Life Sciences 85 (2009) 77–84

Contents lists available at ScienceDirect

Life Sciences

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

Cardiac impairment in rabbits fed a high-fat diet is counteracted bydehydroepiandrosterone supplementation

M. Aragno a,⁎, G. Meineri b, I. Vercellinatto a, P. Bardini a, S. Raimondo e, P.G. Peiretti c, A. Vercelli d, G. Alloatti f,C.E. Tomasinelli a, O. Danni a, G. Boccuzzi g

a Department of Experimental Medicine and Oncology, University of Turin, Corso Raffaello 30, 10125 Turin, Italyb Department of Animal Production, Epidemiology and Ecology, University of Turin, Via L. da Vinci 44, 10095 Grugliasco, Italyc Institute of Food Science and Production, NRC, Via L. da Vinci 44, 10095 Grugliasco, Italyd Associated Veterinary Ambulatory, Corso Traiano 99, 10135, Turin, Italye Department of Clinical and Biological Sciences, University of Turin, Regione Gonzole 10, San Luigi Hospital, 10043 Orbassano, Italyf Department of Animal and Human Biology, University of Turin, Via Accademia Albertina 13, 10123 Torino, Italyg Department of Clinical Pathophysiology, University of Turin, Via Genova 3, 10126 Turin, Italy

High fat diet

⁎ Corresponding author. Tel.: +39 0116707758; fax: +E-mail address: manuela.aragno@unito.it (M. Aragno

0024-3205/$ – see front matter © 2009 Elsevier Inc. Adoi:10.1016/j.lfs.2009.04.020

a b s t r a c t

a r t i c l e i n f o

Article history:

Aims: The biochemical and Received 16 December 2008Accepted 30 April 2009

Keywords:Oxidative stressDehydroepiandrosteroneCardiac impairmentRabbitsMyosin

structural cardiac oxidative-dependent damage induced by high-fat (HF) diet wasexamined in a rabbitmodel, togetherwith the role of dehydroepiandrosterone (DHEA) in contrasting tissue damage.Main methods: New Zealand white rabbits fed a HF diet supplemented or not with DHEA (0.02%) were utilized for12 weeks. Oxidative stress, inflammatory and necrosis parameters, fatty deposition, heavy-chain myosin isoforms(MHC) expression and papillary muscle functionality were examined in the left ventricle of rabbits.Key findings: Rabbits fed a HF diet that showed hyperglycemia, insulin resistance and dyslipidemia together withincrease of oxidative stress and of advanced end-glycation product levels have been observed. Concerning pro-inflammatory insults, therewas increased p65-NFkB activation and increased tumor necrosis factor-alpha and C-reactive protein expressions. Cellular damage induced by the HF diet was detected through the switch of

expression of MHC isoforms, indicating impairment of cardiac contractility, confirmed by altered of basalparameters of papillary muscle functionality. Rabbits fed the HF diet supplemented with DHEA showed a partialreduction of oxidative stress and the inflammatory state. Cardiac necrosis, the shift of MHC isoforms, and cardiacfunctionality, were also partially counteracted.Significance:Rabbits fedwith aHFdiet showeda beneficial effectwhen low-doseDHEAwas added to the diet. Thesteroid,without affectinghighplasmaglucose level or insulin resistance, restoredoxidative balance, lowered lipidlevels and inflammation insults, preventing cellular and functional alterations of cardiac tissue and thus delayingthe onset of cardiac damage.

© 2009 Elsevier Inc. All rights reserved.

Introduction

Diabetic cardiomyopathy is characterized by systolic and diastolicdysfunction and has been reported in diabetic patients with noischemic, valve or hypertensive heart disease. Its developmentincludes metabolic disturbances, small-vessel diseases, autonomicdysfunction, insulin resistance and myocardial fibrosis (Brownlee2005; Farhangkhoee et al. 2005). Recently, an important role in itspathophysiology has been attributed to the generation of reactiveoxygen species, which activates a number of secondary-messengerpathways, eventually leading to cardiac dysfunction (Hartog et al.2007; Masoudi and Inzucchi 2007; Wold et al. 2005).

39 0116707753.).

ll rights reserved.

We recently showed that free-radical overproduction appearsearly in human type 2 diabetes (Brignardello et al. 2007) and that, in arat model of type 1 diabetes, oxidative damage plays a key role in theearly development of cardiomyopathy (Aragno et al. 2006). Antiox-idants might counteract insulin resistance associated with type 2diabetes and cardiovascular diseases (Robertson 2006; Davì et al.1999; Minamiyama et al. 2008).

The role of dehydroepiandrosterone (DHEA), a compound thatpossesses multi-targeted antioxidant properties (Simoncini andGenazzani 2007; Tchernof and Labrie 2004; Aragno et al. 2004,2005; Brignardello et al. 2000), in the cardiovascular system has beenhighlighted by recent reports showing that the human heartsynthesizes DHEA, that its production is suppressed in the failingheart, and that plasma levels of the sulfate conjugate of DHEAdecreasein patients with chronic heart failure in proportion to the severity ofthe condition (Moriyama et al. 2000). We reported elsewhere that, in

78 M. Aragno et al. / Life Sciences 85 (2009) 77–84

streptozotocin (STZ) rats, treatment with DHEA, prevents myocardialdamage induced by oxidative stress, while avoiding impairment ofcardiac myogenic factors and the switch to myosin heavy-chainexpression (Aragno et al. 2008). DHEA also counteracts oxidativeimbalance and advanced glycation end-product (AGE) formation intype 2 diabetic patients (Brignardello et al. 2007).

The present study targets diabetic cardiomyopathy, in a model oftype 2 diabetes induced by a high-fat (HF) diet in rabbits, a speciesthat is highly susceptible to cardiovascular damage (Russell andProctor 2006). It also examined the downstream signaling activatedby oxidative stress in the left ventricle of STZ diabetic rats, and theeffects of DHEA treatment on biochemical and structural changes inthe left ventricle and on myocardial dysfunction.

Materials and methods

Reagents

Unless otherwise indicated, all compounds were purchased fromSigma Chemical Co. (St. Louis, MO, USA), antibodies from Chemicon(Millipore, Billerica, MA, USA), and primers from Tib MolBiol s.r.l.(Genoa, Italy). Lipid profile was determined using reagents kits fromHospitex Diagnostics (Florence, Italy).

Animal treatment

Male New Zealand white rabbits, 15 weeks old (Harlan-Italy, Udine,Italy) weighing 3.0–3.5 kg were cared for in compliance with theDeclaration of Helsinki as revised in 1983, the Italian Ministry of HealthGuidelines (no. 86/609/EEC) and with the Principles of LaboratoryAnimal Care (NIH no. 85-23, revised 1985). The rabbits were acclima-tized for 2 weeks prior to the experiment in a roomwith 12 h light darkcycle, individually housed in stainless steel cages in a temperature- andhumidity-controlled room (23±3 °C, 50±5%) and fed 100 g per day ofstandard rabbit diet (sec. Sherman, cod. 57 with appropriate certificatefrom the Association of Official Agricultural Chemists,1975, and relativeprocedure analyses, Laboratorio Dottori Piccioni, Gessate Milanese,Italy) containing 16.2% (w/w) crude protein, 3.0% (w/w) crude fat, 7.3%(w/w) crude ash, 14.8% (w/w) fiber, 12% (w/w) moisture, 2.1% (w/w)added mineral, 46% (w/w) carbohydrate with water, and 40 IU kg−1

vitamin E. Rabbits were randomly distributed into four groups. Thecontrol group (C) (n=6) received standard rabbit diet; the DHEA-alonetreated group (C-D) (n=6) received the same diet supplemented with0.02% (w/w) DHEA (Sigma Aldrich, Milan, Italy); the high-fat group(HF) (n=6) received a high-fat diet, consisting of standard rabbit diet,containing 16.2% (w/w) crude protein, 7.3% (w/w) crude ash,14.8% (w/w) fiber, 12% (w/w) moisture, 2.1% (w/w) added mineral, 46% (w/w)carbohydrate with water, 40 IU kg−1 vitamin E plus 10% (w/w) addedfat (6.7%w/wcorn oil and 3.3%w/w lard) (Carroll and Tyagi 2005); theHF plus DHEA group (HF-D) (n=6) received the high-fat dietsupplemented with 0.02% (w/w) DHEA. Two days before being killed,the rabbits were fasted overnight and the glucose tolerance test wasperformed. The rabbits were killed at 3 months from the start of theexperiment, by aortic exsanguination after anesthetization with Zoletil100 (Tiletamine-Zolazepam,Virbac, Carros, France). Bloodwas collectedand plasma and serumwere isolated. The heart was rapidly excised andweighed, and portions of left ventricle were taken to obtain cytosolic,nuclear and total extracts. Other heart portions were utilized forhistological microscopy. The papillary muscles were immediatelyremoved for functional parameter detection.

General parameters

Body weight and length of rabbits were measured at time zero andimmediately prior to death.

Oral glucose tolerance test

After a 12 h fasting period, a 50% (w/v) glucose solutionwas orallyadministered to the rabbits at 1.5 g/kg. Blood samples were collectedvia the auricular artery before, and 15, 30, 45, 60, 90, 120 and 240 minafter glucose loading. Glucose levels were tested using an Accu-CheckCompact kit (Roche Diagnostics Gmbh, Mannheim, Germany).

Parameters in plasma

Triglyceride (TG), total cholesterol (TC), low-density-lipoprotein(LDL) and high-density-lipoprotein (HDL) cholesterol were deter-mined by standard enzymatic procedures using reagents kits.Aspartate aminotransferase (AST) and lactate dehydrogenase (LDH)were determined using an enzymatic kit (DiaSys Diagnostic SystemsGmbH, Holzheim, Germany). Plasma insulin was measured with anultrasensitive insulin enzyme-linked immunosorbent assay kit fromDRG Diagnostics (Marburg, Germany). Insulin sensitivity was calcu-lated using the homeostasis model assessment (HOMA): fastingglucose (mmol l−1)×fasting insulin (µg l−1)/22.5. Plasma DHEAwasdetermined by specific radioimmunoassay DSL-9000 (DiagnosticSystem Laboratories, Oxford, U.K.).

Tissue extracts (cytosolic, nuclear and total extracts)

Cytosolic and nuclear fractions from rabbit left ventricle wereprepared by the modified Meldrum et al. method (Meldrum et al.1997). Briefly, left ventricle (100 mg) was homogenized at 10% (w/v)in a Potter Elvehjem homogenizer (Wheaton Science Products,Millville, NJ, USA) using a homogenization buffer containing20 mmol l−1 HEPES, pH 7.9, 1 mmol l−1 MgCl2, 0.5 mmol l−1 EDTA,1% NP-40, 1 mmol l−1 EGTA,1 mmol l−1, DTT 0.5 mmol l−1 PMSF, 5 μgml−1 aprotinin, and 2.5 μg ml−1 leupeptin. Homogenates werecentrifuged at 1000 g for 5 min at 4 °C. Supernatants were removedand centrifuged at 15,000 g at 4 °C for 40 min to obtain cytosolicfraction. The pellets were resuspended in extraction buffer containing20 mmol l−1 HEPES, pH 7.9, 1.5 mmol l−1 MgCl2, 300 mmol l−1 NaCl,0.2 mmol l−1 EDTA, 20% (v/v) glycerol, 1 mmol l−1 EGTA, 1 mmol l−1

DTT, 0.5 mmol l−1 PMSF, 5 μg ml−1 aprotinin, 2.5 μg ml−1 leupeptinand incubated on ice for 30 min for high-salt extraction, followed bycentrifugation at 15,000 g for 20min at 4 °C. The resulting supernatantscontaining nuclear proteins were carefully removed and samples werestored at −80 °C until use.

Total extract was obtained by homogenizing at 10% (w/v) directlywith extraction buffer andwere centrifuged at 1000 g for 5min at 4 °C.Supernatants (total extract) were stored at −80 °C until use. Proteincontent was determined using the Bradford assay (Bradford 1976).

Oxidative biochemical parameters

The redox state was determined in the cytosolic fraction bymonitoring hydrogen peroxide (H2O2) generation adding peroxidasefromhorseradish andacetylated ferrocytochrome c to cytosolic fractions.H2O2 release was evaluated as the increase of the acetylated ferrocyto-chrome c oxidation rate and was monitored at 550 nm minus 540 nmusing an absorption coefficient of 19.9 mmol l−1 cm−1, as described byZoccorato et al. (1989). Reduced and oxidized glutathione contents weremeasured in cytosolic fractions by Owens' method (Owens and Belcher1965). 4-Hydroxynonenal (HNE)was detectedwith an HPLC procedure:the total extract sample was directly injected for HPLC (Waters Assoc.,Milford,MA,USA) usinganRP-18 column(Merck,Darmstadt, Germany).The mobile phase used was 42% (v/v) acetonitrile/bidistilled water.Serial concentrations of HNE (0.5–10 μmol l−1) were used to prepare astandard curve (Esterbauer et al.1991). Catalase activitywas evaluated inthe cytosolic fraction following Aebi's method (Aebi 1984). Totalsuperoxide dismutase (SOD) activity was assayed in the cytosolic

Table 1General characteristics of rabbits at the end of the study (12 weeks).

C C-D HF HF-D

Body weight (g) 3123.0±122.3 3020.3±116.1 3453.0±222.1** 3449.5±80.8**Abdomen length (cm) 32.20±1.47 30.7±1.37 36.8±2.14** 35.0±1.27**Heart weight/body weight (% w/w) 0.22±0.04 0.23±0.03 0.24±0.03 0.22±0.05Heart weight (g) 6.90±0.89 7.10±0.41 8.10±0.69** 7.70±0.99Glucose (mmol l−1) 6.42±0.41 6.11±0.37 9.84±1.09** 9.37±1.10**Insulin (µg l−1) 2.31±0.86 2.01±0.51 13.63±2.78** 12.74±3.14**HOMA 0.67±0.14 0.55±0.16 6.44±1.13** 6.41±1.62**TG (mmol l−1) 1.06±0.08 1.03±0.14 1.62±0.18** 1.20±0.10*#

TC (mmol l−1) 1.41±0.27 1.38±0.14 2.07 ±0.18** 1.64±0.21#

LDL (mmol l−1) 0.55±0.03 0.41±0.08 1.27±0.07** 0.76±0.05*#

HDL (mmol l−1) 0.64±0.05 0.56±0.05 0.48±0.11** 0.64±0.10#

Pentosidine (µmol l−1) n.d. n.d. 0.99±0.19 0.58±0.17#

Catalase (µmol mg protein−1) 61.88±10.15 59.80±7.78 90.94±15.07** 70.81±14.61#

SOD (IU mg protein−1) 1.47±0.19 1.49±0.14 1.97±0.42* 1.53±0.19#

DHEA (ng ml− l) 1.25±0.07 2.75±1.19 1.47±0.27 2.70±0.83

Glucose, insulin, triglyceride (TG), total cholesterol (TC), low-density-lipoprotein cholesterol (LDL), high-density-lipoprotein cholesterol (HDL) and dehydroepiandrosterone(DHEA) levels were measured in plasma.Pentosidine, catalase and total superoxide dismutase (SOD) were determined in the cytosol of left ventricle of rabbits. Data are means±SD (n=6). Statistical significance: *Pb0.05vs C; **Pb0.01 vs C; #Pb0.05 vs HF.

79M. Aragno et al. / Life Sciences 85 (2009) 77–84

fraction by the method described by Flohé and Otting (1984). Since theactivity of xanthine-oxidase may vary, sufficient enzyme was used toproduce a rate of cytochrome c reduction of at least 0.025absorbance IU/min in the assay without SOD.

AGE in term of pentosidine detection with HPLC–MS

Cytosol fractions were treated with 6 mol l−1 hydrochloric acid for2 h at 40 °C and then centrifuged (1860 g) (Miyata et al. 2001); only thesupernatant was utilized. A Thermo-Finnigan Surveyor instrument(ThermoElectron, Rodano,Milan, Italy) equippedwith autosampler andPDA-UV 6000 LP detector was used. Mass spectrometry analyses wereperformed using an LCQ Deca XP plus spectrometer, with electrosprayinterface and ion trap as mass analyzer. The chromatographic separa-tions were run on a Varian Polaris C18-A column (150×2 mm, 3 μmparticle size) (Varian, Leinì, Turin, Italy). Flow rate 200 μl min−1.Gradient mobile phase composition was adopted: 95/5 to 0/100 v/v5mmol l−1 heptafluorobutanoic acid inwater/methanol in 13min. TheLC column effluent was delivered to a UV detector (200–400 nm) andthen to the ion source, using nitrogen as sheath and auxiliary gas (ClaindNitrogen Generator apparatus, Lenno, Como, Italy). The source voltagewas set to 4.5 kV in the positive mode. The heated capillary wasmaintained at 200 °C. The acquisitionmethod used had previously beenoptimized in the tuning sections for pentosidine quasi-molecular ion(capillary, magnetic lenses and collimating octapole voltages) tomaximize sensitivity. Collision energy (CE) was chosen to maintainabout 10% of the precursor ion. The tuning parameters adopted for theESI source were: source current 80.00 μA, capillary voltage 3.00 V, tubelens offset 15 V; for ion optics, multipole 1 offset −5.25 V, intermultipole lens voltage−16.00 V, and multipole 2 offset−9.00 V. Massspectra were collected in tandem MS mode: MS2 of (+) 379 m/z with33% CE in the range 100–400 m/z.

Western blotting

p65-NFkB nuclear and cytosolic extracts of left ventricle weredetected by Laemmli's method (Laemmli 1970). Anti-Lamin-B1 servedas loading control for nuclear p65-NFkB and anti-α-actinin antibodiesfor cytosolic proteins. Specific bands were quantified by densitometryusing analytical software (Bio-Rad, Multi-Analyst, Munchen, Germany)and the net intensity of bands in each experiment was normalized forthe intensity of the corresponding lamin-B1 or α-actinin bands, beforecomparison between control and each sample.

RNA isolation and semi-quantitative RT-PCR

Total RNAwas isolated using the RNA fast kit (Molecular Systems, SanDiego, CA, USA). Total DNA was amplified using sense and antisenseprimers specific for the C-reactive protein (CRP) gene (sense 5′-AGGATCAGGATTCGTTTG-3′ and antisense 5′-CACCACGTACTTGATATGTC-3′), the tumor necrosis factor-alpha (TNF-α) gene (sense 5′-AGGAA-GAGTCCCCAAACAACCT-3′ and antisense 5′-GGCCCGAGAAGCTGAT CTG),the alpha-myosin heavy-chain (α-MHC) gene (sense 5′-GCCAAGGT-GAAGGAGATGAA-3′ and antisense 5′-CTCTCCTGGGTCAGCTTCAG-3′), thebeta-myosin heavy-chain (β-MHC) gene (sense 5′-GGTCGAATACGTTAC-CATCTG-3′ and antisense 5′-AATCGCTGTCCACAGTGGTCG-3′) or for theglyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene (sense 5′-CGCCTGGAGAA AGCTGCTA-3′ and antisense 5′-CCCCAGCATCGAAGGTAGA-3′).

The PCR reaction system contained 1 μl of RT product, 200 μmol l−1

dATP, dGTP, dCTP and dTTP (Finnzymes, Espoo, Finland), 1.25 units ofTaq DNA polymerase (Finnzymes, Espoo, Finland) and 50 pmol ofsense and antisense primers in a total volume of 50 μl. All experimentswere performed on at least three independent cDNA preparations.

PCR products were electrophoresed on 2% w/v agarose gels andamplification products were stainedwith GelStar nucleic acid gel stain(FMC BioProducts, Rockland, ME, USA). Gels were photographed andanalyzed with Kodak 1D Image Analysis software. The net intensity ofbands in each experiment was normalized for the intensity of thecorresponding GAPDH band before comparison between control andeach samples.

Histological staining

For standard histology, portions of left ventricle were fixed in 4%(v/v) neutralized formalin. Fixed material was processed for hema-toxylin and eosin staining. Six-micron paraffin-wax sections of leftventricle were used.

Isolated papillary muscle and contractility determination

Papillarymusclesweredriven at constant frequency (120beats/min)with a pair of electrodes connected to a 302 T Anapulse Stimulator via a305-R Stimulus Isolator (W.P. Instruments, New Haven, CT, USA)operating in constant current mode. Isometric twitches were evaluatedbyaHarvard transducer (60-2997), visualizedon aTektronix 2211digitalstorage oscilloscope and continuously acquired and recorded in a PowerMac computer, using Labview Software (National Instruments Corp.,Austin, Texas, USA). The same software was used to measure developed

Fig. 1. Glucose tolerance test in rabbits fed with control (C), control-DHEA (C-D), high-fat (HF) and high-fat-DHEA (HF-D) diet for 12 weeks. Data are means±SD (n=4) andare expressed as mmol l−1. Statistical significance: *Pb0.05 vs C; **Pb0.01 vs C.

Fig. 2. Content of H2O2 and HNE detected in the total extract of the left ventricle ofcontrol (C), control-DHEA (C-D), high fat (HF) and high fat-DHEA (HF-D). DHEA wasgiven to C and HF rabbits in the diet for 12 weeks. Data are means±SD, (n=6).Statistical significance: **Pb0.01 vs C; #Pb0.05 vs HF.

80 M. Aragno et al. / Life Sciences 85 (2009) 77–84

peak mechanical tension (T max), maximum rate of rise and fall ofdevelopedmechanical tension (+dT/dtmaxand−dT/dtmax), time-to-peak mechanical tension (TPT) and duration of contraction.

Statistical analysis

All valueswere expressed asmeans±SD, andwere analyzed by one-way analysis of variance (ANOVA) followed by Bonferroni's multiplecomparison test. Pb0.05 was considered statistically significant.

Results

General features

Body weight, abdomen length, glucose and insulin were signifi-cantly increased in rabbits fed a HF diet vs control values (Pb0.01)after three months' treatment. DHEA supplementation did not modifythese values (Table 1). Homeostasis model assessment (HOMA) wasalso significantly higher in HF rabbits than in controls (Pb0.01) andnot different from those of HF plus DHEA rabbits. HOMAvalues, whichreflect whole body insulin resistance, have also been validated in otheranimal models (Buettner et al. 2006) The heart/bodyweight ratio wasthe same in all groups, whereas the heart weight of rabbits fed the HFdiet increased vs controls (Pb0.01); DHEA supplementation reducedthis increase. The results of the glucose tolerance test are reported inFig. 1. After oral loading, glucose levels remained high for up to240 min in both the HF and HF-DHEA groups. Control and C-DHEAshowed a peak of glucose concentration between 15 and 30 min afterglucose loading, after which it rapidly returned to normal levels.

TG, TC and LDL concentrations in plasma were significantly higherin HF rabbits vs controls (Pb0.01) and were reduced in HF plus DHEArabbits vs the HF alone group (Pb0.05). HDL was significantly lower(Pb0.01) in the HF group vs controls while in the HF-DHEA group, theHDL value was similar to that of control animals (Pb0.05). BloodDHEA level after 3 months of treatment was unchanged vs the controlgroup, and reached values similar to those found in normal humans.

Oxidative parameters in the left ventricle

Rabbits fed a HF diet for 12 weeks showed a significant increase ofH2O2 and HNE levels in total extract of left ventricle vs the control group(Pb0.01) (Fig. 2). In rabbits fedwith HF-DHEA, the H2O2 andHNE levels(HNE being an end-product of lipid peroxidation) were significantlylower than in the HF rabbits (Pb0.05). The level of pentosidine, incytosolic fraction,was significantly lower in theHF-DHEA than in theHFalone group (Pb0.05) (Table 1). In samples from both control and C-DHEA groups, gas–mass–HPLC analysis failed to detect any peak for

pentosidine, apparently indicating its absence in these animals. More-over, catalase and total SOD activities were increased in the HF rabbits,and HF-DHEA supplementation partially restored these activities tocontrol levels (Pb0.05). No significant difference in the GSSG/GSH ratiowas observed among groups (data not shown). The GSSG/GSH ratio, inHF and HF-DHEA groups, was significantly increased (Pb0.05) vs thecontrol group, indicating a loss of GSH level and thus a decrease ofantioxidant defenses. There was no statistically significant differencebetween the HF-DHEA and the HF group (data not shown).

Inflammatory parameters

p65-NFkB protein in the nuclear and cytosolic fractions of leftventricle was detected by Western blot analysis (Fig. 3, panel A andpanel B, respectively). Nuclear p65-NFkB level in HF rabbits wasincreased vs the control groups (Pb0.05) (panel A). The cytosolic levelof p65-NFkB was reduced in HF rabbits (Pb0.05) (panel B). In the HF-DHEA rabbits, cytosolic p65-NFkB protein level was less reduced thanin the HF group, while the nuclear p65-NFkB level was increased vsthe HF alone group, indicating a lower activation of the p65-NFkBtranscription factor in HF-DHEA group vs HF (Pb0.05).

Both indices of the pro-inflammatory state, TNF-α and CRPexpression, were significantly increased in the left ventricle of rabbitsfed theHFdiet (Pb0.01) (TNF-α: Fig. 3, panel C and CRP: Fig. 3, panel D).The HF-DHEAdiet significantly decreased expression of both TNF-α andCRP vs the HF group (Pb0.05).

Myosin expression

PCR analysis was used to evaluate expression of two isoforms (αand β) of the MHC protein (Fig. 4) in the left ventricle of control, C-DHEA, HF and HF-DHEA animals. The HF diet determined asignificantly decreased expression of α-MHC and an increasedexpression of β-MHC (Pb0.01). When DHEA was added to the HFdiet, α-MHC was brought closer to the control value and β-MHC wasreduced vs the HF group (Pb0.05).

Necrosis markers

LDH and AST releases were evaluated in the plasma (Fig. 5). BothLDH (panel A) and AST (panel B) significantly increased in HF rabbits(Pb0.01). In the HF-DHEA group, the levels of LDH and AST weresignificantly lower than in the HF alone group (Pb0.05).

Histological analysis

In the left ventricle of the HF rabbits (Fig. 6, panel C), histologicalpreparations clearly showed extensive and diffuse lipid deposition.

Fig. 3. Representative gel profiles of p65-NFkB content obtained by Western blot analyses on nucleus (panel A) and on cytosol (panel B) of left ventricle from control (C), control-DHEA (C-D), high fat (HF) and high fat-DHEA (HF-D). Quantitative results of these bands are indicated in the histogram, which shows the net intensity ratio with lamini1 for nucleusand with α-actinin for cytosol. Data are expressed as percentage variations vs the control value. Panel C and panel D represent the level of tumor necrosis factor-α (TNF-α and C-reactive protein (CRP) mRNA in left ventricle from control (C), control-DHEA (C-D), high fat (HF) and high fat-DHEA (HF-D) obtained by RT-PCR analyses. Quantitative results ofthese bands are indicated in histograms, which show the net intensity ratio vs glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Data aremeans±SD, (n=6) and are expressedas percentage variations vs control. Statistical significance: *Pb0.05 vs C; **Pb0.01 vs C; #Pb0.05 vs HF; ##Pb0.01 vs HF.

81M. Aragno et al. / Life Sciences 85 (2009) 77–84

This lipid infiltration was not observed in either the control or the C-DHEA group (panel A and panel B). Tissues obtained from HF-DHEArabbits showed rare areas of slight lipid deposition (panel D).

Cardiac function

The contractile force developed by electrically-driven papillarymuscles was evaluated in basal conditions (Table 2). Basal contrac-

Fig. 4. Alpha-myosin heavy-chain (α-MHC) and beta-myosin heavy-chain (β-MHC) expressDHEA (HF-D). Quantitative results of these bands are indicated in histograms, which show tare expressed as percentage variations vs control. Data are means±SD, (n=6). Statistical s

tility was weaker in papillary muscles from HF rabbits vs controls; thiswas evident not only for maximal developed mechanical tension(Tmax: Pb0.01), but also for maximum rate of rise (+dT/dt max;Pb0.01), maximum rate of fall of developed mechanical tension(−dT/dt max; Pb0.01) and time to peak of mechanical tension (TPT;Pb0.05). In contrast, no significant difference was found betweenpapillary muscles from control and HF rabbits in regard to duration ofcontraction. Treatment with DHEA significantly reduced the effects of

ion in left ventricle from control (C), control-DHEA (C-D), HIGH fat (HF) and HIGH fat-he net intensity ratio vs glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and dataignificance: **Pb0.01 vs C; #Pb0.05 vs HF.

Fig. 5. Level of lactate dehydrogenase (LDH) (panel A) and of aspartate aminotransferase(AST) (panel B) evaluated in plasma of control (C), control-DHEA (C-D), high fat (HF)and high fat-DHEA (HF-D). Data are expressed as units/liter (U/l). Data are means±SD(n=6). Statistical significance: *Pb0.05 vs C; **Pb0.01 vs C; #Pb0.05 vs HF.

82 M. Aragno et al. / Life Sciences 85 (2009) 77–84

the HF diet (Pb0.05). However, DHEA did not per se affect contractileproperties of the papillary muscles.

Discussion

In the left ventricle of male New Zealand rabbits fed a HF diet for12 weeks we observed unbalanced oxidative status, impairment ofcardiac myogenic factors and a switch in MHC gene expression andcontractility dysfunction, all of which were prevented by the additionof DHEA in the HF. We believe that DHEA's key action is againstoxidative imbalance (Brignardello et al. 2007; Aragno et al. 2006),which could be critical in inducing myocardial dysfunction. A greater

Fig. 6.Histochemical analysis of left ventricle sections of control (A), control-DHEA (B), high fin 4% v/v neutralized formalin. Fixed material was processed for hematoxylin and eosin sta

propensity for oxidative stress after myocardial infarction is associatedwith the development of heart failure (Smith et al. 2005) and acorrelation between systolic and diastolic myocardial dysfunction andoxidative stress has been reported in a highly-selected group ofuncomplicated type 2 diabetic patients (González-Vílchez et al. 2005).Here we show that increased H2O2, increased end-products of lipidperoxidation (specifically HNE) and increased activation of thetranscription factor p65-NFkB in the cardiac tissue of HF rabbits, areall counteracted by DHEA. Moreover, in the plasma of rabbits fed a HFdiet, we observed that DHEA counteracted the increase in glycox-idative products, in agreement with previous rat studies (Aragno et al.2004, 2005; Brignardello et al. 2000; Aragno et al. 2008).

Here we show that, by decreasing oxidative stress, DHEA reducesactivation of p65-NFkB transcription factor, determining a reduction ofinflammation in cardiac tissue. Indeed, after DHEA supplementation,expression of the pro-inflammatory cytokines TNF-α and of CRP wasdecreased. The anti-inflammatory effects of DHEA due to cytokinereductionmight also be amplified by the reduction of cholesterol levels,which in turn directly decreases CRP release from tissues (Ridker et al.2002). It has been shown that high dietary cholesterol intake canincrease the production of inflammatory cytokines and that reduction ofdietary cholesterol concentration leads to a reduction in CRP production(Han et al. 2002). When DHEA was added to the rabbits' HF diet,alongside total cholesterol, LDL and triglycerides were also reduced.Some studies have demonstrated a lowering effect of DHEA ontriglyceridemia (Mauriège et al. 2003) whereas others have not doneso (Igwebuike et al. 2008). However, it must be noted that in the latterstudies, lipid metabolism was assessed in the fasted state, that is at atime when lipid flux to the liver is minimal. Thus, the consequences ofDHEA treatment might be underestimated, since it is conceivable thattheybecomemoreevident in thepostprandial state,when themetabolichandlingof lipidsbecomes fullyactive. It has been reported that theanti-obesity effect of DHEAmay inpart be related to changes in lipase activityand beta-adrenergic receptor density (Shepherd and Cleary 1984).Moreover, DHEA accelerates lipid catabolism by direct regulation ofhepatic lipid metabolism (Tang et al. 2007) and our other preliminary

at (C) and high fat-DHEA (D). For standard histology, portions of left ventriclewere fixedining. Six-micron paraffin sections of left ventricle were used. Magnification ×10.

Table 2Basal values for papillary muscle at the end of the study (12 weeks).

C C-D HF HF-D

T max (mN mm−2) 2363±201 3159±335 753±89** 1605±68*#

+dT/dt max (mN s−1) 7536±1356 8515±1788 2713±407** 6481±778**#

−dT/dt max (mN s−1) 2864±173 3752±1163 1343±158** 3895±615**#

TPT (ms) 250.6±21.0 217.0±24.4 205.0±16.7* 240.0±19.8Duration (ms) 716.7±19.6 570.5±70.6 559.0±49.2 646.2±58.3

T max: peak mechanical tension; +dT/dt max: maximum rate of rise of developedtension; −dT/dt max: maximum rate of fall of developed tension; TPT: time-to-peaktension; duration of contraction. Data are means±SD (n=4). Statistical significance:*Pb0.05 vs C; **Pb0.01 vs C; #Pb0.05 vs HF.

83M. Aragno et al. / Life Sciences 85 (2009) 77–84

data on rats suggest that DHEA participates in controlling expression ofgenes involved inboth triglyceride and cholesterol synthesis and storagein the liver (data not reported here). In addition, it was hypothesizedthat DHEA could play a role in the inhibition of lipogenesis andadipogenesis aswell as in the activation of lipolysis in the adipose tissuepossibly through the its effects on adipokines (Pérez-de-Heredia et al.2008). However, the detailed mechanisms whereby DHEA exerts itsanti-lipid effects and modulates triglyceride (TG) storage and mobiliza-tion have not yet been fully elucidated. In agreement with severalhuman studies showing only slight or no effect of DHEA on glucosehomeostasis (Hunt et al. 2000; Callies et al. 2001; Christiansen et al.2004), we found that DHEA supplementation did not affect the highplasma glucose levels induced by the HF diet, nor did it protect againsthyperinsulinemia or improve the HOMA index.

The HF rabbits showed a switch of cardiac heavy-chain myosinfrom the alpha to the beta isoform: this event comprises the heart's“molecular motor” because contractile properties depend to a greatextent on the isoform composition of MHC proteins. A switch in MHCisoform composition has been reported to cause reduced contractilevelocity and energy expenditure (Ramamurthy et al. 2001). It hasbeen reported that inman, as in animals, a reduced content ofα-MHC,which is expressed exclusively in the myocardium, is responsible forthe reduced myocardial contractility during heart failure (Gupta et al.2003) and in diabetes (Razeghi et al. 2002;Wang et al. 2004). Hereweshow that DHEA modulates MHC expression in HF rabbits: in theDHEA treated rats, neither isoform, alpha nor beta, was significantlydifferent from the control group levels.

Rabbits on a HF diet clearly showed alterations in basal papillarymuscle contractile properties, including reduced maximal developedtension, maximum rate of rise and maximum rate of fall of developedtension, a sign of diastolic dysfunction. As in the case of rat cardiacmuscle (Aragno et al. 2008), these alterations of the mechanismscontrolling intracellular calcium handling within cardiac myocytes areprobably related to structural damage caused by the HF diet.

Histological analysis of tissue from HF rabbits showed extensiveand diffuse lipid deposition, as reported in other animal models(Wang et al. 2008; Deng et al. 2005). The increased release of LDH andAST in the plasma of HF rabbits are in keeping with the myocardialdamage. Treatment with DHEA protects the cardiac tissue from thisaltered basal contractility, as well as minimizing histological changesand reducing cell damage caused by the HF diet.

The roleofDHEA in the cardiovascular systemhasbeenhighlightedbythe recent finding, in the human heart, of DHEA production andcytochrome P450-17 gene expression, encoding cytochrome P450 17α-hydroxylase, a key factor inDHEA synthesis (Nakamura et al. 2004).Moreinterestingly, it has been reported that DHEA production is suppressed inthe failing human heart (Moriyama et al. 2000). The vascular protectiveeffect of DHEA might be dependent on G-α-GTP-binding proteinmediated activation of the phosphatidyl-inositol 3-kinase/Akt signalingpathway (Liu et al. 2007). Moreover, DHEA can regulate calciumhomeostasis (Zylinska et al. 2009): it has been reported that DHEAinduces an acute (nongenomically-mediated) vasorelaxing effect on thehumanumbilical arterywhichmaybemediated bya decrease in external

Ca2+ influx by inactivating Ca2+ channels (Perusquía et al. 2007). Severalexplanations have been put forward for multi-targeted antioxidanteffects of DHEA, including its effect on catalase expression (Yildirim et al.2003), and its up-regulation of the thioredoxin system (Gao et al. 2005),of the fatty-acid composition of cellular membranes and of cytokineproduction (Aragno et al. 2000). However, the precise mechanisms areyet to be clarified, and additional non-antioxidant effects cannot yet beruled out. Whether the effect of DHEA is due to DHEA itself, to itsmetabolites, or to a combination of both remains unclear. Hayashi et al.have claimed that the athero-protective effect of DHEAmay bemediatedthrough its conversion to estrogen (Hayashi et al. 2000). However, theamount of DHEA added to the diet wasmany times higher than the dosewe used and, furthermore, we found negligible variations of either 17β-estradiol or testosterone concentrations in rats treated with 4 mg DHEA.Nevertheless, as we report elsewhere, DHEA, but not a variety of othersteroids including β-estradiol, 17 beta-diol and dihydrotestosterone,protects bovine retinal capillary pericytes against glucose-induced lipidperoxidation (Brignardello et al. 1998).

Conclusion

We show that DHEA supplementation can prevent molecular andfunctional alterations of the cardiac muscle, restoring oxidativebalance and lowering lipid levels, in rabbits fed a high-fat diet.These data, together with our recent observations on type 2 diabetespatients (Brignardello et al. 2007) suggest that DHEA treatment mightprevent many events that lead to the cellular damage induced byhyperglycemia, thus delaying the onset or progression of cardiaccomplications in type 2 diabetes.

Acknowledgements

This work was supported by the Regione Piemonte (ProgettoFinalizzato 2006 and 2007), MIUR (ex 60%, 2006) and CRT Foundation(Cardiomiopatia diabetica: individuazione di nuove strategie tera-peutiche - Alfieri Project).

References

Aebi H. Catalase in vitro. Methods in Enzymology 105, 121–126, 1984.Aragno M, Parola S, Brignardello E, Mauro A, Tamagno E, Manti R, Danni O, Boccuzzi G.

Dehydroepiandrosterone prevents oxidative injury induced by transient ischemia/reperfusion in the brain of diabetic rats. Diabetes 49, 1924–1931, 2000.

Aragno M, Mastrocola R, Catalano MG, Brignardello E, Danni O, Boccuzzi G. Oxidativestress impairs skeletal muscle repair in diabetic rats. Diabetes 53, 1082–1088, 2004.

AragnoM, Mastrocola R, Medana C, Restivo F, CatalanoMG, Pons N, Danni O, Boccuzzi G.Up-regulation of advanced glycated products receptors in the brain of diabetic ratsis prevented by antioxidant treatment. Endocrinology 146, 5561–5567, 2005.

Aragno M, Mastrocola R, Medana C, Catalano MG, Vercellinatto I, Danni O, Boccuzzi G.Oxidative stress-dependent impairment of cardiac-specific transcription factors inexperimental diabetes. Endocrinology 147, 5967–5974, 2006.

Aragno M, Mastrocola R, Alloatti G, Vercellinatto I, Bardini P, Geuna S, Catalano MG,Danni O, Boccuzzi G. Oxidative stress triggers cardiac fibrosis in the heart of diabeticrats. Endocrinology 149, 380–388, 2008.

Bradford M. A rapid and sensitive method for the quantitation of microgram quantitiesof protein utilizing the principle of protein dye binding. Analytical Biochemistry 72,248–254, 1976.

Brignardello E, Beltramo E, Molinatti PA, Aragno M, Gatto V, Tamagno E, Danni O, PortaM, Boccuzzi G. Dehydroepiandrosterone protects bovine retinal capillary pericytesagainst glucose toxicity. Journal of Endocrinology 158, 21–26, 1998.

Brignardello E, Gallo M, Aragno M, Manti R, Tamagno E, Danni O, Boccuzzi G.Dehydroepiandrosterone prevents lipid peroxidation and cell growth inhibitioninduced by high glucose concentration in cultured rat mesangial cells. Journal ofEndocrinology 166, 401–406, 2000.

Brignardello E, Runzo C, Aragno M, Catalano MG, Cassader M, Perin PC, Boccuzzi G.Dehydroepiandrosterone administration counteracts oxidative imbalance andadvanced glycation end product formation in type 2 diabetic patients. DiabetesCare 30, 2922–2927, 2007.

Brownlee M. The pathobiology of diabetic complications: A unifying mechanism.Diabetes 54, 1615–1625, 2005.

Buettner R, Parhofer KG, Woenckhaus M, Wrede CE, Kunz-Schughart LA, Scholmerich J,Bollheimer LC. Defining high fat diet rat models: Metabolic and molecular effects ofdifferent fat types. Journal of Molecular Endocrinology 36, 485–501, 2006.

84 M. Aragno et al. / Life Sciences 85 (2009) 77–84

Callies F, Fassnacht M, van Vlijmen JC, Koehler I, Huebler D, Seibel MJ, Arlt W, Allolio B.Dehydroepiandrosterone replacement inwomenwith adrenal insufficiency: Effectson body composition, serum leptin, bone turnover, and exercise capacity. Journal ofClinical Endocrinology and Metabolism 86, 1968–1972, 2001.

Carroll JF, Tyagi SC. Extracellular matrix remodeling in the heart of the homocystei-nemic obese rabbit. American Journal of Hypertension 18, 692–698, 2005.

Christiansen JJ, Gravhol CH, Fisker S, Svenstrup B, Bennett P, Veldhuis J, Andersen M,Christiansen JC, Jørgensen JOL. Dehydroepiandrosterone supplementation inwomen with adrenal failure: Impact on twenty-four hour GH secretion and IGF-related parameters. Clinical Endocrinology 60, 461–469, 2004.

Davì G, Ciabattoni G, Consoli A, Mezzetti A, Falco A, Santarone S, Pennese E, Vitacolonna E,Bucciarelli T, Costantini F, Capani F, PatronoC. Invivo formationof8-iso-prostaglandin-alpha and platelet activation in diabetes mellitus: Effects of improved metaboliccontrol and vitamin E supplementation. Circulation 99, 224–229, 1999.

Deng QG, She H, Cheng JH, French SW, Koop DR, Xiong S, Tsukamoto H. Steatohepatitisinduced by intragastric overfeeding in mice. Hepatology 42, 905–914, 2005.

Esterbauer H, Schaur RJ, Zollner H. Chemistry and biochemistry of 4-hydroxynonenal,malonaldehyde and related aldehydes. Free Radical Biology and Medicine 11,81–128, 1991.

Farhangkhoee H, Khan ZA, Kaur H, Xin X, Chen S, Chakrabarti S. Vascular endothelialdysfunction in diabetic cardiomyopathy: Pathogenesis and potential treatmenttargets. Pharmacology and Therapeutics 111, 384–399, 2005.

Flohé L, Otting F. Superoxide dismutase assays. Methods Enzymology 105, 93–104,1984.Gao J, Sun HY, Zhu ZR, Ding Z, Zhu L. Antioxidant effects of dehydroepiandrosterone are

related to up-regulation of thioredoxin in SH-SY5Y cells. Acta Biochimica andBiophysica Sinica 37, 119–125, 2005.

González-Vílchez F, Ayuela J, Ares M, Pi J, Castillo L, Martín-Durán R. Oxidative stressand fibrosis in incipient myocardial dysfunction in type 2 diabetic patients.International Journal of Cardiology 101, 53–58, 2005.

Gupta M, Sueblinvong V, Raman J, Jeevanandam V, Gupta MP. Single-stranded DNA-binding proteins PURalpha and PURbeta bind to a purine-rich negative regulatoryelement of the alpha-myosin heavy chain gene and control transcriptional andtranslational regulation of the gene expression. Implications in the repression ofalpha-myosin heavy chain during heart failure. Journal of Biological Chemistry 278,44935–44948, 2003.

Han SN, Leka LS, Lichtenstein AH, Ausman LM, Schaefer EJ, Meydani SN. Effect ofhydrogenated and saturated, relative to polyunsaturated, fat on immune andinflammatory responses of adults with moderate hypercholesterolemia. Journal ofLipid Research 43, 445–452, 2002.

Hartog JW, Voors A, Bakker SJ, Smit AJ, van Veldhuisen DJ. Advanced glycation end-products (AGEs) and heart failure: Pathophysiology and clinical implications.European Journal of Heart Failure 9, 1146–1155, 2007.

Hayashi T, Esaki T, Muto E, Kano H, Asai Y, Thakur NK, Sumi D, Jayachandran M, Iguchi A.Dehydroepiandrosterone retards atherosclerosis formation through its conversion toestrogen: The possible role of nitric oxide. Arteriosclerosis, Thrombosis, and VascularBiology 20, 782–792, 2000.

Hunt PJ, Gurnell EM, Huppert FA, Richards C, Prevost AT, Wass JA, Herbert J, ChatterjeeVK. Improvement in mood and fatigue after dehydroepiandrosterone replacementin Addison's disease in a randomized, double blind trial. Journal of ClinicalEndocrinology and Metabolism 85, 4650–4656, 2000.

Igwebuike A, Irving BA, Bigelow ML, Short KR, McConnell JP, Nair KS. Lack ofdehydroepiandrosterone effect on a combined endurance and resistance exerciseprogram in postmenopausal women. Journal of Clinical Endocrinology and Metabo-lism 93, 534–536, 2008.

Laemmli UK. Cleavage of structural proteins during the assembly of the head ofbacteriophage T4. Nature 227, 680–685, 1970.

Liu D, Si H, Reynolds KA, Zhen W, Jia Z, Dillon JS. Dehydroepiandrosterone protectsvascular endothelial cells against apoptosis through a Galphai protein-dependentactivation of phosphatidylinositol 3-kinase/Akt and regulation of antiapoptotic Bcl-2 expression. Endocrinology 148, 3068–3076, 2007.

Masoudi FA, Inzucchi SE. Diabetes mellitus and heart failure: Epidemiology, mechan-isms, and pharmacotherapy. American Journal of Cardiology 99, 113–132, 2007.

Mauriège P, Martel C, Langin D, Lacaille M, Després JP, Bélanger A, Labrie F, Deshaies Y.Chronic effects of dehydroepiandrosterone on rat adipose tissue metabolism.Metabolism 52, 264–272, 2003.

Meldrum DR, Shenkar R, Sheridan BC, Cain BS, Abraham E, Harken AH. Hemorrhageactivates myocardial NFkB and increases TNF-α in the heart. Journal of Molecularand Cellular Cardiology 29, 2849–2854, 1997.

Minamiyama Y, Takemura S, Bito Y, Shinkawa H, Tsukioka T, Nakahira A, Suehiro S,Okada S. Supplementation of alpha-tocopherol improves cardiovascular risk factorsvia the insulin signalling pathway and reduction of mitochondrial reactive oxygenspecies in type 2 diabetic rats. Free Radical Research 42, 261–271, 2008.

Miyata T, van Ypersele de Strihou C, Imasawa T, Yoshino A, Ueda Y, Ogura H, KominamiK, Onogi H, Inagi R, Nangaku M, Kurokawa K. Glyoxalase I deficiency is associatedwith an unusual level of advanced glycation end products in a hemodialysis patient.Kidney International 60, 2351–2359, 2001.

Moriyama Y, Yasue H, Yoshimura M, Mizuno Y, Nishiyama K, Tsunoda R, Kawano H,Kugiyama K, Ogawa H, Saito Y, Nakao K. The plasma levels of dehydroepiandroster-one sulfate are decreased in patients with chronic heart failure in proportion to theseverity. Journal of Clinical Endocrinology and Metabolism 85, 1834–1840, 2000.

Nakamura S, Yoshimura M, Nakayama M, Ito T, Mizuno Y, Harada E, Sakamoto T, Saito Y,Nakao K, Yasue H, Ogawa H. Possible association of heart failure status withsynthetic balance between aldosterone and dehydroepiandrosterone in humanheart. Circulation 110, 1787–1793, 2004.

Owens CWI, Belcher RV. A colorimetric micro-method for the determination ofglutathione. Biochemical Journal 94, 705–709, 1965.

Pérez-de-Heredia F, Sánchez J, Priego T, Nicolás F, Portillo Mdel P, Palou A, Zamora S,Garaulet M. Adiponectin is involved in the protective effect of DHEA againstmetabolic risk in aged rats. Steroids 73, 1128–1136, 2008.

Perusquía M, Navarrete E, González L, Villalón CM. The modulatory role of androgensand progestins in the induction of vasorelaxation in human umbilical artery. LifeSciences 81, 993–1002, 2007.

Ramamurthy B, Höök P, Jones AD, Larsson L. Changes in myosin structure and functionin response to glycation. FASEB Journal 15, 2415–2422, 2001.

Razeghi P, Young ME, Cockrill TC, Frazier OH, Taegtmeyer H. Downregulation ofmyocardial myocyte enhancer factor 2C and myocyte enhancer factor 2C-regulatedgene expression in diabetic patients with nonischemic heart failure. Circulation106, 407–411, 2002.

Ridker PM, Rifai N, Rose L, Buring JE, Cook NR. Comparison of C-reactive protein andlow-density lipoprotein cholesterol levels in the prediction of first cardiovascularevents. New England Journal of Medicine 347, 1557–1565, 2002.

Robertson RP. Oxidative stress and impaired insulin secretion in type 2 diabetes.Current Opinion in Pharmacology 6, 615–619, 2006.

Russell JC, Proctor SD. Small animal models of cardiovascular disease: Tools for thestudy of the roles of metabolic syndrome, dyslipidemia, and atherosclerosis.Cardiovascular Pathology 15, 318–330, 2006.

Shepherd A, Cleary MP. Metabolic alterations after dehydroepiandrosterone treatmentin Zucker rats. American Journal of Physiology 246, 123–128, 1984.

Simoncini T, Genazzani AR. Dehydroepiandrosterone, the endothelium, and cardiovas-cular protection. Endocrinology 148, 3065–3067, 2007.

Smith HM, HamblinM, Hill MF. Greater propensity of diabetic myocardium for oxidativestress after myocardial infarction is associated with the development of heartfailure. Journal of Molecular and Cellular Cardiology 39, 657–665, 2005.

Tang X, Ma H, Zou S, Chen W. Effects of dehydroepiandrosterone (DHEA) on hepaticlipid metabolism parameters and lipogenic gene mRNA expression in broilerchickens. Lipids 42, 1025–1033, 2007.

Tchernof A, Labrie F. Dehydroepiandrosterone, obesity and cardiovascular disease risk:A review of human studies. European Journal of Endocrinology 151, 1–14, 2004.

Wang Y, Guo X, Dhalla NS. Modification of myosin protein and gene expression infailling hearts due to myocardial infarction by enalapril or losartan. Biochimica etBiophysica Acta 1690, 177–184, 2004.

Wang Y, Ausman LM, Russel RM, Greenberg AS, Wang XD. Increased apoptosis in high-fat diet induced non-alcoholic steatohepatitis in rats is associated with c-jun NH2terminal kinase activation and elevated proapoptotic bax. Journal of Nutrition 138,1866–1871, 2008.

Wold LE, Ceylan-Isik AF, Ren J. Oxidative stress and stress signaling: Menace of diabeticcardiomyopathy. Acta Pharmacologica Sinica 26, 908–917, 2005.

Yildirim A, Gumus M, Dalga S, Sahin YN, Akcay F. Dehydroepiandrosterone improveshepatic antioxidant systems after renal ischemia–reperfusion injury in rabbits.Annals of Clinical and Laboratory Science 33, 459–464, 2003.

Zylinska L, Kowalska I, Ferenc B. Calmodulin effects on steroids-regulated plasmamembrane calcium pump activity. Cell Biochemistry and Function 27, 111–117,2009.

Zoccorato F, Deana R, Cavallini L, Alexandre A. Generation of hydrogen peroxide bycerebral cortex synaptosomes. European Journal of Biochemistry 180, 473–478,1989.