Testosterone protects cardiac myocytes from superoxide injury via NF-κB signalling pathways

Life Sciences 133 (2015) 45–52

Contents lists available at ScienceDirect

Life Sciences journal homepage: www.elsevier.com/locate/lifescie

Testosterone protects cardiac myocytes from superoxide injury via NF-κB signalling pathways Fu-Ying Xiao a,⁎, Lina Nheu b, Paul Komesaroff b, Shanhong Ling b,# a b

Department of Biotechnology, Guilin Medical University, Guilin, Guangxi 541001, PR China Department of Medicine, Monash University Central Clinical School, Prahran, Melbourne, Victoria 3181, Australia

a r t i c l e

i n f o

Article history: Received 25 December 2014 Received in revised form 3 March 2015 Accepted 5 May 2015 Available online 30 May 2015 Chemical compounds studied in this article: Testosterone (PubChem CID 6013) Hydrogen peroxide (PubChem CID 784) Flutamide (PubChem CID 3397) Keywords: Testosterone Cardiac myocytes Superoxide injury Androgen receptor (AR) Nuclear factor-κB (NF-κB)

a b s t r a c t Aims: Cellular and molecular mechanisms underlying the effects of androgenic hormone testosterone on the heart remain unclear. This study examined the impact of testosterone on viability of cardiac myocytes and the role of NF-κB signalling pathways. Materials and methods: Rat H9c2 myocytes were cultured in steroid-free media and incubated with hydrogen peroxide (H2O2, 200 μM, 6 h). NF-κB expression was knocked down by RelA (p65) siRNA interference. Testosterone (5–100 nM, 24–48 h) was provided into the media and androgen receptor (AR) blocked by flutamide (100 nM). Cell apoptotic/necrotic death was determined by morphological examination and flow-cytometric analysis. Gene expression was examined by Western blotting analysis. Key findings: Testosterone supplements reduced the superoxide-induced apoptotic/necrotic death, stimulated NF-κB (RelA) expression, activated Akt activity, and inhibited Caspase-3 expression in the cardiac myocytes. The hormonal effects were abolished by either AR blocker flutamide or NF-κB-knockdown. Testosterone also induced ERK1/2 activation, which was not affected by flutamide or NF-κB knockdown, and blocking the ERK activity did not affect the protective effect of the hormone on the cells. Significance: This study demonstrates that exogenous testosterone supplementation protects cardiac myocytes from superoxide injury via AR mediation and dependent on normally functional canonical NF-κB (RelA/p50) signalling pathways. The NF-κB signalling may be an important key molecular basis for myocardial benefits of hormone (testosterone) therapy. © 2015 Elsevier Inc. All rights reserved.

1. Introduction Deficiency of androgenic hormones is associated with reduced quality of life and longevity and a higher risk of cardiovascular disease (CVD) [12,14]. Androgen replacement therapy, which restores the physiological hormonal levels through supplementation of exogenous testosterone, can improve clinical symptoms associated with androgen deficiency. The therapy has been available to hypogonadal men for over three decades and more recently has occasionally also been given to postmenopausal women following oophorectomy. Although epidemiological data suggest that testosterone therapy may reduce components of the metabolic syndrome, the extent to which it provides direct cardiac protection remains uncertain [3] and to date, the clinical application for the prevention of CVD remains controversial. Heart failure is a major contributor to death from CVD. The progression from cardiac injury to symptomatic heart failure is largely ⁎ Corresponding author at: Department of Biotechnology, Guilin Medical University, Guilin, Guangxi 541001, PR China. E-mail address: [email protected] (F.-Y. Xiao). # Guest Professor of Guilin Medical University, Guangxi, PR China.

http://dx.doi.org/10.1016/j.lfs.2015.05.009 0024-3205/© 2015 Elsevier Inc. All rights reserved.

attributable to loss of functional cardiomyocytes through pathways of cell death, resulting in replacement of myocytes by scar tissue. Nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB), a family of phylogenetically conserved proteins that act as nuclear transcription factors, is known to play key roles in the regulation of cardiac survival, apoptosis, hypertrophy and pathological remodelling during ischaemic injury [10]. There are two – canonical (RelA/p50 dimer) and noncanonical (RelB/p52 dimer) – signalling pathways of NF-κB, with RelA/p50 the predominant complex in the heart. Studies have documented the role of NF-κB signalling in the regulation of cardiac survival through repression of apoptotic cell death undergoing disease conditions [2,16,18,19,22]. However, chronic or long-term activation of NF-κB may induce expression of inflammatory cytokines and produce cardiac cell death [4,9]. Testosterone is the most important androgen for the physiology in man. Most (over 90%) testosterone directly binds and activates intracellular androgen receptor (AR). It can also act directly on specific cell membrane binding sites (SMBS), a non-AR-mediated pathway [12]. Cardiac myocytes contain functional AR and SMBS and are therefore targets for hormonal action. Epidemiological studies demonstrate an increased risk of heart failure due to deficiency of physiological androgens, as occurred in ageing men [14], prostate cancer patients with androgen deprivation therapy [15], as well as in women with bilateral

46

F.-Y. Xiao et al. / Life Sciences 133 (2015) 45–52

Fig. 1. Testosterone improves viability of cardiac myocytes undergoing superoxide injury. H9c2 cells were incubated with testosterone (T, 5–100 nM) for 24 h, suffered to hydrogen peroxide (H2O2) treatment (200 μM) for 6 h, then stained with Annexin-V/PI and assessed by FACS. Dotplots: from a representative experiment; apoptotic cells characterised by Annexin-V positive and PI negative staining (indicated in red colour). Bar graph: Mean ± SD of the percentage of survival and apoptotic cells from three independent experiments; *p b 0.05, **p b 0.01, vs. C (T 0).

oophorectomy [12]. In castrated rats [1,20], the androgen deficiency worsened myocardial injury and reduced ejection fraction and diastolic dysfunction, and the cardiac function was improved with testosterone replacement. Testosterone can enhance cardiac hypertrophy and remodelling of the heart, resulting in either improvement of cardiac function [23] or increasing the risk of acute cardiac rupture then worsening of cardiac function [5]. The effects of testosterone on cardiac cells are many and varied, with different molecular signalling pathways involved and to date, remain unclear. In this study, we have examined the impact of testosterone supplementation on the viability of cardiac myocytes, with particularly focusing on the NF-κB signalling pathway. 2. Materials and methods 2.1. Cell cultures Myocardiac H9c2 cell (Catalog. No. CRL-1446), a clonal line derived from embryonic rat heart, was from American Type Culture Collection (ATCC, Manassas, VA). The cells were routinely cultured in Dulbecco's modified Eagle's medium (DMEM, Sigma Co., St. Louis, MI) with Dglucose at 4.5 g/L, 10% foetal bovine serum (FBS), 10,000 U/L penicillin, and 10 mg/L streptomycin, in an incubator with 5% CO2, at 37 °C and the medium was changed every 2 days. When confluence reached the cells (between passages 3 and 5) they were subcultured by detaching with 0.25% trypsin–EDTA solution (Sigma Co.) and re-seeding into new plates at a ratio of 1:5 in DMEM with 10% steroid-free FBS (Sigma Co.). Cells at ~75% confluence were cultured in serum-free DMEM, treated with the hormone and its receptor blocker (see below), and subjected to hydrogen peroxide (H2O2, 200 μM, 6 h) treatment. 2.2. Testosterone and androgen receptor (AR) blocker Testosterone (cat# T1500) and AR blocker flutamide (cat# F9397) were purchased from Sigma Co. and dissolved in 100% ethanol for storage. For the study they were diluted with PBS freshly before application in each experiment and added into the cell culture media with the final

concentration of ethanol at 0.1%. Testosterone was added into the media at concentrations of 5 to 100 nM for 24 to 48 h and flutamide (100 nM) added into the media at 3 h before the hormone.

2.3. Knockdown of NF-κB gene expression Expression of nuclear NF-κB RelA (p65) in H9c2 myocytes was knocked down by RNA interference using a RelA/p65 siRNA Transfection Kit (GeneResearch Co. Australia) that contained SignalSilence® NF-κB p65 siRNA I (2 μM, mouse specific) and Thermo Scientific DharmaFECT® 1 siRNA transfection reagent. As testing in cultured H9c2 cells by the manufacturer, siRNA (100 nM) complexed with the reagent resulted in silencing (reduction) of mRNA (in QuantiGene® branched DNA/RNA assay) at N 90% with little toxicity (cell viability N95% in the alamarBlue® assay). In this study siRNA transfection was performed in accordance with the manufacturer's protocol optimised for use with H9c2 cells in culture. In brief, cells were washed with serum-free DMEM, incubated with the transfection reagent (7 μL in 343 μL DMEM) containing RelA/p65 siRNA (100 nM) for 2–4 h, and continued to culture in serum (10%, steroid-free)-enriched DMEM with the siRNA for 48 h before being ready for the hormonal testing.

2.4. Giemsa staining and morphological analysis H9c2 cells cultured on cover glasses were washed with ice-cold PBS, fixed with methanol for 5 min, stained with Giemsa staining solution that was freshly prepared by 1:20 dilution of Giemsa stock solution (Sigma Co.) for 15–20 min, checked by microscopy until an ideal stain was obtained, and mounted with glycerol gel. The cells were then examined under a light microscope (×200 and ×400); both normal survival and apoptotic/dead cells, characterised by nuclear fragmentation and condensation (at a relatively early stage) or by nuclear disruption (at a late stage) were counted in five different fields randomly in each experiment and the percentage was calculated in each group.

F.-Y. Xiao et al. / Life Sciences 133 (2015) 45–52

47

Fig. 2. The protective effect of testosterone on cardiac myocytes is mediated by classical androgen receptor (AR). H9c2 cells were incubated with flutamide (F, 100 nM, 24 h), testosterone (T, 50 or 100 nM, 18 h), then H2O2 (H, 200 μM, 6 h), and stained with Giemsa and assessed under a light microscope. Photos: images (100×) from a representative experiment; F(−)/(+): cells without/with flutamide treatment. Bar graph: Mean ± SD of the percentage of survival/alive cells from three independent experiments; *p b 0.05, vs. H or H + F.

2.5. Annexin-V/propidium iodide (PI) staining and fluorescence-activated cell sorting scanning (FACScan) H9c2 cells were harvested by 0.1 M EDTA digestion, washed in serum-free DMEM, and stained by incubation with Annexin-V-Fluos label solution at 37 °C for 15 min. The label solution was prepared shortly before use by prediluting 20 μL Annexin-V-fluorescein isothiocyanate (FITC) (Sigma Co.) and 20 μL PI (50 μg/mL) (Sigma Co.) in 1 mL Hepes buffer (10 mmol/L Hepes/NaOH, pH 7.4; 140 mmol/L NaCl; 5 mmol/L CaCl2). Normal cells unstained (FITC −/PI −), typical apoptotic cells stained only by Annexin-V but not PI (FITC+/PI−), and atypical apoptotic or necrotic cells stained by both Annexin-V and PI (FITC+/PI+) were evaluated by using a FACScalibur flow cytometer and CellQues software (Becton Dickinson, CA). 2.6. Western blotting analysis Protein expression of interesting genes, including NF-κB RelA (p65), Caspase-3, serine threonine kinase (Akt), extracellular signal-regulated kinase (ERK) 1/2, Bcl-2 and Bcl-xL, and protein kinases (PK) A and C were assessed by Western blotting analyses. H9c2 cells were lysed by

incubation on ice for 30 min with lysis buffer (20 mmol/L Tris-base pH 7.7, 250 mmol/L NaCl, 2 mmol/L EDTA, 2 mmol/L EGTA, 0.5% NP-40, 10% glycerol, 20 mmol/L β-glycerophosphate, 1 mmol/L Na-vanadate). And 10 μL/mL leupeptin, 5 μL/mL aprotinin, 1 μmol/L pepstatin, 1 mmol/ L AEBSF and 10 mmol/L DTT were added before use. Total proteins were isolated by centrifuging at 5000 rpm for 15 min, and 30 μg protein electrophoresed on 10% SDS-polyacrylamide gels and transferred to Hybond ECL filters (Sigma Co.). The filters were blocked with 5% nonfat dry milk in TBS (20 mmol/L Tris, pH 7.5, 50 mmol/L NaCl, and 0.1% Tween-20) overnight, then washed and incubated with a primary antibody for 1 h to overnight. After washing (5 min, twice), blots were incubated with HRP-conjugated secondary antibody against the primary antibody (DAKO, Denmark) for 1 h. The blots were washed (5 min, twice), incubated for 1 min with enhanced chemiluminescence reagents (Amersham Biosciences Co., Piscataway, NJ), and exposed to X-ray films for 1 to 10 min to obtain ideal exposure. Protein signals were scanned with a PowerLook Scanner and relative levels estimated by densitometry. The (primary) antibodies against RelA, Caspase-3, phosphorylated Akt (pAkt) and Akt, Bcl-2 and Bcl-xL, phosphorylated ERK1/2 (pERK) and ERK1/2, phosphorylated AKA (pAKA) and AKA, and phosphorylated AKC (pAKC) and AKC were obtained from Oncogene Research Products, Inc. (San Diego, CA).

48

F.-Y. Xiao et al. / Life Sciences 133 (2015) 45–52

alive cells at only ~ 25% (p b 0.01); testosterone supplementation did not improve the survival in these cells though it did in the wild-type cells (Fig. 4). Thus, normal functional canonical NF-κB signalling activity is necessary for the survival and hormonal protection in these myocytes. 3.3. Testosterone enhances Akt activity and attenuates overexpression of caspase 3 in cardiac myocytes via AR mediation Superoxide (H2O2) treatment significantly stimulated Akt activation, as an increment of pAkt protein expression, in H9c2 myocytes, and the activation was significantly minimised in the cells with RelA knockdown, implicating an action dependent on the NF-κB signalling; testosterone significantly enhanced the Akt activation in normal but not RelAknockdown myocytes (Fig. 5). This hormonal action could be blocked by flutamide treatment, thus mediated via AR (data not shown). There was a relatively much high expression of gene Caspase-3 in H9c2 myocytes with NF-κB (RelA) knockdown, with its protein level at over 3 folds of that in the wild-type cells, and the Caspase-3 expression was inhibited by testosterone treatment (Fig. 6). The hormonal effect was blocked by flutamide (Fig. 6), indicating an AR-mediated action. Although Caspase 3 is an important gene involved in the development of cell apoptosis, in this study the superoxide treatment did not directly induce the protein production (expression) of this gene (data not shown). Fig. 3. Protein expression of NF-κB RelA in normal or RelA siRNA-treated cardiac myocytes and the effect of testosterone. H9c2 cells were incubated with or without NF-кB RelA siRNA (48 h), flutamide (F, 100 nM, 24 h), then testosterone (T, 50 or 100 nM, 18 h), and RelA proteins assessed by Western blotting analysis. Photos: Western blots from a representative experiment. Bar graph: Mean ± SD of relative RelA protein levels from three independent experiments; *p b 0.05, vs. C; #p b 0.05, vs. T50. All p values between normal and RelA-knockdown cells were less than 0.01.

2.7. Statistical analysis All data are presented as mean ± SD. Data from three or more treatments such as different hormonal concentrations were compared by using two-way ANOVA, with post hoc testing by the Student–Neumann–Keul's test, and comparison of two different treatments by Student's t test. p b 0.05 was considered statistically significant. 3. Results 3.1. Testosterone protects cardiac myocytes from superoxide injury via AR mediation In the study treatment of H2O2 (50–500 μM, 6 h) significantly reduced survival and induced apoptotic death in cultured H9c2 myocytes, in a concentration-dependent manner (data not shown). The cells appeared ~ 50% alive and ~ 45% dead with the superoxide treatment at H2O2 concentration of 200 μM; testosterone (5–100 nM) supplements reduced such damage in a dose-dependent manner, with dead cells at about 20% and over 70% cells survival at 50–100 nM T treatments (Fig. 1). The hormonal protection was significantly minimised with a pretreatment with flutamide (Fig. 2), indicating an AR-mediated action. 3.2. Myocardial protection of testosterone is dependent on a normal expression of NF-κB in the myocytes H9c2 myocytes in culture regularly expressed NF-κB RelA (p65) protein, which was enhanced by testosterone via AR-mediation and significantly decreased by transfection of RelA siRNA (100 nM, 48 h) (Fig. 3). Such low levels of production of RelA protein following the RNA interference, in our experience, could not maintain a normally functional canonical NF-κB (RelA/p50 dimer) signalling activity. Compared to normal (wild-type) H9c2 myocytes, the cells with RelA-knockdown were significantly more suffered from the superoxide injury, with

3.4. Assessment of ERK signalling pathway Testosterone stimulated ERK activation, as an increment of pERK1/2 protein levels, in the H9c2 myocytes. The hormonal action appeared a dose-dependent manner and was affected by neither AR blocker flutamide nor NF-κB knockdown (Fig. 7), implicating an AR-independent non-genomic effect. To examine whether or not this non-genomic action contributes to the cardiomyocyte protection by testosterone, H9c2 cells were incubated with UO123 (Cell Signalling Tech. Australia), an inhibitor of ERK activation. UO123 treatment (1 μM for 24 h) significantly abolished ERK phosphorylation (activation) in the cells (data not shown) but did not affect the protective effect of testosterone (Fig. 8), thus the hormonal protection of the cardiac myocytes from superoxide injury was independent on the non-genomic ERK pathway. 4. Discussion This study shows that exogenous supplement of testosterone, via an AR-mediated manner, protects H9c2 cardiac myocytes from superoxide injury, with a significant reduction of cell apoptotic death and an increase in cell survival. The study found, for the first time, that the direct improvement of cell viability by the hormone needs the presence of a normally functional NF-κB signalling in the cells. Several previous studies provided evidence suggesting a beneficial effect of testosterone on myocardial survival at the cellular level. Testosterone has showed to induce heat shock protein (HSP) 70 in cardiac myocytes and play important role in preconditioning for delayed cardiac protection in settings of ischaemia [13]. This preconditioning might reduce the chance of cellular death during additional ischaemic stress in hearts with relative ischaemia. This hormonal effect is genomic and abolished by AR blockers. Induction of ATP-sensitive potassium channels in the cardiac mitochondrial inner membrane may be another mechanism of cytoprotection conferred by this hormone [6]. Acute administration of testosterone increases contractility of mammalian myocytes, an effect that explains the hormonal benefit for performance enhancement in athletes [8]. NF-κB RelA/p50 is an important molecular signalling pathway in the heart and differential outcomes through NF-κB activation are essential for the myocardial progression from cardiac injury to heart failure in various disease conditions. The cardiac NF-κB signalling pathway includes a large number of genes and factors. Some of them such as

F.-Y. Xiao et al. / Life Sciences 133 (2015) 45–52

49

Fig. 4. Knockdown of NF-κB RelA decreases the survival and abolishes the protective action of testosterone in cardiac myocytes under oxidative stress. Normal or NF-κB RelA-knockdown H9c2 cells were incubated with testosterone (T, 5-100 nM, 24 h) and H2O2 (200 μM, 6 h), then stained with Giemsa and observed under light microscope. Photos: images (100×) from a representative experiment; apoptotic cells characterised by nuclear condensation and membrane rupture and cytoplasma disappear. Line graph: Mean ± SD of the percentage of survival/ alive cells from three independent experiments; *p b 0.05, vs. C (T 0); #p b 0.01, vs. RelA-knockdown cells.

PKA, PKC, and ERK induce cytosol RelA/p50/IκBα formation and phosphorylation, thus activating the signalling pathway, and others such as Bcl-2, Bcl-xL, Akt, and Caspases, known to control cell survival and death, are regulated (targeted) by NF-κB. Activation of NF-κB suppresses apoptosis and favours survival in isolated rat ventricular myocytes [18]. Mice transgenic with NF-κB inhibitor displayed a 50% greater infarct size with significantly higher levels of postinfarct apoptosis [16]. NF-κB knockout mice appear to suffer from increasingly severe cardiac dysfunction following coronary ischaemia [22]. Hypoxia-induced apoptosis in cultured cardiac myocytes can be overcome by forcing NF-κB gene expression possible via repressing the expression of Bnip3, a member of Bcl-2 family [2,19]. The roles of NF-κB in the regulation of myocardial survival are through repression of apoptotic cell death triggered in variant disease conditions including acute ischaemia and hypoxia, as well as oxidative stress [11]. A recent study [7] demonstrated that decreased NF-κB-dependent transcriptional activity by knockdown of Grx1 gene in isolated and cultured cardiac myocytes, including H9c2 cells, significantly increased apoptotic death in the cells following superoxide stimulation (H2O2, 400 μM, 5 min). Knockdown of Grx1 also reduced expression of the anti-apoptotic NFκB target genes Bcl-2 and Bcl-xL [7], while overexpression of Grx1 in cardiac myocytes was cytoprotective, diminishing H2O2-induced apoptosis likely via redox regulation of anti-apoptotic gene Akt [17]. Our study has consistently found that in cultured H9c2 myocytes with silencing of NF-κB (RelA) expression increases superoxide-induced

apoptotic death and testosterone enhances RelA expression and reduces apoptosis only in the cells with normal functional NF-κB (RelA). In the study Akt activity, as indicated by an increment of pAkt protein expression, significantly increased in the myocytes undergoing oxidative stress, which was further enhanced by testosterone via AR mediation (Fig. 5). The activation of Akt activity may be a protective response of the cells under superoxide conditions, which needs a support by normally functional NF-κB signalling because the activation (pAkt level) was much less in RelA-knockdown than in wild-type cells. In comparison, Caspase 3, another NF-κB-regulated gene that mediates cell apoptosis, was expressed at relatively much higher levels in NF-κB Rel A-knockdown than wild-type cardiac myocytes, which was reduced by testosterone treatment via AR mediation (Fig. 6). Thus, both Akt and Caspase-3 may be involved in the NFκB-mediated hormonal action on cardiac myocytes. Bcl-2 and Bcl-xL are two important genes in regulation of myocardial apoptosis. Down-regulation of Bcl-2 and Bcl-xL, via either a direct knockout or NF-κB-mediated inhibition of them, has demonstrated to lead apoptotic death in cardiac myocytes including H9c2 cells, and clinical therapies may be through up-regulation of the activity of these genes [7,24]. By using Western blotting analysis, our study also tried to assess protein expression of Bcl-2 and Bcl-xL in the myocytes, but did not succeed. The Nrf2/Keap1 (nuclear factor [erythroid-derived 2]like 2/Kelch-like ECH-associated protein 1) axis plays a crucial role as a regulator of cellular response against endogenous or exogenous

50

F.-Y. Xiao et al. / Life Sciences 133 (2015) 45–52

Fig. 5. Testosterone increases oxidative stress-stimulated Akt activation in normal but not NF-κB-knockdown cardiac myocytes. H9c2 cells were incubated with or without NF-кB RelA siRNA (48 h), testosterone (T, 50 or 100 nM, 24 h), then H2O2 (H, 200 μM, 6 h), and phosphorylation Akt (pAkt) protein assessed by Western blotting analysis. Photos: Western blots from a representative experiment. Bar graph: Mean ± SD of relative levels of pAkt protein from three independent experiments; *p b 0.05, **p b 0.01, vs. C; p values (normal vs. RelA-knockdown cells) less than 0.01.

Fig. 6. Knockdown of NF-κB expression induces Caspase-3 expression in cardiac myocytes and testosterone has partly inhibited such expression via an AR mediated manner. H9c2 cells were incubated with or without NF-кB RelA siRNA (48 h), flutamide (F, 100 nM, 24 h) then testosterone (T, 50 or 100 nM, 18 h), and Caspase-3 proteins assessed by Western blotting analysis. Photos: Western blots from a representative experiment. Bar graph: Mean ± SD of relative levels of Caspase-3 protein from three independent experiments; *p b 0.05, vs. normal cells; #p b 0.05, vs. C; p values (normal vs. RelA-knockdown cells) less than 0.01.

electrophilic assaults under superoxide conditions; a crosstalk of NFκB and Nrf2 regulates the cell physiology including apoptosis under oxidative stress [21]. Our study could not access this molecular signal due to technical limitation. To date, data concerning the possible roles of these molecules in the hormonal action on the heart remain limited. Like other steroid hormones, testosterone, particularly exogenous testosterone, may have a rapid non-genomic action via SMBS on the cell membrane [12]. This action is not mediated by classical AR and acts through activation of membrane kinase (e.g. ERK, PK) signalling pathways, resulting in a series of biological activities. In our study, testosterone supplement activated ERK activity, assessed by increased levels of pERK1/2 protein, in the myocytes, which could not be blocked by flutamide treatment or silencing NF-κB gene, indicating an ARdependent non-genomic action. Activation of ERK signalling may result in many bioactions including enhancement of cytoplasmic RelA/p50 dimer formation and NF-κB activation [10] as well as mediation of Nrf2/Keap1 signalling and initiate the transcription of antioxidant genes [21]. However, in our study blocking ERK activity does not significantly affect the protective effect of testosterone on H9c2 myocytes under oxidative stress (Fig. 8), indicating that the hormonal action does not depend on this membrane kinase signalling pathway. The present study provides some information concerning the molecular mechanisms underlying myocardial protection of androgens and the possible benefit of exogenous testosterone supplements to the heart. Physiological actions of androgenic hormones on the human body are complex and clinical effects of the hormonal therapy may be many and varied. Exogenous supplementation of testosterone is not recommended for patients with prostate cancer as it can reverse any gains from androgen ablation therapy and worse still cause the cancer to metastasize. The translational significance of this mechanistic study down the road is therefore limited to patients with cardiovascular history who have no prostate cancer co-morbidity.

Fig. 7. Testosterone supplements stimulate phosphorylation of ERK protein in an AR- and NF-кB-independent manner. H9c2 cells were incubated with or without NF-кB RelA siRNA (48 h), flutamide (F, 100 nM, 24 h), then testosterone (T, 50 or 100 nM, 18 h), and phosphorylation ERK1/2 (pERK) protein assessed by Western blotting analysis. Photos: Western blots from a representative experiment. Bar graph: Mean ± SD of relative levels of pERK protein from three independent experiments; *p b 0.05, **p b 0.01, vs. C.

F.-Y. Xiao et al. / Life Sciences 133 (2015) 45–52

51

Fig. 8. Inhibition of the ERK activation does not affect the protective effect of testosterone on cardiac myocytes undergoing superoxide injury. H9c2 cells were incubated with UO123 (1 μM, 24 h), testosterone (T, 0 to 100 nM, 18 h), and H2O2 (H, 200 μM, 6 h), then stained with Annexin-V/PI and analysed by FACS. FACS maps: from a representative experiment; apoptotic cells positively stained by Annexin-V and sorted in M1 area. Bar graph: Mean ± SD of the percentage of apoptotic cells from three independent experiments; *p b 0.05, vs. H.

5. Conclusion This in vitro study demonstrates that exogenous testosterone supplements protect cardiac myocytes under oxidative stress, via a classical AR mediation manner but not non-genomic membrane action. The hormonal protection is dependent on normally functional canonical NF-κB (RelA/p50) signalling pathways and accompanied with an enhancement of activation of Akt activity and reduction of Caspase 3 expression. The data provide information, at molecular level, to the basis of theoretical benefit of testosterone therapy in myocardial injury and heart failure. Further studies need to explore the NF-κB-mediated hormonal protection under in vivo conditions. Conflict of interest statement The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Acknowledgements This study was supported by an International Cooperation/Exchange Scholarship (to FY Xiao) of Guangxi Education Department (Nanning, P.R. China) and Lab Supporting Funds of Monash University (Melbourne, Australia). References [1] P. Alexandersen, J. Haarbo, I. Byrjalsen, H. Lawaetz, C. Christiansen, Natural androgens inhibit male atherosclerosis. A study in castrated, cholesterol-fed rabbits, Circ. Res. 84 (1999) 813–819. [2] D. Baetz, K.M. Regula, K. Ens, J. Shaw, S. Kothari, N. Yurkova, L.A. Kirshenbaum, Nuclear factor-kappaB-mediated cell survival involves transcriptional silencing of the mitochondrial death gene BNIP3 in ventricular myocytes, Circulation 112 (2005) 3777–3785. [3] N. Bassil, S. Alkaade, J.E. Morley, The benefits and risks of testosterone replacement therapy: a review, Ther. Clin. Risk Manag. 5 (2009) 427–448.

[4] M. Brown, M. McGuinness, T. Wright, X. Ren, Y. Wang, G.P. Boivin, H. Hahn, A.M. Feldman, W.K. Jones, Cardiac-specific blockade of NF-κB in cardiac pathophysiology: differences between acute and chronic stimuli in vivo, Am J Physiol Heart Circ Physiol. 289 (2005) H466–H476. [5] M.A. Cavasin, S.S. Sankey, A.L. Yu, S. Menon, X.P. Yang, Estrogen and testosterone have opposing effects on chronic cardiac remodeling and function in mice with myocardial infarction, Am J Physiol Heart Circ Physiol. 284 (2003) H1560–H1569. [6] F. Er, G. Michels, N. Gassanov, F. Rivero, U.C. Hoppe, Testosterone induces cytoprotection by activating ATP-sensitive K + channels in the cardiac mitochondrial inner membrane, Circulation 110 (2004) 3100–3107. [7] M.M. Gallogly, M.D. Shelton, S. Qanungo, H.V. Pai, D.W. Starke, C.L. Hoppel, E.J. Lesnefsky, J.J. Mieyal, Glutaredoxin regulates apoptosis in cardiomyocytes via NFκB targets Bcl-2 and Bcl-xL: implications for cardiac aging, Antioxid. Redox Signal. 12 (2010) 1339–1353. [8] K.L. Golden, J.D. Marsh, Y. Jiang, J. Moulden, Acute actions of testosterone on contractile function of isolated rat ventricular myocytes, Eur. J. Endocrinol. 152 (2005) 479–483. [9] T. Hamid, S.Z. Guo, J.R. Kingery, X. Xiang, B. Dawn, S.D. Prabhu, Cardiomyocyte NF-κB p65 promotes adverse remodelling, apoptosis, and endoplasmic reticulum stress in heart failure, Circ. Res. 89 (2011) 129–138. [10] W. Joseph, J.W. Gordon, L.A. Kirshenbaum, Multiple factors of NF-κB in the heart: to be or not to NF-κB, Circ. Res. 108 (2011) 1122–1132. [11] P. Kratsios, M. Huth, L. Temmerman, E. Salimova, M. Al Banchaabouchi, A. Sgoifo, M. Manghi, K. Suzuki, N. Rosenthal, F. Mourkioti, Antioxidant amelioration of dilated cardiomyopathy caused by conditional deletion of NEMO/IKKgamma in cardiomyocytes, Circ. Res. 106 (2010) 133–144. [12] S. Ling, P.A. Komesaroff, K. Sudhir, Cardiovascular physiology of androgens and androgen testosterone therapy in postmenopausal women, Endocrine, Metabolic & Immune Disorders — Drug Targets 9 (2009) 29–37. [13] J. Liu, S. Tsang, T.M. Wong, Testosterone is required for delayed cardioprotection and enhanced heat shock protein 70 expression induced by preconditioning, Endocrinology 147 (2006) 4569–4577. [14] P.Y. Liu, A.K. Death, D.J. Handelsman, Androgens and cardiovascular disease, Endocr. Rev. 24 (2003) 313–340. [15] E. Martin-Merino, S. Johansson, T. Morris, L.A. Carcia Rodriguez, Androgen deprivation therapy and the risk of coronary heart disease and heart failure in patients with prostate cancer: a nested case–control study in UK primary care, Drug Saf. 34 (2011) 1061–1077. [16] A. Misra, S.B. Haudek, P. Knuefermann, J.G. Vallejo, Z.J. Chen, L.H. Michael, N. Sivasubramanian, E.N. Olson, M.L. Entman, D.L. Mann, Nuclear factor-kappaB protects the adult cardiac myocyte against ischemia-induced apoptosis in a murine model of acute myocardial infarction, Circulation 108 (2003) 3075–3078. [17] H. Murata, Y. Ihara, H. Nakamura, J. Yodoi, K. Sumikawa, T. Kondo, Glutaredoxin exerts an antiapoptotic effect by regulating the redox state of Akt, J. Biolumin. Chemilumin. 278 (2003) 50226–50233.

52

F.-Y. Xiao et al. / Life Sciences 133 (2015) 45–52

[18] S. Mustapha, A. Kirshner, D. De Moissac, L.A. Kirshenbaum, A direct requirement of nuclear factor-kappaB for suppression of apoptosis in ventricular myocytes, Am J Physiol Heart Circ Physiol. 279 (2000) H939–H945. [19] K.M. Regula, D. Baetz, L.A. Kirshenbaum, Nuclear factor-kappaB represses hypoxiainduced mitochondrial defects and cell death of ventricular myocytes, Circulation 110 (2004) 3795–3802. [20] J. Scheuer, A. Malhotra, T.F. Schaible, J. Capasso, Effects of gonadectomy and hormone replacement on rat hearts, Circ. Res. 61 (1987) 12–19. [21] T.M. Stępkowski, M.K. Kruszewski, Molecular cross-talk between the NRF2/KEAP1 signaling pathway, autophagy, and apoptosis, Free Radic. Biol. Med. 50 (2011) 1186–1195.

[22] L. Timmers, J.K. Keulen, I.E. Hoefer, M.F.L. Meijs, B. van Middelaar, K. den Ouden, C.J.A. van Echteld, G. Pasterkamp, D.P.V. de Kleijn, Targeted deletion of nuclear factor κB p50 enhances cardiac remodelling and dysfunction following myocardial infarction, Circ. Res. 104 (2009) 699–706. [23] J.M. Vicencio, C. Ibarra, M. Estrada, M. Chiong, D. Soto, V. Parra, G. Diaz-Araya, E. Jaimovich, S. Lavandero, Testosterone induces an intracellular calcium increase by a nongenomic mechanism in cultured rat cardiac myocytes, Endocrinology 147 (2006) 1386–1395. [24] Q. Zhang, W. Huang, X.Y. Lu, Y. Yang, Ghrelin protects H9c2 cells from hydrogen peroxide-induced apoptosis through NF-κB and mitochondria-mediated signalling, Eur. J. Pharmacol. 654 (2011) 142–149.