Phenyl-\u0001-tert-butyl-nitrone and Benzonidazole Treatment Controlled the Mitochondrial Oxidative Stress and Evolution of Cardiomyopathy in Chronic Chagasic Rats

$Page 1: Phenyl-\u0001-tert-butyl-nitrone and Benzonidazole Treatment Controlled the Mitochondrial Oxidative Stress and Evolution of Cardiomyopathy in Chronic Chagasic Rats$
Ccadnt

FCfDsI

a

Journal of the American College of Cardiology Vol. 55, No. 22, 2010© 2010 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00P

PRE-CLINICAL RESEARCH

Phenyl-�-tert-butyl-nitrone andBenzonidazole Treatment Controlled theMitochondrial Oxidative Stress and Evolutionof Cardiomyopathy in Chronic Chagasic Rats

Jian-Jun Wen, PHD,* Shivali Gupta, PHD,* Zhangjun Guan, MD,† Monisha Dhiman, PHD,*David Condon,* Charles Lui, MD,† Nisha Jain Garg, PHD*‡§

Galveston, Texas

Objectives The purpose of this study was to determine the pathological importance of oxidative stress–induced injuriousprocesses in chagasic heart dysfunction.

Background Trypanosoma cruzi–induced inflammatory pathology and a feedback cycle of mitochondrial dysfunction and oxi-dative stress may contribute to Chagas disease.

Methods Sprague-Dawley rats were infected with T. cruzi and treated with phenyl-�-tert-butylnitrone (PBN), an anti-oxidant, and/or benzonidazole (BZ), an antiparasitic agent. We monitored myocardial parasite burden, oxidativeadducts, mitochondrial complex activities, respiration, and adenosine triphosphate synthesis rates, and inflam-matory and cardiac remodeling responses during disease development. The cardiac hemodynamics was deter-mined for all rats.

Results Benzonidazole (not PBN) decreased the parasite persistence and immune adverse events (proinflammatory cyto-kine expression, �-nicotinamide adenine dinucleotide phosphate oxidase and myeloperoxidase activities, andinflammatory infiltrate) in chronically infected hearts. PBN � BZ (not BZ alone) decreased the mitochondrial re-active oxygen species level, oxidative adducts (malonyldialdehyde, 4-hydroxynonenal, carbonyls), hypertrophicgene expression (atrial natriuretic peptide, B-type natriuretic peptide, �-skeletal actin), and collagen depositionand preserved the respiratory chain efficiency and energy status in chronically infected hearts. Subsequently, LVdysfunction was prevented in PBN � BZ-treated chagasic rats.

Conclusions BZ treatment after the acute stage decreased the parasite persistence and inflammatory pathology. Yet, oxida-tive adducts, mitochondrial dysfunction, and remodeling responses persisted and contributed to declining cardiacfunction in chagasic rats. Combination treatment (PBN � BZ) was beneficial in arresting the T. cruzi–induced in-flammatory and oxidative pathology and chronic heart failure in chagasic rats. (J Am Coll Cardiol 2010;55:2499–508) © 2010 by the American College of Cardiology Foundation

ublished by Elsevier Inc. doi:10.1016/j.jacc.2010.02.030

gap

ppamaoidt

hagas disease is an emerging parasitic disease in developedountries. In endemic countries, �16 to 18 million peoplere infected with Trypanosoma cruzi and at risk of diseaseevelopment (1). In most patients, acute infection results inonspecific clinical symptoms. Several years after that, 30%o 40% of seropositive individuals develop progressive cha-

rom the Departments of *Microbiology and Immunology, †Internal Medicine-ardiology, and ‡Pathology, §Institute for Human Infections and Immunity, Center

or Biodefense and Emerging Infectious Diseases, and Sealy Center for Vaccineevelopment, University of Texas Medical Branch, Galveston, Texas. This work was

upported by a grant (AI054578) from the National Institutes of Health/Nationalnstitute of Allergy and Infectious Disease to Dr. Garg.

tManuscript received July 28, 2009; revised manuscript received February 9, 2010,

ccepted February 15, 2010.

asic cardiomyopathy with ventricular dilation, arrhythmia,nd contractile dysfunction, leading to cardiac arrest andatient’s death (2).Several lines of evidence establish that a low level of

arasites remains in the host during the chronic diseasehase (3). Direct detection of parasites in autopsy specimensnd myocardial biopsy specimens has been shown by im-unohistochemical and immunofluorescence techniques (4)

nd polymerase chain reaction (PCR) (5). The reactivationf acute infection and parasitemia has been noted aftermmunosuppression due to acquired immunodeficiency syn-rome (6) or drug therapy (7). Blood transfusion andransplantation of infected organs obtained from asymp-
omatic individuals are major causes of T. cruzi transmission

iii

iociberttmgptbc

M

DAo

cwidmdtsCRdudsAwm(RLm(didMwhAIcoaxo(g

ktHd(pttci3cKat

2500 Wen et al. JACC Vol. 55, No. 22, 2010Effects of PBN and Benzonidazole in Chagas Disease June 1, 2010:2499–508

in nonendemic countries (8).These studies indicated that par-asite persistence results in consis-tent activation of inflammatoryresponses and the developmentand/or propagation of patholog-ical lesions in the heart.

We and others (9) have shownthat chagasic myocardia sustainoxidative injuries. Phagocytes(macrophages and neutrophils),activated to control T. cruzi in-fection, are a major source ofoxidative stress in the acutely in-fected host. Further, myocardialoxidative stress is enhanced dueto increased mitochondrial pro-duction of reactive oxygen species(ROS) during Chagas disease(10). A compromised antioxi-dant status in T. cruzi–infectedexperimental animals (11) wasassociated with enhanced oxida-tive modification of cellular(11,12) and mitochondrial pro-teins and lipids (13). The plasmaand intraerythrocyte levels ofglutathione and glutathione per-oxidase are decreased (14), andglutathione disulfide and malo-nyldialdehyde (MDA) contents

ncreased (15) in human chagasic patients. These studiesndicated that oxidative injurious processes are of patholog-cal significance in Chagas disease.

In this study, we aimed to evaluate the relative patholog-cal importance of parasite persistence and oxidative injuri-us processes in the progression of Chagas disease andardiac contractile dysfunction. Sprague-Dawley rats werenfected with T. cruzi and treated with phenyl-�-tert-utylnitrone (PBN), a nitrone-based antioxidant that scav-nges a wide variety of free radical species and inhibits freeadical generation (16). Some of the infected rats werereated with benzonidazole (BZ), which is currently thereatment of choice for chagasic patients (17). We deter-ined the efficacy of PBN intervention in reducing the

eneration or the effects of ROS and of BZ in controllingarasite persistence and subsequently evaluated whetherhese treatments (individually or in combination) wereeneficial in preventing progressive chagasic pathology andardiac dysfunction in chronically infected rats.

ethods

etails are provided in the Online Appendix.nimals and parasites. Sprague-Dawley rats (4 to 5 weeks

Abbreviationsand Acronyms

�sk actin � �-skeletalactin

ANOVA � analysis ofvariance

ANP � atrial natriureticpeptide

ATP � adenosinetriphosphate

BNP � B-type natriureticpeptide

BZ � benzonidazole

HNE � 4-hydroxynonenal

LV � left ventricular

MDA � malonyldialdehyde

MPO � myeloperoxidase

NOX � �-nicotinamideadenine dinucleotidephosphate oxidase

PBN � phenyl-�-tert-butylnitrone

PCR � polymerase chainreaction

P-V � pressure-volume

pyr/mal � pyruvate/malate

ROS � reactive oxygenspecies

XOD � xanthine oxidase

ld, Harlan Laboratories, Indianapolis, Indiana) were ac- n

ustomed to animal facility for 1 month and then infectedith T. cruzi (SylvioX10/4, 50,000 trypomastigotes per rat,

ntraperitoneally). Rats were given PBN (1.3 mM, beginningay 0 throughout the course of infection) and/or BZ (0.7M, beginning day 40 post-infection, for 3 weeks) in

rinking water. Tissue parasite burden was determined byraditional and real-time PCR using T. cruzi 18SrDNA–pecific oligonucleotides (18). The institutional Animalare and Use Committee approved all animal experiments.OS and adenosine triphosphate (ATP). The myocar-ial and mitochondrial levels of ROS were determinedsing dihydroethidium and amplex red/horseradish peroxi-ase fluorescent probes (19). Cryostat tissue sections weretained with dihydroethidium to detect in situ ROS. TheTP level in tissue homogenates and isolated mitochondriaas determined by a luciferin/luciferase bioluminescenceethod (19). Mitochondria were energized with pyr/mal

pyruvate/malate) and succinate to monitor the rates ofOS and ATP production.ipid and protein oxidation. The MDA level was deter-ined by a thiobarbituric acid reactive substances assay

20). Protein carbonyls were derivatized with 2,4-initrophenylhydrazine and estimated by enzyme-linked

mmunosorbent assay (21) using a polyclonal anti-2,4-initrophenylhydrazine antibody (Chemicon, Billerica,assachusetts). Paraffin-embedded tissue sections (5 �m)

ere subjected to immunostaining with an anti– 4-ydroxynonenal (HNE) antibody (Alpha Diagnostics, Sanntonio, Texas), and images analyzed by light microscopy.

nflammation and tissue pathology. Formalin-fixed myo-ardial tissue sections were stained with hematoxylin/eosinr Masson’s trichrome and scored for myocarditis, fibrosis,nd parasitic lesions (22). The enzymatic activities ofanthine oxidase (XOD), �-nicotinamide adenine dinucle-tide phosphate oxidase (NOX), and myeloperoxidaseMPO) were monitored by spectrophotometry and/or in-el catalytic staining.

The myocardial mRNA levels for proinflammatory cyto-ines and hypertrophy markers were determined by real-ime reverse transcriptase PCR (18).

eart function. Rats were monitored for heart functionuring the acute (25 to 40 days post-infection) and chronic�150 days post-infection) stages of infection. After intra-eritoneal injection of heparin (1,000 IU/kg) and adminis-ration of anesthesia, tracheotomy was performed to facili-ate breathing, and a conductance pressure-volume (P-V)atheter (Scisense, Inc., London, Ontario, Canada) wasnserted into the left ventricle. Respiration (80 breaths/min,

l/min O2) was controlled through a tracheotomy cannulaonnected to a pressure-controlled respirator (RSP1002,ent Scientific, Litchfield, Connecticut). The conductance

nd pressure signals were digitized, and the volume calibra-ion of the conductance system was performed. Hemody-
amic measurements were made using emka cardiac P-V

aCDaacta(Kdu(

R

PnaTtuntw

nqOafi(sprRRticwc5�tecw(

2501JACC Vol. 55, No. 22, 2010 Wen et al.June 1, 2010:2499–508 Effects of PBN and Benzonidazole in Chagas Disease

nalysis software (emka TECHNOLOGIES USA, Fallshurch, Virginia).ata analysis. Data were expressed as mean � SD (9

nimals per group, triplicate observations per experiment),nalyzed by SPSS software version 14.0 (SPSS, Inc., Chi-ago, Illinois). Normally distributed data were analyzed byhe Student t test (comparison of 2 groups) and 1-waynalysis of variance (ANOVA) with Tukey’s post-hoc testcomparison of multiple groups). Mann-Whitney andruskal-Wallis tests were used when data were not normallyistributed. Significance is shown by �normal-versus-infected/ntreated and #infected/untreated-versus-infected/treated* #p � 0.05, ** ##p � 0.01, *** ###p � 0.001).

esults

arasite burden. Semiquantitative PCR showed a sig-ificant level of Tc18SrDNA signal in the myocardium ofcutely and chronically infected rats (Figs. 1A and 1B).he extent of Tc18SrDNA amplification was similar in

he heart tissue of infected/PBN-treated and infected/ntreated rats. In comparison, Tc18SrDNA-specific sig-al was barely detectable in the myocardium of BZ-reated/chronically infected rats. Specificity of the PCRas confirmed by no amplification of Tc18SrDNA with

Figure 1 Tissue Burden of Trypanosoma cruzi in Infected Rats

Myocardial total DNA was used as a template, and T. cruzi 18SrDNA amplified bycation of Tc18SrDNA PCR amplicons (28 cycles), normalized to rat glyceraldehyde-as mean � SD (n � 9; ###p � 0.001). AH � acutely infected/untreated (25 dayschronically infected/untreated (�150 days post-infection); CHP � chronically infecPBN � BZ treated; NH � normal heart; PBN � phenyl-�-tert-butylnitrone.

ormal rat DNA. Real-time PCR was used to obtain auantitative measure of parasite burden in infected rats.ur data showed a 15-fold higher parasite burden in

cutely infected hearts compared with chronically in-ected hearts (Fig. 1C). BZ treatment (� PBN) resultedn 55% decrease in Tc18SrDNA signal in infected ratsFig. 1C) (p � 0.01 [ANOVA-Tukey’s]). These resultshowed that BZ was effective in controlling the parasiteersistence, and PBN alone had no effect on parasiteeplication and/or survival in the heart.OS and oxidative damage. Fluorometric evaluation ofOS showed 72% and 63% increases in H2O2 levels in

he myocardial homogenates of acutely and chronicallynfected rats (Fig. 2A) (p � 0.01 [t test]) that was notontrolled in infected/BZ-treated rats. When treatedith PBN (� BZ), the enhanced level of H2O2 was

ontrolled by 96% in acute myocardium and by 49% to3% in chronically infected hearts (Fig. 2A) (all p values0.01 [t test/ANOVA-Tukey’s]). In situ studies verified

hese results. We noted a substantial DHE oxidation tothidium red in the myocardial sections of acutely andhronically infected rats that were not treated or treatedith BZ only (Fig. 2B). Infected rats treated with PBN

� BZ) exhibited a significant decrease in ethidium

N/BZ)

, B) and real-time polymerase chain reaction (PCR) (C). Densitometric quantifi-phate dehydrogenase (GAPDH), is shown in B. Data in all figures are presentedinfection); AHP � acutely infected/PBN treated; BZ � benzonidazole; CH �

N treated; CHB � chronically infected/BZ treated; CHBP � chronically infected/

(� PB

PCR (A3-phospost-

ted/PB

fli

o1aarcast(pecr((dslMfiqd

d(iRmic07amStd(pBac

ri8mtdiAdr[ss4nafi

hwcWssLinBMaoIam


uorescence during the acute and chronic stages ofnfection and disease development (Fig. 2B).

In agreement with ROS levels, the myocardial contentsf MDA and carbonyl adducts were increased by 75% and89%, respectively, in the acutely infected rats and by 66%nd 168%, respectively, in chronically infected rats (Figs. 3And 3B) (p � 0.001 [t test]). The PBN (� BZ) treatmentesulted in a decrease in myocardial MDA and carbonylontents by 62% and 77%, respectively, in the acute stagend by 84% to 88% and 100%, respectively, in the chronictage (all p values �0.001 [t test/ANOVA-Tukey’s]). Ratsreated with BZ only exhibited a slight control of MDA31% decrease) and no decrease in carbonyl adducts com-ared with infected/untreated rats. Immunostaining showedxtensive HNE adducts (intensity score: 2 to 3) in acutely andhronically infected rats and in chronically infected/BZ-treatedats (Fig. 3C). HNE adducts were significantly decreasedintensity score: 0 to 1) in the myocardium of PBN-treated� BZ) rats at both acute and chronic stages of infection andisease development (Fig. 3C). These data (Figs. 2 and 3)howed that PBN (but not BZ) treatment decreased the ROSevel and oxidative damage in the myocardium of infected rats.

itochondrial dysfunction. Cardiac mitochondria of in-ected rats exhibited 41% to 43% and 54% to 58% decreasesn the CI (reduced nicotinamide adenine dinucleotide–uinone oxidoreductase) and CIII (cytochrome-c oxi-

Figure 2 Reactive Oxygen SpeciesDetermination in Chagasic Rats (� PBN/BZ)

(A) The myocardial H2O2 levels were determined by the amplex red/horserad-ish peroxidase method. (B) Frozen myocardial sections were loaded with dihy-droethidium (oxidizes to fluorescent ethidium), and micrographs obtained at�20 magnification. ***p � 0.001; ## **p � 0.01; #p � 0.05. Abbreviationsas in Figure 1.

oreductase) respiratory complex activities, respectively, c

uring the course of infection and disease developmentFigs. 4A and 4B) (p � 0.01 [t test]). The respiratory chainnefficiency was associated with enhanced mitochondrialOS production. The pyr/mal- and succinate-stimulateditochondrial ROS production was distinctly increased in

nfected myocardium (acute: 2-fold and 95%, respectively;hronic: 1.8-fold and 91%, respectively) (Fig. 4C) (p �.001 [t test]). Treatment with PBN (� BZ) resulted in0% and 43% to 88% improvement in CI activity in acutelynd chronically infected rats, respectively (Fig. 4A) and nor-alization of CIII activity in infected myocardium (Fig. 4B).

ubsequently, pyr/mal- and succinate-dependent ROS forma-ion was significantly controlled in PBN-treated (� BZ) ratsuring acute (82% and 96% decrease, respectively) and chronic57% and 32% decrease, respectively) stages (Fig. 4C) (all

values �0.001 [t test/ANOVA-Tukey’s]). Treatment withZ alone failed to preserve the respiratory complex activities orvert the increased rate of ROS production in cardiac mito-hondria of infected rats.

The pyr/mal- and succinate-dependent mitochondrialate of ATP synthesis was decreased by 54% to 57% innfected myocardium (Fig. 4D), and associated with 68% to0% and 52% to 62% decreases in mitochondrial andyocardial ATP levels, respectively (Figs. 4E and 4F). PBN

reatment (� BZ) improved the pyr/mal- and succinate-ependent mitochondrial ATP production by 49% to 61%n infected hearts (Fig. 4E) (all p values �0.01 [t test/NOVA-Tukey’s]). Subsequently, mitochondrial and car-iac ATP contents were improved in PBN-treated (� BZ)ats by 35% to 46% (Figs. 4E and 4F) (all p values �0.01t test/ANOVA-Tukey’s]). Treatment with BZ alone re-ulted in no significant improvement in mitochondrial ATPynthesis and myocardial ATP levels in infected rats (Figs.D to 4F). These data (Fig. 4) demonstrated that PBN (butot BZ) preserved the mitochondrial respiratory chainctivity and ATP production and prevented enhanced ROSormation in the myocardium of acutely and chronicallynfected rats.

BZ kills T. cruzi via oxidative stress (23) and thereby mayave standby effects on host mitochondria. To determinehether such is the case, we compared myocardial mito-

hondrial respiration in normal and normal/BZ-treated rats.e observed no significant differences in the rate of

ubstrate-stimulated oxygen consumption in left ventriclelices (in vivo) of normal and normal/BZ-treated rats.ikewise, state 3 respiration and state 3/state 4 ratio of

solated cardiac mitochondria from normal/BZ-treated wereot compromised (Online Fig. 1, Online Table 2). Further,Z-treated/normal rats exhibited no significant increase inDA and carbonyl adducts (data not shown). Thus, BZ

lone did not contribute to mitochondrial dysfunction andxidative stress in the host myocardium.nflammatory pathology. Histological studies showed par-sitic foci and intense inflammatory infiltrate composed ofononuclear (macrophages, T cells) and polymorphonu-

lear (neutrophils) cells in acute myocardium (histological

sm(flaitom3IemtfcBpnl

nSN

dc(lt�rL9a((aattismsih


core 2 to 4) (Fig. 5A). The inflammatory sequel wasoderate and diffused in chronically infected hearts

histological score 1 to 2). The mRNA levels for proin-ammatory cytokines interleukin-1�, interferon gamma,nd tumor necrosis factor-� were significantly increasedn acutely and chronically infected myocardium (Figs. 5Bhrough 5D; all p values �0.001 [t test]). PBN treatmentf infected rats resulted in a moderate decrease inyocardial inflammatory infiltrate (histological score 2 to

) and marginal to no change in cytokine mRNA levels.n comparison, infected rats treated with BZ (� PBN)xhibited a considerable decrease in myocardial inflam-atory infiltrate (histological score 0 to 1). Accordingly,

he myocardial mRNA levels for interleukin-1�, inter-eron gamma, and tumor necrosis factor-� were signifi-antly decreased in chronically infected rats treated withZ only (100%, 99%, and 89.5% decrease, respectively;� 0.01 [ANOVA-Tukey’s]) or treated with a combi-

ation of PBN and BZ (normalized to normal controlevel; p � 0.01 [ANOVA-Tukey’s]) (Fig. 5).

NOX, MPO, and XOD are indicators of macrophage,eutrophil, and endothelial cell activation, respectively.pectrophotometric studies showed a substantial increase in

Figure 3 Myocardial Oxidative Damage in Trypanosoma cruzi–In

(A) Thiobarbituric acid reactive substances assay determination of myocardial malwere measured by enzyme-linked immunosorbent assay. (C) Immunohistochemicacation �40). *** ###p � 0.001; #p � 0.05.

OX, MPO, and XOD activities in the infected myocar- a

ium (acute: 7.6-, 3.2-, and 43.1-fold increase, respectively;hronic: 110%, 2.2-fold, and 108% increase, respectively)Figs. 5E to 5G) (p � 0.01 [t test]). PBN treatment (� BZ)ed to a significant control of NOX and MPO activities inhe acute (60% to 70% and 63%, respectively, all p values

0.01 t test]) and chronic (55% to 95% and 58% to 71%,espectively; all p values �0.01 [ANOVA-Tukey’s]) hearts.ikewise, BZ treatment was effective and resulted in 16% to5% decrease in NOX and 25% to 71% decrease in MPOctivities, respectively, in chronically infected myocardiump � 0.01 [ANOVA-Tukey’s]). PBN and BZ treatmentsindividually or in combination) did not alter the XODctivity in infected rats (Fig. 5G). In-gel catalytic stainingssays validated the spectrophotometric results and showedhat PBN and BZ (individually or in combination) con-rolled the enhanced activities of NOX and MPO innfected myocardium (Online Fig. 2). These data demon-trated that the parasite-dependent infiltration of inflam-atory infiltrate and cytokine expression in the heart were

ignificantly controlled by BZ. PBN was primarily effectiven limiting the inflammatory oxidative stress in chagasicearts as evidenced by a decrease in NOX and MPO

ed Rats (� PBN/BZ)

ldehydes (MDA). (B) 2,4-dinitrophenylhydrazine–derivatized carbonyl proteinsing of heart tissue sections with anti–4-hydroxynonenal (HNE) antibody (magnifi-

fect

onlydial stain

ctivities.

CsaasHcaA6d(B

w((wdmsegHaT


ardiac remodeling. We performed Masson’s trichrometaining of heart sections to detect collagen deposition. Incutely infected rats, scattered foci of myocyte necrosisssociated with inflammatory cells were visible; however, noignificant evidence of fibrosis was detectable (Fig. 6A).istological observation of extensive collagen deposition in

hronically infected rats (histological score 2 to 4) wasssociated with a 10.4-, 4.4-, and 11.2-fold increase inNP, BNP, and �sk actin mRNAs, respectively (Figs. 6B toD) (all p values �0.001 [t test]). We observed a moderateecrease (histological score 2 to 3) in collagen depositionFig. 6A), and a distinct decrease in the enhanced ANP,

Figure 4 Mitochondrial Functional Analysis in Chagasic Rats (�

Spectrophotometric measurement of complex I (A) and complex III (B) activities in(D) production was monitored in pyruvate/malate (pyr/mal)- and succinate (succ)-elevels were determined by a luciferin/luciferase bioluminescence method. *** ##

NP. and �sk actin mRNA levels in infected rats treated d

ith PBN (82%, 77%, and 92%, respectively) or PBN/BZ79.7%, 72.5%, and 74.8%, respectively) (Figs. 6B to 6D)all p values �0.00 [t test/ANOVA-Tukey’s]). Treatmentith BZ alone resulted in a marginal decrease in collageneposition (histological score 3 to 4) and BNP and �sk actinRNAs in chronically infected myocardium. These results

uggested that adjunct therapy with PBN and BZ wasffective in limiting cardiac remodeling responses in cha-asic rats.emodynamic measurements. P-V loops were analyzed

t steady state and after inferior vena cava occlusion (Onlineable 3). Baseline measurements showed no significant

N/BZ)

ted cardiac mitochondria. The rate of H2O2 (C) and adenosine triphosphate (ATP)ed isolated cardiac mitochondria. The mitochondrial (E) and myocardial (F) ATP.001; ##p � 0.01. Abbreviations as in Figure 1.

PB

isolanergiz

#p � 0

ifferences in the hemodynamics of normal and acute rats

(bispscfr

r[ahwosb


data not shown). The pressure shifts of the P-V loops fromaseline (mm Hg: normal, 82.3; chronically infected, 48.1)ndicated a decrease in cardiac contractility in the chronictage. The left ventricular (LV) �dP/dtmax (peak rate ofressure rise), �dP/dtmax (peak rate of pressure decline),troke volume (blood volume pumped with each beat),ardiac output (blood volume pumped per minute), ejectionraction (fraction of blood ejected/contraction), and heart

Figure 5 Cardiac Inflammatory Response in Trypanosoma cruzi

(A) Hematoxylin and eosin staining of heart sections (blue nuclear and pink muscand tumor necrosis factor (TNF)-� (D) mRNAs, determined by real-time reverse tradinucleotide phosphate oxidase (NOX) (E), myleoperoxidase (MPO) (F), and xanthi##p � 0.01; #p � 0.05. Abbreviations as in Figure 1.

ate were decreased by 28%, 43%, 13%, 46%, 34%, and 38%, v

espectively, in chronically infected rats (all p values �0.01Mann-Whitney]). Notable elevation in end-systolic volumend end-diastolic volume occurred in chronically infectedearts with 174% and 44% increase, respectively, comparedith controls. No significant gain in cardiac function wasbserved in chronically infected/BZ-treated rats. In compari-on, a gain in cardiac function with PBN treatment was evidenty a significant elevation of �dp/dtmax, �dp/dtmax, stroke

ted Rats (� PBN/BZ)

plasm/keratin). Myocardial levels of interleukin (IL)-1� (B), interferon (IFN)-� (C),ase polymerase chain reaction. Specific activities of �-nicotinamide adeninease (XOD) (G) were measured by spectrophotometry. *** ###p � 0.001;

–Infec

le/cytonscriptne oxid

olume, cardiac output, ejection fraction, and heart rate in

c9W9Wvi�

tpPcwbPr

dc

D

IsepCed

oCp


hronically infected, PBN-treated rats (48%, 59%, 26%, 70%,3%, and 87%, respectively; all p values �0.01 [Kruskal-

allis]) and PBN/BZ-treated rats (52%, 53%, 100%, 96%,1%, and 90%, respectively; all p values �0.001 [Kruskal-

allis]). Subsequently, end-systolic volume and end-diastolicolume elevation noted in chronic chagasic hearts was normal-zed to control level by PBN (� BZ) treatment (all p values

0.01 [Kruskal-Wallis]).P-V loops with inferior vena cava occlusion showed that

he end-systolic P-V relationship (indicator of systolicerformance) was decreased by 48% and the end-diastolic-V relationship (indicator of diastolic stiffness) was in-reased by 68% in chronically infected rats, and no benefitas afforded by BZ treatment. PBN treatment (� BZ) waseneficial as evidenced by �10% decrease in end-systolic-V relationship and �15% increase in end-diastolic P-V

Figure 6 Myocardial Remodeling in Trypanosoma cruzi–Infected

(A) Masson’s trichrome staining of heart sections (blue collagen, red muscle/cytomyocardial mRNA levels for atrial natriuretic peptide (ANP) (B), brain-natriuretic-pe##p � 0.01. Abbreviations as in Figure 1.

elationship compared with normal controls. Thus, hemo- i

ynamic results showed that PBN markedly prevented thehagasic cardiac dysfunction.

iscussion

n this study, we demonstrate that decreasing the oxidativetress-induced alterations in myocardial structure, genexpression, and mitochondrial function is beneficial inreventing the chronic evolution of cardiac dysfunction inhagas disease. We also provide molecular and biochemical

vidence of the inefficacy of BZ during chronic Chagasisease.We used a rodent model of infection and disease devel-

pment that exhibited the classic hallmarks of chronichagas disease. Infected rats showed a high level of acutearasite burden (Fig. 1), resulting in extensive infiltration of

s (� PBN/BZ)

). Real-time reverse transcriptase polymerase chain reaction measurement ofBNP) (C), and �-skeletal actin (�sk-Actin) (D). *** ###p � 0.001; #p � 0.05;

Rat

plasmptide (

nflammatory infiltrate and expression of proinflammatory

cddde(pop(cwp

pgaircmsFfficopotinotrMcamti(ceBpItarLmhro

od(TPnnflIsbdcdarwsat

hctotf(aaoesfpt(hPhosoiPwssccSrsF


ytokines in the heart (Fig. 5). During the chronic stage,iffused inflammatory response persisted in the heart. Car-iac remodeling in chronically infected rats was evi-enced by an increased expression of hypertrophy mark-rs (ANP, BNP, and �sk actin) and collagen depositionFig. 6). Subsequently, ventricular performance was com-romised, as evidenced by 34% to 46% decrease in cardiacutput and ejection fraction and alterations in otherarameters proposed as a fair index of contractilityOnline Table 3) (24). The baseline hemodynamics ofhronically infected rats, obtained with the P-V loops,ere in accordance with the results from human chagasicatients (25).A systematic review of the literature indicates that the

athogenesis of Chagas disease is dependent on a low-rade, systemic infection with documented immune systemdverse reactions (discussed in Sack et al. [3]). Only acutelynfected patients, irrespective of their ages, are shown toespond to treatment with an antiparasite drug (BZ) and beured, defined by the control of acute parasitemia andyocarditis (17). No convincing evidence exists demon-

trating the efficacy of BZ in chronically infected patients.urther, it is not known whether drug therapies are inef-

ective in chronically infected patients because either BZails to kill tissue parasites or other host events persistndependent of parasites and result in the evolution ofhronic heart failure. To investigate this, we examined inur experimental model the effect of BZ treatment onarasites and host events. Our data showed that oral deliveryf BZ for 3 weeks during an indeterminate phase precludedhe tissue parasite persistence and associated inflammationn chronically infected hearts (Figs. 1 and 5). Activatedeutrophils and macrophages express NOX, inducible nitricxide synthase, and MPO and participate in parasite controlhrough a release of cytotoxic ROS, nitric oxide, and HOCl,espectively (26). Our observation of a decrease in NOX and

PO activities (Fig. 5, Online Fig. 2) in BZ-treatedhronically infected rats provided further evidence thatntiparasite therapy was effective in averting the parasite-ediated inflammatory pathology. Nonetheless, BZ-

reated/chronically infected rats exhibited cardiac remodel-ng (Fig. 6) and deterioration of ventricular contractilityOnline Table 3) comparable to that exhibited by untreated/hronically infected rats. These data provide molecularvidence that a lack of control of chronic Chagas disease byZ is not due to its inefficacy in controlling parasiteersistence and parasite-dependent inflammatory responses.mportantly, the observations in BZ-treated rats of consis-ent oxidative stress–induced myocardial adducts (Fig. 3)nd mitochondrial inefficiency (Fig. 4) strongly indicate aole of oxidative stress and mitochondrial/cellular injuries inV dysfunction during progressive Chagas disease. Theitochondrial production of ROS was significantly en-

anced in BZ-treated and untreated chronically infectedats (Fig. 4), thus, providing evidence of the mitochondrial
rigin of pathological oxidative stress in chagasic hearts. (
PBN treatment of infected rats provided direct evidencef the detrimental effects of oxidative stress and mitochon-rial dysfunction in Chagas disease. Oral delivery of PBN� BZ) resulted in preservation of LV function (Onlineable 3). It is important to note that beneficial effects ofBN in preserving cardiac function were observed despiteo decline in parasite persistence (Fig. 1) and moderate too change in myocardial inflammatory infiltrate and proin-ammatory cytokines in chronically infected hearts (Fig. 5).nstead, PBN-treated/chronically infected rats exhibited aubstantial increase in mitochondrial function as evidencedy improved complex activities and ATP synthesis andecreased ROS production (Fig. 4). The decrease in mito-hondrial ROS levels was associated with a significantecrease in the myocardial accumulation of MDA, HNE,nd carbonyls adducts in PBN-treated/chronically infectedats (Fig. 3). The maximal benefits were obtained when ratsere treated with PBN and BZ in combination; thus,

uggesting that combinatorial therapies, including antipar-sitic drugs and antioxidants, will be effective in preservinghe LV function in chagasic hearts.

Cardiac hypertrophy is a key phenotype of the failingeart and a major independent risk factor predictive ofardiovascular mortality and morbidity. In both experimen-al models and human chagasic patients, clinical evolutionf heart failure is preceded by anatomopathological changeshat include ventricular thickness and interstitial fibrosis,ollowed by ventricular dilation and reduced contractility27). The re-expression of fetal genes (ANP, BNP, �sk actin,nd �-MHC) is a hallmark of hypertrophic remodeling, andconsiderable body of evidence shows the redox regulationf remodeling responses in cardiac diseases of varioustiologies (28). Besides ROS, experimental studies havehown that the inflammatory cytokines, tumor necrosisactor-�, interleukin-1�, and monocyte chemoattractantrotein 1, also promote myocardial hypertrophy and con-ribute to the development and progression of heart failure29). Our observation of a decrease in the expression ofypertrophic markers and collagen deposition in response toBN treatment (� BZ) (Fig. 6) suggests that ROS signalsypertrophic remodeling in chagasic myocardium. The rolef ROS of mitochondrial, but not of inflammatory, origin inignaling hypertrophy in chagasic hearts is provided by thebservation that NOX and MPO, the classic mediators ofnflammatory ROS, were equally depressed in BZ- andBN-treated rats (Fig. 5), and yet hypertrophic phenotypeas depressed in PBN-treated rats only (Fig. 6). Further

tudies would identify whether inflammatory cytokinesynergistically enhance the ROS-mediated signaling cas-ades involved in activation of hypertrophic responses inhagasic hearts.tudy limitations. We acknowledge the limitations of theodent model system used in the current study, and thehort time-period during which animals were monitored.urther long-term studies are required in larger mammals

e.g. dogs, baboons) that more accurately mimic the human

spe

C

Wphibthrpcpipmmoicd

R3BE

R

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

Km

F


ymptoms and progression of Chagas disease to validate theotential benefits of adjunct therapies in preventing chronicvolution of Chagas disease.

onclusions

e have shown that complex inflammatory and oxidativerocesses in the heart contribute to the evolution of chronicypertrophy and LV dysfunction in Chagas disease. Specif-

cally, PBN treatment enhanced the antioxidant/oxidantalance, preserved the mitochondrial respiratory chain ac-ivity and ATP synthesis, and averted the ROS-inducedypertrophy and LV dysfunction in chronically infectedats. BZ treatment was effective in controlling parasiteersistence and associated inflammation, but the feedbackycle of mitochondrial dysfunction and oxidative stressersisted, and, therefore, LV function was not significantlymproved in BZ-treated/chronically infected rats. Thus, weropose that a combination of antioxidants capable ofodulating or delaying the onset of oxidative insult anditochondrial deficiencies and anti-parasite drugs capable

f abolishing parasites and parasite-associated inflammationn the myocardium would prove beneficial in preventingardiac pathology and loss of LV function in Chagasisease.

eprint requests and correspondence: Dr. Nisha Jain Garg,.142C Medical Research Building, University of Texas Medicalranch, 301 University Boulevard, Galveston, Texas 77555-1070.-mail: [email protected].

EFERENCES

1. World Health Organization. Report of Scientific Working Group onChagas Disease. Geneva: UNDP/World Bank/WHO, 2006:4.

2. Rassi A Jr., Rassi A, Little WC. Chagas heart disease. Clin Cardiol2000;23:883–9.

3. Zacks MA, Wen JJ, Vyatkina G, Bhatia V, Garg NJ. An overview ofchagasic cardiomyopathy: pathogenic importance of oxidative stress.An Acad Bras Cienc 2005;77:695–715.

4. Mortara RA, daSilva S, Patricio FR, et al. Imaging Trypanosoma cruziwithin tissues from chagasic patients using confocal microscopy.Parasitol Res 1999;85:800–8.

5. Salomone OA, Juri D, Omelianiuk MO, et al. Prevalence of circulat-ing Trypanosoma cruzi detected by polymerase chain reaction inpatients with Chagas’ cardiomyopathy. Am J Cardiol 2000;85:1274–6.

6. Sartori AM, Lopes MH, Caramelli B, et al. Simultaneous occurrenceof acute myocarditis and reactivated Chagas disease in an AIDSpatient. Clin Infect Dis 1995;21:1297–9.

7. Jardim E, Takayanagui OM. Chagasic meningoencephalitis withdetection of Trypanosoma cruzi in cerebrospinal fluid of an immunode-pressed patient. J Trop Med Hyg 1994;97:367–70.

8. Leiby DA, Rentas FJ, Nelson KE, et al. Evidence of Trypanosoma cruziinfection (Chagas disease) among patients undergoing cardiac surgery.Circulation 2000;102:2978–82.

9. Locksley RM, Klebanoff SJ. Oxygen-dependent microbicidal systemsof phagocytes and host defense against intracellular protozoa. J CellBiochem 1983;22:173–85.

0. Wen J-J, Garg NJ. Mitochondrial generation of ROS is enhanced at
Qo site of complex-III in myocardium of Trypanosoma cruzi-infected t
mice: beneficial effects of an antioxidant. J Bioenerg Biomembr2008;40:587–98.

1. Wen J-J, Vyatkina G, Garg NJ. Oxidative damage during chagasiccardiomyopathy: Role of mitochondrial oxidant release and inefficientantioxidant defense. Free Radic Biol Med 2004;37:1821–33.

2. Wen JJ, Dhiman M, Whorton EB, Garg NJ. Tissue-specific oxidativeimbalance and mitochondrial dysfunction during Trypanosoma cruziinfection in mice. Microbes Infect 2008;10:1201–9.

3. Wen J-J, Garg NJ. Oxidative modifications of mitochondrial respira-tory complexes in response to stress of Trypanosoma cruzi infection.Free Radic Biol Med 2004;37:2072–81.

4. deOliveira TB, Pedrosa RC, Filho DW. Oxidative stress in chroniccardiopathy associated with Chagas disease. Int J Cardiol 2007;116:357–63.

5. Wen J-J, Yachelini PC, Sembaj A, Manzur RE, Garg NJ. Increasedoxidative stress is correlated with mitochondrial dysfunction in cha-gasic patients. Free Radic Biol Med 2006;41:270–6.

6. Floyd RA, Hensley K, Forster MJ, Kelleher-Anderson JA, Wood PL.Nitrones as neuroprotectants and anti-aging drugs. Ann N Y Acad Sci2002;959:321–9.

7. Caldas IS, Talvani A, Caldas S, et al. Benznidazole therapy duringacute Chagas disease reduces parasite load but does not preventchronic cardiac lesions. Parasitol Res 2008;103:413–21.

8. Garg NJ, Bhatia V, Gerstner A, deFord J, Papaconstantinou J. Geneexpression analysis in mitochondria from chagasic mice: alterations inspecific metabolic pathways. Biochemical J 2004;381:743–52.

9. Wen J-J, Bhatia V, Popov VL, Garg NJ. Phenyl-alpha-tert-butylnitrone reverses mitochondrial decay in acute Chagas disease.Am J Pathol 2006;169:1953–64.

0. Ohkawa H, Ohishi N, Yagi K. Assay for lipid peroxides in animaltissues by thiobarbituric acid reaction. Analytical Biochem 1979;95:351–8.

1. Buss IH, Winterbourn CC. Protein carbonyl measurement by ELISA.Methods Mol Biol. 2002;186:123–8.

2. Garg NJ, Popov VL, Papaconstantinou J. Profiling gene transcriptionreveals a deficiency of mitochondrial oxidative phosphorylation inTrypanosoma cruzi-infected murine hearts. Biochim Biophys Acta2003;1638:106–20.

3. Nogueira FB, Ruiz JC, Robello C, Romanha AJ, Murta SM. Molec-ular characterization of cytosolic and mitochondrial tryparedoxinperoxidase in Trypanosoma cruzi populations susceptible and resistantto benznidazole. Parasitol Res 2009;104:835–44.

4. Zabalgoitia M, Ventura J, Lozano JL, et al. Myocardial contrastechocardiography in assessing microcirculation in baboons with Cha-gas disease. Microcirculation 2004;11:271–8.

5. Kawaguchi AT, Sugimachi M, Sunagawa K, Bergsland J, Koide S,Batista RJ. Improved LV contraction and energetics in a Chagasicpatient undergoing partial left ventriculectomy. J Card Surg 2001;16:30–3.

6. Forlenza M, Scharsack JP, Kachamakova NM, Taverne-Thiele AJ,Rombout JH, Wiegertjes GF. Differential contribution of neutrophilicgranulocytes and macrophages to nitrosative stress in a host-parasiteanimal model. Mol Immunol 2008;45:3178–89.

7. Marin-Neto JA, Cunha-Neto E, Maciel BC, Simoes MV. Pathogen-esis of chronic Chagas heart disease. Circulation 2007;115:1109–23.

8. Liaudet L, Vassalli G, Pacher P. Role of peroxynitrite in redoxregulation of cell signal transduction pathways. Front Biosci 2009;14:4809–14.

9. Gullestad L, Aukrust P. Review of trials in chronic heart failureshowing broad-spectrum anti-inflammatory approaches. Am J Cardiol2005;95:17C–23C.

ey Words: antioxidant y benzonidazole y Chagas disease yitochondrial oxidative stress y Trypanosoma cruzi.

APPENDIX

or details of the methods and supplemental figures and tables, please see
he online version of this article.
mailto:[email protected]

Documents

Phenyl-\u0001-tert-butyl-nitrone and Benzonidazole Treatment Controlled the Mitochondrial Oxidative Stress and Evolution of Cardiomyopathy in Chronic Chagasic Rats