Life Sciences 102 (2014) 88–97

Contents lists available at ScienceDirect

Life Sciences

j ourna l homepage: www.e lsev ie r .com/ locate / l i fesc ie

Cardiomyocyte specific peroxisome proliferator-activated receptor-αoverexpression leads to irreversible damage in ischemic murine heart

Georg D. Duerr a,⁎, Jan C. Heinemann a, Vanessa Arnoldi a, Andreas Feisst a, Julian Kley a, Alexander Ghanem b,Armin Welz a, Oliver Dewald a

a Department of Cardiac Surgery, University Clinical Center Bonn, Sigmund-Freud-Str. 25, 53105 Bonn, Germanyb Department of Cardiology, Pneomology and Angiology – Medizinische Klinik II, University Clinical Center Bonn, Sigmund-Freud-Str. 25, 53105 Bonn, Germany

⁎ Corresponding author at: Department of Cardiac SurUniversity Hospital Bonn – Universitätsklinikum Bonn, SBonn, Germany. Tel.: +49 228 287 11136; fax: +49 228

E-mail address: daniel.duerr@ukb.uni-bonn.de (G.D. D

http://dx.doi.org/10.1016/j.lfs.2014.03.0190024-3205/© 2014 Elsevier Inc. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 12 January 2014Accepted 10 March 2014Available online 21 March 2014

Keywords:DiabetesPPAR-αIschemia/reperfusionApoptosisChemokinesInflammationRemodeling

Aims: Peroxisome proliferator-activated receptor (PPAR)-α is downregulated in ischemic myocardium resultingin substrate switch from fatty acid oxidation to glucose utilization. Pharmacological PPAR-α activation leads toincreased fatty acid oxidation and myocardial lipotoxicity. The aim of our study was to investigate the role ofcardiomyocyte specific PPAR-α overexpression in myocardial adaptation to repetitive ischemic injury withoutmyocardial infarction.Main methods: Repetitive, brief I/R was performed in male and female MHC-PPAR-α overexpressing andwildtype-C57/Bl6 (WT)-mice, age 10–12 weeks, for 3 and 7 consecutive days. After echocardiography, theirhearts were excised for histology and gene/protein-expression measurements (Taqman/Western blot).Key findings: MHC-PPAR-α mice developed microinfarctions already after 3 days of repetitive I/R in contrast tointerstitial fibrosis in WT-mice. We found higher deposition of glycogen, increased apoptosis and dysfunctionalregulation of antioxidative mediators in MHC-PPAR-α mice. MHC-PPAR-α mice presented with maladaptationof myosin heavy chain isoforms andworse left ventricular dysfunction thanWT-mice.We found prolonged, che-

mokine-driven macrophage infiltration without induction of proinflammatory cytokines in MHC-PPAR-α mice.Persistent accumulation of myofibroblasts in microinfarctions indicated active remodeling resulting in scar for-mation in contrast to interstitial fibrosis without microinfarctions in WT-mice. However, MHC-PPAR-α heartshad only a weak induction of tenascin-C in contrast to its strong expression in WT-hearts.Significance: Cardiomyocyte-specific PPAR-α overexpression led to irreversible cardiomyocyte loss with deterio-rated ventricular function during brief, repetitive I/R episodes. We identified higher glycogen deposition, in-creased apoptosis, deranged antioxidative capacity and maladaptation of contractile elements as majorcontributors involved in the modulation of post-ischemic inflammation and remodeling.

© 2014 Elsevier Inc. All rights reserved.

Introduction

Diabetes mellitus is one of the major factors associated with cardio-vascular morbidities, e.g. coronary artery disease and heart failure(Asghar et al., 2009; Khavandi et al., 2009). At the same time, cardiovas-cular complications seem to be the leading cause ofmorbidity andmor-tality in diabetic patients (Spector, 1998; Kannel et al., 1974). Still, thepathophysiology of diabetic heart and its adaptive mechanisms arenot well understood. Depending on dietary and physiologic conditions,the substrate utilization in adult heart can change between fatty acidsand glucose in order to meet its energy demands (Neely et al., 1972;

gery – Klinik für Herzchirurgie,igmund-Freud-Str. 25, D-53105287 15591.uerr).

Finck et al., 2002). This switch becomes constrained in diabetic myocar-dium due to a dysregulation in insulin-controlled myocardial glucoseuptake and metabolism (Finck et al., 2002). In uncontrolled diabeticconditions, the heart utilizes almost exclusively mitochondrial fattyacid oxidation for ATP generation (Stanley et al., 1997; Rodrigueset al., 1995). This is associated with increased cardiac oxygen consump-tion and reactive oxygen species (ROS) production due to excessivelyhigh mitochondrial oxidative flux (Finck et al., 2002, 2003; Finck,2007). In addition, diabetic hearts import fatty acids, resulting in myo-cardial lipid accumulation, which predisposes to cardiomyocyte deathand cardiomyocyte dysfunction (Lopaschuk and Spafford, 1989; Chiuet al., 2001; Zhou et al., 2000). The repetitive ischemia and reperfusion(I/R) induced a reversible myocardial dysfunction and an increased glu-cose metabolism in rat hearts (Kim et al., 2003). However, streptozocininduced impaired glucose intolerance led to abnormal glycogen storagein rat myocardium resulting in apoptosis and ventricular dysfunction(Liang et al., 2011). Furthermore, a recent study reported that high

89G.D. Duerr et al. / Life Sciences 102 (2014) 88–97

dose glucose treatment of cardiomyoblasts (H9c2) and of isolatedcardiomyocytes results in cell death in vitro, whichwas partlymediatedby C-C chemokine ligand 2 (CCL2) production and inflammatory re-sponse (Younce et al., 2010).

Peroxisome proliferator-activated receptor-α (PPAR-α) is a ligand-activated nuclear transcription factor, which regulates many enzymesand proteins involved in fatty acid metabolism (Barger and Kelly,2000). Of all PPAR-isoforms (α, β/δ and γ), PPAR-α is highly expressedin the myocardium and plays an important role in transcriptional regu-lation of cardiac metabolism (Finck, 2007; Barger and Kelly, 2000). Inconditions of ischemia or pressure overload, the heart utilizesmore glu-cose and less fatty acids (Young et al., 2001; Razeghi et al., 2001a). Thisswitch in substrate preference is induced by downregulation of PPAR-αgene expression and activity (Barger and Kelly, 2000). There is evidencethat PPAR-αdownregulation and the switch in substrate utilization pre-serve myocardial function, while reactivation of PPAR-α leads to dys-function (Finck et al., 2002; Young et al., 2001; Dewald et al., 2005).PPAR-α seems not only to be involved in modulation of inflammatoryprocess (Smeets et al., 2007), but also to be related to enhanced ROSproduction in response to I/R in vivo and in vitro (Dewald et al., 2005;Sharifpanah et al., 2008).

We established a mouse model where brief, repetitive episodes of I/R lead to development of reversible ischemic cardiomyopathy (Dewaldet al., 2003). This model shows characteristics of human hibernatingmyocardium: transient inflammatory reaction, reversible interstitial fi-brosis and reversible regional ventricular dysfunction in the absenceofmyocardial infarction (Schulz and Heusch, 2000; Heusch et al., 2005).

We already described, that PPAR-α and its related genes are down-regulated during repetitive I/R, and that its activation with WY-14,643results in lipotoxicity and loss of function (Dewald et al., 2005). Otherstudies also described that activation of PPAR-α leads to elevated fattyacid oxidation and causes cardiac dysfunction with poor post-ischemicrecovery (Hafstad et al., 2009; Sambandam et al., 2006). However, con-troversial studies exist suggesting that the use of PPAR-α agonists maybe cardioprotective during I/R (Hafstad et al., 2009; Bulhak et al., 2006;Yue et al., 2003; Wayman et al., 2002; Devchand et al., 1996).

In order to avoid the potential pharmacological bias related to thePPAR-α agonists, we utilized cardiomyocyte specific PPAR-α overex-pressing (MHC-PPAR-α) mice (Finck et al., 2002) for investigation ofits impact on myocardial function, glycogen storage, apoptosis and in-flammatory response in our model of brief, repetitive I/R.

Materials and methods

MHC-PPAR-α and WT-mice

In the present study we used transgenic mice overexpressing PPAR-α under the control of cardiomyocyte specific promoter Myosin-6(Myh6, or α-MHC) on C57Bl/6J background (MHC-PPAR-α (Fincket al., 2002) These mice where kindly provided by Daniel P. Kelly(Sanford-Burnham Medical Research Institute, Orlando, FL, USA). TheirPPAR-α expression was reported to be approximately 130-fold higherthan in the wildtype (WT) mice. We used C57BL/6J WT-mice (CharlesRiver, Sulzfeld, Germany) as controls. All mouse experiments were per-formed on 20–25 g and 10–12 week old mice of both genders in accor-dance with an animal protocol approved by the local governmentalauthorities and according to the EU Directive 2010/63/EU for animalexperiments.

Model of brief, repetitive I/R in mouse hearts

Briefly, an 8–0 Prolene suture (Ethicon, Norderstedt, Germany) wasplaced around the left descending coronary artery (LAD) and storedsubcutaneously in an initial surgery as previously described (Nossuliet al., 2000). Mice recovered from initial surgery trauma for 7 days be-fore first I/R episode. One episode of 15 min ischemia (LAD occlusion)

was followed by reperfusion until the next day over 3 or 7 consecutivedays (scheme, Fig. 1A) (Dewald et al., 2003). Echocardiography wasperformed on the 7-day group. For histological experiments, wholehearts were excised, bloodwas rinsed out with ice-cold cardioplegic so-lution and hearts where fixated in zinc-paraformaldehyde (Z-fix, 4%;Anatech, Battle Creek, MI, USA). Hearts were excised, dissected freefrom great vessels and atria and immediately stored in RNA-later(Qiagen, Hilden, Germany) at 4 °C for no longer than 1 week until fur-ther processing for mRNA-studies.

Echocardiography

Echocardiography was performed with ultrasound system (HDI-5000, ATL–Phillips, Oceanside, CA, USA) using a linear array 15 MHztransducer (CL15-7). 2D-echo studies were performed on anesthetizedanimals (2% isoflurane for induction and 0.8–1.2% in 100% O2 for main-tenance of anesthesia) as described (Velten et al., 2012). Heart rate wasmonitored continuously in order to minimize cardiodepression-relatedeffects. Briefly, M-mode data was acquired in the parasternal short-axisview at the level of the papillary muscle. Fractional shortening (FS) as aglobal left ventricular function parameter, anterior wall thickening(AWT) as a regional parameter of left ventricular function and end dia-stolic diameter were measured and calculated.

Histology

Parafinized hearts were cut from basis to apex and 5 μm thick sec-tions from the region of the papillary muscles were stained. Afterdeparafinization and rehydration, the sections were stained by usinghematoxylin and eosin (HE, VWR, Darmstadt, Germany) for basic histo-pathology. Quantitative planimetric analysis of total collagen stainedarea as percentage of the total wall area was performed on picrosiriusred stained sections (SR, Sigma-Aldrich, Steinheim, Germany), as al-ready published (Dewald et al., 2003). We also performed quantitativeanalysis of microinfarctions by using picrosirius red staining by delin-eating the microinfarcted area and relating it to the collagen stainedarea, i.e. collagen in microinfarctions as a percentage of total collagenarea. Glycogen was stained with periodic acid Schiff (PAS) reagents(Sigma-Aldrich) according to themanufacturer's protocol. The specific-ity of glycogen stainingwas confirmed by pretreatmentwithα-amylasebefore PAS staining. Briefly, planimetric evaluation of collagen and PASpositive area was performed on two pictures taken from the anteriorleft ventricular wall with a magnification of 200× of one slide permouse.

Immunohistochemistry

Primary antibodies used in this study were alpha smooth muscleactin (α-SMAC) mouse monoclonal antibody (clone 1A4; Sigma,St. Louis, MO, USA), MAC-2 clone 3/38 rat antibody for macrophages(AXXORA, Lörrach, Germany), MCA771G (Ly-6B.2) clone 7/4 rat anti-mouse monoclonal antibody for neutrophils (MorphoSys, Oxford, UK),tenascin C rabbit anti-chicken polyclonal antibody (Chemicon, Temecula,CA,USA) andpolyclonal rabbit anti-mouse cleavedASP175 caspase-3 an-tibody (Cell signaling, Danvers, MS, USA). M.O.M. immunodetection kit(AXXORA) was used for mouse-derived antibodies. Vectastain Elite ABCkits (Vector, Burlingame, CA, USA) and diaminobenzidine (AXXORA)were used. For cleaved caspase-3 staining biotinylated secondary anti-body was used (DAKO, Hamburg, Germany) and its counterstainingwas performed by using Quick hemalaun kit (Vector). Briefly, quantita-tive planimetric analysis of α-SMAC positive area as percentage to ante-rior left ventricular wall was performed on two pictures taken from theanterior left ventricular wall with a magnification of 200× of one slideper mouse, as published before (Frangogiannis et al., 2000). Macrophageas well as neutrophil density were obtained bymanual cell count on fourpictures (magnification 400×) taken from the left ventricle of one slide

Fig. 1. PPAR-α overexpression leads to cardiomyocyte loss and pathological glycogen storage, andmaladaptation of contractile elements. (A) Brief scheme of the study design providing overviewon I/R-schedule and subsequent measurements: mice recover for 7 days (d) after implantation of the LAD ligature (LAD: red, Prolene® suture: blue, PE tube: gray), until repetitive I/R isperformed over 1, 3 or 7 days. HE-staining of representative left ventricular sections from7-days I/R show(B) interstitial cellular infiltration inWT-hearts, in contrast to (C) cardiomyocyteloss and microinfarctions (arrow) with increased cellularity in MHC-PPAR-α hearts. (D) PPAR-αmRNA expression in sham, 3 and 7-day I/R MHC-PPAR-αmice shows ~80-fold upregu-lation when compared to expression in WT shams. Periodic acid Schiff (PAS) staining reveals (E) low glycogen storage in WT-hearts, when compared to (F) a high glycogen storage inMHC-PPAR-α hearts after 7s I/R; (G) Planimetric evaluation of glycogen content. Scale bars represent 50 μm. RT-qPCR using Taqman®, mRNA-expression related to GAPDH andnormalized to sham using comparative ΔΔCt-method. * P ≤ 0.05 between the genotypes, # P ≤ 0.05 vs. respective sham; In (A): # P ≤ 0.05 vs. WT sham. WT: n = 8/group; MHC-PPAR-α: n = 6/group.

90 G.D. Duerr et al. / Life Sciences 102 (2014) 88–97

permouse. Cells showingno typicalmonocyte/macrophage or neutrophilmorphology were excluded.

TUNEL assay

TUNEL kit (In Situ Cell Death Detection Kit, POD, Roche, Mannheim,Germany) was used and TUNEL staining was performed according tothe manufacturer's protocol. Fluorescence signal was converted byusing diaminobenzidine (AXXORA) and counterstainingwas performedwith Quick hemalaun kit (Vector).

mRNA isolation and Taqman® RT-qPCR

mRNA-expression was determined by using Taqman® real-timequantitative PCR system (RT-qPCR, Applied Biosystems, Foster City,CA, USA). Total RNA was isolated by using a standard protocol byphenole/chloroform extraction (Trizol, Invitrogen). First-strand cDNAwas synthesized by using high capacity cDNA transcription kit (AppliedBiosystems) with random hexameric primers as described in themanufacturer's protocol. RT-qPCR was performed with 1/10 dilutedcDNA according to the manufacturer's instructions on an ABI Prism7900HT Sequence Detection System and SDS2.2 Software (Applied

Biosystems). All primers weremeasured by using FAM-TAMRA® chem-istry and the relative standard curve method. Dissociation curve analy-sis was performed to ascertain the amplification of a single PCR product.Target gene expression was normalized to shams and GAPDH.

Western blot analysis

Total protein was extracted from organic phase of Trizol-samplesafter isolatingmRNA fromaqueous phase bywashingwith ethanol, chlo-roform and deionized water (Life Technologies, Darmstadt, Germany),subsequent spinning (13,000 rpm) and dissolving pellet in 4% SDS. Pro-tein concentration was measured by bicinchoninic acid assay (BCA,Thermo Scientific Pierce, Rockford, IL, USA). Proteins where loaded onNuPAGE® 4–12% Bis-Tris protein gel (Life Technologies) and blottedon nitrocellulose membrane using iBlot technique (both Life Technolo-gies). After blocking with 5% milk powder in TBST-buffer, membraneswhere incubated with primary antibodies for GAPDH (1:1000 in 5%BSA in deionized water, Cell Signalling, Boston, MA, USA), HMOX-1(1:2000 in 5% BSA in deionized water, Abcam, Cambridge, UK) andRAC-1 (1:1000, 5% BSA in deionized, Abcam) overnight at 4 °C. After in-cubation with secondary antibody (HRP-linked, 1:2000 in 5% milk

91G.D. Duerr et al. / Life Sciences 102 (2014) 88–97

powder in TBST, Cell Signalling) for 1 h, proteins where detected by en-hanced chemiluminescence (Thermo Scientific Pierce).

Statistical analysis

Data were normally distributed and presented asmean± S.E.M. Sta-tistical analysiswas performedby one-wayANOVAwithNewman–Keulspost hoc testing using GraphPad PRISM (Version 5.0, Graphpad Inc., LaJolla, CA, USA). Unpaired t-test was used when comparing two groups.Differences with P ≤ 0.05 were considered statistically significant.

Results

Loss of cardiomyocytes, glycogen accumulation andmaladaptation to I/R inMHC-PPAR-α mice

Basic histology usingHE staining revealed predominantly interstitialcellular infiltrationwithout cardiomyocyte loss inWTmyocardiumafter7 days of repetitive I/R (Fig. 1B). MHC-PPAR-α mice presented at thesame time with cardiomyocyte loss in small delineated areas havinghigh cellularity – microinfarctions – in the ischemic left ventricularwall (Fig. 1C). We measured PPAR-α mRNA expression in our MHC-PPAR-α mice and found the expected high level in sham animals,which interestingly decreased during the repetitive I/R protocol(Fig. 1D).

PAS staining for glycogen showed discrete deposition in shams ofboth genotypes and in ischemic WT-hearts after 7-days I/R (Fig. 1E)when compared to its strong deposition in the areas of microinfarctionsin MHC-PPAR-α mice (Fig. 1F). Planimetric evaluation of PAS stainingrevealed significantly more glycogen deposition in ischemic myocardi-um of MHC-PPAR-α hearts (Fig. 1G), which may contribute to the lossof cardiomyocytes.

Increased apoptosis and antioxidative mediators during I/R

Cardiomyocyte apoptosis was evaluated by using cleaved caspase-3staining, which demonstrated only a few apoptotic cardiomyocytes inWT-hearts after 3 days of I/R (Fig. 2A). In contrast, MHC-PPAR-α heartspresented with an increased number of apoptotic cardiomyocytes,especially in microinfarcted areas already after 3 days of I/R (Fig. 2B).At the same time caspase-3 mRNA levels were significantly higher inMHC-PPAR-α hearts when compared to their respective shams andto WT samples (Fig. 2C). We found a similar expression pattern ofcaspase-8 (Fig. 2F). BCL2, an additional marker of mitochondrial in-volvement in apoptosis, was significantly upregulated after 3 days of I/R in MHC-PPAR-α mice, but then its expression decreased after 7 daysof I/R, and was comparable between the genotypes (Fig. 2I). In orderto evaluate progression of injured cardiomyocytes to apoptosis, weused TUNEL assay. TUNEL staining shows no dead cardiomyocytesin sections from WT-mice (Fig. 2D), but several nuclei with DNA-fragmentation indicating apoptosis of cardiomyocytes in MHC-PPAR-αhearts after 7-days I/R (Fig. 2E).

The mRNA-expression of antioxidative mediator heme oxygenase 1was comparable between the genotypes during I/R, while being signifi-cantly upregulated after 3 days of I/R when compared to the corre-sponding shams (Fig. 2G). A similar mRNA-expression pattern wasfound for glutathione peroxidase 1 (data not shown). In contrast, West-ern blot analysis showed diminished HMOX-1 protein level in MHC-PPAR-α hearts after 3 days of I/Rwhen compared to shams orWThearts(Fig. 2J). We also measured significantly higher mRNA-expression ofNADPH-oxidase regulator RAC-1 after 3 days of I/R in WT-hearts thanin the MHC-PPAR-α hearts (Fig. 2H). This was accompanied by higherRAC-1 protein production in WT-myocardium after 3 days of I/R com-pared to the MHC-PPAR-α hearts (Fig. 2K). These data suggest a dys-functional regulation of antioxidative protection mechanisms in MHC-

PPAR-a mice in response to ischemic stress, thus being one of possibleexplanations for cardiomyocyte loss and microinfarctions.

Left ventricular dysfunction

Echocardiography revealedworse left ventricular dysfunction inMHC-PPAR-α mice (Fig. 3A). Left ventricular end-diastolic diameter (LVEDD)was significantly increased in MHC-PPAR-α mice after 7s I/R (Fig. 3B).Both fractional shortening, as parameter of global function, and anteriorwall thickening, as regional function parameter, showed significantlymore dysfunction in MHC-PPAR-αmice after 7 days of I/R (Fig. 3C, D).

In order to evaluate the adaptation of contractile elements incardiomyocytes during repetitive ischemic episodes, we measuredmRNA-expression ofMHC iso-genes.WT-micemarkedly downregulatedthe higher ATP-consuming α-MHC isoform after 7 days, whereas theirMHC-PPAR-α counterparts increased its expression (Fig. 3E). In contrastto the WT-mice, the MHC-PPAR-α mice were unable to induce theenergeticallymore efficient β-MHC during I/R (Fig. 3F), thereby showingimpaired adaptation of contractile elements expression in these hearts.

Inflammatory response during repetitive I/R

We evaluated cellular infiltration and molecular markers of inflam-matory response in order to investigate the potential link betweenPPAR-α and inflammation. MAC-2 staining revealed interstitial macro-phage infiltration of ischemic WT myocardium after 7 days of I/R(Fig. 4A). In contrast, a strong macrophage accumulation was found inthe areas of cardiomyocyte loss in MHC-PPAR-α hearts after 7 days ofI/R (Fig. 4B). Quantitative evaluation of MAC-2 staining revealed signifi-cantly higher macrophage density after 3 and 7 days in MHC-PPAR-αhearts, when compared to the WT-hearts (Fig. 4C). We performed a dif-ferential analysis of the macrophage infiltration and found significantlyhigher cell density in interstitial space than in rarely foundmicroinfarctedareas in WT-hearts after 3 days of I/R (Fig. 4D). In contrast, the MHC-PPAR-α mice revealed only a few macrophages in interstitial space,while showing significantly higher density in microinfarctions after 3and – by trend – 7 days of I/R (Fig. 4E). In order to further investigateacute inflammatory reaction, we performed neutrophil staining. We ob-served significantly enhanced neutrophil density in ischemic areas ofMHC-PPAR-α hearts after 3-days I/R compared to sham and WT, beingstill higher after 7 days of I/R (Fig. 4F).

We measured expression of macrophage-related chemokines. CCL2presented with significantly lower induction in MHC-PPAR-α heartsafter 3 days of I/R (Fig. 5A). Its closely related CCL4 showed a comparableexpression pattern from theWT-mice during I/R (Fig. 5B). Also,mRNA in-duction of neutrophils chemotactic chemokine CCL3was comparable be-tween the genotypes, even though its expressionwas significantly higherbetween shams and 3days of I/R inMHC-PPAR-αmice (Fig. 5C). Interest-ingly, the significant upregulation of pro-inflammatory cytokine TNF-αin WT animals after 3 days was paired by the absence of its inductioninMHC-PPAR-αmice (Fig. 5D).We found a comparable expression pat-tern for IL-1β without a significant difference (data not shown). Anti-inflammatory cytokine IL-10 presented also with a comparable expres-sion pattern between the genotypes, thus showing potential for timelyresolution of inflammatory response in both mice groups (Fig. 5E). But,osteopontin-1 (OPN-1) expression was significantly increased after3 days of I/R in WT-mice. This is suggesting an early onset of macro-phage maturation in WT-mice associated with timely resolution of in-flammatory reaction and myocardial remodeling. Consequently, theminor induction of OPN-1 in MHC-PPAR-α hearts supports the persis-tent macrophage activity of the microinfarctions (Fig. 5F).

Myofibroblast differentiation and remodeling

In order to investigate myocardial remodeling process, heart sectionsstained for collagen producing myofibroblasts showed only weak signals

Fig. 2.MHC-PPAR-α hearts show enhanced apoptosis and partially preserved antioxidative protection during I/R. Cleaved caspase-3 stainings of representative left ventricular sections show(A) low apoptosis inWT-hearts in comparison to (B) the enhanced apoptosis inMHC-PPAR-α hearts after 3 days (d) of I/R.mRNAexpression of apoptosis-relatedmediators (C) caspase-3,(F) caspase-8, and (I) BCL2. TUNEL staining shows no dead cardiomyocytes in sections from (D) WT-mice, but several nuclei with DNA-fragmentation (arrows) indicating apoptosis ofcardiomyocytes in (E) MHC-PPAR-α hearts after 7-days I/R. (G) mRNA expression and (J) Western blot analysis of heme oxygenase 1 (HMOX-1) and RAC-1 (H,K) during repetitive I/Rprotocol. Scale bars represent 50 μm in A–E, 25 μm in the inset of E. RT-qPCR using Taqman®, mRNA-expression related to GAPDH and normalized to sham using comparative ΔΔCt-method. * P ≤ 0.05, # P ≤ 0.05 vs. respective sham; WT: n = 8/group; MHC-PPAR-α: n = 6/group.

92 G.D. Duerr et al. / Life Sciences 102 (2014) 88–97

inWT-mice after 7 days of I/R (Fig. 6A). In contrast, themicroinfarctions ofMHC-PPAR-α hearts contained numerous myofibroblasts (Fig. 6B), in-dicating active remodeling in these mice. Planimetric evaluation of α-SMAC-stained area showed a significantly higher percentage of thetotal area inMHC-PPAR-α hearts after 3 and 7 days of I/R (Fig. 6C). Sub-sequently, interstitial collagen deposition was found in WT-mice after7 days of I/R (Fig. 6D), whereas fibrotic depositions were concentratedto the microinfarctions in MHC-PPAR-α hearts (Fig. 6E). Quantificationof the picrosirius red staining revealed a constant and comparable in-crease in total collagen area in both genotypes during the I/R protocol(Fig. 6F). Differential planimetric analysis of microinfarctions showed~18% of collagen stained area inMHC-PPAR-α hearts being significantlyhigher than ~3% in WT-hearts (Fig. 6G).

Staining for early remodeling marker tenascin C presented fineinterstitial signals in ischemic left ventricle in WT-hearts after 7 daysof I/R (Fig. 6H), which could not be detected in MHC-PPAR-α mice at

the same time (Fig. 6I). Tenascin mRNA-expression was significantlylower and its induction shorter inMHC-PPAR-αmice (Fig. 6J). Taken to-gether,MHC-PPAR-αmice seem to have a prolonged, active remodelingdue to the occurrence of microinfarctions.

Discussion

In this study, we provide novel insights into cardioprotective mech-anisms involving PPAR-α in ischemic murine myocardium. The cardio-myocyte specific overexpression of PPAR-α was associated withincreased glycogen deposition and apoptosis, dysfunction of antioxida-tive mediators and maladaptation of contractile elements expression,subsequently leading to irreversible loss of cardiomyocytes with pro-found left ventricular dysfunction. In consequence, the inflammatoryresponsewas prolonged and accompanied by activemyocardial remod-eling due to the scar formation.

Fig. 3.Deteriorated cardiac function inMHC-PPAR-αmice. (A) Representative short axisM-mode echocardiographs fromWT- (red frame) andMHC-PPAR-αmice (green frame) after 7 days(d) of I/R. Functional parameters of (B) left ventricular end-diastolic diameter (LVEDD), (C) fractional shortening (FS), and (D) anterior wall thickening (AWT) demonstrate a significantlyworse global and regional left ventricular dysfunction in MHC-PPAR-αmice. mRNA-expression of myosin heavy chain (MHC) iso-genes (E) α and (F) β during the I/R protocol. RT-qPCRusing Taqman®,mRNA-expression related to GAPDH and normalized to sham using comparativeΔΔCt-method. * P≤ 0.05, # P≤ 0.05 vs. respective sham;WT: n= 8/group; MHC-PPAR-α: n = 6/group.

93G.D. Duerr et al. / Life Sciences 102 (2014) 88–97

MHC-PPAR-α cardiomyocytes fail to adapt to repetitive ischemic stress

MHC-PPAR-α mice have ~130-fold increase in PPAR-α expressionleading to a condition comparable with diabetic heart (Finck et al.,2002). They show increasedmyocardial uptake of fatty acids, decreasedexpression of genes involved in glucose utilization pathways, and sub-stantially lower uptake of 18F-desoxyglucose. Our brief, repetitive I/R

Fig. 4.MHC-PPAR-α hearts show a different pattern of macrophage infiltration.MAC-2 staining ofstitial macrophage infiltration (arrow) inWT-hearts, in contrast to (B) increased macrophage dMAC-2+ cells revealed increased macrophage density after 3 and 7-days I/R. Macrophage disinto microinfarctions in (D) WT-hearts vs. (E) MHC-PPAR-α hearts. (F) Quantification of M50 μm. * P ≤ 0.05, # P ≤ 0.05 vs. respective sham; WT: n = 8/group; MHC-PPAR-α: n = 6/gro

caused downregulation of PPAR-α in these mice, which has been dem-onstrated tomodulate the switch from fatty acid to glucose utilization inorder to preserve cardiac function in WT-mice during repetitive I/R, inrat hearts subjected to pressure-overload and in failing humanmyocar-dium (Young et al., 2001; Dewald et al., 2005; Razeghi et al., 2001b). Atthe same time we could confirm the detrimental effect of PPAR-α acti-vation in ischemic myocardium with microinfarctions, which were

representative left ventricular sections after 7 days (d) of I/R show (A) predominant inter-ensity inmicroinfarctions (white star) inMHC-PPAR-αmyocardium. (C) Quantification oftribution in ischemic myocardium differentiated by infiltration into interstitial space orCA771G+ cells showed neutrophil density after 3 and 7-days I/R. Scale bars representup.

Fig. 5.mRNA expression profile of inflammatory mediators.mRNA-expression of macrophage-related chemokines (A) CCL2 and (B) CCL4, and of (C) neutrophil-related chemokine CCL3 inWT and MHC-PPAR-α hearts after 3 and 7 days (d) of I/R. Corresponding mRNA-expression of cytokines (D) TNF-α, (E) IL-10 and (F) Osteopontin 1 (OPN-1). RT-qPCR using Taqman®,mRNA-expression related to GAPDH and normalized to sham using comparativeΔΔCt-method. * P≤ 0.05, # P≤ 0.05 vs. respective sham;WT: n= 8/group;MHC-PPAR-α: n= 6/group.

94 G.D. Duerr et al. / Life Sciences 102 (2014) 88–97

accompanied by lipotoxicity when using PPAR-α agonist WY-14,643(Dewald et al., 2005). Activation of PPAR-α has been associated withincrease in myocardial triglyceride deposition, elevated fatty acidoxidation, decreased glucose utilization and impaired cardiac function

Fig. 6. Prolonged myocardial remodeling in MHC-PPAR-αmice. Representative left ventricular secinterstitialmyofibroblasts inWT-hearts (arrowhead indicatesα-SMACpositive staining in arterand microinfarctions (*) of MHC-PPAR-α hearts. (C) Planimetric analysis of α-SMAC stained a(D) interstitial fibrosis (arrow; arrowhead: vessel) in WT-heart, in contrast to (E) a dense fibMHC-PPAR-α hearts. While planimetric analysis of picrosirius red staining shows (F) compdeposition in (G) microinfarcted areas is significantly higher in MHC-PPAR-α hearts. Represeninterstitial distribution in WT-heart (arrow), and (I) no interstitial staining or signals in micro3 and 7 days I/R (J). Scale bars: A–B 100 μm, D–E 100 μm, H–I 50 μm. RT-qPCR using Taqman®

method. * P ≤ 0.05, # P ≤ 0.05 vs. respective sham; WT: n = 8/group; MHC-PPAR-α: n = 6/g

(Dewald et al., 2005; Hafstad et al., 2009; Sambandam et al., 2006). Inaddition, we present here increased glycogen deposition in ischemicareas and predominantly in microinfarctions of MHC-PPAR-α hearts,which seems to present an additional burden on the already stressed

tions stained for myofibroblast markerα-SMAC after 7 days (d) of I/R shows (A) only fewioles), whereas (B) numerousmyofibroblasts (arrow)were found in ischemicmyocardiumrea after 3 and 7 days I/R. Representative picrosirius red stainings after 7 days I/R showsrotic deposition in microinfarctions (arrow, dotted line indicates microinfarcted area) inarable total collagen stained area between the genotypes after 7 days I/R, the collagentative staining for early remodeling marker tenascin C after 7 days I/R reveals (H) its fineinfarcted areas (star) in MHC-PPAR-α myocardium. mRNA expression of tenascin C after, mRNA-expression related to GAPDH and normalized to sham using comparative ΔΔCt-roup.

95G.D. Duerr et al. / Life Sciences 102 (2014) 88–97

cardiomyocytes. Previous studies reported that elevated cardiac glyco-gen is associated with increased cardiomyocyte apoptosis (Liang et al.,2011; Younce et al., 2010). In consequence we examined apoptosispathways and found evidence for increased cell death in MHC-PPAR-αmice resulting in microinfarctions already after 3 days of repetitive I/R.Therefore not only the lipotoxicity, but also glycogen deposition seemsto contribute to the irreversible loss of cardiomyocytes, thus preventingthe development of reversible ischemic cardiomyopathy in our model,which shares characteristics of human hibernating myocardium.

Furthermore, ischemic stress and oxidative burst following reperfu-sion induce apoptosis of cardiomyocytes due to the generation of toxicROS by initiating formation of single strand DNA breaks (Yeh et al.,2003; Akki et al., 2009). Since repetitive I/R is also leading to increasedproduction of ROS (Dewald et al., 2003),wemeasured expression of an-tioxidative mediators and found a preserved mRNA-induction of themin MHC-PPAR-α mice. In contrast, protein expression data of HMOX-1revealed a dysfunctional antioxidative protection of ischemic MHC-PPAR-α cardiomyocytes. RAC-1, a GTPase of the Rho family, is involvedin pleiotropic cellular events including regulation of cell growth, cellshape, adhesion, cytoskeletal reorganization, and activation of theNAD(P)H oxidase resulting in ROS release (Akki et al., 2009; Aktories,1997; Hall, 1998; Dusi et al., 1995; Rinckel et al., 1999). In the ischemicmurine heart, inhibition of RAC-1 significantly reduced ROS productionand was associated with cardioprotection (Wang et al., 2012). Our datashow an absence in RAC-1 mRNA- and protein expression in MHC-PPAR-α mice in contrast to their increase in WT-hearts during I/R,which could be interpreted as a compensatory mechanism for suppres-sion of ROS production in MHC-PPAR-α hearts.

The expression pattern of the contractile elements, e.g. MHC-isoforms, also determines cardiomyocyte survival in pathological condi-tions. In conditions of hypoxia, I/R or pressure overload there is a switchin isoform expression resulting in an increase in the ratio of β-MHC toα-MHC expression (Dewald et al., 2005; Razeghi et al., 2001b). MHC-PPAR-αmice are unable to switch their expression from high ATP con-sumingα-MHC to the less ATP demanding and thereforemore econom-ical β-MHC isoform, thus indicating cardiomyocyte maladaptationduring repetitive I/R episodes. In contrast to WT-hearts, the MHC-PPAR-α hearts even induced α-MHC expression in response to I/R,while their β-MHCwas not induced at all. Taken together, the combina-tion of all these effects contributed to the cardiomyocyte loss and result-ed in significantly worse left ventricular function of MHC-PPAR-αmice.We assume, that MHC-PPAR-αmice have no perspective for functionalrecovery after discontinuation of the I/R-protocol, which was found inWT-mice (Dewald et al., 2003), or after revascularization of humanhibernating myocardium (Frangogiannis et al., 2002a).

PPAR-α overexpression modulates inflammation and its cellular response

Since repetitive I/R leads to a transient inflammatory reaction(Dewald et al., 2003), we further elucidated the molecular and cellularevents of inflammatory response associated with development ofmicroinfarctions in MHC-PPAR-αmice. We found a significantly higherand prolonged macrophage accumulation in MHC-PPAR-α hearts after7 days of I/Rwhen compared toWT, indicating a persistent inflammato-ry reaction. Interestingly, the macrophages showed a strong affiliationto invade the microinfarctions, due to the apoptotic cardiomyocytesand the need for their removal. Even though we did not evaluate themacrophage subpopulations, onemay assume that the classically activat-edmacrophages are the predominant subtype during our observationpe-riod. This assumption is based on the continuous loss of cardiomyocytesduring repetitive I/R and on results of other groups (Nahrendorf et al.,2007). Another aspect of inflammatory reaction, i.e. neutrophil infiltra-tion of ischemic myocardium, also supports this explanation. Interest-ingly, their most potent chemoattractant chemokine CCL2 showedsignificantly lower induction inMHC-PPAR-α hearts, but the closely re-lated CCL4 seems to have compensated for this fact. The induction of

CCL3 acts only as a weak monocytic chemoattractant while attractingpredominantly neutrophils, which are also needed for removal of celldebris. Possible interactions of these chemokines with PPAR-α stillneed to be investigated.

Moreover, pro-inflammatory cytokines TNF-α and IL-1β, which nor-mally initiate the inflammatory cascade in infarcted myocardium(Frangogiannis et al., 2002b), were not induced in response to repetitiveI/R in MHC-PPAR-α mice, thus suggesting a potential suppression ofthemwhen PPAR-α system is activated. Still, the initiation of resolutionof the inflammatory response, beingmostly driven by IL-10 expression,seems to be intact under cardiomyocyte specific PPAR-α overexpres-sion. On the other hand, significantly lower expression of OPN-1 inMHC-PPAR-α hearts, a marker of macrophage maturation prior totheir apoptosis and matricellular protein (Rittling, 2011; Renkl et al.,2005), may be associated with prolonged macrophage activity inmicroinfarctions, and points to prolonged myocardial remodeling. Toour knowledge, this is the first report on cellular and molecular inflam-matory response in ischemic hearts of MHC-PPAR-α mice (Fig. 7).

PPAR-α overexpression leads to prolonged remodeling

Different aspects ofmyocardial remodelingwere investigated in thisstudy. Myofibroblasts are responsible for synthesis and depositionof extracellular matrix components such as collagen types I and III.(Dewald et al., 2003, 2004; Gabbiani, 2003) The replacement fibrosis,e.g. after myocardial infarction, leads to myofibroblasts differentiationand strong accumulation in damaged area during scar formation(Duerr et al., 2011). At a time point where WT-mice already showed acompleted interstitial collagen deposition, the MHC-PPAR-α heartspresented with significantly more myofibroblasts, being located inmicroinfarctions and indicating an active myocardial remodeling. Thestrong concentration of collagen deposition in microinfarction ofMHC-PPAR-α hearts was accompanied by less interstitial collagen de-position, when compared with WT-mice bearing only a few spotsbeing identified as microinfarctions. These findings are solely relatedto the development of microinfarctions and subsequent scar formation.A possible direct link between cardiomyocyte specific PPAR-α activa-tion and prolonged myocardial remodeling may be assumed based ona weak expression of tenascin C, which is an early remodeling marker.Still, this remains speculative until further studies have been done to in-vestigate this.

This study did not investigate long-term effects of cardiac specificPPAR-α overexpression after discontinuation of repetitive I/R. Still, itcan be assumed that in contrast to full normalization of ventricular func-tion inWT-mice (Dewald et al., 2003), this is not to be expected inMHC-PPAR-α mice due to the microinfarctions and the diabetic condition oftheir hearts. Even though we did not report protein data, most of themeasured genes are tightly regulated at the transcription level, andthis should well reflect their action on protein level. In addition, the im-munohistochemistry provided supportive information in this respect.

Conclusions

Application of brief, repetitive I/R in MHC-PPAR-α mice providednovel insights in complex mechanisms underlying cardioprotectionandmyocardial remodeling. Overexpression of PPAR-α led to increaseddeposition of glycogen, higher apoptosis rate, deranged capacity of anti-oxidative enzymes and changes in expression pattern of contractileelements in ischemic myocardium and resulted in irreversible loss ofcardiomyocytes with deteriorated left ventricular function. MHC-PPAR-α hearts presented with a chemokine driven, prolonged macro-phage infiltration and with timely resolution of inflammatory response.The subsequent remodeling was prolonged and associated with a weakinduction of tenascin C. These findings suggest a new possible link be-tween PPAR-α system, inflammatory reaction andmyocardial remodel-ing in a diabetic heart.

Fig. 7. Proposed interactions of PPAR-α and apoptosis, inflammation and myocardial remodeling in a mouse model of brief, repetitive I/R. The findings provided in this study (black) are placedinto context of the previously published data (blue) and the adaptation mechanisms to the reactive oxygen species (ROS) production (orange) during repetitive I/R.

96 G.D. Duerr et al. / Life Sciences 102 (2014) 88–97

Funding

None.

Conflict of interest statement

The authors declare that there are no conflicts of interest.

Acknowledgment

Wewould like to thank Daniel P. Kelly, MD, Sanford-BurnhamMed-ical Research Institute at Lake Nona, Orlando, FL, USA, for providing uswith the MHC-PPAR-α mice, and Christine Peigney for her excellenttechnical assistance.

References

Sambandam N, et al. Chronic activation of PPARalpha is detrimental to cardiac recoveryafter ischemia. Am J Physiol Heart Circ Physiol 2006;290(1):H87–95.

Aktories K. Bacterial toxins that target Rho proteins. J Clin Invest 1997;99(5):827–9.Asghar O, et al. Diabetic cardiomyopathy. Clin Sci (Lond) 2009;116(10):741–60.Barger PM, Kelly DP. PPAR signaling in the control of cardiac energy metabolism. Trends

Cardiovasc Med 2000;10(6):238–45.Bulhak AA, et al. Protection against myocardial ischaemia/reperfusion injury by PPAR-

alpha activation is related to production of nitric oxide and endothelin-1. Basic ResCardiol 2006;101(3):244–52.

Chiu HC, et al. A novel mouse model of lipotoxic cardiomyopathy. J Clin Invest 2001;107(7):813–22.

Devchand PR, et al. The PPARalpha-leukotriene B4 pathway to inflammation control. Na-ture 1996;384(6604):39–43.

Dewald O, et al. Development of murine ischemic cardiomyopathy is associated with atransient inflammatory reaction and depends on reactive oxygen species. Proc NatlAcad Sci U S A 2003;100(5):2700–5.

Dewald O, et al. A murinemodel of ischemic cardiomyopathy induced by repetitive ische-mia and reperfusion. Thorac Cardiovasc Surg 2004;52(5):305–11.

Dewald O, et al. Downregulation of peroxisome proliferator-activated receptor-alphagene expression in a mouse model of ischemic cardiomyopathy is dependent onreactive oxygen species and prevents lipotoxicity. Circulation 2005;112(3):407–15.

Duerr GD, et al. Comparison of myocardial remodeling between cryoinfarction and reper-fused infarction in mice. J Biomed Biotechnol 2011:961298.

Dusi S, Donini M, Rossi F. Mechanisms of NADPH oxidase activation in human neutro-phils: p67phox is required for the translocation of rac 1 but not of rac 2 from cytosolto the membranes. Biochem J 1995;308(Pt 3):991–4.

Finck BN. The PPAR regulatory system in cardiac physiology and disease. Cardiovasc Res2007;73(2):269–77.

Finck BN, et al. The cardiac phenotype induced by PPARalpha overexpression mimics thatcaused by diabetes mellitus. J Clin Invest 2002;109(1):121–30.

Finck BN, et al. A critical role for PPARalpha-mediated lipotoxicity in the pathogenesis ofdiabetic cardiomyopathy: modulation by dietary fat content. Proc Natl Acad Sci U S A2003;100(3):1226–31.

Frangogiannis NG, Michael LH, Entman ML. Myofibroblasts in reperfused myocardialinfarcts express the embryonic form of smooth muscle myosin heavy chain(SMemb). Cardiovasc Res 2000;48(1):89–100.

Frangogiannis NG, et al. Active interstitial remodeling: an important process in the hiber-nating human myocardium. J Am Coll Cardiol 2002a;39(9):1468–74.

Frangogiannis NG, Smith CW, Entman ML. The inflammatory response in myocardialinfarction. Cardiovasc Res 2002b;53(1):31–47.

Gabbiani G. The myofibroblast in wound healing and fibrocontractive diseases. J Pathol2003;200(4):500–3.

Hafstad AD, et al. Cardiac peroxisome proliferator-activated receptor-alpha activationcauses increased fatty acid oxidation, reducing efficiency and post-ischaemic func-tional loss. Cardiovasc Res 2009;83(3):519–26.

Hall A. Rho GTPases and the actin cytoskeleton. Science 1998;279(5350):509–14.Heusch G, Schulz R, Rahimtoola SH. Myocardial hibernation: a delicate balance. Am J

Physiol Heart Circ Physiol 2005;288(3):H984–99.Akki A, et al. NADPH oxidase signaling and cardiac myocyte function. J Mol Cell Cardiol

2009;47(1):15–22.Kannel WB, Hjortland M, Castelli WP. Role of diabetes in congestive heart failure: the

Framingham study. Am J Cardiol 1974;34(1):29–34.Khavandi K, et al. Diabetic cardiomyopathy—a distinct disease? Best Pract Res Clin

Endocrinol Metab 2009;23(3):347–60.Kim SJ, et al. Persistent stunning induces myocardial hibernation and protection: flow/

function and metabolic mechanisms. Circ Res 2003;92(11):1233–9.Liang JL, et al. Effect of impaired glucose tolerance on cardiac dysfunction in a rat model of

prediabetes. Chin Med J (Engl) 2011;124(5):734–9.Lopaschuk GD, Spafford M. Response of isolated working hearts to fatty acids and carni-

tine palmitoyltransferase I inhibition during reduction of coronary flow in acutelyand chronically diabetic rats. Circ Res 1989;65(2):378–87.

Nahrendorf M, et al. The healing myocardium sequentially mobilizes two monocyte sub-sets with divergent and complementary functions. J ExpMed 2007;204(12):3037–47.

Neely JR, Rovetto MJ, Oram JF. Myocardial utilization of carbohydrate and lipids. ProgCardiovasc Dis 1972;15(3):289–329.

97G.D. Duerr et al. / Life Sciences 102 (2014) 88–97

Nossuli TO, et al. A chronic mousemodel of myocardial ischemia–reperfusion: essential incytokine studies. Am J Physiol Heart Circ Physiol 2000;278(4):H1049–55.

Razeghi P, et al. Hypoxia in vivo decreases peroxisome proliferator-activated receptoralpha-regulated gene expression in rat heart. Biochem Biophys Res Commun2001a;287(1):5–10.

Razeghi P, et al. Metabolic gene expression in fetal and failing human heart. Circulation2001b;104(24):2923–31.

Renkl AC, et al. Osteopontin functionally activates dendritic cells and induces their differ-entiation toward a Th1-polarizing phenotype. Blood 2005;106(3):946–55.

Rinckel LA, et al. Rac1 disrupts p67phox/p40phox binding: a novel role for Rac in NADPHoxidase activation. Biochem Biophys Res Commun 1999;263(1):118–22.

Rittling SR. Osteopontin in macrophage function. Expert Rev Mol Med 2011;13:e15.Rodrigues B, Cam MC, McNeill JH. Myocardial substrate metabolism: implications for

diabetic cardiomyopathy. J Mol Cell Cardiol 1995;27(1):169–79.Schulz R, Heusch G. Hibernating myocardium. Heart 2000;84(6):587–94.Sharifpanah F, et al. Peroxisome proliferator-activated receptor alpha agonists enhance

cardiomyogenesis of mouse ES cells by utilization of a reactive oxygen species-dependent mechanism. Stem Cells 2008;26(1):64–71.

Smeets PJ, et al. Peroxisome proliferator-activated receptors and inflammation: take it toheart. Acta Physiol (Oxf) 2007;191(3):171–88.

Spector KS. Diabetic cardiomyopathy. Clin Cardiol 1998;21(12):885–7.

Stanley WC, Lopaschuk GD, McCormack JG. Regulation of energy substrate metabolism inthe diabetic heart. Cardiovasc Res 1997;34(1):25–33.

Velten M, et al. Priming with synthetic oligonucleotides attenuates pressure overload-induced inflammation and cardiac hypertrophy in mice. Cardiovasc Res 2012;96(3):422–32.

Wang X, et al. Loss of the miR-144/451 cluster impairs ischaemic preconditioning-mediated cardioprotection by targeting Rac-1. Cardiovasc Res 2012;94(2):379–90.

Wayman NS, et al. Ligands of the peroxisome proliferator-activated receptors (PPAR-gamma andPPAR-alpha) reducemyocardial infarct size. FASEB J 2002;16(9):1027–40.

Yeh CH, et al. Continuous tepid blood cardioplegia can preserve coronary endotheliumandameliorate the occurrence of cardiomyocyte apoptosis. Chest 2003;123(5):1647–54.

Younce CW, Wang K, Kolattukudy PE. Hyperglycaemia-induced cardiomyocyte death ismediated via MCP-1 production and induction of a novel zinc-finger protein MCPIP.Cardiovasc Res 2010;87(4):665–74.

Young ME, et al. Reactivation of peroxisome proliferator-activated receptor alpha is asso-ciated with contractile dysfunction in hypertrophied rat heart. J Biol Chem 2001;276(48):44390–5.

Yue TL, et al. Activation of peroxisome proliferator-activated receptor-alpha protects theheart from ischemia/reperfusion injury. Circulation 2003;108(19):2393–9.

Zhou YT, et al. Lipotoxic heart disease in obese rats: implications for human obesity. ProcNatl Acad Sci U S A 2000;97(4):1784–9.