Male Wistar rats weighing 250–330 g (11–12 weeks of age) from the Central Animal Facilities of our institution were used. The animals were housed under a 12-h light/dark cycle, at 22–23°C and 54–55% humidity. Rats were fed a pellet rodent diet (Nuvilab CR1, manufactured by Nuvital, Curitiba, Brazil), ad libitum, and had free access to water. Surviving male rats (n=105) were randomly assigned to one of three groups: Sham (SH; n=26), Myocardial infarction (MI; n=45) and submitted to Hyperbaric therapy (HBO; n=34). Animals in the SH and MI groups were maintained under oxygenation at ambient pressure for the same time as the treated group.

All procedures were performed according to the principles of ethics of animal experimentation adopted by the Brazilian College of Animal Experimentation (COBEA-www.cobea.org.br), in conformity with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication no. 85-23, revised 1996). The protocol of the present study was approved by the Research Ethics Committee (CEP 0891/09) of the Federal University of São Paulo.

The rats were anesthetized with halothane, intubated and ventilated with a ventilator for rodents (model 683; Harvard Apparatus Co., South Natik, MA, USA). The rats were randomized between the Sham (SH), myocardial infarction (MI) and hyperbaric therapy (HBO) groups. MI was produced in the animals of the MI and HBO groups by ligation of the anterior descending coronary artery, as described previously (31,32). Sham rats underwent a similar procedure, without coronary ligation.

All experiments were initiated at the same time (13:00) in order to avoid influence of the different biological rhythms.

Immediately after recovery from anesthesia (approximately 2 min), animals from the HBO group were exposed to 100% O₂ under a pressure of 2.5 atmospheres absolute (ATA) for 60 min. within a hyperbaric chamber for small rodents, as previously described (18). The gas in the chamber was continuously vented to avoid CO₂ retention (<0.1%) and the temperature was maintained in the range of 25–27°C. Compression and decompression were performed at the rate of 0.2 ATA/min (20 kPa/min) and the pressure inside the chamber was stable over a period of 12 to 15 min.

Animals in the SH and MI groups were maintained under oxygenation at ambient pressure for the same time as the treated group. At the end of the protocol, the surviving rats (SH=26, MI=45, and HBO=34). were sacrificed under halothane anesthesia and each heart was removed and sectioned transversely at the middle of the left ventricle (LV). The basal half was used to measure the infarct size, with triphenyltetrazolium (TTZ), according to a previously described protocol (18). The apical half was prepared in gel (OCT-Sakura, Japan) and frozen with liquid nitrogen for cryostat sectioning. The samples were stored at −80°C for further analysis by microscopy.

Proteins were extracted as previously described (33). Homogenate protein samples of 30 µg were subjected to SDS-PAGE on 12% polyacrylamide gel. The separated proteins were transferred onto hydrophobic polyvinylidene difluoride membranes (Hybond-P, Amersham Biosciences), and the transfer efficiency was examined with 0.5% Ponceau S. The membranes were soaked in a blocking buffer (5% nonfat dry milk and 0.1% Tween 20 in PBS, pH 7.5) for 1 h at room temperature and then incubated overnight at 4°C with primary antibodies: Mouse monoclonal anti-catalase (dilution 1:1,000, Sigma-Aldrich; Merck KGaA, catalog number C0979), rabbit polyclonal anti-nitrotyrosine (dilution 1:2,000, EMD Millipore, catalog number 06-284), rabbit polyclonal anti-superoxide dismutase 1 (dilution 1:2,000; Abcam, catalog number ab16831), and rabbit polyclonal anti-peroxiredoxin (dilution 1:2,000, PrxSO_2/3, Abcam, catalog number ab16830), which recognizes sulfinic (-SO₂) and sulfonic (SO₃) forms of Prx I to IV; mouse monoclonal anti-GAPDH (dilution 1:2,000, Abcam, catalog number ab8245). After incubation, the membranes were washed three times and then incubated for 1 h at room temperature with horseradish peroxidase-conjugated secondary antibodies (dilution 1:5,000; Abcam, catalog number ab97051). Detection was performed with chemiluminescence reagents (GE Healthcare) and values for target protein were normalized to GAPDH. Dot blot method for nitrotyrosine was analyzed as previously described (34). Samples were transferred onto a nitrocellulose membrane by vacuum for dot blot fluorescence detection performed using Odyssey (LI-COR Biosciences); intensities of dots were determined using Image J 1.43u software (available at http://imagej.nih.gov/ij/, developed by Wayne Rasband, National Institutes of Health).

Samples were homogenized in 100 mM phosphate buffer (pH 7.0; 250 µl/10 mg tissue) and centrifuged (12,000 × g for 10 min, at 4°C). To the supernatant (200 µl), 5% of sulphosalicylic acid (200 µl) was added to precipitate the proteins. Detect X^® kit (Arbor Assays, Ann Arbor, MI, USA) was then utilized to quantify the total concentration of glutathione (GSH). After incubating the mixture with ThioStar^® (Detect X, Arbor Assays) for 15 min, at room temperature, the fluorescence was determined (excitation 390 nm, emission 510 nm), using a spectrophotometer (U-2810; Hitachi), and the reduced GSH concentration measured.

Subsequently, a reaction mixture was added to convert all oxidized glutathione (GSSG) into free GSH, which then reacted with the excess ThioStar^® to yield the signal related to the total GSH content. A standard curve was used to calculate the total and reduced glutathione concentrations. The oxidized glutathione concentration was obtained using the following calculation: GSSG=(total GSH-reduced GSH)/2. Detection limits ranged between 38 nM, for free GSH, and 42 nM, for the total GSH. Results were normalized to the muscle fragment weight for between-animal comparisons. All assays were performed in triplicate.

In situ microfluorotopography of dihydroethidium (DHE) oxidation products was performed as previously described (35), with 3 µmol/l final DHE concentration. Slides were analyzed by confocal microscopy (Zeiss LSM510) with laser excitation at 488 nm and emission at 610 nm. Controls, performed by incubating slides for 30 min with Peg-SOD (500 U/ml), indicated preferential detection of superoxide with DHE. Quantitative analysis of fluorescence images was performed with Leica Qwin Plus (Leica Microsystems Ltd., Switzerland) software.

Myocardial homogenates were prepared under liquid N₂. After centrifugation (13,400 × g for 20 min, at 4°C), 20 µl aliquots were injected into NOA (Nitric Oxide Analyzer model 280; Sievers Instruments, USA), with VCl₃ and HCl (at 95°C) used as reductants, as previously described (36). Nitric oxide (•NO), nitrite (NO₂⁻) and nitrate (NO₃⁻) by-products were normalized for protein concentration.

The data are expressed as mean ± SEM. The Student t-test was used for comparisons of infarct sizes and the Chi-square test was used to compare mortality. Two-way ANOVA was applied to parametric data using Newman-Keuls to identify statistical differences. Kruskal-Wallis was performed on non-parametric data, associated with the Dunn's test to identify statistical differences. The statistical program used was GraphPad Prism 6.0 (GraphPad Software Inc., San Diego, CA, USA). Differences with P≤0.05 were considered significant.

Immediate mortality was established when it occurred in the period between the introduction of the animal into the HBO chamber until the end of the decompression (~90 min). In the MI group, 27 animals died during this period, whereas HBO therapy was effective in improving post-infarction survival, with only 6 immediate deaths noted. Fig. 1 shows the percentage of mortality for all animals submitted to the protocol, showing a clearly lower rate of mortality in the HBO group (15%) than that in the MI group (37.5%). Additionally, no significant difference in infarct size was observed (Fig. 1B) between the experimental groups (MI 38±2.0% vs. HBO 43±2.5%).

In order to investigate whether redox processes are involved in the protective effect induced by HBO therapy, we first addressed the redox status in LV homogenates, by measuring the amount of reduced and oxidized glutathione. The infarcted groups showed (Fig. 1C) a reduction in the GSH redox buffer (MI, 97±11 and HBO, 68±5 µM/mg) when compared to the SH group (126±10 µM/mg). However, disulfide (Fig. 1D) was significantly increased only in the HBO group (15±1 µg/mg) when compared to that in the SH (9±1 µM/mg) and MI (12±1 µM/mg) groups. The GSH/GSSG ratio showed that the values in the HBO group (10±1) underwent a greater alteration of this intracellular redox buffer, when compared to those of SH (30±4) and MI (17±3) groups (Fig. 1E). There was no statistical significance in the comparison of SH and MI groups (Fig. 1E). Therefore, these results indicate that myocardial infarct decreases the amount of reduced glutathione and HBO therapy accentuates such a pro-oxidizing effect.

The higher oxidative environment promoted by MI may be caused by increased oxidant generation or by reduced antioxidant defense. First, we assessed the potential to generate ROS by addressing the total amount of DHE-oxidation by-products by microfluorotopography on slides of the myocardium in the SH group (Fig. 2A), the MI group (Fig. 2B), the HBO group (Fig. 2C) and the MI group (Fig. 2D). The fluorescence of DHE-oxidation products, measured in units/area, were significantly lower in the HBO group (0.86±0.08) compared to the MI group (1.87±0.08) and the SH group (1.40±0.11) (Fig. 2E). Although we could not exclude the potential interference by other peroxides (37), the pre-incubation with pegylated SOD was able to inhibit the oxidation of DHE (0.09±0.06), showing greater specificity of the technique to the superoxide anion. Thus, hyperbaric therapy was able not only to reduce the production of superoxide induced by infarction, but also to minimize myocardial levels in relation to non-infarcted animals (Fig. 2E).

Due to the higher levels of superoxide promoted by MI, combined with the protective effect induced by HBO, the antioxidant capacity was investigated by assessing the expression of SOD, whose values showed that HBO therapy significantly increased enzyme expression in the HBO group (1.0±0.06 AU/µg) relative to the MI (0.79±0.04 AU/µg) and SH (0.69±0.08 AU/µg) groups, with no significant difference between the SH and MI groups (Fig. 3A). In addition, expression of catalase (CAT) was investigated due to the intriguing effect observed in the HBO group concerning the higher levels of oxidized glutathione. Similarly to that observed with the SOD enzyme, HBO therapy significantly increased CAT enzyme expression in the HBO group (0.97±0.06 AU/µg) in relation to the MI (0.73±0.07 AU/µg) and SH (0.66±0.04 AU/µg) groups. There was no significant difference between the SH and MI groups (Fig. 3B). Thus, HBO therapy was able to upregulate enzymatic antioxidants. In order to better investigate the antioxidant activity, the expression of peroxiredoxin (Prx) hyperoxidation was assessed. Importantly, the expression of the sulfonylated form of Prx (SO_2/3) protein was lower in the HBO group (0.65±0.05 AU/µg), when compared to MI (1.24±0.18 AU/µg) and SH (1.45±0.26 AU/µg) values, corroborating with better redox control. There was no significant difference between the MI group and the SH group (Fig. 3C). Therefore, not only the total expression of antioxidants was increased such as SOD and CAT, but also the Prx hyperoxidation levels were decreased by HBO therapy.

Nitric oxide and its oxidation-derived by-products show a dual role during normal and pathological conditions, including MI (38,39). Coronary occlusion caused a significant increase in nitrate (NO₃⁻) (nM/mg protein) in the MI (1.37±0.26) and HBO (1.58±0.41) groups in relation to the SH group (0.56±0.08). No significant difference was found between the two groups with MI (Fig. 4A), which represents a greater availability of nitric oxide in an environment with increased oxidative stress. No significant difference was found in levels of nitrite (NO₂⁻) (nM/mg protein) between the SH (0.51±0.08), MI (0.69±0.13) and HBO (0.74±0.20) groups (Fig. 4B). In addition, the positive labeling of 3-nitrotyrosine (AU/µg prot) by dot blot analysis (Fig. 4C) showed a difference between groups. A lower amount of nitrated protein was found in the HBO group (2.40±0.18) compared to the MI (3.08±0.16) and SH (3.36±0.20) groups. There was no significant difference between the SH and MI groups. Therefore, these data suggest that MI induces increased production of NO while HBO is effective to prevent such NO-related modification in global target proteins, as reflected by the lower levels of 3-nitrotyrosine.