The heart was quickly excised, and the posterior papillary muscle was secured utilizing a set of rings. One of these ends was affixed to maintain stability, while the other was linked to a force transducer (TSD125 – Byopac Systems, Inc; CA), which was further linked to an amplifier (DA100C Byopac Systems, Inc; CA). Both the transducer and the amplifier were housed within a glass container brimming with 20 mL of nutrition solution with specific proportions of electrolytes and glucose (NaCl 120 mM; KCl 5.4 mM; CaCl2 1.25 mM; MgCl2 1.2 mM; NaH2PO4 2 mM; Na2SO4 1.2 mM; NaHCO3 18 mM and glucose 11 mM). This solution was subjected to a 95% O2 and 5% CO2 mixture at 30°C, pH 7.4. Subsequently, the papillary muscles underwent electrical stimulation provided by a pair of silver electrodes, which delivered rectangular pulses of 15 ms duration at a voltage 1.5 times the threshold (10 – 20 mV, 0.5 Hz). The papillary muscles were stretched to the point of maximum active tension muscle length (Lmax). Following this, a stabilization period of 30 minutes was allowed before the beginning of the experimental protocols. Data was captured using an acquisition system (MP100 Byopac Systems, Inc; CA), registering at an impressive rate of 500 samples per second. At the end, the papillary muscles were weighed, and the developed isometric contraction force was adjusted according to the weight of the muscles. These included the isometric contraction, presented as force expressed in grams per milligram of muscle weight (F; g/mg), the activation and relaxation times (time from contraction initiation to maximum strength peak and maximum peak time to 50% of isometric relaxation respectively), as well as values obtained from positive and negative force time derivatives (+ and - dF/dt; g/mg/s). The assessment of inotropic response to calcium was analyzed using a calcium concentration-response curve, applying varied concentrations of extracellular calcium: 0.62, 1.25, 2.5, and 3.75 mM CaCl₂ in the perfusion solution. The resulting force variation was weight-corrected for the muscles and expressed in g/mg.

SSM and IFM were isolated according to the method of Palmer et al. (20), with minor modification (19 – 22). Briefly, the LV was rinsed in ice-cold Chappel-Perry buffer (100 mM KCl, 50 mM MOPS, 5 mM MgSO₄, 1 mM ATP, 1 mM EGTA, 2 mg/ml BSA), blotted dry and then weighed. The ventricles were minced and homogenized in 1:10 (wt/vol) ice-cold Chappel-Perry buffer. The homogenates were centrifuged at 700 × g for 10 min. The supernatant containing SSM was extracted and centrifuged again at 10,000 x g to isolate SSM. The remaining pellet from the 700 x g spin was resuspended in KCl-MOPS-EGTA buffer containing 100 mM KCl, 50 mM MOPS, and 0.5 mM EGTA at pH 7.4, and treated with trypsin (5 mg/g) for 10 min at 4°C. The samples were incubated with trypsin inhibitor (albumin) and spun down at 700 x g for 10 min. The IFM-containing supernatant was then spun down at 10,000 x g for 10 min. The pellets were washed twice in ice-cold Chappel-Perry buffer and spun down at 10,000 x g for 10 min and then resuspended in ice KME buffer. The concentration of mitochondrial protein was measured by the Lowry method using bovine serum albumin as a standard.

Mitochondrial respiration was assessed in both IFM and SSM as described previously (18 – 23). Isolated mitochondria (0.5 mg/mL) were respired in respiration buffer containing 100 mM KCl, 50 mM 3-(N-morpholino)propanesulfonic acid buffer (MOPS), 5 mM KH₂PO₄, 1 mM EGTA and 1mg/mL BSA/Fraction V, using a polarographic oxygen-sensing systems for measurement of oxygen consumption in liquid suspension (Qubit System, Kingston, ON, Canada). States 3 and 4 respiration were measured with glutamate + malate (10 and 5 mM, respectively), pyruvate + malate (10 and 5 mM, respectively), palmitoyl-L-carnitine (40 µM), succinate (20 mM) + rotenone (7.5 µM) or ascorbate (0.2 mM) + N,N,N’,N’-tetramethyl-p-phenylenediamine (TMPD 0.5 mM) to assess respiration through the electron transport chain (ETC), exclusively. State 3 respiration (ADP-stimulated) was measured in the presence of 2 mM ADP. State 4 respiration (ADP-limited) was assessed after ADP consumption. Respiratory control ratio, the ratio of state 3 to state 4 was calculated to assess the control of oxygen consumption by phosphorylation. The ratio of ADP added in the chamber to the total amount of oxygen consumed in state 3 (ADP:O ratio) was calculated as an index of the efficiency of oxidative phosphorylation.

Mitochondrial size and membrane potential were measured as previously described (24, 19). Briefly, isolated IFM and SSM were stained with MitoTracker Green FM (Molecular Probes, Carlsbad, CA) and assessed using a flow cytometer (FACSCanto II, BD Biosciences). To confirm differences in absolute mitochondria size, we included size calibration beads composed of microsphere suspensions ranging in size from 0.5 to 6 µm to serve as reliable size references. The arithmetic mean output from the forward scatter detector was used as an index of mitochondrial size. For membrane potential, mitochondria were incubated with 5,5’,6,6’-tetrachloro-1,1’,3,3’-tetraethylbenzimidazol carbocyanine iodide (JC-1) (Molecular Probes, Carlsbad, CA) at a final concentration of 0.3 µM. The shift to orange is due to dye aggregates forming upon polarization, causing a shift in emitted light from 530 nm (green) to 590 nm (orange).

Ca²⁺-induced MPTP was evaluated in SSM and IFM from LV myocardium using previously described method (18, 19). The capacity for Ca²⁺ uptake was evaluated in isolated mitochondria by following the extramitochondrial [Ca²⁺] during a progressive, escalating exposure to Ca²⁺. Briefly, using a 96-well fluorometric plate reader at 37°C, 250 µg/mL of mitochondrial protein was suspended in calcium free buffer (100 mM KCl, 50 mM MOPS, 5 mM KH₂PO₄, 1 mM MgCl₂, 5 mM EGTA) with glutamate + malate as substrate (5 and 2.5 mM). A bolus injection of 25 nmol free Ca²⁺ was injected every 7 min, and extramitochondrial [Ca²⁺] was recorded every 40-s using 750 nM Calcium Green-5N.

Serum levels of testosterone and estrogen were measured according to the manufacturer. Testosterone levels were measured using the Testosterone ELISA kit (#29K116 42K034DRG Instruments GmbH, Germany). The assay detection limit for testosterone was 0.083 ng/mL. The intra-assay coefficient of variation for each assay was between 3.28% and 4.16%. The inter-assay coefficient of variation for each assay was between 4.73% and 9.94%. Estrogen levels were measured using the Estrogen ELISA kit (#42K026-3DRG Instruments GmbH,Germany) (25, 26). The assay detection limit for estradiol was 10.60 pg/mL. The intra-assay coefficient of variation for each assay was between 8.70% and 9.23%. The inter-assay coefficient of variation for each assay was between 6.87% and 14.91%.

Western blot was performed as previously described (27). Samples (25-80 µg for whole heart sample and 10-30 µg for isolated mitochondria) were separated using 7.5%, 10% or 12% SDS-PAGE. Proteins were transferred to nitrocellulose membranes and incubated with specific antibodies: goat polyclonal anti-Serca-2 (1:1000; Santa Cruz Biotechnology-USA), rabbit polyclonal anti-p-PLB^Ser16 (1:2500; Bradrilla-UK), mouse monoclonal anti-PLB (1:3000; Santa Cruz Biotechnology-USA), rabbit polyclonal anti-PGC-1α (1:1000; Santa Cruz Biotechnology-USA), rabbit polyclonal anti-NRF1 (1:500; Santa Cruz Biotechnology-USA), mouse monoclonal anti-Mfn-1 (1:200; Santa Cruz Biotechnology-USA), mouse monoclonal anti-Mfn-2 (1:200; Santa Cruz Biotechnology-USA), rabbit polyclonal anti-p-AMPK-α (1:1000, Cell Signaling, USA), rabbit polyclonal anti-AMPK-α (1:1000, Cell Signaling, USA), mouse monoclonal anti-total OXPHOS (CI subunit NDUFB8, CII-30kDa, CIII-Core protein 2, CIV subunit I and CV alpha subunit, 1:2500, Abcam, USA), rabbit polyclonal anti-MCU (1:1000, from Sigma-Aldrich-USA), rabbit polyclonal anti-SOD-1 (1:6000; Sigma-Aldrich-USA), mouse monoclonal anti gp91phox (1:2500 from Transduction Laboratories-USA), rabbit polyclonal anti SOD-1 (1:2000, Sigma-Aldrich-USA), mouse monoclonal anti-Cyclophilin D (1:2000, Mitoscience, USA), mouse monoclonal anti-TFAM (1:1000, Santa Cruz Biotechnology-USA), mouse monoclonal anti-catalase (1:12000, Sigma-Aldrich-USA), rabbit polyclonal anti-SOD-Mn (1:5000, Millipore-USA), rabbit polyclonal anti-NOX-1 (1:1000, Sigma-Aldrich-USA), rabbit polyclonal anti-ACE-1 (1:200, Santa Cruz Biotechnology-USA).

After washing, the membranes were incubated with anti-mouse (1:5.000, Sigma-Aldrich-USA), anti-goat (1:5.000, Sigma-Aldrich-USA) or anti-rabbit (1:5000, Sigma-Aldrich-USA) immunoglobulin antibodies conjugated to horseradish peroxidase. The immunocomplexes were detected using ECL Prime (Amersham International-UK) and the bands were detected using a Chemidoc system (Bio-Rad, USA). The same membranes were then used to determine GAPDH expression using a mouse monoclonal antibody (1:5000; Santa Cruz Biotechnology-USA) or ponceau staining for isolated mitochondria.

Hydrogen peroxide production in isolated mitochondrial subpopulations IFM and SSM was determined using the oxidation of the fluorogenic indicator Amplex Red (Invitrogen) in the presence of horseradish peroxidase as previously described (28). The concentrations of horseradish peroxidase and Amplex Red in the incubation were 0.1 unit/mL and 50μM, respectively, and detection of fluorescence was assessed on a Molecular Devices Flex Station 3 fluorescence plate reader (Molecular Devices, Sunnyvale, CA) with 530-nm excitation and 590-nm emission wavelengths. Standard curves were obtained by adding known amounts of H₂O₂ to the assay medium in the presence of the substrates amplex red and horseradish peroxidase. H₂O₂ production was initiated in mitochondria using glutamate + malate (G + M) and succinate + rotenone (S + R) as substrates.

To identify carbonyl groups that are introduced into the amino acid side chains after oxidative modification of proteins, 2D-oxyblot analysis was performed (29). The derivative that was produced by reaction with 2,4-dinitrophenylhydrazine (DNPH) was immunodetected by an antibody specific to the attached DNP moiety of proteins using a commercial kit (Oxyblot, Millipore).

The DNPH derivatization was carried out on 20 μg of protein for 15 minutes according to the manufacturer’s instructions. One-dimensional electrophoresis was carried out on a 12% SDS/polyacrylamide gel after DNPH derivatization. Proteins were transferred to nitrocellulose membranes which were then incubated in the primary antibody solution (anti-DNP 1: 150) for 2 h, followed by incubation in the secondary antibody solution (1:300) for 1 h at room temperature. The washing procedure was repeated eight times within 40 minutes. Immunoreactive bands were visualized by enhanced chemiluminescence (ECL; Amersham Biosciences) and immunodetection was carried out using a ChemiDoc system (Bio-Rad). To determine specificity, the oxidized proteins provided by the kit were included as a positive control. Treatment of samples with a control solution served as a negative control to the DNPH treatment. As an additional control, the anti-DNP antibody was omitted.

Values are shown as mean ± standard error of the mean (SEM). Differences were analyzed using one or two-way ANOVA followed by a Fisher test. P < 0.05 was considered significant.

As shown in Table 1 , castration of male animals for 12 weeks did not change body weight, left or right ventricle weight, demonstrating that endogenous testosterone does not affect the heart mass. As expected, castration decreased seminal vesicle weight after 12 weeks when compared to SHAM and testosterone replacement was able to restore this value.

Testosterone levels decreased after 12 weeks in OQT animals and (SHAM: 2.09 ± 0,18 ng/mL; OQT: 0,53 ± 0.17 ng/mL^* and OQT + T: 2.16 ± 0.32 ng/mL^# n = 6 animals per group) and testosterone replacement restored to the SHAM level. Estradiol was also measured and was not different between groups (SHAM: 71.6 ± 5.2 pg/mL; OQT: 76 ± 5 pg/mL; OQT + T: 68 ± 7.71 pg/mL n = 6 animals per group).

Myocardial contractility was analyzed using isolated papillary muscles at baseline and after exposure to different extracellular calcium concentrations. As expected, increasing extracellular calcium concentration resulted in a positive inotropic response in papillary muscles of all groups. However, the inotropic response to calcium at the concentrations of 0.62, 1.25, 2.5 and 3.75 mM were lower in the OQT group. ( Figures 1A–C ). Testosterone replacement prevented the reduction in contractile force because papillary muscles from OQT + T group presented similar inotropic responses as the SHAM group ( Figures 1A–C ). Figure 1C shows representative contractile responses recorded from papillary muscles paced at 0.5 Hz.

In myocardial cells, calcium handling depends upon Serca-2a and PLB function (30, 31). Serca2a and PLB protein expressions were measured in the hearts of our animals. As shown in Figure 1D , Serca2a protein expression was significantly decreased in the OQT group and this decrease was prevented by testosterone replacement. PLB protein expression was significantly increased in the OQT group and testosterone replacement prevented this increase in PLB protein expression ( Figure 1E ). We also measured phosphorylated phospholamban (P-PLB) at ser¹⁶, the main target for PKA phosphorylation. P-PLB levels were significantly reduced in the OQT group, indicating that PLB activity was enhanced in the OQT group ( Figure 1F ).

Yield refers to the amount of mitochondrial protein per gram of tissue. The yield for IFM was similar between the groups ( Figure 2A ); however, the yield for SSM was significantly increased in the OQT group when compared to SHAM ( Figure 2D ). Testosterone replacement prevented the increase in mitochondrial yield in the castrated animals. ( Figure 2D ).

To determine mitochondrial membrane potential and morphological differences between SHAM, OQT and OQT + T mitochondrial subpopulations, we used a gated flow cytometry method. Forward-scattered light (FSC) was used to estimate mitochondrial size, whereas assessment of side-scattered light (SSC) was used to estimate mitochondrial complexity, both of which were based on a logarithmic scale. Membrane potential of the mitochondrial subpopulations was analyzed by flow cytometry. Overall, baseline mitochondrial membrane potential (with endogenous substrate) was significantly increased in the IFM from OQT animals ( Figure 2B ), whereas the membrane potential was unchanged in the SSM ( Figure 2E ). We next incubated the samples with glutamate + malate (10 and 5 mM, respectively) to stimulate the ETC. While there was an initial increase in membrane potential in all groups when the substrates were added, membrane potential was significantly increased in the IFM, but not the SSM from the OQT groups ( Figures 2C, F . For both substrates, the increase in IFM membrane potential was prevented by testosterone replacement ( Figures 2B, C ).

Mitochondrial morphology was assessed by flow cytometry to show relative morphological differences between the two subpopulations and confirm the success of the isolation. SSM were larger in size (FSC) and possessed greater internal complexity (SSC) compared with IFM, which were smaller and more compact ( Figures 3A–D ). Castration significantly increased the size and complexity of IFM compared to SHAM ( Figures 3A, C ). These increases were prevented by testosterone replacement. Conversely, the size and complexity of SSM were not affected by castration ( Figures 3B, D ).

To further investigate the mechanisms involved in mediating the changes in mitochondrial morphology, we developed western blots for mitofusins 1 and 2 which are involved in the regulation of mitochondrial fusion. Figures 3E, F show the expression of mitofusins 1 and 2, respectively. Mfn-1 protein expression was not altered between the groups. Mfn-2 was decreased significantly in the OQT group ( Figure 3F ). As mitochondrial biogenesis is regulated by PGC-1α and PGC-1α translocation to mitochondria is regulated by AMPK-α, we also measured the expression and phosphorylation of these two proteins. PGC-1α protein expression was not altered after twelve weeks of hormone deprivation ( Figure 3G ), whereas AMPK-α protein phosphorylation was decreased in the OQT group and restored to control levels by testosterone replacement ( Figure 3H ).

To assess the effects of testosterone deficiency on mitochondrial bioenergetics in the IFM and SSM, of the SHAM, OQT and OQT + T groups, we measured mitochondrial respiration with different substrates (glutamate + malate, pyruvate + malate, palmitoyl-L-carnitine, succinate + rotenone and ascorbate + TMPD) that use distinct oxidative pathways, mitochondrial transport and provide specific substrates to Complex I, Complex II and Complex IV.

The maximal rate of mitochondrial respiration (State 3) in IFM expressed as nanomoles of oxygen per mg of mitochondrial protein was lower in the OQT group when compared to SHAM and OQT+T with glutamate + malate, pyruvate + malate, rotenone + succinate, palmitoyl-L-carnitine and ascorbate + TMPD as substrates ( Figures 4A–E ). To further investigate if the reduction in state 3 in the OQT group was due to low expression of the ETC complexes, we developed western blots for complex I-V, which showed that testosterone deficiency increased complex I and II protein expression in IFM ( Figure 4F ). These increases were prevented by testosterone replacement.

We also measured state 3 with all substrates in SSM. State 3 was reduced with pyruvate + malate ( Figure 5B ) and rotenone + succinate ( Figure 5C ), but not with glutamate + malate ( Figure 5A ), palmitoyl-L-carnitine ( Figure 5D ) or ascorbate + TMPD ( Figure 5E ). Western blots for complex I-V did not reveal any differences in protein expression for the SSM ( Figure 5F ).

Two standardized methods were used to assess Ca²⁺-induced MPTP in LV mitochondria. MPTP was assessed from mitochondrial swelling induced by high [Ca²⁺] as reflected by the decrease in absorbance at 540 nm following the addition of a bolus of Ca²⁺ to isolated mitochondria. Measurement of baseline absorbance before the addition of Ca²⁺ was performed to determine basal values. There was a decrease in absorbance with addition of either 100 or 500 nmol Ca²⁺/mg protein in all groups, but no difference was found between the groups (data not shown).

Ca²⁺-induced MPTP was also assessed by measuring the ability of isolated mitochondria to take up added Ca²⁺. As shown in Figure 6 (top panel), IFM from the OQT group had significantly enhanced Ca²⁺ retention capacity compared to the SHAM and OQT + T groups. The Ca²⁺ uptake in SSM was the same in all groups ( Figure 6 , bottom panel).

To further elucidate whether these alterations in mitochondrial Ca²⁺ retention capacity were due to changes in cyclophilin D or MCU, we developed western blots for these two proteins. As shown in Figure 6 , there were no differences in protein expression in either IFM (top panel) or SSM (bottom panel).

PGC-1α trans-activates nuclear respiratory factor 1, which in turn, activates mtDNA transcription factor A (TFAM) that regulates mtDNA transcription and replication. Subsequently, NRF-1 promotes transcription of mitochondrial transcription factor TFAM, which targets mtDNA-encoded genes. TFAM protein expression was significantly reduced after castration in IFM ( Figure 7A ), but not in SSM ( Figure 7B ). Again, the reduction in TFAM could be prevented by testosterone replacement ( Figure 7A ).

In the IFM, the expression of mitochondrial Mn-SOD ( Figure 7C ) and catalase ( Figure 7E ) were significantly decreased after castration and this reduction was prevented by testosterone replacement. In SSM, mitochondrial SOD-Mn expression ( Figure 7D ) was unchanged by castration, while catalase levels ( Figure 7F ) were significantly increased. The PGC-1a protein expression decreased in both IFM ( Figure 7G ) and SSM ( Figure 7H ) OQT group.

It is well known that NADPH oxidase transfers an electron to molecular oxygen to form superoxide anion. We examined the role of NADPH oxidase isoforms and mitochondrial protein oxidation in the hearts from our groups. Western blots showed significant increases in NOX-1 and gp91^phox in the hearts of the OQT group and testosterone replacement restored this increase to SHAM levels ( Figures 8A, B ). We next analyzed protein expression for SOD-Cu-Zn ( Figure 8C ) and angiotensin converting enzyme (ACE-1) ( Figure 8D ). SOD-Cu-Zn in the heart was not changed after castration; however, ACE-1 protein expression was significantly increased in the OQT group. This increase was blocked by testosterone replacement. As reactive oxygen species can promote protein carbonylation, we developed western blots to detect DNPH-derivatized proteins in both mitochondrial subpopulations. As shown in Figures 9A, B , orchidectomy significantly increased mitochondrial protein carbonylation in IFM, but not in SSM. The increased protein carbonylation in IFM was prevented by testosterone replacement ( Figure 9A ).

In order to understand the influence of oxidative stress, a separate group of OQT rats (n=8) was treated with apocynin (1.5 mM in drinking water) for 12 weeks after castration. Myocardial contractility was analyzed using isolated papillary muscles at baseline ( Figure 10A ) and with different extracellular calcium concentrations ( Figure 10B ). Increasing extracellular calcium concentration elicited a positive inotropic response in the papillary muscles of all groups; however, the inotropic response was significantly smaller in the OQT group. Treatment of the castrated rats with apocynin prevented the reductions in the inotropic responses to calcium (0.62, 1.25, 2.5 and 3.75 mM) ( Figures 10A, B ). Treatment with apocynin also prevented the castration-related decreases in state 3 oxidative phosphorylation, independent of substrate ( Figures 10C–F ).

The ability of mitochondria to generate H₂O₂ was evaluated in isolated interfibrillar mitochondria (IFM) subpopulations (Figures 10G, H) and subsarcolemmal mitochondria (SSM) subpopulations (Figures 10I, J) from the left ventricular (LV) myocardium using the amplex red assay. Testosterone deficiency resulted in an increase in hydrogen peroxide production in IFM ( Figure 10G ) when using the glutamate + malate substrate. Treatment with apocynin prevented oxidative stress-induced mitochondrial H₂O₂ production in the IFM subpopulation when using the glutamate + malate substrate.