue velocities were marginally increased in thePDTC-treated group (1.42±0.37 vs 2.06±0.36, pb0.26). Thus, all threesystolic parameters showed a change in the same direction.

Effect of PDTC on NF-κB signaling componentsTo establish an association between NF-κB activation and cardiac

remodeling, NF-κB activity was inhibited by using PDTC, a potentNF-κB inhibitor, administered to the mice in drinking water at100 mg/kg/day for a period of 12 weeks. We analyzed the NF-κBsignaling components as described below:

NF-κB activation. Two groups ofmice (n=5)were treatedwith PDTC foraperiodof 12 weeksandwere sacrificedafterward. The results are shownin Figs. 5A andB. PDTC treatment suppressedNF-κB/DNAbinding activity(181.44±7.8 to 124.08±5.05 Arbitrary Units (A.U.)/10 μg nuclearextract; pb0.001) significantly in Myo-Tg mice but not in WT mice.

IκBα phosphorylation and total protein level. The appearance of totalIκBα cytosolic proteinwas investigated by immunoblot analysis. The dataare shown in Fig. 5C. A basal level of IκBαwas detected inWTmice. PDTCtreatment significantly reduced IκBα level in Myo-Tg mice, whereas asubstantial upregulation was observed in untreated Myo-Tg mice(162±3.24 to 86±2.61 A.U./50 μg cytoplasmic extract; pb0.001;Fig. 5C). The phosphorylation of IκBα in PDTC-treated Myo-Tg mice wasalso determined. The data showed that PDTC treatment significantlyinhibited phosphorylation of IκBα (142.77±4.76 to 68.45±3.86 A.U./50 μg cytoplasmic extract; pb0.001) in Myo-Tg mice, compared tountreated Myo-Tg mice (Fig. 5D).

Effect of PDTC treatment on gene expression by quantitative RT-PCR arrayAltered gene expression profile in the hearts was observed

between WT, Myo-Tg, and PDTC-treated Myo-Tg animals and the

Fig. 2. (A) Collagenase I and II (MMP-2 and MMP-9) activity using zymography. (B) Western blot showing the expression of collagenase I and II (MMP-2 and MMP-9) using theirspecific antibodies. (C) Normalized image intensity quantification (fold change) of theMMP-2 expression. (D) Normalized image intensity quantification (fold change) of theMMP-9expression. (E) Expression of TIMP-1, TIMP-2, TIMP-3, and TIMP-4 in 4-, 16-, and 36-weekMyo-Tgmice compared with age-matchedWT. GAPDHwas used as loading control for theexperiment. (F, G, H, and I) Normalized band intensity quantification of TIMP-1 to TIMP-4 in 4-, 16-, and 36-week Myo-Tg mice compared with WT. Data represent means±SEMfrom three experiments. *pb0.05 compared to the respective age-matched WT.

210 S. Kumar et al. / Free Radical Biology & Medicine 50 (2011) 206–215

fold of differential regulation has been tabulated in Table 1. Amongthe 84 assessed genes, 46 genes were significantly altered with a foldchange larger than 2.0 (38 upregulated genes and 8 downregulatedgenes). The genes that were upregulated by greater than fivefoldwereMmp9, Col2a1, Col1a1, Mmp3, Itgax, Tnc, Spp1, Fn1, Cdh1, Mmp8,Selp, Mmp14, Postn, Thbs1, Ctgf, Itgae, Itgb3, Ncam1, and Icam1. Thegenes that were downregulated by greater than fivefold were Sgce,Vtn, and Mmp15. Of the 46 differentially regulated genes, 19 geneswere associated with transmembrane cell adhesion molecules and 27genes were associated with cell–matrix adhesionmolecules and otheradhesion molecules. Seven genes were associated with basementmembrane constituents, 18 genes were associated with collagens andECM structural constituents and proteases, and 12 genes wereassociated with ECM protease inhibitors and other ECM molecules.(Note. Functions of some genes overlap each other.) These genes areresponsible for profibrotic events, collagen deposition, and maladap-tive remodeling. PDTC-treatedMyo-Tg animals showed a reduction inprofibrotic gene expression and reduced the collagen synthesis.

Validation of genes responsive to PDTC treatment by quantitativereal-time PCR

Statistical significance was obtained for seven important genes inPDTC-treated Myo-Tg (n=3) vs Myo-Tg-untreated animals (n=3)compared with theWT animals (n=3) as shown in Fig. 6. QuantitativePCR showed significant (pb0.001) upregulation in gene expression ofCTGF (14.2±0.2-fold), Col I (6.0±0.7-fold), Col III (1.2±0.1-fold),Tgf-β2 (2.5±0.6-fold), and fibronectin (9.9±0.2-fold). There profi-brotic genes were significantly downregulated and reverted to near oreven below the WT expression in the PDTC-treated Myo-Tg animals.

Effect of PDTC treatment on collagen expression in Myo-Tg miceTo evaluate the collagen expression in PDTC-treated Myo-Tg mice,

Northern blotting was performed using WT, Myo-Tg, and PDTC–Myo-Tgmouse hearts. Our data showed a significant downregulation of theexpression level of Col I (37.1%; pb0.05) and Col III (47.2%; pb0.05) inthe PDTC-treated Myo-Tg mice compared to untreated Myo-Tg mice(Figs. 7A and B). In addition, we checked the protein expression of

Fig. 3. (A) mRNA expression of LOX, PHD, and GAPDH in 4-, 16-, and 36-weekMyo-Tg mice compared with age-matchedWT. (B) Normalized densitometry quantification of the LOXand PHDmRNA expression. (C) Protein expression of LOX and CTGF inWT vs Myo-Tg mice at 4, 16, and 36 weeks. (D and E) Normalized band intensity quantification of the LOX andCTGF protein expression, respectively. Data represent means±SEM from three experiments. *pb0.05 compared to respective age-matched WT.

211S. Kumar et al. / Free Radical Biology & Medicine 50 (2011) 206–215

both Col I and Col III and observed similar findings, viz., decreases inthe levels of Col I and Col III by 37 (pb0.05) and 29% (pb0.05),respectively (Fig. 7C). Also, there was a reduction at the translationallevel in the expression of CTGF by 28% (pb0.05) and MMP-9 by 23%(pb0.05; Fig. 7C). β-Actin was used as loading control for theexperiment. The changes in the expression levels of Col I, Col III, CTGF,and MMP-9 are graphically depicted in Fig. 7D.

Fig. 4. Effect of PDTC on cardiac mass and cardiac function in Myo-Tg mice. (A) Heart we*pb0.001 compared with WT, #pb0.001 compared with untreated Myo-Tg mice. These resuM-mode tracings of the left ventricle obtained from sham-treated and PDTC-treated animarrows). Improved LV fractional shortening and decreased left ventricle internal dimensioncavity; PW, posterior wall (n=3, pb0.06).

Discussion

The novel findings of this study are: (1) a complete profiling ofECM genes and adhesion molecules in WT, Myo-Tg, and PDTC–Myo-Tg animals, (2) a complete profile of LOX, PHD, collagen, MMP, andTIMP expression during the progression of cardiac hypertrophy andits transition to heart failure, (3) blockade of NF-κB signaling cascade

ight:body weight ratio in PDTC-treated Myo-Tg mice. Values represent means±SEM;lts are presented as the mean±SEM and represent six observations. (B) Representativeals. Arrows indicate endocardial borders in diastole (broken arrows) and systole (fullare evident in the PDTC-treated mouse. IVS, interventricular septum; LV, left ventricle

Fig. 5. Inhibition of NF-κB signaling components in PDTC-treated Myo-Tg mice.(A) Nuclear protein was extracted from the hearts of 24-week-old WT, 12-week PDTC-treatedWT, 24-week untreated Myo-Tg mice, and 12-week PDTC-treated Tg mice. DNAbinding reactions or EMSA was performed. (B) Quantification of NF-κB binding activity.(C) Cytoplasmic protein extracts were made from 24-week-old WT, PDTC-treated WT,untreated Myo-Tg, and PDTC-treated Myo-Tg mice and Western blot was performedusing IκBα antibody as a probe. (D) Cytoplasmic protein extracts were made from24-week treated WT, PDTC-treated WT, untreated Myo-Tg, and PDTC-treated Myo-Tgmice and Western blot was performed using phospho-IκBα antibody as a probe.These results are presented as the means±SEM and represent five experiments withdifferent mice (pb0.001 compared with the untreated Myo-Tg mice). ⁎ means pb0.05compared to the WT while # means pb0.001 compared to the Myo-Tg.

Table 1Relative fold change of the differentially expressed genes from RT2 ECM array

Gene symbol Myo-Tg vs WT PDTC-Myo-Tg vs Myo-Tg

Upregulated in Myo-TgMmp9 77.9 2.9Col2a1 61.1 1.2Col1a1 56.2 13.8Mmp3 43.8 1.9Itgax 36.6 6.3Tnc 28.9 14.6Spp1 25.7 1.9Fn1 18.7 2.5Cdh1 18.1 −1.1Mmp8 17.1 2.1Selp 14.2 1.8Mmp14 12.3 8.5Postn 11.5 4.8Thbs1 9.2 −1.1Ctgf 8.8 2.1Itgae 6.6 5.3Itgb3 6.3 1.6Ncam1 6.2 5.1Icam1 5.2 5.6Adamts8 5.0 3.1Itga3 5.0 1.3Itga5 4.6 1.8Fbln1 4.5 4.7Vcan 3.9 2.3Itgam 3.0 −1.5Ctnna2 2.9 1.4Sparc 2.7 1Cd44 2.6 1.3Lamc1 2.6 5.6Adamts2 2.6 1.2Timp2 2.6 1.9Col3a1 2.5 2.2Col5a1 2.4 1.8Emilin1 2.3 2Itga4 2.2 1.5Thbs3 2.2 1.6Ecm1 2.1 −1.5Col4a3 2.0 1.8

Downregulated in Myo-TgSgce −285 −1.4Vtn −7.4 −1.9Mmp15 −5 2.3Lama2 −3.8 −1.7Hapln1 −3.1 −1.7Cdh2 −3.0 −1.8Ctnna1 −2.1 −1.7Mmp13 −2.1 1.3Mmp2 −1.9 −1.2Thbs2 −1.8 1.1Col6a1 −1.7 −1.3Lamb2 −1.6 −1.6Pecam1 −1.4 1.1Sele −1.4 1.4Cdh4 −1.3 7.5Ctnnb1 −1.3 1.7Mmp1a −1.2 1.4Itgb1 −1.1 −1.1Lamb3 −1.1 1.1Adamts5 −1.1 −1.1Mmp12 −1.1 2.7

Column 1 lists the gene symbols for the genes analyzed using the RT2 ECM array.Columns 2 and 3 show fold change values of the differentially regulated genes in Myo-Tg vs the WT and PDTC-treated Myo-Tg vs untreated Myo-Tg animals, respectively.

212 S. Kumar et al. / Free Radical Biology & Medicine 50 (2011) 206–215

with PDTC regressed cardiac hypertrophy and improved cardiacfunction, and (4) PDTC reduced collagen expression andMMP activity.

Collagens I and III are the essential components of myocardium,maintaining its structural and functional integrity. As Col I and Col IIIhave different physical properties, the changes in expression have amajor impact on systolic and diastolic function of the heart. Col Iprovides the tensile strength and its increase would therefore imposemyocardial stiffness. Col III forms an elastic network that stores kineticenergy. Increase in Col III levels would therefore influence flaccidity ofthe heart, which ultimately compromises diastolic dysfunction. Ourstudies showed the maximum upregulation of both Col I and Col III atthe end stage of heart failure, supporting the existing theory [31]. It isalso evident that at the initiation phase of cardiac hypertrophy (4 weeksof age) there were no significant changes in collagen expression as thecardiac function was not compromised, but, during the progressionphase (16 weeks), there was a significant enhancement of Col I and ColIII, which gradually increased toward the end stage of heart failure.

Previous studies from our and other laboratories suggestedincreased induction of MMP-2 and MMP-9, either in infarcted ratmyocardium [8,32,33] or in dilated cardiomyopathy in the humanheart [7,10,30]. It has also been reported that targeted deletion ofMMP-2 or MMP-9 in mice resulted in a better survival rate, lowerincidence of LV rupture, less LV cavity dilation, attenuation of collagendeposition, and improved fractional shortening compared toWTmice

after myocardial infarction (MI) [8,33], suggesting a potential role inthe heart failure process. One of the key regulators of ECM turnover isMMPs and their endogenous inhibitors, TIMPs. In this study, wecharacterized the expression ofMMPs (includingMMP-2 andMMP-9)and TIMPs (including TIMP-1, -2, -3, and -4) during the progression ofhypertrophy in Myo-Tg mice. Our data showed an enhancement ofboth MMP-2 and MMP-9 during the progression of cardiac

Fig. 6. Validation of selected candidate genes responsible for cardiac hypertrophy byquantitative real-time PCR. Graph shows relative fold change for gene transcriptssetting the expression ofWT as 1. Data represent means±SEM of at least threemice pergroup. *pb0.05 compared to WT; #pb0.05 compared to the Myo-Tg group.

213S. Kumar et al. / Free Radical Biology & Medicine 50 (2011) 206–215

hypertrophy and were more robust as heart failure progressed, atboth the mRNA and the protein level. Compared with WT, the TIMPlevel during progression of cardiac hypertrophy in the Myo-Tg miceshowed maximum upregulation at the compensated stage of heartfailure (Figs. 2 and 3). There was a marked elevation of TIMP-1 andTIMP-2 at the heart failure stage, suggesting an important role incardiac remodeling.

Numerous studies have demonstrated that cardiac hypertrophycan cause inflammation and development of fibrosis that is charac-terized by an imbalance between deposition and breakdown of ECMcomponents [2]. Recent experimental evidence suggests that inflam-mation may not be necessary or sufficient for progression to fibrosis.The overexpression of the potent profibrotic mediator, TGF-β1, for

Fig. 7. (A) mRNA expression of collagen I, collagen III, and 18S (loading control) in WT, WT-quantification (fold change) of the Col I and Col III expression. (C) Protein expression of Coperformed in duplicate. (D) Normalized band intensity quantification of the Col I, Col III, CTG*pb0.05 compared to WT; #pb0.05 compared to the Myo-Tg group.

example, leads to progressive fibrosis in mice, without any significantinflammatory component [34].

The regulation of ECM is a complex biological process. It is wellaccepted that cardiac fibrosis is the end result of bothmatrix synthesisand degradation, including MMPs and collagens. Matrikines (adegraded product of matrix protein) also serve as a stimulator forcollagen synthesis that ultimately results in deposition of fibrotictissues in the myocardium [35]. Pharmacological intervention usingMMP inhibition has also been shown to reduce ventricular remodel-ing after MI in rat and mouse models, demonstrating the beneficialeffect during progression of chronic heart failure [11,32]. Based on thisfinding, we further speculate that the ECM remodeling (includingMMP-2 and MMP-9, TIMP-1 and TIMP-2) process, which differenti-ates the compensated (16 weeks of age) phase from the decom-pensated (36 weeks of age), would be an important phase for whichto design possible therapeutic targets.

To establish a cause-and-effect relationship of NF-κB activationand cardiac hypertrophy in this Myo-Tg mouse model, we firstexamined the effect of inhibition of NF-κB by using its inhibitor PDTCin Myo-Tg mice. Our data demonstrated a regression of cardiac massassociated with improved cardiac function in PDTC-treated mice. Thissuggests a causal link between NF-κB activation and cardiacremodeling in Myo-Tg mice (Figs. 4A and B). In addition, our datarevealed that treatment with PDTC significantly inhibited NF-κBactivation associated with reduction of total IκBα protein content,phosphorylation of IκBα, and attenuation of the transcriptional levelof IκBα mRNA in Myo-Tg mice (Figs. 5A, C, D, and E). The exactmechanism of action of PDTC is not well understood but has beenshown to be an antioxidant. Paradoxically, the pro-oxidant andmetal-chelating properties of PDTC could also be involved in the ability toinhibit NF-κB activation [36]. In our study, the inhibition of NF-κBsignaling components by PDTC may be due to the interference ofreactive oxygen metabolism [37], chelation of divalent metal ions[38], or influence of thiol levels [39]. Therefore, the cardiac remodel-ing in Myo-Tg may be due to an intense antioxidant effect and PDTC

PDTC, Myo-Tg, and PDTC-treated Myo-Tg (PDTC) mice. (B) Normalized image intensityl I, Col III, CTGF, and MMP-9 in WT, Myo-Tg, and PDTC-treated Myo-Tg (PDTC) animalsF, and MMP-9 protein expression. Data represent means±SEM for three experiments.

214 S. Kumar et al. / Free Radical Biology & Medicine 50 (2011) 206–215

possibly has a protective role in this process. Taken together, the datademonstrate that significant changes can be readily detected aftertreatment with PDTC in the expression of profibrotic genes, factorsthat can adversely influence cardiac function. These changes in geneexpression could partially contribute to the ROS scavenging proper-ties of PDTC.

Recently, it has been reported that cardiac-specific overexpressionof an IκBα triple mutant (S32A, S36A, and Y42F; 3 M) completelyblocked NF-κB activation induced by cytokines or in response toTNFα-induced cardiomyopathy [40,41]. These 3 M mice furthersuggest that the serine and tyrosine phosphorylation pathways aredifferentially activated in different cardiac pathophysiological pro-cesses and that NF-κB play an important role in this process.Furthermore, a genetic breed was produced from TNFα-overexpres-sing mice and p50 knockout mice to assess whether the cardiotoxiceffects of proinflammatory cytokines were mediated through NF-κB[42,43]. Blockade of NF-κB activation attenuated the myocardialinflammation and improved cardiac function in male TNFα transgenicmice. All these results corroborate with our findings presented here.

Inaddition, our studydemonstrates that PDTC reducedbothCol I andCol III activity inMyo-Tgmice, suggesting aprotective role inmyocardialdamage and fibrosis. The exact mechanism of collagen inhibition byPDTC is not known. It has been suggested that the Col I promoter has anoncanonical NF-κB binding element at the−271 and−235 positions,whichmight be responsible for regulating the Col I gene [44]. Moreover,the Col I and Col III promoters have an AP-1 binding site and thus it isreasonable to hypothesize that NF-κB can regulate collagen expressionvia interaction with another transcription factor such as AP-1 or an as-yet-unidentified NF-κB element. In this regard, Stein et al. showed thatNF-κB can potentiate transcription from a promoter lacking an NF-κBelementvia interactionwithAP-1 [45]. Therefore, a complex interplayoftranscription factors has been implicated in regulating collagenexpression, opening up a new area of further research.

Furthermore, we observed downregulation of both MMP-2 andMMP-9 in PDTC-treated Myo-Tg mice. It was previously reported thatNF-κB regulates MMP-9 activity, as the MMP-9 promoter has anNF-κB binding site [46,47], and it is therefore more plausible topropose the downregulation of MMP-9 by PDTC. The regulation ofMMP-2 by PDTC is not known, as there is no consensus for an NF-κBbinding element present in the MMP-2 promoter region [48]. It ispossible that MMP-2 regulation by PDTC could be via an interactionwith another membrane-type MMP that contains an NF-κB consensusin the promoter region [49,50]. Therefore, it is possible that NF-κBdirectly modulates MMP-2 expression and activity via its interactionwith other transcription factors that regulate MMP-2 expression.

Myocardial ECMremodeling andfibrosis seen both in animalmodelsof heart failure and in patientswith this disease are implicated in cardiacdysfunction andmay have prognostic value. Our data demonstrate newinformation, including theupregulation of LOX and PHD,which regulatecollagen cross-linking and biosynthesis during the progression ofcardiac hypertrophy. PDTC treatment in Myo-Tg mice inhibits theNF-κB signaling cascade and collagen expression and improves cardiacfunction. Thesedata suggest a therapeutic benefit in cardiac remodeling,in which the NF-κB signaling pathway plays a significant role. Our studyfurther indicates that for effective treatment of cardiac fibrosis, acomplete inhibition of NF-κB may be necessary.

Supplementarymaterials related to this article can be found onlineat doi:10.1016/j.freeradbiomed.2010.10.711.

Acknowledgments

This studywas supported in part by an American Heart AssociationNational Scientist Development Grant (0835227 N) and start-upfunds from the Texas A&MHealth Science Center, College of Medicine,to S.G. The authors acknowledge the kind gift of IκBα cDNA probefrom Dr. Allan Brasier (The University of Texas Medical Branch,

Galveston, TX, USA). The authors acknowledge the generosity of Dr.Phillip Trackman (Boston University, Boston, MA, USA) for providingthe Lox antibody.

References

[1] Creemers, E. E.; Cleutjens, J. P.; Smits, J. F.; Daemen, M. J. Matrix metalloproteinaseinhibition after myocardial infarction: a new approach to prevent heart failure?Circ. Res. 89:201–210; 2001.

[2] Fielitz, J.; Philipp, S.; Herda, L. R.; Schuch, E.; Pilz, B.; Schubert, C.; Gunzler, V.;Willenbrock, R.; Regitz-Zagrosek, V. Inhibition of prolyl 4-hydroxylase preventsleft ventricular remodelling in rats with thoracic aortic banding. Eur. J. Heart Fail.9:336–342; 2007.

[3] Cleutjens, J. P.; Kandala, J. C.; Guarda, E.; Guntaka, R. V.; Weber, K. T. Regulation ofcollagen degradation in the rat myocardium after infarction. J. Mol. Cell. Cardiol.27:1281–1292; 1995.

[4] Trackman, P. C. Diverse biological functions of extracellular collagen processingenzymes. J. Cell. Biochem. 96:927–937; 2005.

[5] Smith-Mungo, L. I.; Kagan, H. M. Lysyl oxidase: properties, regulation andmultiplefunctions in biology. Matrix Biol. 16:387–398; 1998.

[6] Acker, T.; Fandrey, J.; Acker, H. The good, the bad and the ugly in oxygen-sensing:ROS, cytochromes and prolyl-hydroxylases. Cardiovasc. Res. 71:195–207; 2006.

[7] Sivakumar, P.; Gupta, S.; Sarkar, S.; Sen, S. Upregulation of lysyl oxidase andMMPsduring cardiac remodeling in human dilated cardiomyopathy. Mol. Cell. Biochem.307:159–167; 2008.

[8] Hayashidani, S.; Tsutsui, H.; Ikeuchi, M.; Shiomi, T.; Matsusaka, H.; Kubota, T.;Imanaka-Yoshida, K.; Itoh, T.; Takeshita, A. Targeted deletion ofMMP-2 attenuatesearly LV rupture and late remodeling after experimental myocardial infarction.Am. J. Physiol. Heart Circ. Physiol. 285:H1229–H1235; 2003.

[9] Kim, H. E.; Dalal, S. S.; Young, E.; Legato, M. J.; Weisfeldt, M. L.; D'Armiento, J.Disruption of the myocardial extracellular matrix leads to cardiac dysfunction.J. Clin. Invest. 106:857–866; 2000.

[10] Thomas, C. V.; Coker, M. L.; Zellner, J. L.; Handy, J. R.; Crumbley III, A. J.; Spinale,F. G. Increased matrix metalloproteinase activity and selective upregulation in LVmyocardium from patients with end-stage dilated cardiomyopathy. Circulation97:1708–1715; 1998.

[11] Rohde, L. E.; Ducharme, A.; Arroyo, L. H.; Aikawa, M.; Sukhova, G. H.; Lopez-Anaya,A.; McClure, K. F.; Mitchell, P. G.; Libby, P.; Lee, R. T. Matrix metalloproteinaseinhibition attenuates early left ventricular enlargement after experimentalmyocardial infarction in mice. Circulation 99:3063–3070; 1999.

[12] Sarkar, S.; Leaman, D. W.; Gupta, S.; Sil, P.; Young, D.; Morehead, A.; Mukherjee,D.; Ratliff, N.; Sun, Y.; Rayborn, M.; Hollyfield, J.; Sen, S. Cardiac overexpression ofmyotrophin triggers myocardial hypertrophy and heart failure in transgenic mice.J. Biol. Chem. 279:20422–20434; 2004.

[13] Baeuerle, P. A.; Baltimore, D. NF-kappa B: ten years after. Cell 87:13–20; 1996.[14] Baldwin Jr., A. S. The NF-kappa B and I kappa B proteins: new discoveries and

insights. Annu. Rev. Immunol. 14:649–683; 1996.[15] May, M. J.; Ghosh, S. Signal transduction through NF-kappa B. Immunol. Today 19:

80–88; 1998.[16] Maulik, N.; Sato, M.; Price, B. D.; Das, D. K. An essential role of NFkappaB in

tyrosine kinase signaling of p38 MAP kinase regulation of myocardial adaptationto ischemia. FEBS Lett. 429:365–369; 1998.

[17] Xuan, Y. T.; Tang, X. L.; Banerjee, S.; Takano, H.; Li, R. C.; Han, H.; Qiu, Y.; Li, J. J.;Bolli, R. Nuclear factor-kappaB plays an essential role in the late phase of ischemicpreconditioning in conscious rabbits. Circ. Res. 84:1095–1109; 1999.

[18] Li, C.; Kao, R. L.; Ha, T.; Kelley, J.; Browder, I. W.; Williams, D. L. Early activation ofIKKbeta during in vivo myocardial ischemia. Am. J. Physiol. Heart Circ. Physiol. 280:H1264–1271; 2001.

[19] Ritchie, M. E. Nuclear factor-kappaB is selectively and markedly activated inhumans with unstable angina pectoris. Circulation 98:1707–1713; 1998.

[20] Obata, H.; Biro, S.; Arima, N.; Kaieda, H.; Kihara, T.; Eto, H.; Miyata, M.; Tanaka, H. NF-kappaB is induced in thenuclei of cultured rat aortic smoothmuscle cells by stimulationof various growth factors. Biochem. Biophys. Res. Commun. 224:27–32; 1996.

[21] Valen, G.; Hansson, G. K.; Dumitrescu, A.; Vaage, J. Unstable angina activatesmyocardial heat shock protein 72, endothelial nitric oxide synthase, andtranscription factors NFkappaB and AP-1. Cardiovasc. Res. 47:49–56; 2000.

[22] Timmers, L.; van Keulen, J. K.; Hoefer, I. E.; Meijs, M. F.; van Middelaar, B.; denOuden, K.; van Echteld, C. J.; Pasterkamp, G.; de Kleijn, D. P. Targeted deletion ofnuclear factor kappaB p50 enhances cardiac remodeling and dysfunctionfollowing myocardial infarction. Circ. Res. 104:699–706; 2009.

[23] Gupta, S.; Sen, S. Role of the NF-kappaB signaling cascade and NF-kappaB-targetedgenes in failing human hearts. J. Mol. Med. 83:993–1004; 2005.

[24] Gupta, S.; Young, D.; Sen, S. Inhibition of NF-kappaB induces regression of cardiachypertrophy, independent of blood pressure control, in spontaneously hyperten-sive rats. Am. J. Physiol. Heart Circ. Physiol. 289:H20–29; 2005.

[25] Gupta, S.; Young, D.; Maitra, R. K.; Gupta, A.; Popovic, Z. B.; Yong, S. L.; Mahajan, A.;Wang, Q.; Sen, S. Prevention of cardiac hypertrophy and heart failure by silencingof NF-kappaB. J. Mol. Biol. 375:637–649; 2008.

[26] Li, Y.; Ha, T.; Gao, X.; Kelley, J.; Williams, D. L.; Browder, I. W.; Kao, R. L.; Li, C. NF-{kappa}B activation is required for the development of cardiac hypertrophy invivo. Am. J. Physiol. Heart Circ. Physiol. 287:H1712–1720; 2004.

[27] Gupta, S.; Purcell, N. H.; Lin, A.; Sen, S. Activation of nuclear factor-kappaB isnecessary for myotrophin-induced cardiac hypertrophy. J. Cell Biol. 159:1019–1028; 2002.

215S. Kumar et al. / Free Radical Biology & Medicine 50 (2011) 206–215

[28] Chomczynski, P.; Sacchi, N. Single-step method of RNA isolation by acidguanidinium thiocyanate–phenol–chloroform extraction. Anal. Biochem. 162:156–159; 1987.

[29] Tian, J.; Pecaut, M. J.; Slater, J. M.; Gridley, D. S. Spaceflight modulates expressionof extracellular matrix, adhesion, and profibrotic molecules in mouse lung. J. Appl.Physiol. 108:162–171; 2010.

[30] Tyagi, S. C.; Kumar, S. G.; Haas, S. J.; Reddy, H. K.; Voelker, D. J.; Hayden, M. R.;Demmy, T. L.; Schmaltz, R. A.; Curtis, J. J. Post-transcriptional regulation ofextracellular matrix metalloproteinase in human heart end-stage failuresecondary to ischemic cardiomyopathy. J. Mol. Cell. Cardiol. 28:1415–1428; 1996.

[31] Martos, R.; Baugh, J.; Ledwidge, M.; O'Loughlin, C.; Conlon, C.; Patle, A.; Donnelly,S. C.; McDonald, K. Diastolic heart failure: evidence of increased myocardialcollagen turnover linked to diastolic dysfunction. Circulation 115:888–895; 2007.

[32] Peterson, J. T.; Hallak, H.; Johnson, L.; Li, H.; O'Brien, P. M.; Sliskovic, D. R.; Bocan,T. M. A.; Coker, M. L.; Etoh, T.; Spinale, F. G. Matrix metalloproteinase inhibitionattenuates left ventricular remodeling and dysfunction in a rat model of progressiveheart failure. Circulation 103:2303–2309; 2001.

[33] Ducharme,A.; Frantz, S.; Aikawa,M.; Rabkin, E.; Lindsey,M.; Rohde, L. E.; Schoen, F. J.;Kelly, R. A.; Werb, Z.; Libby, P.; Lee, R. T. Targeted deletion of matrix metallopro-teinase-9 attenuates left ventricular enlargement and collagen accumulation afterexperimental myocardial infarction. J. Clin. Invest. 106:55–62; 2000.

[34] Sage, H.; Vernon, R. B.; Funk, S. E.; Everitt, E. A.; Angello, J. SPARC, a secretedprotein associated with cellular proliferation, inhibits cell spreading in vitro andexhibits Ca2+-dependent binding to the extracellular matrix. J. Cell Biol. 109:341–356; 1989.

[35] Li, Y. Y.; McTiernan, C. F.; Feldman, A. M. Interplay of matrix metalloproteinases,tissue inhibitors of metalloproteinases and their regulators in cardiac matrixremodeling. Cardiovasc. Res. 46:214–224; 2000.

[36] Pinkus, R. L.; Weiner, L. M.; Daniel, V. Role of oxidants and antioxidants in theinduction of AP-1, NF-κB and glutathione S-transferase gene expression. J. Biol.Chem. 271:13422–13429; 1996.

[37] Satriano, J.; Schlondorff, D. Activation and attenuation of transcription factorNF-κB in mouse glomerular mesenglial cells in response to TNFα, immunoglobinG and adenosine 3′5′0 cyclic monophosphate: evidence for involvement ofreactive oxygen species. J. Clin. Invest. 94:1629–1636; 1994.

[38] Sunderman Sr., F. W. Therapeutic properties of sodium diethyldithiocarbamate: itsrole as an inhibitor in the progression of AIDS. Ann. Clin. Lab. Sci. 21:70–81; 1991.

[39] Mihm, A.; Ennen, J.; Pessara, U.; Kurth, R.; Druge, W. Inhibition of HIV-1replication and NF-κB activity by cysteine and cysteine derivatives. AIDS 5:497–505; 1991.

[40] Brown, M.; McGuinness, M.; Wright, T.; Ren, X.; Wang, Y.; Boivin, G. P.; Hahn, H.;Feldman, A. M.; Jones, W. K. Cardiac-specific blockade of NF-kappaB in cardiacpathophysiology: differences between acute and chronic stimuli in vivo. Am. J.Physiol. Heart Circ. Physiol. 289:H466–476; 2005.

[41] Higuchi, Y.; Chan, T. O.; Brown, M. A.; Zhang, J.; DeGeorge Jr., B. R.; Funakoshi, H.;Gibson, G.; McTiernan, C. F.; Kubota, T.; Jones, W. K.; Feldman, A. M.Cardioprotection afforded by NF-kappaB ablation is associated with activationof Akt in mice overexpressing TNF-alpha. Am. J. Physiol. Heart Circ. Physiol. 290:H590–598; 2006.

[42] Kawamura, N.; Kubota, T.; Kawano, S.; Monden, Y.; Feldman, A. M.; Tsutsui, H.;Takeshita, A.; Sunagawa, K. Blockade of NF-kappaB improves cardiac function andsurvival without affecting inflammation in TNF-alpha-induced cardiomyopathy.Cardiovasc. Res. 66:520–529; 2005.

[43] Kawano, S.; Kubota, T.; Monden, Y.; Tsutsumi, T.; Inoue, T.; Kawamura, N.; Tsutsui,H.; Sunagawa, K. Blockade of NF-{kappa}B improves cardiac function and survivalafter myocardial infarction. Am. J. Physiol. Heart Circ. Physiol. 291:H1337–1344;2006.

[44] Kouba, D. J.; Chung, K. Y.; Nishiyama, T.; Vindevoghel, L.; Kon, A.; Klement, J. F.;Uitto, J.; Mauviel, A. Nuclear factor-kappa B mediates TNF-alpha inhibitory effecton alpha 2(I) collagen (COL1A2) gene transcription in human dermal fibroblasts.J. Immunol. 162:4226–4234; 1999.

[45] Stein, B.; Cogswell, P. C.; Baldwin Jr., A. S. Functional and physical associationsbetween NF-kappa B and C/EBP family members: a Rel domain–bZIP interaction.Mol. Cell. Biol. 13:3964–3974; 1993.

[46] Sanceau, J.; Boyd, D. D.; Seiki, M.; Bauvois, B. Interferons inhibit tumor necrosisfactor-alpha-mediated matrix metalloproteinase-9 activation via interferonregulatory factor-1 binding competition with NF-kappa B. J. Biol. Chem. 277:35766–35775; 2002.

[47] Fini, M. E.; Bartlett, J. D.; Matsubara, M.; Rinehart, W. B.; Mody, M. K.; Girard,M. T.; Rainville, M. The rabbit gene for 92-kDa matrix metalloproteinase: roleof AP1 and AP2 in cell type-specific transcription. J. Biol. Chem. 269:28620–28628;1994.

[48] Westermarck, J.; Kahari, V. M. Regulation of matrix metalloproteinase expressionin tumor invasion. FASEB J. 13:781–792; 1999.

[49] Han, Y. P.; Tuan, T. L.; Wu, H.; Hughes, M.; Garner, W. L. TNF-alpha stimulatesactivation of pro-MMP2 in human skin through NF-(kappa)B mediated inductionof MT1-MMP. J. Cell Sci. 114:131–139; 2001.

[50] Murphy, G.; Stanton, H.; Cowell, S.; Butler, G.; Knauper, V.; Atkinson, S.;Gavrilovic, J. Mechanisms for pro matrix metalloproteinase activation. APMIS107:38–44; 1999.