Journal of Molecular and Cellular Cardiology 50 (2011) 426–432

Contents lists available at ScienceDirect

Journal of Molecular and Cellular Cardiology

j ourna l homepage: www.e lsev ie r.com/ locate /y jmcc

Original article

Critical role for white blood cell NAD(P)H oxidase-mediated plasminogen activatorinhibitor-1 oxidation and ventricular rupture following acute myocardial infarction☆

Udit Agarwal a,b, Xiaorong Zhou a, Kristal Weber a, Alisher R. Dadabayev a, Marc S. Penn a,⁎a Department of Cardiovascular Medicine and Stem Cell Biology and Regenerative Medicine, Cleveland Clinic, Cleveland, OH, USAb School of Biomedical Sciences, Kent State University, Kent, OH, USA

☆ This work was funded by the Skirball Foundation.⁎ Corresponding author. Skirball Laboratory for Cardio

Departments of Cardiovascular Medicine and Stem Cell B9500 Euclid Ave, Cleveland, OH 44195, USA. Tel.: +1 21

E-mail address: Pennm@CCF.ORG (M.S. Penn).

0022-2828/$ – see front matter © 2010 Elsevier Ltd. Aldoi:10.1016/j.yjmcc.2010.08.024

a b s t r a c t

a r t i c l e i n f o

Article history:Received 29 March 2010Received in revised form 9 August 2010Accepted 23 August 2010Available online 31 August 2010

Keywords:InflammationMyocardial ruptureOxidation

Plasminogen activator inhibitor-1 (PAI-1) is an oxidant-sensitive protease inhibitor that is inactivated byoxidation and has a critical role in ventricular remodeling post myocardial infarction (MI). PAI-1 knockout (KO)mice die within 7 days of myocardial infarction post MI due to increased plasmin activity leading to ventricularrupture. The goal of this study was to assess the relevant pathways of leukocyte-derived oxidants post MI thatalter PAI-1 activity. Transplantation ofwild-type (WT)bonemarrow into PAI-1 nullmice prolonged survival afterMI (WT marrow: 41.66% vs. PAI-1 KO marrow: 0% in PAI-1 KO mice at day 7 (pb0.02). To determine relevantenzyme systems, we transplanted marrow from mice with specific deletions relevant to leukocyte-derivedoxidants (NAD(P)H oxidase, iNOS, myeloperoxidase (MPO)) to determine which deletion controls PAI-1oxidative inactivation and prolongs survival. MI was induced by ligation of the left anterior descending artery(LAD) and the incidence of cardiac rupturewasmonitored. PAI-1KO transplantedwithMPOKO, or iNOSKObonemarrow died within 9 days after MI. PAI-1 KO mice transplanted with p47phox KO marrow exhibited prolongedsurvival 21 days afterMI (30% survival, pb0.03, n=10) compared toWTmarrow (8.3%, n=12). Three days afterMI, PAI-1 KOmice transplanted with p47phox KOmarrow had increased PAI-1 activity and decreased nitration ofPAI-1 in myocardial tissue compared to PAI-1 KO mice transplanted with WT marrow. These data suggest thatmodulating O2

•− generation by NAD(P)H oxidase appears to be a therapeutically relevant target for increasingmyocardial PAI-1 levels after MI, whereas downstream enzymes like MPO and iNOS may not be.

vascular Cellular Therapeutics,iology, Cleveland Clinic, NE30,6 445 1932.

l rights reserved.

© 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Acute myocardial infarction (AMI) and the subsequent leftventricular remodeling is the most common cause of chronic heartfailure. In the near term post AMI (b2 weeks), the remodeling processcan lead to left ventricular wall rupture, papillary muscle rupture, orventricular septal defects. These complications, although rare, arelethal and account for 5–30% of in hospital mortality post AMI [1]. Thepathways central to left ventricular remodeling, in part to minimizemechanical complications of myocardial infarction and optimizecardiac function, are under active investigation.

Several studies have emphasized the critical role of plasminogen/plasmin/uPA axis, its inhibitor plasminogen activator inhibitor-1 (PAI-1),and leukocyte-derived oxidants in left ventricular remodeling followingAMI [2–4]. PAI-1, a serine protease inhibitor, plays an important role in

ventricular rupture as PAI-1 null mice die of ventricular rupture within6 days of AMI [5]. Studies have also reported that PAI-1 is an oxidantsensitive protease and inhibition of uPA (urokinase-like plasminogenactivator), an activator of plasminogen plasmin axis, by PAI-1 is redox-sensitive [5–7].

Multiple studies have demonstrated an important contribution ofvarious leukocyte-derived oxidant-producing enzyme systems (NAD(P)H oxidase, myeloperoxidase, iNOS) in ventricular remodeling[5,8,9]. While PAI-1 is known to be oxidant sensitive, the relativecontribution of each of these leukocyte-generating oxidants on PAI-1function following AMI is unknown.

In the present study, we systematically assessed the contribution ofindividual oxidant generating pathways on PAI-1 activity and ventric-ular rupture. We demonstrate that NAD(P)H oxidase generation of O2

•−

upstream of MPO, and iNOS, is required for PAI-1 inactivation and has asignificant role in ventricular rupture. Since statin therapy inhibits theformation of NAD(P)H oxidase [10], our findings suggest a potentialmechanism for the improvement in ventricular function followingmyocardial infarction. Our findings further suggest that inhibition ofdownstream enzymes such as myeloperoxidase and iNOS may beinsufficient to significantly inhibit the consequences of oxidation ininflammation.

427U. Agarwal et al. / Journal of Molecular and Cellular Cardiology 50 (2011) 426–432

2. Methods and materials

All animal protocols were approved by the Animal ResearchCommittee, and all animals were housed in the AAALAC-approvedanimal facility of the ClevelandClinic. The time course of the experimentis depicted in Fig. 1.

2.1. Bone marrow transplantation

2.1.1. Preparation of total bone marrow cellsFour- to six-week-old male mice of each PAI-1 KO, wild-type,

p47phox (NAD(P)H oxidase subunit) KO, MPO KO, and iNOS KOgenotype were sacrificed by intraperitoneal injection of xylazine andketamine. Femur and tibia were stripped of the muscles and cut fromboth ends. The bonemarrow cells were flushed in 50 ml of falconwithα-MEM. The total bonemarrow cells were subjected to red blood cellslysis buffer (155 mM NH4Cl, 12 mM NaHCO3, recipe 0.1 mM EDTA) tolyse the RBCs. Finally, the total bone marrow cells were resuspendedat a concentration of 3 million cells/10 μl of PBS, and 10 μl wastransplanted into each femur of the radiated mice.

2.1.2. Preparation of animals for bone marrow transplantationSix- to eight-week-old PAI-1 KO and wild-type mice were radiated

with two divided doses of 425 rad (total dose 850 rad) each 4 hoursapart after keeping them on acidified water (pH=2.5–3) for 7 dayssupplemented with neomycin 1.1 g/l and Polymixin 1 million U/l.Various groups of mice for the experimental study were generated bytransplanting the PAI-1 KOmice with 3 million total bone marrow cellsof wild-type, p47phox KO, MPO KO, and iNOS KO mice, respectively, viaintrafemoral route as described [11]. Control groups of mice werecreatedby transplantingPAI-1KOmicewithPAI-1KObonemarrowandwild-type mice with wild-type bone marrow, respectively. Thetransplanted mice were kept on acidified water (pH=2.5–3.5) withantibiotics for another 4 weeks in cages which were changed everysecond day.

2.2. Left anterior descending (LAD) artery ligation

Anterior wall MI was performed after 6 weeks of transplantationas previously described [5,12]. Briefly, AMI was induced in PAI-1 KOmice transplanted with PAI-1 KO, wild-type, p47phox KO, MPO KO, andiNOS KO bone marrow by ligation of LAD. The animals wereanesthetized with xylazine/ketamine, intubated, and ventilated with

Fig. 1. Schematic diagram of the

room air at 105 breaths per minute using a rodent ventilator (HarvardApparatus). Sternotomy was performed, and LAD was identified withthe help of surgical microscope (LeicaM500). LAD ligated by using 7-0prolene. Immediate blanching and anterior wall dysfunction revealeda successful ligation. The chest and skinwere closed using 6-0 prolene.The animals were removed from the ventilator and kept under oxygenuntil they recover from anesthesia. Only the animals which survivedfirst 24 hours of ligation were considered for the study.

2.3. Hematoxylin and eosin staining

Five days post MI, the hearts from PAI-1 KOmice andWTmicewereharvested and fixed in formalin. The tissues were embedded in paraffin,and 6-μm sections were generated from below the level of the LADligation. The cross sections were stained with hematoxylin and eosin.

2.4. Immunostaining

The hearts from WT bone marrow and p47phox KO bone marrow-transplanted PAI-1 KO mice were harvested 3 days after MI. Thehearts were fixed in formalin and embedded in paraffin. Six-micrometer sections were cut and then deparaffinized as described[5]. The sections were immunostained with primary p47phox (NCF-1)antibody (Novus Biologicals; NBP1-40132) overnight at 4 °C followedby secondary biotinylated goat–IgG–Antibody (Novus Biologicals;NB710-B) for 1 hour. The reactionwas quenched using Vectastain ABCand DAB kit (Vector SK-4100). The sections were stained withhematoxylin as described above. The images were analyzed underupright microscope at 40x and 100x magnifications.

2.5. ElLISA for PAI-1 activity

Total PAI-1 and active PAI-1 levels in tissue were compared amongtwo groups: PAI-1 KO mice transplanted with wild-type bone marrowand PAI-1 KO mice transplanted with p47phox KO bone marrow. Theanimals were sacrificed at day 3 post AMI, and the heart was perfusedwith10 mlof saline to removeblood fromthe tissue. The infarcted tissueof the left ventricle was cut with scissors just below the ligation of LADartery. Tissue from these groups for total PAI-1 and active PAI-1 ELISAassays were homogenized in PBSwith 0.1% Triton X-100 supplementedwith PMSF (100 mM), leupeptin (10 μg/ml), and aprotinin (10 μg/ml)[13]. The solutions were then centrifuged at 5000 rpm for 10 min. Thesupernatant was collected and frozen at −80 °C until used. Assay was

time course for the study.

428 U. Agarwal et al. / Journal of Molecular and Cellular Cardiology 50 (2011) 426–432

performedasper theguidelines onELISAplates for total and active PAI-1from Molecular Innovations. One hundred fifty micrograms of proteinwas loaded in each well. The absorbance was measured at 450 nm, andthe active PAI-1/total PAI-1 ratio was analyzed.

2.6. ELISA for PAI-1 nitration

Tissue lysates were prepared as mentioned above and levels ofnitrotyrosine residues on PAI-1 molecule were compared among twogroups: PAI-1 KO mice transplanted with wild-type bone marrow andPAI-1 KO mice transplanted with p47phox KO bone marrow. Onehundred fifty micrograms of protein lysate was subjected to Total PAI-1ELISA plates. Nitrotyrosine residues on total PAI-1 attached on total PAI-1 ELISA plateswere detected using nitrotyrosine primary and secondaryantibody from OxiSelect™ Nitrotyrosine ELISA Kit (Catalog # STA-305)and analyzedwith the standard curve obtained fromnitrated BSA of thesamekit. The ratio of total nitrotyrosine concentration and total PAI-1 inthe both the groups was compared.

2.7. Statistical analysis

All numerical data are expressed as mean±SEM. Comparisonamong two groups was done by t-test. Survival curves were derivedusing the Kaplan–Meier method and compared using a log-rank test.The differences were considered significant at a p value of b0.05.

3. Results

3.1. Hemorrhagic Infarct in PAI-1 KO mice

PAI-1, an inhibitor of uPA/tPA driven plasminogen/plasmin axis,when deficient, causes a robust increase of extracellular matrixdegradation. Moreover, in the blood compartment, PAI-1 deficiencyleads to rapid degradation of clot and bleeding. On performing theautopsy on 5-day-old PAI-1 KO mice, we found a hemorrhagic infarct

Fig. 2. Hemorrhagic infarct in PAI-1 KOmice 5 days post MI. (A) Figure Gross myocardialhemorrhage in the myocardium just below the level of the LAD ligation (arrow). (B) H&Estaining of myocardium 5 days post MI demonstrates bleeding into the myocardium(inset).

(Fig. 2A),whichwas confirmed byH&E staining (Fig. 2B). These findingsare consistentwith accelerated tissue degradation in the PAI-KOmouse.

3.2. Effect of WT marrow in rescuing PAI-1 null mouse

We performed bone marrow transplantation (BMT) with PAI-1 KOmice using bone marrow harvested from PAI-1 KO or WT mice. At4 weeks post transplantation, both groups of mice were found to havesimilar levels of bonemarrow reconstitution based on complete bloodcounts. Six weeks after BMT, acute myocardial infarction was induced,and the survival was monitored for 21 days. All PAI-1 null micetransplanted with PAI-1 KO marrow died of ventricular rupturewithin 7 days of myocardial infarction (Fig. 2). As in previous studies,ventricular rupture was defined as the animal appearing well the dayprior to death and presence of blood in the chest post-mortem [5]. Ofthe PAI-1 KO mice transplanted with WT marrow, 41.66% survived to7 days and 8.33% of the mice survived to 21 days post AMI. The wild-type mice transplanted with wild-type marrow had an 83% survival at21 days post MI (Fig. 3).

3.3. Effect of leukocyte-derived oxidant-generating enzyme systems inlong-term rescue of PAI-1 null mice

To study the effect of leukocyte-derived oxidant-generatingsystem on PAI-1 oxidation after AMI, we generated the PAI-1-deficient mice with bone marrow derived from p47phox KO(Fig. 3A), MPO KO (Fig. 3B), and iNOS KO (Fig. 3C), respectively.These different groups of mice had white blood cell PAI-1 expressionbut lack NAD(P)H oxidase, MPO, or iNOS function in their leukocytes.Following BMT, WBC reconstitution at 28 days was within normallimits of WT mice regardless of the source of bone marrow. After AMIinduction, we monitored the ventricular rupture of these mice for21 days and compared it with the PAI-1 KO mice transplanted withWTmarrow.We found that PAI-1 KOmice that received bonemarrowfrom p47phox KO (n=10)mice had a significant increase in survival to30% at 21 days (Fig. 3A) compared to an 8.3% survival at 21 days ofPAI-KO mice that received WT marrow (n=12, pb0.03). MPO nullmarrow (n=11) (Fig. 3B) and iNOS null marrow (n=6) (Fig. 3C)transplanted mice died of ventricular rupture by day 9 after AMI andwere statistically no different than WT marrow (pN0.05).

3.4. Effect of p47phox, MPO and iNOS in the initial post infarct healingphase (b7 days)

The transplantation of WT marrow into PAI-1 KO mice led to adelay in the onset of myocardial rupture from 3 to 4 days after AMI ascompared to PAI-1 KOmice transplanted with PAI-1 KO bonemarrow.Therefore, we wanted to determine if transplantation of p47phox,MPO, or iNOS null bone marrow lead to a further delay of the onset ofrupture. Our results demonstrated that transplantation of p47phox,MPO, and iNOS null marrow delayed the onset of myocardial ruptureto 6 days after AMI. Furthermore, at day 7, when all PAI-1 KO micewere dead, 90% of p47phox KO bone marrow-transplanted, 54.5% ofMPO KO bone marrow-transplanted, and 33.3% of iNOS KO bonemarrow-transplanted PAI-1 KO animals were living, as compared to41.66% of WT bone marrow-transplanted PAI-1 KO animals (Table 1).At day 7 post MI, among all the groups when compared to WTtransplanted PAI-1 KO mice, only p47phox-deficient bone marrow-transplanted PAI-1 KO mice had a significant difference in survivalcompared to mice that received WT marrow (p valueb0.03).

3.5. Characterization of p47phox expression

To further verify and investigate the role ofWBC p47phox expression,we performed immunostaining for the C-terminal of p47phox subunit ofNADPH oxidase complex 3 days afterMI. As seen in Fig. 4, those animals

Fig. 3.Mortality due to ventricular rupture was monitored as a function of time after LAD ligation. The survival data for WT bone marrow transplantedWTmice, grey squares, n=6;PAI-1 KO mice transplanted with PAI-1 KO bone marrow, open circles, n=7; and WT bone marrow, grey circles, n=12, are repeated on each figure to aid comparison. The blacksquares represent data from PAI-1 KO mice transplanted with (A) p47phox KO bone marrow, n=10; (B) MPO KO bone marrow, n=11; and (C) iNOS KO bone marrow, n=6,respectively. At day 21, the p47phox KO marrow transplanted PAI-1 KO animals have a significantly better survival, pb0.03, as compared to the WT bone marrow transplanted PAI-1KO animals. All other groups were not statistically different than PAI-1 KO mice that received WT bone marrow.

429U. Agarwal et al. / Journal of Molecular and Cellular Cardiology 50 (2011) 426–432

that received WT bone marrow transplantation had intense staining ofp47phox in thewhitebloodcells and significantp47phox expression in thesurrounding cardiac myocytes. Conversely, there was no p47phox

expression in the PAI-1 null mice that received p47phox null bonemarrow. Importantly, in the mice that received p47phox null bonemarrow, there was still significant p47phox expression in the cardiacmyocytes. This observation further supports our hypothesis that whiteblood cell NADPH oxidase is critical for PAI-1 oxidation.

3.6. PAI-1 activity in vivo

To confirm the effects of p47phox on PAI-1 activity, we wanted toverify that the PAI-1 KO mice transplanted with p47phox KO marrowhad increased PAI-1 activity as compared to PAI-1 KO mice thatreceived WT marrow. Therefore, we examined the PAI-1 activity inthe infarcted region of the heart in both groups 3 days after AMI using

Table 1Survival 7 days post MI in PAI-1 KO mice that received bone marrow transplantationfrom different mice strains that lacked specific oxidant-generating enzyme systems.

Bone marrow transplanted into PAI-1 KO mice

PAI-1 KO WT MPO KO p47phox KO iNOS KOn=7 n=12 n=11 n=10 n=6

7-day survival (%) 0 41.7 54.5 90.0⁎ 66.7

⁎ Pb0.03 compared to WT.

ELISA and activity assays. The total PAI-1 in both groups was similar;however, the amount of active PAI-1 present in p47phox KO bonemarrow-transplanted PAI-1 KOmice was 2.6-fold higher as comparedto WT bone marrow-transplanted PAI- KO mice. As shown in Figs. 5Aand B, while N90% of PAI-1 is inactive 3 days after LAD ligation, therewas significantly greater PAI-1 activity in those animals lacking WBCNAD(P)H oxidase compared to WTWBC, sufficient enough to prolongthe survival of these animals.

3.7. Oxidation of PAI-1 in transplanted mice

We wanted to confirm that the PAI-1 derived from the blood cellsin WT transplanted PAI-1 KO mice was modified in vivo 3 days postMI. Therefore, we estimated the change in total nitration of PAI-1protein by performing ELISA assay. We have previously demonstratedthat changes in PAI-1 nitrotyrosine levels correlate with changes inoxidation of PAI-1 [5]. Therefore, total PAI-1 in tissue lysates wascaptured onto the ELISA plates coated with PAI-1 antibody. Followingthis, the total PAI-1 present on the plate was interrogated withnitrotyrosine antibody. A standard curve made from nitrated BSA wasestimated to compare the total nitrotyrosine concentration on PAI-1molecules. The results demonstrated that total nitrotyrosine concen-tration per nanogram of total PAI-1 obtained from WT bone marrow-transplated PAI-1 KO mice (n=5) was 2.7-fold greater (pb0.04)compared to PAI-1 obtained from p47phox KO bone marrow-transplanted PAI-1 KO mice (n=7) (Fig. 6). These findings demon-strate that NAD(P)H oxidase deletion had a significant impact indecreasing PAI-1 oxidation and thereby increasing PAI-1 activity.

Fig. 4. Immunostaining for NAD(P)H oxidase subunit p47phox expression: (A) NAD(P)H oxidase subunit p47phox is expressed in the cardiac myocyte andwhite blood cells ofWT bonemarrow-transplanted PAI-1 KO mice. (B) NAD(P)H oxidase subunit p47phox is expressed in the cardiac myocyte but not in the white blood cells of p47phox KO bone marrow-transplanted PAI-1 KO mice. Inset shows the 100× image.

430 U. Agarwal et al. / Journal of Molecular and Cellular Cardiology 50 (2011) 426–432

4. Discussion

This study adds to the growing evidence for a role for leukocyte-derived oxidants in regulating ventricular rupture after myocardialinfarction. In addition, these data further refine the critical enzymesystems that regulate protease activation and suggest the relevantenzymes systems that could be targets for therapeutic benefit.

The propeptide of zymogen plasminogen is cleaved to plasmin byuPA/tPA [14]. PAI-1 regulates the activation of uPA [15] and as weshow here this reaction is redox-sensitive. In the setting ofuninhibited uPA activity, plasmin is generated from plasminogenleading to degradation of the extra cellular matrix as well as thesynergistic activation of MMP-9 [16,17]. The increased proteaseactivity can increase the risk of ventricular rupture as evidenced bythe inevitable myocardial rupture in PAI-1-deficient mice [5]. Ourobservation that transplantation of WT marrow into PAI-1 null miceled to the incomplete delay and inhibition of myocardial ruptureallowed us to develop the system through which we couldsystematically investigate the relative oxidant-generating enzymesystems involved inmodulating PAI-1 activity. To achieve our goal, weused the PAI KO model as a host to study the relative contribution ofeach oxidant-generating enzyme in oxidizing PAI-1 and subsequentprevention of ventricular rupture in PAI-1 KO mice.

We tested the relative importance of NAD(P)H oxidase, MPO, andiNOS in inactivating PAI-1 activity following AMI.We rationalized thatusing the PAI-1 KO mouse as a host and knocking these enzymes outof the white blood cells that expressed PAI-1 [18–20] at the site ofinflammation in the heart, we could determine those enzyme systems

that had a significant role in the functional consequences of PAI-1oxidation. After complete bone marrow reconstitution, the blood ofthese mice had PAI-1 but lacked the specific oxidant-generatingenzyme system. Interestingly, only the deletion of NAD(P)H oxidaseled to any significant prolongation of survival following AMI. Theseresults suggested that NAD(P)H oxidase derived free radicals have amajor rate limiting role in oxidizing PAI-1.

Leukocyte infiltration following AMI begins within hours of AMIand peaking at days 2–4 following AMI [21]. In PAI-1-deficient mice,the volume of neutrophilic infiltration is increased at day 3 post AMI[22] and their hearts begin to rupture at an earlier time point ascompared to thewild-typemice. Our results demonstrated that PAI-1-deficient mice begin to rupture their ventricle 3 days after AMIcompared to 4 days in the PAI-1 null mice that received WT bonemarrow. Interestingly, ventricular rupture was delayed to at least6 days after AMI in the PAI-1 KO mice transplanted with marrowdeficient in any of the oxidant-generating enzymes. Although theventricular rupture was delayed during the initial phase of inflam-mation, long-term survival was only affected by deletion of NAD(P)Hoxidase. This finding may be attributed to the fact that NAD(P)Hoxidase is upstream of MPO, and iNOS. Therefore, in the setting ofdepletion of NAD(P)H oxidase-derived superoxide ion, the down-stream effects of MPO and iNOS are also inhibited, leading to a near-to-complete inhibition of leukocyte-generated oxidants generation.Our data further suggest that, in the setting of the lack of MPO, or iNOSbut replete NAD(P)H oxidase, there is sufficient PAI-1 oxidation fromthe remaining enzyme systems that tissue degradation is notinhibited and ventricular rupture still occurs.

Fig. 5. Total PAI-1 and active PAI-1 was analyzed using ELISA assay. Figure demonstratesno difference in total PAI-1 concentration; however, active PAI-1 in the hearts of PAI-1KO mice transplanted with p47phox KO bone marrow was significantly 2.6-fold higheras compared to PAI-1 KO mice transplanted WT total bone marrow (n=5) 3 days afterLAD ligation. Data represent mean±SEM, p valueb0.03.

431U. Agarwal et al. / Journal of Molecular and Cellular Cardiology 50 (2011) 426–432

4.1. Limitations

There are limitations associated with the methodologies imple-mented in our studies that should be mentioned. Specifically, LVfunction and remodeling in these mice was not assessed in our

Fig. 6. Nitrotyrosine residues on total PAI-1 were determined by ELISA assay at day 3post MI. Figure shows 2.7-fold increase in total nitration of PAI-1 protein in WT bonemarrow-transplanted PAI-1 KOmice (n=5), as compared to p47phox KO bone marrow-transplanted PAI-1 KO mice (n=7). Data represent mean±SEM, p valueb0.04.

studies. Handling of mice induces stress which can be associated withincreased cardiac rupture. Therefore, to minimize potentially con-founding variables, these mice were not manipulated following theLAD ligation protocol. Another limitation is that multiple cell types aretransplanted following whole bone marrow transplantation includingstem cells. Therefore, it is theoretically possible that diminished stemcell function secondary to deletion of oxidant generating systemscould be responsible for some our observations. Finally, otherproteases and protease inhibitors are oxidant-sensitive. For example,inhibition of NAD(P)H oxidase has been shown to decrease MMP-2activity [23] which preserves ventricular function. Thus, our principalfinding that NAD(P)H oxidase is themost relevant oxidant-generatingsystem for ventricular rupture may serve as an important startingpoint for future studies on the effects of oxidation on the function ofother proteases and protease inhibitors.

Consistent with our findings recent studies suggest that short-termtreatmentwith statins in patients improve ventricular remodeling afterAMI [10,24–26]. The subsequent decrease in oxidative stress at the siteof inflammation has been shown to be a potential reason. However, ourfindings demonstrate a plausible mechanism for these observations ofimproved LV remodeling in these patients due to NAD(P)H oxidaseinhibition, preserved PAI-1 oxidation, anddecreased protease activationand tissue degradation.

In conclusion, our study demonstrates that PAI-1 has an importantrole to play in ventricular remodeling and rupture post MI and offers anovel mechanistic link between NAD(P)H oxidase and PAI-1 oxida-tion. These data further suggest inhibition of PAI-1 oxidation andprotease activation as a potential target for the early administration ofstatin therapy following AMI.

Disclosures

None of the authors have any conflicts of interest related to thesestudies or their findings.

References

[1] Przyklenk K, Connelly CM, McLaughlin RJ, Kloner RA, Apstein CS. Effect of myocytenecrosis on strength, strain, and stiffness of isolated myocardial strips. Am Heart J1987;114:1349–59.

[2] Penn MS. The role of leukocyte-generated oxidants in left ventricular remodeling.Am J Cardiol 2008;101:30D–3D.

[3] Heymans S, Luttun A, Nuyens D, Theilmeier G, Creemers E, Moons L, et al.Inhibition of plasminogen activators or matrix metalloproteinases preventscardiac rupture but impairs therapeutic angiogenesis and causes cardiac failure.Nat Med 1999;5:1135–42.

[4] Moriwaki H, Stempien-Otero A, Kremen M, Cozen AE, Dichek DA. Overexpressionof urokinase by macrophages or deficiency of plasminogen activator inhibitor type1 causes cardiac fibrosis in mice. Circ Res 2004;95:637–44.

[5] Askari A, Brennan ML, Zhou X, Drinko J, Morehead A, Thomas JT, et al.Myeloperoxidase and plasminogen activator inhibitor-1 play a central role inventricular remodeling after myocardial infarction. J Exp Med 2003;197:615–24.

[6] Strandberg L, Lawrence DA, Johansson LB, Ny T. The oxidative inactivation ofplasminogen activator inhibitor type 1 results from a conformational change in themolecule and does not require the involvement of the P1' methionine. J Biol Chem1991;266:13852–8.

[7] Sigurdardottir O, Wiman B. Identification of a PAI-1 binding site in vitronectin.Biochim Biophys Acta 1994;1208:104–10.

[8] Doerries C, Grote K, Hilfiker-Kleiner D, Luchtefeld M, Schaefer A, Holland SM, et al.Critical role of the NAD(P)H oxidase subunit p47phox for left ventricularremodeling/dysfunction and survival after myocardial infarction. Circ Res2007;100:894–903.

[9] Liu YH, Carretero OA, Cingolani OH, Liao TD, Sun Y, Xu J, et al. Role of induciblenitric oxide synthase in cardiac function and remodeling in mice with heartfailure due to myocardial infarction. Am J Physiol Heart Circ Physiol 2005;289:H2616–23.

[10] Ito M, Adachi T, Pimentel DR, Ido Y, Colucci WS. Statins inhibit beta-adrenergicreceptor-stimulated apoptosis in adult rat ventricular myocytes via a Rac1-dependent mechanism. Circulation 2004;110:412–8.

[11] Mazurier F, Doedens M, Gan OI, Dick JE. Rapid myeloerythroid repopulation afterintrafemoral transplantation of NOD–SCID mice reveals a new class of humanstem cells. Nat Med 2003;9:959–63.

432 U. Agarwal et al. / Journal of Molecular and Cellular Cardiology 50 (2011) 426–432

[12] Vasilyev N, Williams T, Brennan ML, Unzek S, Zhou X, Heinecke JW, et al.Myeloperoxidase-generated oxidants modulate left ventricular remodeling butnot infarct size after myocardial infarction. Circulation 2005;112:2812–20.

[13] Arndt PG, Young SK, Poch KR, Nick JA, Falk S, Schrier RW, et al. Systemic inhibitionof the angiotensin-converting enzyme limits lipopolysaccharide-induced lungneutrophil recruitment through both bradykinin and angiotensin II-regulatedpathways. J Immunol 2006;177:7233–41.

[14] Thorsen S, Mullertz S, Suenson E, Kok P. Sequence of formation of molecular formsof plasminogen and plasmin–inhibitor complexes in plasma activated byurokinase or tissue-type plasminogen activator. Biochem J 1984;223:179–87.

[15] Kohler HP, Grant PJ. Plasminogen-activator inhibitor type 1 and coronary arterydisease. N Engl J Med 2000;342:1792–801.

[16] Gong Y, Hart E, Shchurin A, Hoover-Plow J. Inflammatory macrophage migrationrequiresMMP-9 activation by plasminogen inmice. J Clin Invest 2008;118:3012–24.

[17] Liu Z, Li N, Diaz LA, Shipley M, Senior RM,Werb Z. Synergy between a plasminogencascade and MMP-9 in autoimmune disease. J Clin Invest 2005;115:879–87.

[18] Barnes JL, Mitchell RJ, Torres ES. Expression of plasminogen activator-inhibitor-1(PAI-1) during cellular remodeling in proliferative glomerulonephritis in the rat. JHistochem Cytochem 1995;43:895–905.

[19] Park YJ, Liu G, Lorne EF, Zhao X, Wang J, Tsuruta Y, et al. PAI-1 inhibits neutrophilefferocytosis. Proc Natl Acad Sci USA 2008;105:11784–9.

[20] Wojta J, Kaun C, Zorn G, Ghannadan M, Hauswirth AW, Sperr WR, et al. C5astimulates production of plasminogen activator inhibitor-1 in human mast cellsand basophils. Blood 2002;100:517–23.

[21] Yang F, Liu YH, Yang XP, Xu J, Kapke A, Carretero OA. Myocardial infarction andcardiac remodelling in mice. Exp Physiol 2002;87:547–55.

[22] Zaman AK, Fujii S, Schneider DJ, Taatjes DJ, Lijnen HR, Sobel BE. Deleterious effectsof lack of cardiac PAI-1 after coronary occlusion in mice and their pathophysi-ologic determinants. Histochem Cell Biol 2007;128:135–45.

[23] Looi YH, Grieve DJ, Siva A, Walker SJ, Anilkumar N, Cave AC, et al. Involvement ofNox2 NADPH oxidase in adverse cardiac remodeling after myocardial infarction.Hypertension 2008;51:319–25.

[24] Teshima Y, Yufu K, Akioka H, Iwao T, Anan F, Nakagawa M, et al. Early atorvastatintherapy improves cardiac function in patients with acute myocardial infarction.J Cardiol 2009;53:58–64.

[25] Kanadasi M, Cayli M, Demirtas M, Inal T, Demir M, Koc M, et al. The effect of earlystatin treatment on inflammation and cardiac events in acute coronarysyndrome patients with low-density lipoprotein cholesterol. Heart Vessels2006;21:291–7.

[26] Gurgun C, Ildizli M, Yavuzgil O, Sin A, Apaydin A, Cinar C, et al. The effects of shortterm statin treatment on left ventricular function and inflammatory markers inpatients with chronic heart failure. Int J Cardiol 2008;123:102–7.