. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

The cellular prion protein counteracts cardiacoxidative stressFilippo Zanetti1†, Andrea Carpi2, Roberta Menabo3, Marco Giorgio2, Rainer Schulz4,Guro Valen5, Anton Baysa5, Maria Lina Massimino3, Maria Catia Sorgato1,3,Alessandro Bertoli1*, and Fabio Di Lisa1*

1Department of Biomedical Science, University of Padova, Padova, Italy; 2Department of Experimental Oncology, European Institute of Oncology, Milano, Italy; 3CNR Institute of Neuroscience,University of Padova, Padova, Italy; 4Institut fur Physiologie, Justus-Liebig Universitat, Gießen, Germany; and 5Department of Physiology, University of Oslo, Oslo, Norway

Received 6 November 2013; revised 4 August 2014; accepted 8 August 2014; online publish-ahead-of-print 18 August 2014

Time for primary review: 33 days

Aims The cellular prion protein, PrPC, whose aberrant isoforms are related to prion diseases of humans and animals, has a stillobscure physiological function. Having observed an increased expression of PrPC in two in vivo paradigms of heart remod-elling, we focused on isolated mouse hearts to ascertain the capacity of PrPC to antagonize oxidative damage induced byischaemic and non-ischaemic protocols.

Methodsand results

Hearts isolated from mice expressing PrPC in variable amounts were subjected to different and complementary oxidativeperfusion protocols. Accumulation of reactive oxygen species, oxidation of myofibrillar proteins, and cell death wereevaluated. We found that overexpressed PrPC reduced oxidative stress and cell death caused by post-ischaemic reperfu-sion. Conversely, deletion of PrPC increased oxidative stress during both ischaemic preconditioning and perfusion(15 min) with H2O2. Supporting its relation with intracellular systems involved in oxidative stress, PrPC was found to in-fluence the activity of catalase and, for the first time, the expression of p66Shc, a protein implicated in oxidative stress-mediated cell death.

Conclusions Our data demonstrate that PrPC contributes to the cardiac mechanisms antagonizing oxidative insults.- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -Keywords Cellular prion protein † PrP † Oxidative stress † Heart † ROS

1. IntroductionThe cellular prion protein (PrPC) is a highly conserved sialoglycoproteinpresent in most mammalian cells, which is attached to the cell surface viaa glycolipid anchor. Upon a conformational change towards ab-enriched structure, PrPC converts into an aberrant isoform (PrPSc).PrPSc is the major constituent of prions, the aetiological agents of agroup of rare and incurable neurodegenerative disorders, named trans-missible spongiform encephalopathies (TSEs) or prion diseases, whichaffect both humans and animals.1,2 Neurodegeneration is the hallmark ofprion diseases. However, it has been shown that cardiac muscles canalso be naturally affected during TSEs. Examples of this kind are the priondeposits observed in the myocardium of a severe dilated cardiomyopathyconcomitant with sporadic Creutzfeldt-Jakob disease (CJD),3 of animalsaffected by chronic wasting disease,4 and of a familial CJD mouse model.5

Knowledge on PrPC physiological function would help elucidating themechanism of prion pathogenesis. However, the absence of grossphenotypic traits in PrP-knockout (PrP-KO) mice6 –8 has hampered sofar full understanding of the role of the protein. A compensatory mech-anism could explain the lack of overt PrP-KO phenotypes, wherebychanges linked to PrPC deficiency would become apparent only underdefined stress conditions.9 Nonetheless, use of several strategies—re-combinant PrP forms, cell model systems, and genetically modifiedanimals—has proposed a wide spectrum of PrPC roles, ranging from dif-ferentiation to cell survival,10,11 and, in particular, to cell protection fromoxidative injury.12 For example, after (transient or permanent) cerebralischaemia involving oxidative stress and cell death,13 the affected brainarea of patients and animal models showed increased PrPC levels.14–16

A role for PrPC in the protective adaptive response against oxidativestress was also suggested by the reduced or exacerbated ischaemic

† Present address. Department of Pathology and Immunology, Centre Medical Universitaire, 1 rue Michel Servet, 1211 Geneva, Switzerland.

* Corresponding author. Tel: +39 049 8276150; fax: +39 049 8073310, Email: alessandro.bertoli@unipd.it (A. Bertoli); Tel: +39 049 8276132; fax: +39 049 8276040. Email: dilisa@bio.unipd.it,dilisa@civ.bio.unipd.it, fabio.dilisa@gmail.com (F. Di Lisa).

Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2014. For permissions please email: journals.permissions@oup.com.

Cardiovascular Research (2014) 104, 93–102doi:10.1093/cvr/cvu194

by guest on Novem

ber 12, 2014D

ownloaded from

damage observed in the brains of animals overexpressing,17,18 or KOfor,14,19,20 PrPC, respectively.

In light of these results and the knowledge that skeletal and cardiacmuscles from PrP-KO mice display higher levels of oxidized lipids andproteins, and reduced antioxidant activity than the wild-type (WT)counterparts,21 we investigated whether PrPC protects the heart fromoxidative stress in analogy to brains. The study was carried out by com-paring hearts isolated from mice expressing different quantities of PrPC-WT and PrP-KO mice, and mice with a three-fold expression of PrPC

(PrP-OE). After applying different, and complementary, ischaemic andnon-ischaemic oxidative stress protocols, hearts were examined forthe accumulation of reactive oxygen species (ROS), oxidation of myofi-brillar proteins, and cell death. To unravel the molecular basis of PrPC

antioxidant action, we also analysed proteins involved in heart antioxi-dant responses. The obtained results indicate that PrPC belongs to thecell mechanisms protecting cardiomyocytes from oxidative insults, pos-sibly by influencing the activity of catalase and the expression of p66Shc.

Of interest, a possible in vivo relevance of PrPC against heart diseasewas suggested by the increased PrPC expression elicited in two animalmodels of heart remodelling (HR) towards heart failure implicating oxi-dative stress.

2. MethodsSee Supplementary material online for a more detailed description.

2.1 AnimalsAll animal experimentation conformed to the Guide for the Care and Use ofLaboratory Animals (US National Institute of Health, publication no. 85–23,revised 1996).

We used 12-month-old male Chinchilla rabbits for rapid left ventricularpacing, and 3-month-old male C57BL/6 mice for permanent coronaryartery ligation provoking myocardial infarction. For all other experiments,we used WT mice with FVB genotype and congenic PrP-KO (line F10),and PrP-OE mice (line Tg37) (both lines kindly provided by the MRCPrion Unit, London, UK).8 The PrP-KO line was obtained by cross-breedingfor 10 generations Zurich I PrP-KO mice (bearing a hybrid SV129X/C57-Bl6genotype)6 with FVB WT (PrP+/+) mice, and then by interbreeding thePrP+/2 littermates (N10) to generate the PrP-KO (F10) line with analmost pure (theoretically .99.9%) FVB genotype.8 PrP-OE Tg37 micewere generated by reintroducing PrP transgenes onto the PrP-KO F10line, in which the three-fold expressionof PrPC did not form prion-like aggre-gates.8 To minimize individual biological variability, 4-month-old male micewere always used. Before thoracotomy, rabbits subjected to left ventricularpacing were anaesthetized by means of i.v. administration of ketamine(50 mg/kg)–xylazine (3 mg/kg), followed by propofol (12–25 mL/h)–fen-tanyl (0.003 mg/kg). Before instrumentation for myocardial infarction,mice were anaesthetized with inhalation of 1.5% isoflurane mixed withpure oxygen, intubated, and ventilated with the same gas mixture duringthe entire procedure. For ex vivo heart perfusion experiments, following an-aesthesia by intraperitoneal injection of a cocktail of tiletamine hydrochlor-ide and zolezepam hydrochloride (Zoletil 100, 30 mg/kg body weight;Virbac, Milan, Italy), mice were euthanized by cervical dislocation beforeheart excision.

2.2 Induction of myocardial infarction in miceThe myocardial infarction model in mice was based on the in vivo permanentocclusion of the left anterior descending coronary artery, as described.22

The permanent occlusion was kept for 24 h, 1, 2, 4, and 6 weeks, and foreach time point, sham-operated mice were used as a reference. At the endof observation, animals were sacrificed, and hearts were explanted andstored at 2808C for subsequent analyses.

2.3 Perfusion protocols of isolated mouse heartsPerfusion of isolated hearts was carried out in the non-recirculating Langen-dorff mode,23 with a perfusion buffer (PB) containing (in mM) 115.0 NaCl,4.75 KCl, 2.15 KH2PO4 (pH 7.4), 25.0 NaHCO3, 0.65 MgSO4, 1.69 CaCl2,and 11.0 glucose, and gassed with O2 (95%) and CO2 (5%). After a 5-min per-fusion stabilization period, hearts were subjected to the following perfusionprotocols: (i) ischaemia/reperfusion (I/R), consisting of 40 min of globalischaemia (achieved by stopping the coronary flow) followed by 15 min ofreperfusion; (ii) ischaemic preconditioning (IPC), i.e. three cycles of 5-minischaemia and 5-min reperfusion, followed, or not, by I/R; and (iii) perfusion(for 15, or 30, min) with hydrogen peroxide (H2O2, 1 mM in PB). Control(normoxic) hearts were subjected only to the 5-min stabilization protocol.For subsequent analyses, 5 mL of samples of the coronary effluent werecollected at 1-min intervals during post-ischaemic reperfusion, or perfusionwith H2O2, as described.24 At the end of the experiments, hearts wereeither immediately used for determining the total heart content of lacticdehydrogenase (LDH), and dihydroethidium (DHE) staining, or stored inliquid nitrogen for further tests.

2.4 Estimation of enzymatic activity2.4.1 Lactic dehydrogenaseMyocardial cell death following the different treatments was quantified bythe presence, i.e. the activity, of LDH in the coronaryeffluent, as described.24

2.4.2 Superoxide dismutaseSuperoxide dismutase 1 (SOD1) (Cu/Zn–SOD) and SOD2 (Mn–SOD) ac-tivities were determined on heart cytosolic and mitochondrial fractions, re-spectively, byestimating the inhibition of (SOD-containing) heart samplesonxanthine/xanthine oxidase-induced reduction of cytochrome c.25 Datawerenormalized to those obtained with 1 U of purified SOD, and reported as thepercentage of the mean value obtained with WT samples.

2.4.3 CatalaseCatalase (CAT) activity of homogenized heart samples was measured by fol-lowing H2O2 consumption at 240 nm.26 Data were calibrated by means of astandard curve generated by using known amounts of purified CAT, andexpressed as the percentage of the mean value obtained with WT samples.

2.5 Western blotting and densitometric analysisAfter determining total protein contents, Western blot (WB) analyses onwhole heart homogenates, or mitochondrial fractions, were carried out bySDS–PAGE protein separation under either non-reducing [for mousePrPC and tropomyosin (TM)], or reducing [for rabbit PrPC, p66Shc, SOD1,SOD2, and succinate dehydrogenase complex subunit A (SDHA)], condi-tions. Proteins were then electroblotted onto nitrocellulose membranes,which were subsequently blocked (1 h, RT) with either bovine serumalbumin [3% (w/v) for PrPC and SDHA immunoblotting], or with non-fatdry milk [5% (w/v) for TM, p66Shc, SOD1, and SOD2 immunoblotting], inphosphate-buffered saline added with 0.1% (v/v) Tween-20 (PBS-T). Mem-branes were incubated (overnight, RT) with the desired primary antibodydiluted in the blocking solution. After extensive washings with PBS-T, mem-branes were incubated (1 h, RT) with horseradish peroxidase-conjugatedanti-mouse, or anti-rabbit, secondary antibody (Santa Cruz Biotechnology,Santa Cruz, CA, USA, 1 : 3000 in the blocking solution). After washings,immunoreactive bands were visualized on a digital image analyser using anenhanced chemiluminescence reagent kit. For quantitative comparisons,densitometric analyses of immunoreactive bands were performed asdetailed in Supplementary material online.

2.5.1 AntibodiesThe followingmono- (m) and polyclonal (p) antibodies (Ab) wereused (dilu-tions are indicated in parenthesis): anti-PrP mouse mAb 8H4 (1 : 7000, a kindgift of Dr M.S. Sy, Case Western University, Cleveland, OH, USA); anti-TM

F. Zanetti et al.94by guest on N

ovember 12, 2014

Dow

nloaded from

mouse mAb CH1 (1 : 2000) (Sigma); anti-Shc rabbit pAb (1 : 2000) (Trans-duction Laboratories); anti-SOD1 rabbit pAb (1 : 1000) (Abcam);anti-SOD2 rabbit pAb (1 : 2000) (Sigma); and anti-SDHA rabbit pAb(1 : 1000) (Sigma).

2.6 In situ superoxide detectionTissue staining with DHE was used to measure heart accumulation of thesuperoxide anion.27 Freshly prepared heart cryosections (10 mm) wereincubated (30 min, 378C) in the presence of DHE (10 mM in DMSO), anddigitalized fluorescence images were subjected to an automatic computer-based analysis. The fluorescence intensity of each tissue section was calcu-lated as the average intensity from four randomly selected fields, and thennormalized to the mean value of the WT samples.

2.7 Statistical analysisAll results are presented as mean+ standard error of the mean (SEM). Thenumber of replicates, n, for each set of results is reported in the figurelegends, and refers to independent experiments, i.e. measurements per-formed on different hearts. Statistical analysis for pair comparison withingroups was performed using the Student’s t-test, whereas analysisbetween groups by one-way ANOVA followed by the Bonferroni’s posthoc test for group differences. Statistics were computed with the Prism soft-ware (GraphPad Software, San Diego, CA, USA). Differences betweenmeans were accepted as statistically significant at the 95% level (P , 0.05).

3. Results

3.1 PrPC expression is transiently increasedin models of HR towards failureInitially, we verified whether hearts responded to ischaemic injury byincreasing theexpressionofPrPC, aswaspreviouslyobserved in ischaemicbrains.14–16 To this end, we analysed by WB already available samplesfrom mice with permanent coronary artery ligation. The finding thatPrPC was transiently increased in the initial phase of HR (Figure 1 andsee Supplementary material online, Figure S1) suggests that the proteinacts as an early adaptive cellular response. This concept was supportedby the augmented PrPC content observed in a non-ischaemic HR model(rabbits subjected to rapid left ventricular pacing, see Supplementarymaterial online, Figure S2).

Notably, although both the above in vivo models involve oxidativedamage,28– 30 they are not suited to establish a direct relationshipbetween a given protein, e.g. PrPC, and cardiac oxidative stress. Wethus investigated this issue by comparing I/R damage in hearts isolatedfrom WT mice with genetically modified congenic mice, i.e. PrP-KOmice and a transgenic PrP-OE line. We first controlled by WB the ex-pression of PrPC in the different hearts (Figure 2; see also Supplementarymaterial online, Figure S3). Though to a lesser extent than in the brain(Figure 2, lane 1), PrPC was readily identified in WT hearts (lane 2),while higher amounts were present in PrP-OE hearts (lane 4) (OE toWT PrPC ratio, 3.0+ 0.2; see also Supplementary material online,Figure S4). Of interest, the increase of PrPC expression genetically deter-mined in PrP-OE hearts is similar to the increase observed in WT heartsin response to 1 week of coronary ligation (Figure 1B). As expected, noimmunosignal was present in PrP-KO hearts (Figure 2, lane 3).

The three PrPC bands evident in Figure 2 reflect the typical glycosyla-tion of the protein, whose different glycoforms (non-, mono-, ordi-glycosylated with increasing molecular mass) correspond to the vari-able occupancy of the two N-glycosylation sites.1 A link between thebiological function and glycosylation states of PrPC is still missing,yet the presence of the fully glycosylated form indicates the correct

post-translational processing of the protein in both WT and PrP-OEhearts (lanes 2 and 4).

3.2 Isolated PrP-OE hearts are protectedagainst ROS accumulation, proteinoxidation, and cell death induced bypost-ischaemic reperfusionIn the I/R protocol, hearts were subjected to a 40-min period ofischaemia followed by 15 min of reperfusion, after which death ofcardiomyocytes was monitored by quantifying the release of LDH inthe coronary effluent. As shown in Table 1 (first line), and in the bardiagram of Figure 3A, I/R produced �27% loss of cell viability in bothWT and PrP-KO hearts, in contrast to the significantly reduced celldeath (�22%) in PrP-OE hearts, indicating that the three-fold-overexpressed PrPC protected from I/R injury (F2,108 ¼ 3.249, P ,

0.05; post hoc test: P , 0.05 for PrP-OE vs. WT and vs. PrP-KO). Theprotection afforded by PrPC overexpression was further supported byevaluating the area of necrosis by triphenyl tetrazolium chloride stainingof heart slices in a limited number of experiments (see Supplementarymaterial online, Figure S5).

Figure1 Myocardial infarction inducesatransient increaseofPrPCex-pression in mouse hearts. PrPC expression was compared betweenhearts from adult male mice either control operated (Sham), or sub-jected to permanent ligation of the left coronary artery [provoking myo-cardial infarction (MI)] for24 h,or1,2,4, and6weeks. (A)Representativeblots showing PrPC immunosignals (upper part) and the loading controlsof Ponceau red-stained nitrocellulose membranes (lower part). The(upper left) arrowhead indicates the di-glycosylated (D) form of PrPC,while MW (on the upper right side) the apparent molecular mass. (B)Densitometric analysis of PrPC-immunoreactive bands, showing thatPrPC is maximally expressed after 1 week from the coronary artery liga-tion. n ¼ 6 for each time point of Sham and treated mice; ***P , 0.001,Student’s t-test;ANOVAcomparisonbetweengroups(MIeffectonPrPC

expression, dark bars): F4,31¼ 5.450, P , 0.005 (Bonferroni’s post hoctest: P , 0.01 for 1 week vs. both 24 h and 2 weeks, P , 0.001 for 1week vs. 6 weeks). For other details, see Methods.

PrPC antioxidant function in the heart 95by guest on N

ovember 12, 2014

Dow

nloaded from

Although prolonged no-flow ischaemia of isolated hearts provokes acomplex set of cellular derangements (e.g. mitochondria dysfunctions,ATP depletion, pH changes, and impairment of ion homeostasis), amajor contribution to heart injury comes from the oxidative damagecausedby massiveROS amountsproduced upon re-establishing the cor-onary flow.28 To provide evidence for the antioxidant capacity of PrPC,we measured both ROS generated by hearts at the end of the I/R proto-col, and the ROS-induced oxidation of heart contractile proteins. Ofthese, we analysed TM, in light of the link between derangement ofheart contractility and quantity of TM dimers that form after the cross-linking of single cysteine residues present in TM molecules.31

By staining heart cryosections with DHE, whose reaction with super-oxide anions generates the fluorogenic 2-hydroxyethidium,27 and byimmunoblot experiments of heart samples probed with an antibodyto TM, we found that PrP-OE samples were significantly less fluorescent(Figure 3B; F2,12 ¼ 10.470, P , 0.005; post hoc test: P , 0.01 PrP-OE vs.WT, P , 0.001 PrP-OE vs. PrP-KO), and generated significantly lessTM dimers (Figure 3C, F2,11 ¼ 4.291, P , 0.05; post hoc test: P , 0.05for PrP-OE vs. WT and vs. PrP-KO; see also Supplementary materialonline, Figure S6), compared with the other two groups. An almost

undetectable fluorescence signal was instead displayed by normoxicheart cryosections (data not shown).

Taken together, these data indicate that only the overexpression ofPrPC protects the heart from I/R oxidative damage. To explain whynormal amounts of PrPC could not antagonize I/R injury, and whymortality was similar in WT and PrP-KO hearts, one should considerthe aggressiveness of the entire I/R procedure on cardiomyocytes,given that post-reperfusion ROS act synergistically with the complexset of the above-mentioned cell derangements provoked by prolongedischaemia.32

3.3 PrPC antagonizes the effects of ROSin isolated heartsROS are known to be involved both in cell damage and in cell signallingand protection.28 To better understand the antioxidant property ofPrPC, we thus investigated these two opposite faces of ROS. First, weused IPC, involving protective ROS signalling; secondly, we applied asevere and direct oxidative stress by means of perfusion with H2O2.

IPC, which consists of repetitive brief episodes of I/R preceding a pro-longed ischaemic period, is suggested to produce sub-lethal ROS thattrigger defence mechanisms eventually antagonizing the large burst ofROS during post-ischaemic reperfusion.33 DHE staining of hearts atthe end of three cycles of short I/R episodes (5 min/5 min) showedthat PrP-KO hearts accumulated significantly higher ROS comparedwith WT and PrP-OE hearts (Figure 4A, F2,11¼ 7.755, P , 0.01; post hoctest: P , 0.01 PrP-KO vs. WT, P , 0.05 PrP-KO vs. PrP-OE). Thesedata indicate that PrPC antagonize IPC-induced ROS accumulation.However, the larger ROS amount generated in the absence of PrPC wasunable to increase the protection observed in WT hearts (Table 1,compare the first and second line, and Figure 4B, compare grey andhatched bars, P , 0.01 for both WT and PrP-KO hearts, Student’st-test). Conversely, no significant effect was observed in PrP-OE hearts,whichdisplayedROSlevelscomparablewith thoseof theWTgroup.Add-ition of the potent ROS scavenger, N-2-mercaptopropionyl-glycine(MPG), abolished the effect irrespective of the PrP genotype (third lineof Table 1; black bars of Figure 4B), confirming that (mild) production ofROS during IPC is necessary for heart protection.

Next, a more direct evidence of PrPC’s antioxidant potential wasobtained using a model (perfusion with H2O2) in which myocardialdamage can be ascribed primarily to oxidative stress. Figure 5A showsthat damage by H2O2 was initially small and that it became moreevident with longer exposure to H2O2 in all hearts. However, a clearPrP-KO phenotype emerged at later time points of perfusion (14 and15 min), when there was both a significantly higher cell death [fourth

Figure 2 PrPC content of WT, PrP-KO, and PrP-OE hearts. Hearthomogenates (lanes 2–4, 15 mg of proteins per lane) were analysedfor PrPC content by WB. Compared with the WT sample (lane 2),the PrP-OE heart displays a more evident immunosignal for PrPC

(lane 4; see also Supplementary material online, Figures S3 and S4),while no signal is present in the PrP-KO sample (lane 3). A WT brainhomogenate (lane 1, 5 mg of proteins) was loaded for comparison.Arrowheads (on the left) indicate the three PrPC glycoforms (di-, D;mono-, M; un-glycosylated, U), while the apparent molecular mass isindicated on the right. Shown lanes were cut from the original blot ofSupplementary material online, Figure S3. For other details, seeMethods and Supplementary material online.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Table 1 Loss of viability of isolated hearts with different PrPC amounts subjected to perfusion protocols

LDH release (%) WT PrP-KO PrP-OE

I/R 27.2+2.1 (n ¼ 38) 27.2+1.8 (n ¼ 38) 21.7+1.4 (n ¼ 35)

IPC + I/R 18.3+2.4 (n ¼ 14) 15.4+2.4 (n ¼ 12) 17.5+2.0 (n ¼ 9)

IPC (+MPG) + I/R 28.6+5.4 (n ¼ 4) 26.4+4.9 (n ¼ 4) 22.5+4.2 (n ¼ 4)

H2O2 6.2+1.1 (n ¼ 11) 12.5+2.3 (n ¼ 11) 6.5+2.2 (n ¼ 11)

Loss of myocardial viability of WT, PrP-KO, and PrP-OE hearts (second, third, and fourth columns, respectively)was evaluated as the percentage of LDH released in the coronaryeffluent overheart total (coronary effluent plus tissue homogenate) LDH, following perfusion with the protocol indicated in the first column. Data are mean+ SEM; n is the number of experiments. Forstatistical analysis, see the text and the legends of Figures 3–5. Other details are described in Methods and also in Supplementary material online.I/R, ischaemia followed by reperfusion; IPC, ischaemic preconditioning before I/R; MPG, N-2-mercaptopropionyl-glycine.

F. Zanetti et al.96by guest on N

ovember 12, 2014

Dow

nloaded from

line of Table 1, and Figure 5A; (100+ 35)% relative increase with respectto WT hearts after 15 min perfusion; for statistical analysis, see Supple-mentary material online, Table S1], and TM oxidation (Figure 5B; F2,22 ¼

4.75, P , 0.05; post hoc test: P , 0.05 for PrP-KO vs. WT,P , 0.01 for PrP-KO vs. PrP-OE), than in PrPC-expressing samples.

As shown (Figure 5A, black line), MPG fully reversed the PrP-KOphenotype.

Extending the perfusion to 30 min resulted in exacerbated injury of allhearts. However, while there was no difference in death between WTand PrP-KO cardiomyocytes, PrPC overexpression significantly

Figure 3 PrP-OE hearts are protected against cell death, ROS accumulation, and protein oxidation induced by I/R. (A) The bar diagram shows that,compared with I/R-treated WT and PrP-KO hearts, I/R-treated PrP-OE hearts have a significantly reduced cell death (estimated as in the legend ofTable 1). (B) Upper and lower-left panels show fluorescence micrographs (each representative of five independent experiments) of cryosectionsfrom I/R-treated WT, PrP-KO, and PrP-OE heart, stained with DHE to quantify superoxide contents. Fluorescence quantification (expressed as thepercentage of the mean values obtained for WT samples, lower-right panel) indicates that PrP-OE hearts accumulate significantly less ROS thanWT and PrP-KO hearts. (C ) Here, dimerization of TM in hearts after I/R is taken as a paradigm of ROS-induced oxidation of myocardial contractileproteins. The left panel shows the WBs (representative of at least four independent experiments) of heart homogenates probed for TM. A sampleof a WT normoxic heart (i.e. not subjected to the I/R protocol) was loaded as a negative control (cntr). Clearly, both the WB and the densitometricanalyses (bar diagram, right panel) indicate that PrP-OE hearts harbour less TM dimers than the other two samples. Shown lanes were cut from theoriginal blots of Supplementary material online, Figure S6. Densitometric data (expressed as the percentage of the WT samples) were calculated bynormalizing the intensity of the signal of TM dimers to that of the corresponding TM monomers. In A, n ¼ 38 for WT and PrP-KO, and n ¼ 35 forPrP-OE; in B, n ¼ 5 for each PrP genotype; in C, n ¼ 5 for WT and PrP-OE, and n ¼ 4 for PrP-KO. Bonferroni’s post hoc test, *P , 0.05, **P , 0.01,***P , 0.001. Other details are described in the legend of Table 1.

PrPC antioxidant function in the heart 97by guest on N

ovember 12, 2014

Dow

nloaded from

protected the myocardium (inset in Figure 5A). These results furthersupport our suggestion that cells that had been impaired by a pro-tracted ischaemic episode (Figure 3A), or, as in this case, by prolongedperfusion with H2O2, need higher PrPC amounts to prevent ROS-mediated cell death.

3.4 Molecular mechanisms of PrPC

antioxidant function in hearts: catalaseactivity and p66Shc expressionBecause of previous indications that PrPC modulates various cellantioxidant systems,12,21 we investigated whether the response to

Figure 4 PrPC reducesboth ROS formationduring IPC, and theprotectiveeffect of IPC against I/R injury. (A)Upperand lower-left panels show fluorescencemicrographs (each representative of at least four independent experiments) of DHE-stained cryosections from WT, PrP-KO, and PrP-OE hearts exposed toIPC. Fluorescencequantification (as thepercentageof theWTsamples, lower-rightbar diagram) indicates that PrP-KOhearts accumulatehigher ROSamountsduring IPC than WT and PrP-OE samples. (B) Isolated hearts were subjected to I/R without (grey bars), or with, a preceding protocol of IPC run in the absence(hatched bars), or in the presence (black bars), of the ROS scavenger MPG (1 mM). The protective effect of IPC is maximal (�45% relative reduction in myo-cardial cell death) in PrP-KO, intermediate (�30%) in WT, and non-significant in PrP-OE, hearts. MPG abrogates the IPC effect in all hearts. In A, n ¼ 5 for WTand PrP-OE, and n ¼ 4 for PrP-KO. In B, grey bars: n ¼ 38 for WT and PrP-KO, and n ¼ 35 for PrP-OE; hatched bars: n ¼ 14 for WT, n ¼ 12 for PrP-KO, andn ¼ 9 for PrP-OE; black bars: n ¼ 4 for each PrP genotype. *P , 0.05, **P , 0.01 (Bonferroni’s post hoc test in A, and Student’s t-test in B). In B, within-groupANOVA comparison between data in grey and hatched bars: WT, F2,53¼ 3.266, P , 0.05 (Bonferroni’s post hoc test, P , 0.05); PrP-KO, F2,51 ¼ 6.322,P , 0.005 (Bonferroni’s post hoc test, P , 0.01); PrP-OE, not significant. Other details are described in the legend of Figure 3.

F. Zanetti et al.98by guest on N

ovember 12, 2014

Dow

nloaded from

oxidative insults in hearts with different PrPC levels could be attributedto variations of myocardial anti-/pro-oxidant resources.

We first tested the activity of CAT. The importance of CAT in the celldetoxification from H2O2 is well recognized, and reduced CAT activity

has already been reported in PrP-KO hearts compared with WThearts.21 In our case, we observed a significant decrease of CAT activity(�15%) in PrP-KO hearts compared with PrPC-expressing counter-parts, but no significant difference between WT and PrP-OE hearts(Figure 6A; F2,27 ¼ 4.742, P , 0.05; post hoc test: P , 0.05 for bothPrP-KO vs. WT and PrP-KO vs. PrP-OE). However, we could notrelate this finding to a reduced expression of the enzyme, as CAT wasnot detected by WB, possibly because of its low expression in thecardiac tissue.34

We also tested SOD, another fundamental component of ROS-scavenging systems in cells. SOD exists in three isoforms, of whichone resides mainly in the cytosol (Cu/Zn-dependent SOD1) andanother in mitochondria (Mn-dependent SOD2). A reduction in totalSOD activity was reported in PrP-KO and in prion-infected brains,21,35

but these data were questioned.36,37 Using cytosolic and mitochondrialfractions to evaluate separately the activity of SOD1 and SOD2, wefound that both enzymes displayed similar activity (Figure 6A) andexpression level (data not shown) in all hearts.

To investigate a further link between PrPC and intracellular mechan-isms involving oxidative stress, we considered p66Shc, a splice variantof the two cytosolic adaptor proteins p52Shc and p46Shc. This choicestemmed from the notion that mitochondria are one of the major cellsources of ROS, and that p66Shc is a key factor in ROS formationbecause, after re-localizing to the mitochondrial intermembranespace, it diverts electrons from the respiratory chain to O2.

38 According-ly, embryonic fibroblasts and isolated hearts from p66Shc-KO mice havebeen shown to be more resistant to apoptotic stimuli and I/R injury,respectively.24,39 As shown in the WB analysis of Figure 6B, PrP-KOtotal heart homogenates contained higher levels (�80%) of p66Shc

than PrPC-expressing samples (F2,43 ¼ 3.293, P , 0.05; post hoc test:P , 0.05 for both PrP-KO vs. WT and PrP-KO vs. PrP-OE). Conversely,all hearts displayed similar amounts of p52Shc and p46Shc (data notshown). Most importantly, in spite of the finding that PrPC ablation didnot affect the heart mitochondrial content (assessed by the WB quanti-fication of the mitochondrial marker SDH, data not shown), we foundthat the mitochondrial fraction of PrP-KO hearts had also higher(�60%) p66Shc content than PrPC-expressing hearts (Figure 6C;F2,13 ¼ 6.989, P , 0.01; post hoc test: P , 0.05 for PrP-KO vs. WT, andP , 0.01 for PrP-KO vs. PrP-OE).

4. DiscussionByusing for the first time in the prion field perfused ex vivohearts to studyPrPC antioxidant features, the presentwork providesevidence that PrPC

modulates cardiac ROS formation, and that its overexpression protectsthe heart from reperfusion injury by reducing oxidative stress.

A close insight into PrPC antioxidant features was accomplished bycomparing the response of hearts isolated from mice with differentPrP genotypes (WT, OE, and KO) to three complementary perfusionprotocols—I/R, preceded or not by IPC, and perfusion with H2O2.Taken together, these models highlighted a likely physiological signifi-cance of PrPC antioxidant properties under stress conditions.

We found that, while ischaemia (40 min) followed by reperfusion(15 min) failed to underscore differences in cell death between WTand PrP-KO hearts, PrP-OE hearts showed a significantly reduced celldamage and lower amounts of ROS and oxidized myofibrillar proteins.This observation is in line with the finding that high quantities of(endogenous or exogenous) PrPC defend against ischaemic braininjury and hypoxia.15,17 It is therefore not surprising the inability of

Figure5 The absence of PrPC exacerbates myocardial cell death andTM oxidation of H2O2-perfused hearts. (A) Loss of myocardial viabilitywas evaluated as the with-time cumulative LDH amounts released inthe coronary effluent during perfusion (15 or 30 min) with H2O2

(1 mM). After 14- and 15-min perfusion, PrP-KO hearts (opensquares) show a significantly higher myocardial cell death than WT(open circles) and PrP-OE (open triangles) hearts. The presence ofMPG (1 mM) in the perfusing medium reduces the cell death ofPrP-KO hearts (filled squares) to that observed in PrPC-expressinghearts. The inset, reporting cell death in the late phases of the30-min perfusion protocol, shows that overexpressed PrPC protectsfrom cell death from 27 min onwards. (B) The upper panel reportsthe WBs of H2O2-perfused (15 min) hearts probed for TM (represen-tative of at least seven independent experiments). A WT heart per-fused without H2O2 was used as a negative control (cntr). Both theupper and the lower (reporting the bar diagram of the densitometricanalyses of TM-immunoreactive bands) panels indicate that, after15-min perfusion with H2O2, PrP-KO hearts have higher levels of (oxi-dized) TM dimers than WT and PrP-OE samples. In A: n ¼ 11 for eachPrP genotype after 15-min perfusion without MPG; n ¼ 3 for PrP-KOhearts after 15-min perfusion with MPG; n ¼ 4 for the 30-min perfusionof each PrP genotype. In B: n ¼ 7 for WT, and n ¼ 9 for PrP-KO andPrP-OE hearts. For the statistical analysis of data reported in (A), see Sup-plementary material online, Table S1. InB, *P , 0.05, **P , 0.01, Bonfer-roni’s post hoc test. Other details are described in the legend of Figure 3.

PrPC antioxidant function in the heart 99by guest on N

ovember 12, 2014

Dow

nloaded from

physiological PrPC levels to oppose the fierce post-ischaemic burst ofROS, which could account for the lack of major differences in I/Rinjury between WT and PrP-KO hearts.

A further insight into the protective property of PrPC came from per-fusion with H2O2, an oxidative challenge devoid of ischaemic insults,

which should set at best the conditions for appreciating the specific con-tribution of PrPC to the antioxidant defences of the heart. Indeed, theovert PrP-KO phenotype disclosed by 14–15 min H2O2-based perfu-sion and the similar response of WT and PrP-OE hearts—with alower extent of cell death and oxidized proteins—supports thenotion that physiological PrPC amounts are sufficient to protect fromoxidative insults. PrPC anti-ROS properties in an intact tissue arefurther corroborated by the response to longer (30 min) exposure toH2O2, after which the oxidative damage to cardiomyocytes must havebeen so profound that only overexpressed PrPC could partly alleviateROS injury.

Another indication of the antioxidant role of PrPC was obtained byapplying IPC, which induced a larger production of ROS in hearts fromPrP-KO mice. However, this larger ROS formation did not improveIPC protection, in line with the notion that no intervention is ableto increase the cardioprotective efficacy of IPC.40 Nor could IPC addfurther protection to that already exerted by three-fold levels of PrPC

in PrP-OE hearts, suggesting that IPC and PrPC overexpression work viathe same mechanism, i.e. by reducing ROS formation upon reperfusion.

Taken together, these results indicate the capacityof PrPC toantagon-ize ROS under a variety of circumstances: when ROS provoke heartinjury (as after I/R- and H2O2-based protocols), or trigger intracellularsignalling (as during IPC). Besides the acute model of perfused heartshighlighting a direct action of PrPC on ROS, one can also envisage adap-tivemechanisms in which an increased expression of PrPC—as observedhere in two (ischaemic and non-ischaemic) in vivo models of HR and pre-viously in cerebral ischaemia14 –16—strengthens the endogenousdefence of the heart against oxidative stress.

To identify the molecularbasisof PrPC protection, wefirst consideredCAT and SOD, whose activity was previously reported to be reduced indifferent PrP-KO cells and tissues, including heart.21 We confirmed thesignificant decrease of CAT activity in PrP-KO hearts compared withPrPC-expressing hearts. However, because we failed to detect theprotein by WB, further studies are needed to clarify whether such a re-duction is consequent to a decreased quantity, or an inhibitory modifi-cation, of the enzyme.

Figure 6 Catalase activity is reduced, and p66Shc expression isincreased, in PrP-KO hearts. (A) Samples of untreated hearts fromWT (grey bars), PrP-KO (hatched bars), and PrP-OE (black bars)mice were tested for CAT, SOD1, and SOD2 activities. Whereas inall hearts a similar SOD1 and SOD2 activity is found, the activity ofCAT is significantly reduced (�15%) in PrP-KO hearts with respectto PrPC-expressing hearts. (B and C) p66Shc content of total homoge-nates (B), or mitochondrial fractions (C), of untreated hearts wasassessed by WB. Upper panels show representative WBs (out of 5to 18 independent experiments) for p66Shc. Lower panels report thebar diagrams of the densitometric analysis of p66Shc-immunoreactivebands (expressed as the percentage of WT samples), indicating thatp66Shc is significantly more abundant in both total homogenates(�80%) and mitochondrial fractions (�60%) of PrP-KO hearts thanin WT and PrP-OE samples. In (A), n ¼ 12 for WT, n ¼ 8 forPrP-KO, and n ¼ 10 for PrP-OE, for CAT activity; n ¼ 4 for each PrPgenotype, for SOD1 and SOD2 activities; in (B): n ¼ 15 for WT,n ¼ 18 for PrP-KO, and n ¼ 13 for PrP-OE; in (C): n ¼ 6 for WT,and n ¼ 5 for PrP-KO and PrP-OE. *P , 0.05, **P , 0.01, Bonferroni’spost hoc test. Further details are described in Methods and Supplemen-tary material online.

F. Zanetti et al.100by guest on N

ovember 12, 2014

Dow

nloaded from

In contrast, hearts with different PrP genotypes displayed similar activ-ities and amounts of both cytosolic and mitochondrial SOD isoforms.SOD regulation by PrPC has been highly debated, and some authorshave rejected this possibility on methodological grounds.36,37 To notethat, we tested SOD activity with the xanthine/xanthine oxidase-basedassay, because the sensitivity and reproducibility of the test are higherthan the alternative nitro-blue tetrazolium-based technique.36

However, we also provided evidence that PrPC-dependent defenceagainst ROS may involve control of p66Shc expression. Following itspartial re-distribution from the cytosol to the intermembrane space ofmitochondria, p66Shc oxidizes cytochrome c and the resulting electronsreduce O2 to H2O2. This is a powerful mechanism that produces asmuch as one-third of the H2O2 pool of cells.41 Accordingly,p66Shc-KO cells are protected against oxidative stress,39 and, in particu-lar, p66Shc-KO hearts are more resistant to I/R injuries.24 Hence, ourfinding that PrP-KO hearts expressed higher amounts of both totaland mitochondrial p66Shc could not only explain the lower capacity ofPrP-KO hearts to antagonize ROS effects, but also add a new actor(PrPC) in p66Shc-dependent oxidative stress and mitochondria-mediated apoptotic mechanisms.

By proposing that PrPC is a part of signal transduction complexes inwhich the protein interacts with several partners, a fewstudies have pro-vided the mechanistic basis to explain how PrPC could modulate intra-cellular events in spite of its location at the cell surface.11,42,43 Bycontrolling different signalling pathways, including those mediated byCa2+, Erk 1/2, or PI3K/Akt,11,42– 46 PrPC could ultimately also affectgene expression, as suggested by different approaches, includinglarge-scale proteomics.42,46 –48 Interestingly, the recent finding that acti-vation of Akt in the heart induces the nuclear E2-related factor 2 (Nrf2),and transcription of cytoprotective genes against I/R injury,49 provides apossible connection between PrPC and the control of antioxidantsystems. Indeed, this mechanism could help explaining why PrP-KOhearts display a reduced activity of CAT, whose expression is underNrf2 control.50 –52 However, additional mechanisms have to be identi-fied to explain the change in p66Shc, which—to our knowledge—hasnever been reported to be controlled by the Nrf2 pathway.

In addition, PrPC-dependent up-regulation of anti- (CAT), and down-regulation of pro- (p66Shc), oxidant systems cannot fully account for theobserved antagonism of PrPC against ROS injury, given that overex-pressed PrPC protects hearts from I/R- and 30-min H2O2-induced injur-ies, but leaves unaltered both the activity of CAT and the expression ofp66Shc with respect to WT hearts. This finding thus entails the existenceof other sufficiently strong (PrPC-dependent) protective mechanismsthat counteract the severe damage provoked by I/R and prolongedH2O2 treatment.

5. ConclusionsIn conclusion, we have provided evidence that, following prolongedI/R, overexpression of PrPC reduces oxidative stress through an asyet unknown mechanism, and protects the heart from irreversibleinjury. On the other hand, abrogation of PrPC expression potentiatedROS formation during the preconditioning phase of IPC, and aggravatedthe oxidative damage and cell death caused by perfusion with H2O2.The protective effects of PrPC are likely to be implemented alsoin vivo, as shown by the increased expression of the protein in twomodels of HR, suggesting that cardiac PrPC might contribute toendogenous defences against oxidative stress. Further studies,however, are needed to fully elucidate the entire array of mechanisms

underlying the antioxidant capacity of PrPC in the heart, and, possibly,in other organs.

Supplementary materialSupplementary material is available at Cardiovascular Research online.

Conflict of interest: none declared.

FundingThis work was supported by grants from the Italian Ministryof University andResearch (MIUR) (PRIN 2008 to M.C.S.), the University of Padova (Progettod’Ateneo CPDA089551/08 to A.B. and CPDA121988/12 to M.C.S.), Fonda-zione Cariparo (to F.D.L.), and CNR (Project of Special Interest on Aging tothe Institute of Neuroscience).

References1. Prusiner SB. Prions. Proc Natl Acad Sci USA 1998;95:13363–13383.2. Aguzzi A, Calella AM. Prions: protein aggregation and infectious diseases. Physiol Rev

2009;89:1105–1152.3. Ashwath ML, Dearmond SJ, Culclasure T. Prion-associated dilated cardiomyopathy. Arch

Intern Med 2005;165:338–340.4. Jewell JE, Brown J, Kreeger T, Williams ES. Prion protein in cardiac muscle of elk (Cervus

elaphus nelsoni) and white-tailed deer (Odocoileus virginianus) infected with chronicwasting disease. J Gen Virol 2006;87:3443–3450.

5. Chiesa R, Pestronk A, Schmidt RE, Tourtellotte WG, Ghetti B, Piccardo P, Harris DA.Primary myopathy and accumulation of PrPSc-like like molecules in peripheral tissuesof transgenic mice expressing a prion protein insertional mutation. Neurobiol Dis 2001;8:279–288.

6. Bueler H, Fischer M, Lang Y, Bluethmann H, Lipp HP, DeArmond SJ, Prusiner SB,Aguet M, Weissmann C. Normal development and behaviour of mice lacking the neur-onal cell-surface PrP protein. Nature 1992;356:577–582.

7. Manson JC, Clarke AR, Hooper ML, Aitchison L, McConnell I, Hope J. 129/Ola mice car-rying a null mutation in PrP that abolishes mRNA production are developmentallynormal. Mol Neurobiol 1994;8:121–127.

8. Mallucci GR, Ratte S, Asante EA, Linehan J, Gowland I, Jefferys JG, Collinge J. Post-natalknockout of prion protein alters hippocampal CA1 properties, but does not result inneurodegeneration. EMBO J 2002;21:202–210.

9. Steele AD, Lindquist S, Aguzzi A. The prion protein knockout mouse: a phenotype underchallenge. Prion 2007;1:83–93.

10. Aguzzi A, Baumann F, Bremer J. The prion’s elusive reason for being. Annu Rev Neurosci2008;31:439–477.

11. Linden R, Martins VR, Prado MA, Cammarota M, Izquierdo I, Brentani RR. Physiology ofthe prion protein. Physiol Rev 2008;88:673–728.

12. Milhavet O, Lehmann S. Oxidative stress and the prion protein in transmissible spongi-form encephalopathies. Brain Res Brain Res Rev 2002;38:328–339.

13. Dirnagl U, Iadecola C, Moskowitz MA. Pathobiology of ischaemic stroke: an integratedview. Trends Neurosci 1999;22:391–397.

14. McLennan NF, Brennan PM, McNeill A, Davies I, Fotheringham A, Rennison KA,Ritchie D, Brannan F, Head MW, Ironside JW, Williams A, Bell JE. Prion protein accumu-lation and neuroprotection in hypoxic brain damage. Am J Pathol 2004;165:227–235.

15. Weise J,CromeO, Sandau R, Schulz-SchaefferW, Bahr M,Zerr I. Upregulationof cellularprion protein (PrPC) after focal cerebral ischemia and influence of lesion severity. Neu-rosci Lett 2004;372:146–150.

16. Mitsios N, Saka M, Krupinski J, Pennucci R, Sanfeliu C, Miguel Turu M, Gaffney J, Kumar P,Kumar S, Sullivan M, Slevin M. Cellular prion protein is increased in the plasma and peri-infarcted brain tissue after acute stroke. J Neurosci Res 2007;85:602–611.

17. Shyu WC, Lin SZ, Chiang MF, Ding DC, Li KW, Chen SF, YangHI, Li H. Overexpression ofPrPC by adenovirus-mediated gene targeting reduces ischemic injury in a stroke ratmodel. J Neurosci 2005;25:8967–8977.

18. Weise J, Doeppner TR, Muller T, Wrede A, Schulz-Schaeffer W, Zerr I, Witte OW,Bahr M. Overexpression of cellular prion protein alters postischemic Erk1/2 phosphor-ylation but not Akt phosphorylation and protects against focal cerebral ischemia. RestorNeurol Neurosci 2008;26:57–64.

19. Spudich A, Frigg R, Kilic E, Kilic U, Oesch B, Raeber A, Bassetti CL, Hermann DM. Aggra-vation of ischemic brain injury by prion protein deficiency: role of ERK-1/-2 and STAT-1.Neurobiol Dis 2005;20:442–449.

20. Weise J, Sandau R, SchwartingS, CromeO,Wrede A, Schulz-Schaeffer W,Zerr I, BahrM.Deletion of cellular prion protein results in reduced Akt activation, enhanced postis-chemic caspase-3 activation, and exacerbation of ischemic brain injury. Stroke 2006;37:1296–1300.

PrPC antioxidant function in the heart 101by guest on N

ovember 12, 2014

Dow

nloaded from

21. Klamt F, Dal-Pizzol F, Conte da Frota ML Jr, Walz R, Andrades ME, da Silva EG,Brentani RR, Izquierdo I, Fonseca Moreira JC. Imbalance of antioxidant defense inmice lacking cellular prion protein. Free Radic Biol Med 2001;30:1137–1144.

22. Czibik G, Gravning J, Martinov V, Ishaq B, Knudsen E, Attramadal H, Valen G. Genetherapy with hypoxia-inducible factor 1 alpha in skeletal muscle is cardioprotective invivo. Life Sci 2011;88:543–550.

23. Langendorff O. Untersuchungen am uberlebenden Saugethierherzen. Pfluegers Archiv1895;61:291–322.

24. Carpi A, Menabo R, Kaludercic N, Pelicci P, Di Lisa F, Giorgio M. The cardioprotectiveeffects elicited by p66(Shc) ablation demonstrate the crucial role of mitochondrialROS formation in ischemia/reperfusion injury.Biochim BiophysActa 2009;1787:774–780.

25. McCord JM, Fridovich I. The utility of superoxide dismutase in studying free radicalreactions. I. Radicals generated by the interaction of sulfite, dimethyl sulfoxide, andoxygen. J Biol Chem 1969;244:6056–6063.

26. Aebi H. Catalase in vitro. Methods Enzymol 1984;105:121–126.27. Oudot A, Martin C, Busseuil D, Vergely C, Demaison L, Rochette L. NADPH oxidases

are in part responsible for increased cardiovascular superoxide production during aging.Free Radic Biol Med 2006;40:2214–2222.

28. Giordano FJ. Oxygen, oxidative stress, hypoxia, and heart failure. J Clin Invest 2005;115:500–508.

29. Di Lisa F, Canton M, Carpi A, Kaludercic N, Menabo R, Menazza S, Semenzato M. Mito-chondrial injury and protection in ischemic pre- and postconditioning. Antioxid RedoxSignal 2011;14:881–891.

30. Heusch P, Canton M, Aker S, van de Sand A, Konietzka I, Rassaf T, Menazza S, Brodde OE,Di Lisa F, Heusch G, Schulz R. The contribution of reactive oxygen species and p38mitogen-activated protein kinase to myofilament oxidation and progression of heartfailure in rabbits. Br J Pharmacol 2010;160:1408–1416.

31. Canton M, Skyschally A, Menabo R, Boengler K, Gres P, Schulz R, Haude M, Erbel R, DiLisa F, Heusch G.Oxidativemodificationof tropomyosin andmyocardial dysfunction fol-lowing coronary microembolization. Eur Heart J 2006;27:875–881.

32. Jennings RB, Murry CE, Steenbergen C, Reimer KA. Development of cell injury in sus-tained acute ischemia. Circulation 1990;82:S2–S12.

33. Pain T, Yan XM, Critz SD, Yue Y, Nakano A, Liu GS, Heusch G, Cohen MV, Downey JM.Opening of mitochondrial K(ATP) channels triggers the preconditioned state by gener-ating free radicals. Circ Res 2000;87:460–466.

34. Ishikawa T, Akerboom TPM, Sies H. In: Cohen GM, ed. Target Organ Toxicity. 1st ed. BocaRaton, Florida, USA: CRC Press Inc.; 1986. p. 129–143.

35. Wong BS, Brown DR, Pan T, Whiteman M, Liu T, Bu X, Li R, Gambetti P, Olesik J,Rubenstein R, Sy MS. Oxidative impairment in scrapie-infected mice is associated withbrain metals perturbations and altered antioxidant activities. J Neurochem 2001;79:689–698.

36. Hutter G, Heppner FL, Aguzzi A. No superoxide dismutase activity of cellular prionprotein in vivo. Biol Chem 2003;384:1279–1285.

37. Waggoner DJ, Drisaldi B, Bartnikas TB, Casareno RL, Prohaska JR, Gitlin JD, Harris DA.Brain copper content and cuproenzyme activity do not vary with prion protein expres-sion level. J Biol Chem 2000;275:7455–7458.

38. Cosentino F, Francia P, Camici GG, Pelicci PG, Luscher TF, Volpe M. Final commonmolecular pathways of aging and cardiovascular disease: role of the p66Shc protein.Arterioscler Thromb Vasc Biol 2008;28:622–628.

39. Migliaccio E, Giorgio M, Mele S, Pelicci G, Reboldi P, Pandolfi PP, Lanfrancone L,Pelicci PG. The p66Shc adaptor protein controls oxidative stress response and lifespan in mammals. Nature 1999;402:309–313.

40. Kloner RA, Bolli R, Marban E, Reinlib L, Braunwald E. Medical and cellular implications ofstunning, hibernation, and preconditioning: an NHLBI workshop. Circulation 1998;97:1848–1867.

41. Giorgio M, Migliaccio E, Orsini F, Paolucci D, Moroni M, Contursi C, Pelliccia G, Luzi L,Minucci S, Marcaccio M, Pinton P, Rizzuto R, Bernardi P, Paolucci F, Pelicci PG. Electrontransfer between cytochrome c and p66Shc generates reactive oxygen species thattrigger mitochondrial apoptosis. Cell 2005;122:221–233.

42. Sorgato MC, Peggion C, Bertoli A. Is, indeed, the prion protein a Harlequin servant of‘many’ masters? Prion 2009;3:202–205.

43. Peggion C, Bertoli A, Sorgato MC. Possible role for Ca2+ in the pathophysiology of theprion protein? Biofactors 2011;37:241–249.

44. Khosravani H, Zhang Y, Tsutsui S, Hameed S, Altier C, Hamid J, Chen L, Villemaire M,Ali Z, Jirik FR, Zamponi GW. Prion protein attenuates excitotoxicity by inhibitingNMDA receptors. J Cell Biol 2008;181:551–565.

45. Stella R, Massimino ML, Sandri M, Sorgato MC, Bertoli A. Cellular prion proteinpromotes regeneration of adult muscle tissue. Mol Cell Biol 2010;30:4864–4876.

46. Lazzari C, Peggion C, Stella R, Massimino ML, Lim D, Bertoli A, Sorgato MC. Cellularprion protein is implicated in the regulation of local Ca2+ movements in cerebellargranule neurons. J Neurochem 2011;116:881–890.

47. Ramljak S, Asif AR, Armstrong VW, Wrede A, Groschup MH, Buschmann A,Schulz-Schaeffer W, Bodemer W, Zerr I. Physiological role of the cellular prionprotein (PrPC): protein profiling study in two cell culture systems. J Proteome Res 2008;7:2681–2695.

48. Stella R, Cifani P, Peggion C, Hansson K, Lazzari C, Bendz M, Levander F, Sorgato MC,Bertoli A, James P. Relative quantification of membrane proteins in wild-type andprion protein (PrP)-knockout cerebellar granule neurons. J Proteome Res 2012;11:523–536.

49. Deng C, Sun Z, Tong G, Yi W, Ma L, Zhao B, Cheng L, Zhang J, Cao F, Yi D. a-Lipoicacid reduces infarct size and preserves cardiac function in rat myocardial ischemia/reperfusion injury through activation of PI3K/Akt/Nrf2 pathway. PLoS ONE 2013;8:e58371.

50. Chan K, Kan YW. Nrf2 is essential for protection against acute pulmonary injury in mice.Proc Natl Acad Sci USA 1999;96:12731–12736.

51. Dreger H, Westphal K, Weller A, Baumann G, Stangl V, Meiners S, Stangl K.Nrf2-dependent upregulation of antioxidative enzymes: a novel pathway for prote-asome inhibitor-mediated cardioprotection. Cardiovasc Res 2009;83:354–361.

52. Dowell JA, Johnson JA. Mechanisms of Nrf2 protection in astrocytes as identified byquantitative proteomics and siRNA screening. PLoS ONE 2013;8:e70163.

F. Zanetti et al.102by guest on N

ovember 12, 2014

Dow

nloaded from