All procedures were conducted under the approval of the University of Louisville IACUC in accordance with the NIH Guide for the Care and Use of Laboratory Animals [DHHS publication No. (NIH) 85-23, rev. 1996] (Dai et al., 2010). Cardiac-specific ecSOD Tg mice heterozygous for the transgene used in experiments were produced by crossing with WT C57BL/6 mice with littermate WT mice used as controls as previously described (Obal et al., 2012).

Adult cardiac myocytes from WT and ecSOD Tg mice were isolated as previously described and ~ 2 × 10⁵ myocytes plated per well in 24-well culture plates with inserts (NUNC) (Hu et al., 2007). After equilibration in myocyte culture media, the culture medium was immediately changed to oxygen depleted (bubbled with N₂ for 10 min) hypoxia buffer (NaCl 118 mM, NaHCO₃ 24 mM, NaH₂PO₄ 1 mM, CaCl₂ 2.5 mM, MgCl₂ 1.2 mM, sodium lactate 20 mM, KCl 16 mM, 2-deoxyglucose 10 mM, pH 6.4) and myocytes maintained under hypoxic conditions, 1% O₂, 5% CO₂, and 94% N₂, for 30 min in a hypoxia chamber (Billups-Rothenberg). Myocytes were then reoxygenated when the hypoxia buffer was replaced with normoxic culture media (Hu et al., 2007). Myocytes were reoxygenated for 120 min. Lactate dehydrogenase (LDH) activity in media collected from myocyte culture media immediately after 120 min in normoxic buffer under normoxia was detected by colorimetric assay by measuring the formazan product at 492 nm as previously described (Roche Applied Science) (Hu et al., 2007). Total LDH release was determined in media after treating cells with 2% triton X100. LDH release in each well was expressed as a percentage of total LDH normalized to protein content determined by Bradford assay.

Mitochondria from WT and ecSOD hearts were isolated as previously described (West et al., 2008). WT and ecSOD hearts were minced and homogenized in buffer A, 225 mM mannitol, 70 mM sucrose, 5 mM MOPS, 2 mM EGTA, and 0.2% fatty acid-free BSA, using 5 strokes of a glass-Col homogenizer. The homogenate was centrifuged at 500×g for 10 min at 4°C and the supernatant filtered through cheesecloth. Mitochondria were obtained after centrifugation at 5000×g for 10 min at 4°C followed by 2 washes. Mitochondria (150–200 μg protein/assay) were then resuspended in 200 μl swelling assay buffer, 40 mM HEPES, 195 mM mannitol, 25 mM sucrose, and 0.01 mM EGTA for mitochondrial permeability transition (MPT) studies.

MPT activation was determined by measuring the change in optical density with Ca²⁺-induced mitochondrial swelling at 520 nm as previously described (West et al., 2008). Mitochondria were loaded into wells of a 96 well plate with 1 μM rotenone and 5 mM succinate and read at 37°C for 30 min. Mitochondria from WT and ecSOD Tg hearts were also treated with cyclosporine A (CsA, 10μM) as a control. Mitochondrial swelling was initiated with the injection of Ca²⁺ into each corresponding well with shaking after recording 5 min of baseline. Swelling was recorded for 20 min.

Subcellular fractions were extracted from isolated WT and ecSOD myocytes using the ProtoExtract Subcellular Proteome Extraction kit (Calbiochem). Myocytes from WT and ecSOD Tg mice were homogenized in Fraction I buffer and gently agitated for 10 min at 4°C and centrifuged at 1000×g for 10 min. The supernatant was removed and stored on ice (Fraction I, cytosolic). Extraction buffer II with protease inhibitor cocktail was added to the pellet and the pellet resuspended and incubated for 30 min at 4°C with gentle agitation. The suspension was centrifuged at 6000×g for 10 min at 4°C and the supernatant collected (Fraction II, membrane) on ice. Extraction buffer III with protease inhibitor and Benzonase Nuclease were added to the pellet and the pellet resuspended and incubated for 10 min with agitation at 4°C. The suspension was centrifuged at 7000×g for 10 min at 4°C and the supernatant (Fraction III, nuclear) collected on ice. The pellet was resuspended with extraction buffer IV with protease inhibitor cocktail (Fraction IV, cytoskeletal). Protein concentration was determined by the Non-interfering Protein Assay Kit (Calbiochem).

Hearts from WT and ecSOD Tg mice were isolated, frozen, pulverized, and homogenized in buffer and homogenates prepared and processed for PAGE and Western analysis as described previously (Hu et al., 2007). Blots were probed with anti-ecSOD (1:1000, Sigma Aldrich, S4946), anti-CytC (1:1000, Abcam, 7H8.2C12), anti-vimentin (1:2000, Sigma Aldrich, V6389), Histone H3 (1:1000, Cell Signaling, 9715), and anti-E-cadherin (1:1000, Cell Signaling, 4065) antibodies, visualized by enhanced chemiluminescence and quantitated either by densitometry of film or by phosphorimager (GE, Typhoon) and analysis with ImageQuant (GE).

ecSOD was detected and colocalized to the mitochondrial via the electron transport protein, cytochrome C (CytC) in 4 μm sections cut from paraffin embedded ecSOD Tg hearts by immunofluorescent confocal microscopy (Zeiss LSM 510). Sections were stained with anti-ecSOD (1:100, Sigma Aldrich, S4946) and counterstained with CytC (1:100, Abcam, 7H8.2C12) for mitochondrial colocalization and DAPI for nuclear staining. Line analysis of ecSOD and CytC channel intensity in confocal images was used to provide further delineation of ecSOD and CytC colocalization using NIH ImageJ(1.48Ver).

Data are presented as the mean ± SEM. Groups were compared by unpaired Student's t-test where a P < 0.05 was considered significant.

Our previous studies have shown that cardiac specific ecSOD overexpression protects the heart from ischemia reperfusion injury supporting earlier studies showing ecSOD is cardioprotective (Obal et al., 2012). Cardiac-specific ecSOD expression also demonstrated that ecSOD could be detected in the intracellular compartment of cardiac myocytes in addition to its prototypic extracellular localization. To further investigate the action of intracellular localization of ecSOD, the effect of overexpression on hypoxia reoxygenation injury was examined in myocytes isolated from ecSOD Tg and WT mice. Myocytes, enzymatically isolated from Tg and WT hearts, were incubated under ambient oxygen conditions for 30 min after isolation. Myocytes were then incubated under 1% O₂, 94% N₂, and 5% CO₂ in hypoxic medium for 30 min followed by 2 h reoxygenation under ambient oxygen with 5% CO₂. Supernatants from WT and ecSOD Tg myocytes were collected to determine LDH release after normoxic incubation and after 120 min reoxygenation. Under normoxic conditions little difference was seen in LDH release between WT and ecSOD Tg myocytes (Figure 1). After hypoxia and reoxygenation, LDH release was significantly elevated in WT myocytes whereas release was unchanged in ecSOD Tg myocytes compared to WT (Figure 1). These results confirm our earlier results and demonstrate that the resistance to hypoxia reoxygenation injury in ecSOD Tg hearts is myocyte, autonomous.

Based on the intracellular localization of ecSOD highlighted by myocyte-specific transgenic overexpression, we further investigated the intracellular localization of ecSOD by subcellular fractionation and determination of ecSOD distribution by Western analysis. WT and ecSOD Tg hearts were processed for isolation of cytosolic, membrane, nuclear, and cytoskeletal fractions. Fractions from WT mice resolved by SDS PAGE and detection by Western analysis showed that ecSOD is mainly associated with the membrane fraction but was also seen at a similar level in the nuclear fraction and to a lesser degree in the cytosol (Cytosol:Membrane:Nuclear: 10%:45%:45%) and was not detectable in the cytoskeletal fraction (Figure 2A). ecSOD was markedly increased in Tg mice in all fractions. ecSOD was also shown to be associated with mitochondria isolated from both WT and ecSOD Tg hearts (Figure 2B). Notably, ecSOD levels could be detected in the mitochondrial fraction in WT mice and these levels were again markedly increased in ecSOD Tg mice compared to WT (Figure 2B). Mitochondrial localization was further examined in sections of ecSOD Tg hearts by colocalization with the mitochondrial transport chain protein, cytochrome C oxidase (CytC) by immunofluorescent confocal microscopy. In Figure 2C, intracellular ecSOD colocalizes extensively with CytC. In line analysis of ecSOD and CytC channel output in representative confocal images of myocytes stained for ecSOD (red), CytC (green), and DNA (blue), registration between ecSOD and CytC channels is very strong (Figure 2D). A fraction of cytosolic ecSOD is also found not to be associated with CytC. These results validate subcellular fraction results which demonstrate ecSOD in transgenic hearts may be found in cytosolic and nuclear (perinuclear) fractions. These results demonstrate ecSOD is capable of cytoplasmic localization with the ability to associate with intracellular organelles, specifically mitochondria.

The enrichment of mitochondrial ecSOD in Tg mice provides an additional mechanism that can protect the myocardium from ischemia reperfusion injury. The localization of ecSOD to intracellular fractions and mitochondrial and increased resistance of ecSOD Tg myocytes to hypoxia reoxygenation injury as described above suggest alternate mechanisms may account for this action. As ecSOD was found to be specifically associated with isolated mitochondria, the effect of this association on MPT was tested by examining Ca²⁺-induced mitochondrial swelling in WT and ecSOD Tg hearts. To investigate the role of increased mitochondrial ecSOD, mitochondria were isolated from WT and ecSOD Tg hearts and swelling was measured as an index of Ca²⁺ induced MPT. Under basal conditions, WT and ecSOD Tg mitochondria exhibited similar and stable baseline OD in the presence of CsA, which binds cyclophilin D to block pore opening (Figure 3). Under control conditions in the absence of Ca²⁺, MPT was greater in WT than in ecSOD Tg mitochondria but did not reach statistical significance. With increasing Ca²⁺, WT mitochondria displayed significantly increased swelling and concomitant increased MPT. In ecSOD mitochondria, swelling was consistently and significantly less compared to WT indicating attenuated MPT. The heterogeneity of transgenic ecSOD expression in myocytes as seen in confocal microscopy images (Figure 2C) would suggest that homogeneous ecSOD expression would further accentuate the prevention of MPT in ecSOD mitochondria. These results demonstrate that MPT in ecSOD Tg mitochondria with increasing Ca²⁺ is attenuated and provides a mechanism by which transgenic ecSOD protects myocytes from hypoxia reoxygenation injury.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.