Available online at www.sciencedirect.com

Journal of Molecular and Cellular Cardiology 44 (2008) 1016–1022www.elsevier.com/locate/yjmcc

Original article

Acrolein consumption exacerbates myocardial ischemic injury and blocksnitric oxide-induced PKCε signaling and cardioprotection

Guang-Wu Wang a,b, Yiru Guo b, Thomas M. Vondriska a, Jun Zhang a, Su Zhang b, Linda L. Tsai a,Nobel C. Zong a, Roberto Bolli b, Aruni Bhatnagar b, Sumanth D. Prabhu b,c,⁎

a Department of Physiology and the Cardiovascular Research Laboratories, David Geffen School of Medicine at UCLA, Los Angeles, CA, USAb Institute of Molecular Cardiology, Department of Medicine/Cardiology, University of Louisville, Louisville, KY, USA

c Medical Service, Louisville VA Medical Center, Louisville, KY, USA

Received 7 December 2007; received in revised form 21 March 2008; accepted 22 March 2008Available online 8 April 2008

Abstract

Aldehydes are common reactive constituents of food, water and air. Several food aldehydes are potentially carcinogenic and toxic; however,the direct effects of dietary aldehydes on cardiac ischemia-reperfusion (IR) injury are unknown. We tested the hypothesis that dietary consumptionof aldehydes modulates myocardial IR injury and preconditioning. Mice were gavage-fed the α, β-unsaturated aldehyde acrolein (5mg/kg) orwater (vehicle) 24h prior to a 30-min coronary artery occlusion and 24-hour reperfusion. Myocardial infarct size was significantly increased inacrolein-treated mice, demonstrating that acute acrolein exposure worsens cardiac IR injury. Furthermore, late cardioprotection afforded by thenitric oxide (NO) donor diethylenetriamine/NO (DETA/NO; dose: 0.1mg/kg × 4, i.v.) was abrogated by the administration of acrolein 2h prior toDETA/NO treatment, indicating that oral acrolein impairs NO donor-induced late preconditioning. To examine potential intracellular targets ofaldehydes, we investigated the impact of acrolein on mitochondrial PKCε signaling in the heart. Acrolein-protein adducts were formed in a dose-dependent manner in isolated cardiac mitochondria in vitro and specific acrolein-PKCε adducts were present in cardiac mitochondrial fractionsfollowing acrolein exposure in vivo, demonstrating that mitochondria are major targets of aldehyde toxicity. Furthermore, DETA/NOpreconditioning induced both PKCε translocation and increased mitochondrial PKCε localization. Both of these responses were blocked byacrolein pretreatment, providing evidence that aldehydes disrupt cardioprotective signaling events involving PKCε. Consumption of an aldehyde-rich diet could exacerbate cardiac IR injury and block NO donor-induced cardioprotection via mechanisms that disrupt PKCε signaling.Published by Elsevier Inc.

Keywords: Acrolein; Diethylenetriamine/NO; Myocardial infarction; PKC epsilon

1. Introduction

The extent of ischemic injury suffered by the heart dependsupon several factors including the duration of ischemia and thetime of reperfusion. In addition, biochemical responses intrinsicto the myocardium modulate the outcome of ischemic insults.Extensive experimental and clinical work has shown that ische-mic preconditioning or pharmacological preconditioning withagents such as adenosine or nitroglycerin decreases myocardial

⁎ Corresponding author. Medicine/Cardiology, University of Louisville, ACB,3rd Floor, 550 South Jackson Street, Louisville, KY 40202, USA. Tel.: +1 502852 7959; fax: +1 502 852 7147.

E-mail address: sprabhu@louisville.edu (S.D. Prabhu).

0022-2828/$ - see front matter. Published by Elsevier Inc.doi:10.1016/j.yjmcc.2008.03.020

ischemia-reperfusion (IR) injury [1,2]. Triggers of precondi-tioning activate complex signaling pathways that ultimatelystrengthen the resistance of the heart to ischemia. The inherentvulnerability of the heart to IR injury and its ability to mount apreconditioning response are, however, susceptible to environ-mental influences. Myocardial resistance to ischemic injury, forinstance, is enhanced by physical exercise [2,3], caloricrestriction [4], and alcohol consumption [5] and preconditioningmechanisms are disrupted by aging [6], caffeine [7] or treatmentwith drugs such as cyclooxygenase-2 inhibitors [8]. However,whether reactive constituents of diet affect myocardial sensi-tivity to ischemia or preconditioning is unknown.

Aldehydes are highly reactive components of food and wa-ter. More than 300 different aldehydes have been identified in

1017G.-W. Wang et al. / Journal of Molecular and Cellular Cardiology 44 (2008) 1016–1022

various foods [9] (see Online Supplement, Table 1). Additionalaldehydes are generated during frying and cooking. Indeed,because they are natural products of lipid peroxidation andglucose oxidation, aldehydes are generated in high concentra-tions by heating and cooking fats, oils, and sugars [9]. At least36 different aldehydes are also present in water [10], and withthe exception of metals, aldehydes are considered to be themajor pollutants in drinking water [10] (Online Supplement,Table 1). By themselves, aldehydes are highly reactive poten-tial carcinogens [9,11]. They form covalent adducts with DNA[12], proteins [13], and lipids [14] and induce cytotoxicity [15].Aldehydes have also been shown to depress myofilamentsensitivity and cardiac contraction [16], inhibit mitochondrialrespiration [17], and alter ion-channel conductance pathways[18]. Nevertheless, the cardiovascular toxicity of ingestedaldehydes has not been directly studied and their effects onmyocardial responses to ischemia are unknown.

The present studywas, therefore, designed to examinewhetherdietary consumption of aldehydes affects myocardial ischemia-reperfusion injury and preconditioning. Because toxicologicalstudies of complex aldehyde mixtures in food are difficult tointerpret, we used acrolein as the prototypical dietary aldehyde.High levels of acrolein have been detected in several foodsincluding cheese, donuts, fish, bread, potatoes, and alcoholicbeverages [9,19] (Online Supplement, Table 1). The concentra-tion of acrolein is particularly high in cigarette smoke and heatedoils [19]. Our study shows that at concentrations comparable tohuman consumption, acrolein worsens infarct size following IRinjury and blocks nitric oxide-induced cardioprotection in mice

Fig. 1. Experimental protocol for studies of myocardial ischemia-reperfusion injury anof coronary occlusion followed by 24 h of reperfusion. In Group 1, 24 h before occluacrolein, the mice were subjected to coronary occlusion and reperfusion as indicated.every 25min (total dose, 0.4 mg/kg) or PBS (vehicle). Mice in Group III received the sbefore DETA/NO or PBS treatment. In both Groups II and III, 24 h after DETA/NO ofollowed by 24-hour reperfusion, or their hearts were excised for PKC studies as ind

via a mechanism that appears to involve disruption of proteinkinaseCε (PKCε) signal transduction. These findings suggest thatfoods rich in aldehydes could increase myocardial susceptibilityto IR injury and abolish cardioprotective signaling.

2. Methods

The experimental protocols described herein were per-formed in accordance with the National Institutes of HealthGuide for the Care and Use of Laboratory Animals (PublicationNo. 86-23).

2.1. Myocardial ischemia/reperfusion surgery and infarct sizeanalysis

ICR mice were subjected to myocardial ischemia andreperfusion as previously described [20]. Individual protocolsare illustrated in Fig. 1. In Group I, mice were fed either acrolein(5mg/kg) or water by gavage, and 24h later were anesthetizedwith pentobarbital sodium (50mg/kg body wt i.p.), intubated,and ventilated with 100% oxygen. The chest was opened by aleft thoracotomy between ribs three and four and a 8-0 silksuture was placed under the left anterior descending coronaryartery 1 to 3mm from the tip of the left atrial appendage.Ischemia was induced by ligation of the suture (a 1 to 2mmsection of PE-10 tubing was placed between the suture and theartery to prevent damage to the vessel). Following a 30-minocclusion, the suture was removed, the chest wall closed, andthe heart was allowed to reperfuse for 24h. The heart was then

d nitric oxide-mediated cardioprotection. Three groups of mice underwent 30minsion (Day 0) the mice were fed acrolein (5 mg/kg, p.o.). Twenty four hours afterIn Group II, the mice received 4 intravenous boluses of DETA/NO (0.1 mg/kg)ame dose of DETA/NO or PBS except that they were also fed 5mg/kg acrolein 2 hr acrolein treatment, the mice were either subjected to 30-min coronary occlusionicated.

1018 G.-W. Wang et al. / Journal of Molecular and Cellular Cardiology 44 (2008) 1016–1022

excised and perfused postmortem as described [20]. Theinfarcted region was delineated by perfusion with a 1% solutionof 2,3,5-triphenyltetrazolium chloride (TTC) in phosphatebuffer (pH 7.4, 37°C). Non-infarcted tissue would take upTTC (red), whereas infarcted regions would be TTC-free(white). To delineate the occluded-reperfused coronary vascularbed, the coronary artery was tied at the site of the previousocclusion and the aortic root perfused with a 5% solution ofphthalo blue dye. Hence, the nonischemic region (region not atrisk) would be stained blue whereas the risk region would notstain blue. Infarct size (white area) was measured by video-planimetry with NIH Image software and expressed as a per-centage of the region at risk (red plus white area). The region atrisk was expressed as a percentage of the total left ventricle(LV).

2.2. Acrolein administration

The dose-range of human aldehyde exposure estimated frommaximal daily consumption was 7mg/kg aldehyde/day (OnlineSupplement, Table 2). The estimate of acrolein consumptionfrom just 8 different types of foods was 0.1mg/kg/day (OnlineSupplement, Table 3). The approximate daily consumption ofunsaturated aldehydes (such as acrolein) was estimated to be5mg/kg, whereas that of saturated aldehydes (such as form-aldehyde and acetaldehyde) was 2mg/kg. Based on theseestimates a 5mg/kg dose of acrolein, representing the expectedunsaturated aldehyde intake, was chosen for this study. Eightweek old male mice (ICR) were gavage-fed acrolein (in 200μLwater) or the same volume of water (vehicle). Free acrolein wasprepared daily by acid hydrolysis (pH 3.0) of diethyl acetalacrolein (Sigma) in 0.1N HCl for 1h at room temperature (RT).Mice were housed in a pathogen-free room at 24°C, 55–65%

Fig. 2. Acute acrolein administration increases infarct size and blocks cardioprotectionmice were gavage-fed 5 mg/kg acrolein 24 h prior to a 30-min left coronary artery occvehicle). B, Cardioprotection by DETA/NO is blocked by acrolein. Preconditioning w(⁎pb0.05 vs. vehicle). However, this protection is blocked by administration of acr

relative humidity, and with a 12:12-hour light:dark cycle. Ani-mals had free access to food and water.

2.3. DETA/NO-induced cardioprotection

This protocol has been previously shown to induce a latephase of pharmacological preconditioning in rabbits [21]and mice [22]. Mice were given four consecutive intravenousbolus doses of either the NO donor diethylenetriamine/NO(DETA/NO, 0.1mg/kg every 25min for a total dose 0.4mg/kg)or PBS (vehicle) on Day 0 (Group II; Fig. 1). In an additionalgroup (Group III), the mice were fed acrolein 2h before DETA/NO or PBS treatment.

2.4. Subcellular fractionation, immunoblotting, andimmunoprecipitation

Western immunoblotting analysis of PKCεwas performed asdescribed [21]. Briefly, 24h after acrolein administration andDETA/NO preconditioning, mouse hearts were homogenized toobtain total myocardial lysates. Cytosolic and particulate ormitochondrial fractions were obtained by differential centrifu-gation as previously described [22,23]. Cytosolic contaminationof the mitochondrial fraction was less than 0.5% as measured bylactate dehydrogenase activity. Proteins were separated on a10% SDS-PAGE gel and immunoblotted using anti-PKCε mo-noclonal antibodies (1:1000; BD Transduction Labs).

2.5. Assay for formation of acrolein-protein adducts in cardiacmitochondria

Mitochondria were isolated from adult mouse hearts by en-zymatic digestion, homogenization, and differential centrifugation

. A, Acrolein exposure increases infarct size in the naive myocardium. Male ICRlusion and 24-hour reperfusion and the infarct size was determined (⁎pb0.05 vs.ith the NO donor DETA/NO (0.1 mg/kg×4 i.v.) significantly reduces infarct sizeolein 2 h prior to DETA/NO treatment (#pb0.05 vs. DETA/NO).

Fig. 3. A, Translocation of PKCε is blocked by acrolein. Cytosolic andparticulate distribution of PKCε was determined by Western immunoblotting(IB) in all groups. When given prior to vehicle, acrolein had no effect on PKCεdistribution at 24 h, but when administered prior to DETA/NO, acrolein sig-nificantly attenuated DETA/NO-induced PKCε translocation (⁎pb0.001 vs.vehicle; #pb0.001 vs. DETA/NO; n=3–6 per group).

1019G.-W. Wang et al. / Journal of Molecular and Cellular Cardiology 44 (2008) 1016–1022

as previously described [22,23]. Isolated mitochondria (50μgprotein) were incubated with different concentrations of acrolein(0, 1, 3, 10, 30, 100, 300, and 600μmol/L) for 30min at 37°C, andacrolein-protein adducts were detected by immunoblotting(1:5000; anti-acrolein-lysine monoclonal antibody was the ge-nerous gift of Dr. Koji Uchida).

To detect the formation of acrolein-PKCε adducts, im-munoprecipitation was carried out as described [24]. Brief-ly, 500μgmitochondrial protein was incubatedwith 1μg of PKCεantibody plus 20μL Protein A/G-Agarose (Santa Cruz) overnightat 4°C. After washing three times with lysis buffer, the sampleswere subjected to SDS-PAGE and immunoblotting (1:2000; anti-acrolein-lysine monoclonal antibody).

2.6. Statistics

Data are reported as mean ± standard error measurement(SEM). Infarct sizes and protein expression were analyzed witha one-way ANOVA followed by Student's t-tests for unpaireddata.

3. Results

3.1. Acrolein exacerbates myocardial IR injury

Mice gavage-fed 5mg/kg acrolein displayed no overt sign oftoxicity. There were no changes in liver enzymes, urinecomposition, blood electrolytes, or echocardiographicallymeasured cardiac function (data not shown). Indeed micecould be maintained on this dose of acrolein for up to one weekwith no mortality. To test the hypothesis that aldehydeconsumption is detrimental to the heart, myocardial infarctsizes were determined in mice subjected to a 30-min coronaryartery occlusion and 24-hour reperfusion following acroleinadministration (Fig. 2A). The data show that acute acroleinexposure significantly increased infarct size as compared tovehicle-treated mice (51.6 ± 1.4% vs. 37.7 ± 6.2%, p b 0.05).The risk region as a percentage of total LV was equivalentbetween the groups (vehicle 49.6 ± 12.8%; acrolein 47.1 ±9.5%, p = NS).

3.2. Acrolein blocks NO donor-induced cardioprotection

We next examined whether acrolein impairs pharmacolo-gical preconditioning using a model of NO donor-inducedcardioprotection (Fig. 2B). Mice administered the NO donorDETA/NO 24h prior to a 30-min coronary artery occlusionand 24-hour reperfusion exhibited significantly reducedinfarct size compared to control mice (22.2 ± 3.1% vs.41.4 ± 8.2%; p b 0.05), demonstrating a late preconditioningeffect. In contrast, administration of acrolein 2h prior toDETA/NO effectively abolished the cardioprotection affordedby the NO donor (infarct size 44.8 ± 6.2%; p b 0.05 vs.DETA/NO alone). The risk region as a percentage of totalLV was comparable between groups (vehicle 40.3 ± 3.1%;DETA/NO 43.2 ± 2.5%; acrolein + DETA/NO 36.7 ± 4.5%,p = NS).

3.3. Acrolein abrogates PKCε translocation and blocks NOdonor-induced increase in mitochondrial PKCε expression

To determine the mechanism by which aldehydes abolishNO-dependent cardioprotection, translocation of PKCε wasexamined in all groups of mice (Fig. 3A). Analogous to ob-servations in rabbits [21,25], DETA/NO induced a significanttranslocation of PKCε 24h after its administration (particu-late fraction increased from 23.4 ± 5.7% to 36.7 ± 3.7%;measurements given as percentage of total PKCε expression,p b 0.001 vs. vehicle). Translocation, however, was blocked bytreatment with acrolein 2h prior to DETA/NO administration(particulate fraction 15.3 ± 2.0%, p b 0.001 vs. DETA/NO).Next, the expression of PKCε was examined in mitochondrialfractions. As seen in Fig. 4A, DETA/NO increased mitochon-drial localization of PKCε (241.6±3.1% of vehicle; pb0.05 vs.vehicle). Analogous to PKCε translocation, the increase inmitochondrial expression of PKCεwas also blocked if acroleinwas administered prior to DETA/NO (28.8±7.4% of vehicle;pb0.05 vs. DETA/NO and vs. vehicle). Fig. 4B demonstratesthe purity of isolated mitochondrial fractions, as assessedby sarcolemmal (Na+/K+ ATPase), mitochondrial (cytochromeC oxidase IV), and cytosolic (aldose reductase) markers.Cytosolic contamination of the mitochondrial fraction wasnegligible. Sarcolemmal contamination was less than 5%.

Fig. 4. Acrolein toxicity targets the cardiac mitochondria. Cardiac mitochondria were isolated and Western immunoblotted (IB) for PKCε expression. A, Acroleintreatment results in decreased mitochondrial PKCε expression. DETA/NO-induced augmentation of mitochondrial PKCε expression at 24 h was blocked by theadministration of acrolein prior to DETA/NO treatment. B, Purity of isolated mitochondrial fractions. Total homogenate and cytosolic and mitochondrial fractions weresubjected to Western blotting for sarcolemmal (Na+/K+ ATPase) mitochondrial (cytochrome C oxidase IV), and cytosolic (aldose reductase) markers. Cytosoliccontamination of the mitochondrial fraction was negligible. Sarcolemmal contamination of the mitochondrial fraction was less than 5%. C, Formation of acrolein-protein adducts in isolated cardiac mitochondria. Isolated mitochondria were incubated with different concentrations of acrolein (0, 1, 3, 10, 30, 300, and 600 μmol/L).A dose-dependent formation of acrolein-protein adducts was observed. D, Formation of acrolein-PKCε adducts in cardiac mitochondria. Cardiac mitochondria wereisolated at the end of the in vivo experiments and immunoprecipitation was performed. Acrolein-PKCε adducts were observed in the mitochondria following acroleinadministration. Furthermore, the adducts formed even in the presence of DETA/NO after acrolein treatment, indicating that DETA/NO had no effect on their formation.

1020 G.-W. Wang et al. / Journal of Molecular and Cellular Cardiology 44 (2008) 1016–1022

3.4. Acrolein-protein adducts form in cardiac mitochondria

Lastly, the ability of acrolein to directly modify cardiac pro-teins was examined in cardiac mitochondria. In vitro, aldehyde-protein adducts formed in isolated cardiac mitochondria in adose-dependent fashion (Fig. 4C) while specific acrolein-PKCεadducts were detected in cardiac mitochondrial fractions afteracrolein exposure in vivo (Fig. 4D), suggesting that the mito-chondria are highly susceptible to aldehyde toxicity.

4. Discussion

The major findings of this study are that oral acroleinconsumption increases myocardial IR injury and abolishes thecardioprotective effects of nitric oxide. These novel results haveseveral important implications. First, the consumption ofacrolein and related aldehydes in aldehyde-rich foods couldsignificantly affect the myocardial sensitivity to ischemia. Thesignaling mechanisms by which these changes occur appear toinvolve attenuated PKCε translocation and decreased mito-chondrial PKCε expression, as well as direct aldehyde-proteinadduct formation in cardiac mitochondria. Second, the dele-terious effects of aldehydes also include the abrogation of NO-dependent pharmacological cardioprotection, suggesting that

aldehyde consumption can interfere with the protective effectsof nitrate drug therapy for coronary artery disease. Third, thesefindings support the notion that acrolein may act at PKCε-dependent foci within cardioprotective signaling networks toblock the development of the infarct-sparing phenotype.

We interpreted our results to signify that acrolein both exa-cerbates infarct size following ischemia and reperfusion andblocks NO donor-induced preconditioning and cardioprotec-tion (Fig. 2). An alternate explanation could be that of an offsetphenomenon, i.e. that DETA/NO attenuated acrolein-inducedinjury. We feel that this was much less likely given for tworeasons. First, there was no statistically significant difference ininfarct size between the group treated with acrolein alone vs.the group treated with acrolein prior to DETA/NO. Second,acrolein administration also suppressed the translocation andmitochondrial localization of PKCε induced by DETA/NO. Asthere is strong evidence that subcellular redistribution of PKCεis a critical event essential for the production of NO donor-induced pharmacological preconditioning [21,26–28], thisobservation indicates that acrolein directly disrupted NO-dependent preconditioning pathways (and argues against anoffset phenomenon). Indeed, as PKCε has a well-establishedrole in cardioprotection [21,24,29], translocation of this iso-form in response to the same dose of DETA/NO that induces

1021G.-W. Wang et al. / Journal of Molecular and Cellular Cardiology 44 (2008) 1016–1022

cardioprotection and PKCε translocation in the rabbit [21] wasnot unexpected. In addition, the finding that translocation ofPKCε was blocked by the same dose of acrolein that preventedcardioprotection further corroborates the critical mechanisticlink between NO, PKCε, and cardioprotection. The data hereinsupport the idea that PKCε plays a necessary role in protectingthe murine heart against infarction, since reduced translocationin response to NO following acrolein treatment was accom-panied by loss of the cardioprotective phenotype.

Our data indicate that NO-induced cardioprotectionincreases localization of PKCε to the mitochondria. Given themultitude of cardioprotective processes known to depend on themitochondria, this finding has significance regarding signalingtransduction by PKCε at this organelle. Previous studies haveshown that PKCε physically interacts with components of themitochondrial permeability pore to inhibit mitochondrialpermeability transition [23]. The observation that cardioprotec-tion provided by DETA/NO was associated with mitochondrialPKCε translocation further supports the view that mitochondrialassociation of PKCε may be an essential mechanism ofpreconditioning. Intriguingly, treatment with acrolein beforeDETA/NO administration was sufficient to block the increase inPKCε in the mitochondria. Finally, the finding that acroleinforms direct adducts with proteins including PKCε in cardiacmitochondria, taken together with previous data demonstratingthat aldehydes interfere with mitochondrial function [15,17,30],supports the notion that this organelle is a target for aldehydetoxicity in the heart.

The observation that aldehyde consumption exacerbatesmyocardial ischemia/reperfusion injury and blocks cardiopro-tection highlights the vulnerability of the ischemic heart toenvironmental influences. Exposure to environmental pollutantsis an emerging risk factor for cardiovascular disease [31] asseveral studies suggest that acute exposure to polluted airincreases myocardial sensitivity to ischemia and arrhythmias[32]. Indeed, in a recent analysis, ischemic heart disease was thelargest specific cause of death associated with pollutant ex-posure, which accounted for a one-quarter of such death [33].Statistically significant associations were also observed forarrhythmias, heart failure and cardiac arrest, but surprisingly nopositive correlations were observed with respiratory diseases,indicating that the diseased heart is highly sensitive to en-vironmental pollutants. Hence, the present data, demonstrating adirect link between a specific ubiquitous dietary component andmyocardial sensitivity to ischemic injury, further underscore theunique vulnerability of the heart to xenobiotic toxicity andsuggest that reactive xenobiotics in the diet may be heretoforeunrecognized modulators of myocardial ischemia-reperfusioninjury.

Even though our data support a strong link between al-dehyde consumption and sensitivity to ischemia in mice, thehuman susceptibility to food aldehydes remains unknown. Thedose-range of acrolein used in this study was similar to theexpected human consumption; however, aldehyde toxicity isalso a function of aldehyde metabolism. Aldehydes such asacrolein are extensively metabolized by cardiovascular tissues,kidney and liver [15,19,34,35], and their toxicity depends upon

the total aldehyde detoxification capacity which itself maybe under the control of environmental factors, such as diseaseor age. Even within experimental animals sensitivity to acro-lein varies. The LD50 of acrolein in mice is 40 mg/kg, whereasin rabbits it is 7 mg/kg [9]. Whether humans are more or lesssensitive to acrolein is not known, but additional work isclearly warranted to ascertain human sensitivity and to deter-mine whether the aldehyde content of food (analogous tothe cholesterol or fat content) is a contributing factor to theseverity of heart disease. In this regard it is interesting to pointout that humans display wide variations in the levels ofglutathione-S-transferase (GST) [36], an enzyme involved inaldehyde detoxification [15,19]. GST-deficient and null-genotypes in human populations are associated with increasedrisk for cancer [36] and dietary regulation of DNA damage[37]. It is possible that similar associations may exist forischemic heart disease.

Acrolein and related aldehydes are not only common foodconstituents but also ubiquitous pollutants present in highamounts in coal, wood, cotton, cigarette smoke, automobileexhaust, and industrial waste [9,19]. Hence, even though itremains unclear whether the effects of inhaled and ingestedaldehydes are similar, our findings raise the possibility thataldehydes present in ambient air might be significant mediatorsof the harmful cardiovascular effects of these pollutants. This isconsistent with the high cardiovascular toxicity associatedwith the aldehyde-containing components of polluted air orcigarette smoke [32]. Further experimental, clinical and epide-miological studies are needed to address these issues.

In summary, we provide the first line of evidence thatreactive aldehydes in diet can influence myocardial sensitivityto ischemia and block NO-mediated cardioprotection related, atleast in part, to the disruption of the activation and mito-chondrial localization of PKCε. The vulnerability of cardiopro-tective signaling to dietary aldehydes points towards a complexinterplay between environment and genetic susceptibility, muchof which remains currently unknown.

Acknowledgments

This study was supported in part by AHA Grant 0465137Y(GWW),NIHGrantsHL-43151 (RB),HL-55757 (RB),HL-78825(RB, AB, and SDP), ES-12062 (AB), ES-11860 (AB and SDP),VA Merit Grant (SDP), and the Laubisch Endowment at UCLA.

Appendix A. Supplementary data

Supplementary data associated with this article can be found,in the online version, at doi:10.1016/j.yjmcc.2008.03.020.

References

[1] Bolli R. The late phase of preconditioning. Circ Res 2000;87:972–83.[2] Yellon DM, Downey JM. Preconditioning the myocardium: from cellular

physiology to clinical cardiology. Physiol Rev 2003;83:1113–51.[3] Yamashita N, Hoshida S, Otsu K, Asahi M, Kuzuya T, Hori M. Exercise

provides direct biphasic cardioprotection via manganese superoxidedismutase activation. J Exp Med 1999;189:1699–706.

1022 G.-W. Wang et al. / Journal of Molecular and Cellular Cardiology 44 (2008) 1016–1022

[4] Shinmura K, Tamaki K, Bolli R. Short-term caloric restriction improvesischemic tolerance independent of opening of ATP-sensitive K+ channelsin both young and aged hearts. J Mol Cell Cardiol 2005;39:285–96.

[5] Miyamae M, Diamond I, Weiner MW, Camacho SA, Figueredo VM.Regular alcohol consumption mimics cardiac preconditioning by protect-ing against ischemia-reperfusion injury. Proc Natl Acad Sci U S A1997;94:3235–9.

[6] Lee TM, Su SF, Chou TF, Lee YT, Tsai CH. Loss of preconditioning byattenuated activation of myocardial ATP-sensitive potassium channels inelderly patients undergoing coronary angioplasty. Circulation 2002;105:334–40.

[7] Riksen NP, Zhou Z, OyenWJ, Jaspers R, Ramakers BP, Brouwer RM, et al.Caffeine prevents protection in two human models of ischemic precondi-tioning. J Am Coll Cardiol 2006;48:700–7.

[8] Shinmura K, Tang XL, Wang Y, Xuan YT, Liu SQ, Takano H, et al.Cyclooxygenase-2 mediates the cardioprotective effects of the late phase ofischemic preconditioning in conscious rabbits. Proc Natl Acad Sci U S A2000;97:10197–202.

[9] Feron VJ, Til HP, de VF, Woutersen RA, Cassee FR, van Bladeren PJ.Aldehydes: occurrence, carcinogenic potential, mechanism of action andrisk assessment. Mutat Res 1991;259:363–85.

[10] Committee on Aldehydes. Formaldehyde and other aldehydes. Board ontoxicology and environmental health hazards, Assembly of Life Sciences.Washington, D.C.: National Academy Press; 1981. p. 234–41.

[11] Janzowski C, Glaab V, Mueller C, Straesser U, Kamp HG, Eisenbrand G.alpha,beta-Unsaturated carbonyl compounds: induction of oxidative DNAdamage in mammalian cells. Mutagenesis 2003;18:465–70.

[12] Kurtz AJ, Lloyd RS. 1,N2-deoxyguanosine adducts of acrolein, croto-naldehyde, and trans-4-hydroxynonenal cross-link to peptides via Schiffbase linkage. J Biol Chem 2003;278:5970–6.

[13] Uchida K, Kanematsu M, Sakai K, Matsuda T, Hattori N, Mizuno Y, et al.Protein-bound acrolein: potential markers for oxidative stress. Proc NatlAcad Sci U S A 1998;95:4882–7.

[14] Guichardant M, Taibi-Tronche P, Fay LB, Lagarde M. Covalent modi-fications of aminophospholipids by 4-hydroxynonenal. Free Radic BiolMed 1998;25:1049–56.

[15] Esterbauer H, Schaur RJ, Zollner H. Chemistry and biochemistry of4-hydroxynonenal, malonaldehyde and related aldehydes. Free RadicBiol Med 1991;11:81–128.

[16] Luo J, Hill BG, Gu Y, Cai J, Srivastava S, Bhatnagar A, et al. Mechanismsof acrolein-induced myocardial dysfunction: implications for environ-mental and endogenous aldehyde exposure. Am J Physiol Heart CircPhysiol 2007;293:H3673–84.

[17] Humphries KM, Szweda LI. Selective inactivation of alpha-ketoglutaratedehydrogenase and pyruvate dehydrogenase: reaction of lipoic acid with4-hydroxy-2-nonenal. Biochemistry 1998;37:15835–41.

[18] Bhatnagar A. Electrophysiological effects of 4-hydroxynonenal, an alde-hydic product of lipid peroxidation, on isolated rat ventricular myocytes.Circ Res 1995;76:293–304.

[19] Ghilarducci DP, Tjeerdema RS. Fate and effects of acrolein. Rev EnvironContam Toxicol 1995;144:95–146.

[20] Guo Y, Wu WJ, Qiu Y, Tang XL, Yang Z, Bolli R. Demonstration of anearly and a late phase of ischemic preconditioning in mice. Am J Physiol1998;275:H1375–87.

[21] Ping P, Takano H, Zhang J, Tang XL, Qiu Y, Li RC, et al. Isoform-selectiveactivation of protein kinase C by nitric oxide in the heart of conscious

rabbits: a signaling mechanism for both nitric oxide-induced and ischemia-induced preconditioning. Circ Res 1999;84:587–604.

[22] Wang G, Liem DA, Vondriska TM, Honda HM, Korge P, Pantaleon DM,et al. Nitric oxide donors protect murine myocardium against infarctionvia modulation of mitochondrial permeability transition. Am J PhysiolHeart Circ Physiol 2005;288:H1290–5.

[23] Baines CP, Song CX, Zheng YT, Wang GW, Zhang J, Wang OL, et al.Protein kinase Ce interacts with and inhibits the permeability transitionpore in cardiac mitochondria. Circ Res 2003;92:873–80.

[24] Ping P, Song C, Zhang J, Guo Y, Cao X, Li RC, et al. Formation of proteinkinase Ce-Lck signaling modules confers cardioprotection. J Clin Invest2002;109:499–507.

[25] Vondriska TM, Zhang J, Song C, Tang XL, Cao X, Baines CP, et al. Proteinkinase C e-Src modules direct signal transduction in nitric oxide-inducedcardioprotection: complex formation as a means for cardioprotectivesignaling. Circ Res 2001;88:1306–13.

[26] Banerjee S, Tang XL, Qiu Y, Takano H, Manchikalapudi S, Dawn B, et al.Nitroglycerin induces late preconditioning against myocardial stunning viaa PKC-dependent pathway. Am J Physiol 1999;277:H2488–94.

[27] Balafanova Z, Bolli R, Zhang J, Zheng Y, Pass JM, Bhatnagar A, et al.Nitric oxide (NO) induces nitration of protein kinase Ce (PKCe),facilitating PKCe translocation via enhanced PKCe-RACK2 interactions:a novel mechanism of NO-triggered activation of PKCe. J Biol Chem2002;277:15021–7.

[28] Xuan YT, Guo Y, Zhu Y, Wang OL, Rokosh G, Bolli R. Endothelial nitricoxide synthase plays an obligatory role in the late phase of ischemicpreconditioning by activating the protein kinase Ce-p44/42 mitogen-activated protein kinase-pSer-signal transducers and activators of tran-scription1/3 pathway. Circulation 2007;116:535–44.

[29] LiuGS, CohenMV,Mochly-RosenD, Downey JM. Protein kinase C-epsilonis responsible for the protection of preconditioning in rabbit cardiomyocytes.J Mol Cell Cardiol 1999;31:1937–48.

[30] Luo J, Shi R. Acrolein induces oxidative stress in brain mitochondria.Neurochem Int 2005;46:243–52.

[31] Brook RD, Franklin B, Cascio W, Hong Y, Howard G, Lipsett M, et al. Airpollution and cardiovascular disease: a statement for healthcare profes-sionals from the Expert Panel on Population and Prevention Science of theAmerican Heart Association. Circulation 2004;109:2655–71.

[32] Bhatnagar A. Environmental cardiology: studying mechanistic linksbetween pollution and heart disease. Circ Res 2006;99:692–705.

[33] Pope III CA, Burnett RT, Thurston GD, ThunMJ, Calle EE, Krewski D, et al.Cardiovascular mortality and long-term exposure to particulate air pollution:epidemiological evidence of general pathophysiological pathways of disease.Circulation 2004;109:71–7.

[34] Ruef J, Liu SQ, Bode C, Tocchi M, Srivastava S, Runge MS, et al.Involvement of aldose reductase in vascular smooth muscle cell growthand lesion formation after arterial injury. Arterioscler Thromb Vasc Biol2000;20:1745–52.

[35] Srivastava S, Chandra A, Wang LF, Seifert Jr WE, DaGue BB, Ansari NH,et al. Metabolism of the lipid peroxidation product, 4-hydroxy-trans-2-nonenal, in isolated perfused rat heart. J Biol Chem 1998;273:10893–900.

[36] McIlwain CC, Townsend DM, Tew KD. Glutathione S-transferasepolymorphisms: cancer incidence and therapy. Oncogene 2006;25:1639–48.

[37] Reszka E, Wasowicz W, Gromadzinska J. Genetic polymorphism ofxenobiotic metabolising enzymes, diet and cancer susceptibility. Br J Nutr2006;96:609–19.