The following reagents were from commercial sources: testosterone, bicalutamide, AIP, and 5-bromo-2-deoxyuridine (BrdU), Sigma-Aldrich (St. Louis, MO, United States); anti-phospho-CaMKII (Thr286, cat. # 3361), anti-MEF2C (cat. # 5030) and anti-phospholamban (PLN, cat. # 14562) antibodies, Cell Signaling Technology (Danvers, MA, United States); CellTracker Green (5-chloromethyl fluorescein diacetate) from Thermo-Fisher Scientific (Rockford, IL, United States); anti-phospho-PLN (Thr17, cat. # sc-24565), anti-CaMKII (cat. # sc-5392) and anti-AR (cat. # sc-815) antibodies and siRNAs targeting CaMKIIδ (cat. # sc-38953), AR (cat. # sc-29204), and MEF2C (cat. # sc-38062), Santa Cruz Biotechnology (Santa Cruz, CA, United States); [³H]-leucine, NEN Radiochemicals Perkin Elmer (Waltham, MA, United States); and collagenase type II, Worthington Biochemical Corporation (Lakewood, CA, United States). All other reagents were of analytical grade and are commercially available.

This study was carried out in accordance with the recommendations of the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The protocol was approved by the local Institutional Animal Care and Use Committee of the Faculty of Medicine, University of Chile (protocol CBA # 0768 FMUCH). Rats were bred in the Animal Breeding Facility of the Faculty of Medicine, University of Chile. Primary cultures of neonatal cardiac myocytes were prepared from the hearts of 1–3-day-old Sprague-Dawley rats, as described previously (Ibarra et al., 2013). The use of this protocol yields cardiac myocyte cultures that are at least 90% pure, and these cultures represent an established cellular model for studying cardiac hypertrophy (Chlopcikova et al., 2001). To prevent fibroblast growth, the growth medium was supplemented with 2.5 μM BrdU. Cardiac myocytes were cultured in growth medium containing DMEM:M-199 (4:1) supplemented with 10% FBS and 1% penicillin-streptomycin, and were deprived of FBS for 24 h before hormone stimulation.

Twelve, 8-week-old, male Sprague-Dawley rats (body weight ∼300 g) were ORX. Additionally, six normal rats without testosterone treatment were used as control animals. For surgery, animals were anesthetized with ketamine (70 mg⋅kg^-1, ip) and xylazine (7 mg⋅kg^-1, ip), and after the surgery, the ORX rats were administered tramadol (0.05 mg⋅kg^-1, sc) for pain control and allowed to recover for 7 days before testosterone supplementation. The animals were allowed free access to food and water and were maintained under a 12/12-h light/dark cycle. The ORX rats were randomly assigned to two groups (n = 6 each): ORX plus vehicle (peanut oil) treatment; and ORX plus supplementation with testosterone (125 mg⋅kg^-1⋅week^-1) for 5 weeks. Normal rats treated with vehicle served as the control group. Plasma testosterone concentrations were evaluated using ELISA (Cayman Chemical, Ann Arbor, MI, United States). After the treatment, the ORX and control rats were weighed and then euthanized by administering an overdose of sodium pentobarbital (200 mg⋅kg^-1), after which the hearts were dissected and weighed to calculate the left-ventricle and heart weight ratio with respect to body weight and tibia length. Moreover, seminal vesicles and prostates were weighed to evaluate systemic effects of the administrated testosterone.

MEF2 transcriptional activity was evaluated by using the plasmid 3XMEF2-Luc (Addgene plasmid #32967), which contains MEF2-binding boxes cloned upstream of the firefly luciferase reporter gene; 3XMEF2-Luc was a gift from Dr. Ron Prywes. Furthermore, cardiac myocytes were transfected with either a plasmid expressing a wild-type isoform of CaMKII (XE117 CAMKII-CS2+; Addgene plasmid #16737), or a plasmid expressing a constitutively active form of CaMKII (XE118 CAMKII-T286D-CS2+; Addgene plasmid #16736). In this active form of CaMKII, Thr286 is mutated to Asp, which mimics the phosphorylation of this site and results in CaMKII activation independently of binding to Ca²⁺/calmodulin; XE118 CAMKII-T286D-CS2+ was a gift from Dr. Randall Moon. A plasmid expressing Renilla luciferase was used as the control for transcriptional activity (Promega, Madison, WI, United States). Transfections were performed using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, United States), according to manufacturer specifications, and the plasmid DNA was used at a final concentration of 1 μg⋅mL^-1 in each experimental condition. Cardiac myocytes were incubated with testosterone for 24 h in the presence or absence of inhibitors, and then the cells were lysed and MEF2-Luc and Renilla luciferase activities were measured after 24 h of testosterone stimulation, to allow accumulation of gene product (Wu et al., 2006), using the dual-luciferase kit Assay Reporter System (Promega, Madison, WI, United States) and a luminometer (Berthold luminometer F12, Pforzheim, Germany).

In addition to the inhibitor experiments, we performed knockdown experiments by transfecting cardiac myocytes with siRNAs specifically targeting CaMKIIδ (siRNA-CaMKIIδ), MEF2C (siRNA-MEF2C), and AR (siRNA-AR). As a control, cardiac myocytes were transfected with a non-targeting siRNA (Control siRNA-A; Santa Cruz Biotechnology, sc-37007). For this set of experiments, cardiac myocytes grown on 60-mm dishes were transfected with 20 nM siRNAs by using Lipofectamine 2000, and then protein downregulation in each experimental condition was confirmed through Western blotting.

For mRNA-expression analysis, total RNA was isolated from lysates prepared from homogenized left-ventricle tissue of both normal and ORX rats; lysates were prepared using TRIzol^® reagent (Invitrogen, Carlsbad, CA, United States). Next, 2 μg of the isolated RNA was reverse-transcribed in a reaction volume of 20 μL containing 1 μM Oligo-dT primer, 0.5 μM dNTPs, 10 U of RNase inhibitor, and SuperScript II Reverse Transcriptase (Thermo-Fisher Scientific, Rockford, IL, United States), according to the manufacturer’s instructions. Quantitative real-time PCR (qPCR) was performed using the StepOnePlus^TM Real-Time PCR System (Applied Biosystems, Foster City, CA, United States); at least three independent qPCR experiments were performed for each time point. The following primer sequences were used: β-MHC: 5′-AAGTCCTCCCTCAAGCTCCTAAGT-3′, 5′-TTGCTTTGCCTTTGCCC-3′; GAPDH: 5′-ACATGCCGCCTGGAGAAAC-3′, 5′-AGCCCAGGATGCCCTTTAGT-3′. Expression values were normalized relative to the mRNA levels of GAPDH, used as the internal control, and are reported in units of 2-ΔΔCT ± SE. The CT values were determined by using MXPro software in cases where the fluorescence was 25% higher than background. PCR products were verified using melting-curve analysis.

Immunofluorescent labeling was performed as described previously (Ibarra et al., 2013). Briefly, cardiac myocytes were stimulated with testosterone (100 nM) for different times, fixed with paraformaldehyde (4%), and then incubated with rabbit anti-MEF2C antibody (1:400; 4°C, overnight). Next, the cells were washed thrice with PBS, and then incubated with goat anti-rabbit Alexa Fluor 488 IgG (H + L) (1:500; room temperature, 1 h), after which nuclei were counterstained with 4′-6-diamidino-2-phenylindole (DAPI) for 10 min. Images were acquired using a fluorescence microscope (Colibri.2 LED system, Carl Zeiss AG, Oberkochen, Germany) and analyzed using ImageJ software (NIH, Bethesda, MD, United States). To quantify fluorescence, the summed pixel intensity was calculated for each section delimited by a region of interest (ROI) within either the nucleus or the cytoplasm of cardiac myocytes that had been stimulated or not stimulated with testosterone.

Lysates were prepared from cardiac myocytes that were grown on 60-mm plates and serum-starved for 24 h before exposure to testosterone for various times (indicated in Results and in figure legends). Cells were washed with ice-cold PBS and scraped off plates in 70 μl harvesting lysis buffer [150 mm NaCl, 1 mm EGTA, 1% Nonidet P-40, 50 mm Tris-HCl (pH 7.4), 5 mm sodium orthovanadate, 20 mm NaF, 1 μg/ml aprotinin, 1 μm pepstatin, 20 μm leupeptin, 1 mM benzamidine, and 0.2 mm 4-(2-aminoethyl)benzene sulfonyl fluoride)]. Furthermore, tissue lysates were prepared by homogenizing left-ventricle samples and centrifuging them at 15,000 ×g for 10 min. Supernatants were removed, divided into portions and protein concentrations were determined by using a bicinchoninic acid assay kit according the manufacturer’s instructions (Thermo-Fisher Scientific, Rockford, IL, United States). Equal amounts of proteins (20 μg) were denatured at 100°C in 30% glycerol, 8% sodium dodecyl sulfate, 10% 2-mercaptoethanol, 25 mm Tris-HCl (pH 6.8), and 0.1% bromophenol blue; resolved by 10 or 15% (for PLN) acrylamide-bisacrylamide gels and transferred to nitrocellulose membranes as previously described (Altamirano et al., 2009). Membranes were blocked in 5% bovine serum albumin in Tris-buffered saline containing 0.05% Tween 20 and then immunoblotted with primary antibodies (1:1000). The protein bands in the blots were visualized using the ECL detection kit Westar Supernova (Cyanagen, Bologna, Italy), and band intensities were determined through densitometric scanning with ImageJ software.

Transfected and non-transfected cardiac myocytes were cultured on gelatin-coated coverslip for 24 h, and then the culture medium was replaced with a medium supplemented with testosterone together with or without inhibitors and the cells were cultured for another 48 h. For cell-size measurement, cardiac myocytes were incubated with the vital fluorescent dye CellTracker Green for 45 min, after which fluorescence images were acquired (using a Colibri.2 LED illumination system, Zeiss) and analyzed and compared using ImageJ software. For the measurements, we used at least eight different fields from five independent cultures in each condition (>100 cells). Cell size measurements were performed blinded.

Cardiac myocytes were cultured for 24 h and then incubated with [³H]-leucine (2.5 μCi⋅mL^-1) for an additional 48 h in the presence or absence of testosterone (Duran et al., 2016). Next, the cells were washed four times with ice-cold PBS and treated with 10% trichloroacetic acid at 4°C for 60 min. The samples were centrifuged for 20 min at 15,000 ×g, and the obtained pellets were washed once with ice-cold absolute acetone and dissolved in 0.2 M NaOH. Aliquots of triplicate samples per group were counted in a liquid scintillation counter (Beckman Instruments, Fullerton, CA, United States).

All data are expressed as means ± SEM and presented as fold-induction relative to the control in each experiment. Sample size for each experimental procedure were calculated by a priori power analyses using the G^∗Power 3 software (Faul et al., 2007). Statistical analysis was performed using t-tests or ANOVA for multiple comparisons, followed by a post hoc Bonferroni test. In each figure, “n” indicates the number of independent experiments performed on different cell cultures or animal groups. P < 0.05 was considered statistically significant. All analyses were conducted using GraphPad Prism 5 software (GraphPad Software, San Diego, CA, United States).

CaMKII phosphorylation at Thr286 is a well-established parameter for monitoring CaMKII activity (Ramirez et al., 1997; Hughes et al., 2001). Previously, we showed that testosterone treatment induced a rapid (<30 min) increase in CaMKII phosphorylation in cardiac myocytes (Wilson et al., 2013). To extend these findings, we examined the kinetics of CaMKII phosphorylation over short (5–60 min) and long (3–24 h) periods of stimulation with 100 nM testosterone. In the short time-course, CaMKII phosphorylation increased at 5 min and peaked at 15 min after testosterone stimulation (Figure 1A), whereas in the longer time-course, CaMKII phosphorylation peaked at 3 h, to a level lower than that at 15 min, and then returned to its basal level within 6–9 h (Figure 1B).

Because the maximal effect of testosterone on CaMKII phosphorylation was reached at 15 min after stimulation, we assessed the CaMKII phosphorylation at 15 min in response to stimulation of cardiac myocytes with 1, 50, 100, 500, or 1000 nM testosterone: CaMKII phosphorylation was increased at 50 nM and was maximally elevated at 100 nM, with higher concentrations of testosterone being relatively less effective in increasing CaMKII phosphorylation (Figure 1C). To test whether this increase in phosphorylation was associated with enhanced CaMKII activity, we measured the levels of Thr17 phosphorylation in PLN, a downstream target for activated CaMKII (Mattiazzi et al., 2005). Testosterone (100 nM) increased PLN phosphorylation and the level peaked at 30 min (Figure 1D).

The transcription factor MEF2 is a crucial downstream target that integrates CaMKII signaling pathways (Zhang et al., 2007). To investigate the ability of testosterone to stimulate MEF2 activity, cardiac myocytes were transfected with a MEF2 luciferase reporter plasmid (MEF2-Luc) and then stimulated with 100 nM testosterone for 6, 12, 24, or 48 h. Testosterone-stimulated MEF2 activity peaked at 24 h and was sustained for at least 48 h (Figure 2A). Next, we evaluated MEF2 activity at 24 h after stimulation with different concentrations of testosterone (1, 50, 100, or 1000 nM), which revealed that maximal MEF2 activity was stimulated by 50–100 nM testosterone (Figure 2B); here, treatment with 10 nM IGF-1 for 24 h served as the positive control for MEF2 activation in cardiac myocytes. Moreover, specific activation of MEF2C by testosterone was confirmed knocking down MEF2C protein using RNA interference. Transfection of siRNA-MEF2C reduced MEF2C protein expression by 63% as compared to transfection with a negative-control siRNA (Figure 2C). Downregulation of MEF2C protein abolished the increase in MEF2 activity measured in response to testosterone treatment of cardiac myocytes (Figure 2D).

MEF2 activity and MEF2C nuclear localization are increased by IGF-I, which is a pro-hypertrophic hormone (Munoz et al., 2009). To determine whether testosterone induces nuclear translocation of MEF2C, we evaluated the subcellular distribution of MEF2C protein in cardiac myocytes. After testosterone stimulation, MEF2C immunofluorescence was localized mainly in the nucleus, as evidenced by the overlap with DAPI counterstaining (Figure 2E). Furthermore, quantification of the fluorescence intensity of MEF2C staining in the nucleus and the cytoplasm (Figure 2F) revealed that the nuclear-to-cytoplasmic fluorescence ratio increased and peaked at 30 min after testosterone stimulation, which indicated nuclear translocation of MEF2C protein.

To study whether MEF2 activation by testosterone is mediated through CaMKII, we either transfected cardiac myocytes with siRNA-CaMKIIδ, the principal CaMKII isoform expressed in heart (Gray and Heller Brown, 2014), or pretreated the cells with 1 μM AIP, a cell-permeable peptide inhibitor of CaMKII. Transfection with siRNA-CaMKIIδ resulted in a 52% reduction in CaMKIIδ protein content in cardiac myocytes (Figure 3A), and, notably, CaMKIIδ downregulation abolished the activation of MEF2 induced by testosterone (Figure 3B). Similarly, CaMKII inhibition by AIP blocked the testosterone-induced increase in MEF2 activity (Figure 3C). These results suggest that CaMKII functions upstream of MEF2 in testosterone-triggered MEF2 activation in cardiac myocytes.

To assess the role of AR in testosterone-induced MEF2 activity, cardiac myocytes were transfected with siRNA-AR. In these transfected cells, AR protein expression was decreased by ∼49% (Figure 4A), and, furthermore, testosterone-stimulated increase in MEF2 activity was abolished (Figure 4B). Similar results were obtained in cells pretreated with the AR inhibitor bicalutamide (1 μM) (Figure 4C).

Next, we investigated whether MEF2 activation by testosterone links the CaMKII and AR signaling pathways. To evaluate the roles of CaMKII and AR in the activation of MEF2, cardiac myocytes were transfected with either constitutively active CaMKII (CA-CaMKII) or wild-type CaMKII (WT-CaMKII). As compared to WT-CaMKII overexpression, CA-CaMKII overexpression increased the basal activity of MEF2, and subsequent stimulation with 100 nM testosterone resulted in an additional increase in MEF2 activity (Figure 4D). Moreover, we found that bicalutamide pretreatment of CA-CaMKII-expressing cardiac myocytes did not affect the elevation of MEF2 basal activity, but it prevented the further activity increase induced by testosterone (Figure 4D). Collectively, these results suggest that CaMKII and AR signaling pathways regulate MEF2 activation in response to testosterone in cardiac myocytes.

To examine the involvement of CaMKII/MEF2 and AR signaling pathways in testosterone-induced cardiac hypertrophy, we measured cellular size and [³H]-leucine incorporation, which are both accepted indicators of cardiac myocyte hypertrophy. First, we determined whether hypertrophy depends on AR activation: we either preincubated cardiac myocytes with 1 μM bicalutamide or transfected the cells with siRNA-AR. Stimulation with testosterone (100 nM) for 48 h increased cardiac myocyte size and [³H]-leucine incorporation, and both of these effects were blocked by bicalutamide (Figures 5A,B) and siRNA-AR (Figures 5C,D). Second, when cardiac myocytes were treated with 1 μM AIP, the CaMKII inhibitor, the testosterone-induced increase in both cardiac myocyte size and [³H]-leucine incorporation was blocked (Figures 5E,F). Furthermore, MEF2C downregulation mediated by siRNA-MEF2C abolished cardiac myocyte hypertrophy stimulated by testosterone (Figures 5G,H). These results together suggest that testosterone links the CaMKII signaling pathway to MEF2C- and AR-regulated transcription in cardiac myocyte hypertrophy.

Lastly, we conducted in vivo studies to investigate the physiological relevance of the findings obtained using cultured cardiac myocytes: we examined the function of the CaMKII/MEF2 and AR signaling pathways in control and ORX rats that were or were not administered testosterone. In the ORX rats treated with testosterone, the following ratios were increased: left-ventricle weight/body weight (Figure 6A) and left-ventricle weight/tibia length (Figure 6B). Plasma testosterone concentrations were significantly higher in the testosterone-treated rats than in untreated rats, and, furthermore, demonstrating that testosterone produced its expected effects in this in vivo hypertrophy model, the weights of seminal vesicles and prostates (which are sensitive indicators of testosterone actions) were markedly higher in the testosterone-administered ORX rats than in ORX or control rats (Table 1). Moreover, testosterone supplementation upregulated β-MHC mRNA expression (Figure 6C) as well as β-MHC and SKA protein levels compared with ORX or control rats (Figures 6D,E).

Additionally, we determined the phosphorylation levels of CaMKII (Thr286) and PLN (Thr17) by immunoblotting extracts of left-ventricle tissue from testosterone-administered ORX and control rats. Testosterone increased the phosphorylation of both CaMKII (Figure 7A) and PLN (Figure 7B). Moreover, in the testosterone-treated ORX rats, MEF2C and AR protein levels were higher than untreated ORX or control rats (Figures 7C,D).