Extracellular superoxide dismutase (EC-SOD) R213G variant reduces mitochondrial ROS and preserves mitochondrial function in bleomycin-induced lung injury: EC-SOD R213G variant and intracellular redox regulation

Abstract

Extracellular superoxide dismutase (EC-SOD) is highly expressed in the lung and vasculature. A common human single nucleotide polymorphism (SNP) in the matrix binding region of EC-SOD leads to a single amino acid substitution, R213G, and alters EC-SOD tissue binding affinity. The change in tissue binding affinity redistributes EC-SOD from tissue to extracellular fluids. Mice (R213G mice) expressing a knock-in of this EC-SOD SNP exhibit elevated plasma and reduced lung EC-SOD content and activity and are protected against bleomycin-induced lung injury and inflammation. It is unknown how the redistribution of EC-SOD alters site-specific redox-regulated molecules relevant for protection. In this study, we tested the hypothesis that the change in the local EC-SOD content would influence not only the extracellular redox microenvironment where EC-SOD is localized but also protect the intracellular redox status of the lung. Mice were treated with bleomycin and harvested 7 days post-treatment. Superoxide levels, measured by electron paramagnetic resonance (EPR), were lower in plasma and Bronchoalveolar lavage fluid (BALF) cells in R213G mice compared to wild-type (WT) mice, while lung cellular superoxide levels in R213G mice were not elevated post-bleomycin compared to WT mice despite low lung EC-SOD levels. Lung glutathione redox potential (E_hGSSG), determined by HPLC and fluorescence, was more oxidized in WT compared to R213G mice. In R213G mice, lung mitochondrial oxidative stress was reduced shown by mitochondrial superoxide level measured by EPR in lung and the resistance to bleomycin-induced cardiolipin oxidation. Bleomycin treatment suppressed mitochondrial respiration in WT mice. Mitochondrial function was impaired at baseline in R213G mice but did not exhibit further suppression in respiration post-bleomycin. Collectively, the results indicate that R213G variant preserves intracellular redox state and protects mitochondrial function in the setting of bleomycin-induced inflammation.

Graphical Abstract

Introduction

Extracellular Superoxide dismutase (EC-SOD), the sole extracellular enzymatic defense against superoxide, protects in diverse vascular and lung diseases. A human variant of EC-SOD (rs1799895) contains a single nucleotide polymorphism (SNP) in the C-terminus which results in a single amino acid substitution in the matrix binding region of the enzyme; this arginine to glycine substitution (R213G) lowers tissue binding affinity, resulting in higher alveolar and serum EC-SOD levels but lower lung levels [1]. Individuals harboring this variant demonstrate a lower risk of lung disease [2, 3]. Mice expressing knock-in of the human SNP (R213G mice) exhibit enhanced resolution of lung inflammation and protection against subsequent fibrosis in response to bleomycin [1, 4, 5].

While EC-SOD catalyzes the dismutation of superoxide to hydrogen peroxide and oxygen in the extracellular compartment, it has also been observed that alterations in the expression level or localization of EC-SOD results in changes in intracellular redox-sensitive molecules such as the tripeptide glutathione, an important antioxidant molecule^6–14. For instance, mice lacking EC-SOD show a decrease in the ratio of reduced to oxidized glutathione (GSH/GSSG) in cigarette smoke (CS) model [6]. Furthermore, the GSH/GSSG ratio was preserved in the protected R213G mice in the fibrosis phase 21 days after bleomycin though GSH levels were not measured in this study during the earlier timepoint associated with inflammation [4]. The extracellular redox environment has been shown to regulate intracellular redox-sensitive signaling in cell culture models and is altered in the setting of aging as well as lung and vascular disease, including the murine model of bleomycin-induced fibrosis [3, 7–9]. We were thus interested in more precisely defining how alterations in local EC-SOD content due to the R213G EC-SOD SNP modulate the redox environment in the extracellular and intracellular compartment during the robust inflammatory phase of lung injury.

To evaluate the impact of EC-SOD on the intracellular redox environment, in addition to the assessments of intracellular thiol state, we considered mitochondrial oxidative stress and function as another major target critical to redox-sensitive signaling and implicated in bleomycin-induced lung injury. Mitochondrial dysfunction is emerging as a major regulator of heart and lung diseases [10]. Increased mitochondrial superoxide due to mitochondrial dysfunction results in oxidation of the inner mitochondrial membrane phospholipid, cardiolipin; oxidized cardiolipin is implicated in bleomycin-induced lung injury [11, 12]. The link between the EC-SOD and mitochondrial function has not been carefully examined.

Since mice expressing the R213G variant of EC-SOD demonstrated a change in the local EC-SOD content and improved resolution of inflammation by 7 days post-bleomycin, we hypothesized that the R213G EC-SOD variant would influence the extracellular redox environment as well as the intracellular redox status at this inflammatory timepoint. To test this hypothesis, we evaluated WT and R213G mice at 7 days post-bleomycin and measured ROS production in extracellular fluids and lung, followed by lung glutathione redox status; mitochondrial oxidative stress and function including mitochondrial ROS production, cardiolipin oxidation; mitochondrial content; and mitochondrial respiration.

Material and Methods

Mouse Model

Animal studies were approved by the University of Colorado Denver (Aurora, CO) Institutional Animal Care and Use Committee. 8–12 week old male and female mice homozygous for the R213G SNP in sod3 (EC-SOD) on the C57BL/6 J mouse background (R213G) and C57BL/6 J WT mice were anesthetized with isoflurane (1–3%), treated with a single dose of intratracheal bleomycin (5U/Kg) or PBS, and euthanized 7 days post-injury as previously described [1, 4, 13, 14]. Since the mouse strain background can influence mitochondrial metabolism, the C57Bl/6 J background strain of the R213G mice was reconfirmed using the Transnetyx miniMUGA (Mouse Universal Genotyping Array) [15, 16]. Blood, bronchoalveolar lavage fluid (BALF) and lung tissue were collected as described below.

Superoxide production by Electron Paramagnetic Resonance Spectroscopy (EPR)

Superoxide production in plasma, bronchoalveolar lavage fluid (BALF) and lung was measured by EPR using compartment-specific EPR spin probes to detect cellular or mitochondrial superoxide [17, 18]. CPH and CMH are both cell-permeable EPR spin probes that react with high sensitivity with superoxide, as well as radicals derived from superoxide including peroxynitrite, but do not react with hydrogen peroxide [19–21]. Mito-TEMPO-H is an EPR spin probe targeted to the mitochondria that reacts with mitochondrial superoxide [22]. Following the oxidation of the spin probes, the resultant nitroxide radicals were detected by EPR using the parameters described below.

Plasma superoxide measurements

Seven-days post bleomycin treatment, mice were injected with CPH, and plasma was collected to determine circulating extracellular superoxide levels. CPH was first dissolved in degassed KHB containing the following ion chelators, 25 μM deferoxamine mesylate salt and 5 μM diethyl dithiocarbamate, for a final stock CPH concentration of 2 mg/ml. CPH was delivered as a 9 mg/kg intraperitoneal bolus followed by a single subcutaneous dose of 13.5 mg/kg as previously described, based on an average weight of 20 g per mouse [23]. One hour after injecting the CPH spin probe, mice were anesthetized, and blood was drawn via right ventricular puncture into a syringe coated with 1000 USP/mL heparin. Plasma was collected by spinning the blood at 3000 rpm for 10 min. 50 μl of plasma was loaded in EPR capillary tube for EPR measurements at room temperature (RT).

Lung cellular superoxide measurements

Lung tissue was also obtained from the mice injected with CPH as described above to detect lung cellular superoxide. Following collection of blood and euthanasia, the chest was opened, and lungs were flushed with 10 ml cold PBS via the right ventricle to remove blood. 10 mg of lung tissue was cut, weighed and placed in a tissue cell, an accessory used to detect EPR signal from tissues (Bruker BioSpin). EPR measurements were performed at RT. The tissue cell accessory was cleaned with ethanol between samples.

BALF cell superoxide measurements

To detect cellular superoxide in BALF cells, a separate cohort of mice were euthanized 7 days post bleomycin and a cannula was surgically placed in the trachea. BALF was collected by slowly instilling and withdrawing via the tracheal cannula five (1 mL) aliquots of PBS containing the metal chelator, DTPA (100 μM). BALF cells were collected by pooling the 5 BALF aliquots and spinning down the cells at 700 g for 7 min. The cells were resuspended in KHB contains 100 μM DTPA and incubated with CMH 0.4 mM final concentration for 50 min at 37C. The samples were loaded in Teflon tubing, flash frozen in liquid nitrogen and measured at 77K [17].

Lung mitochondrial superoxide

To detect lung mitochondrial superoxide, a separate cohort of mice were euthanized 7 days post bleomycin. Lungs were lavaged, flushed and fresh lung tissue was weighed (100 mg) and homogenized in 300 μl Tris-EDTA buffer containing 0.25 M sucrose using the BEAD RUPTOR (OMANI). The tissue homogenate was immediately placed on ice and homogenate transferred to a 1.5 mL Eppendorf tube. 30 μl homogenate was added to 165 μl KHB + 100 μM DTPA, treated with 5 μl of the mitochondrial specific EPR spin probe, mito-TEMPO-H (9.5 mM stock) and incubated for 1h at 37 °C. After 1h, 150 μl was loaded into PTFE tubing, flash frozen in liquid nitrogen and measured at 77K.

EPR acquisition parameters and determination of nitroxide radical concentration.

EPR measurements at RT used the following EPR acquisition parameters: microwave frequency = 9.65 GHz; center field = 3432 G; modulation amplitude = 2.0 G; sweep width = 80 G; microwave power = 19.9 mW; total number of scans = 10; sweep time = 12.11 s; and time constant = 20.48 ms.

EPR measurements at 77k used the following EPR acquisition parameters: microwave frequency = 9.65 GHz; center field = 3438 G; modulation amplitude =4.0 G or 6 G; sweep width = 150 G; microwave power = 0.316 mW; total number of scans = 2; sweep time = 60 s; and time constant = 1.28 ms.

The nitroxide radical concentration was obtained by simulating the spectra using the SpinFit module incorporated in the Xenon software of the bench-top Bruker EMXnano EPR spectrometer or by double integration followed by the SpinCount module (Bruker).

Thiol couple concentration and redox potential by HPLC and fluorescence

Reduced glutathione, (GSH) and oxidized glutathione (GSSG) were determined using a previously described protocol [24, 25]. Briefly, lung tissue was immediately placed in PCA solution, then sonicated until cloudy. Lung homogenate was kept on ice until samples were centrifuged. The supernatant was stored at −80°C until derivation. Protein concentration was measured using a BCA protein assay (Thermo Fisher Scientific). After derivatization, samples were analyzed by HPLC with fluorescence detection. The metabolites were identified by co-elution with standards, and quantification was achieved by integration relative to the internal standard, γ-Glu–Glu [24]. The redox potential (E_h) of lung thiol/disulfide pool was calculated using the Nernst equation, E_h=E_o+RT/nF ln[disulfide]/[thiol]², where E_o is the standard redox potential for the specific couple, R is the gas constant, T is the absolute temperature, n is 2 for the number of electrons transferred, and F is Faraday’s constant. The E_o value used for the GSH/GSSG redox potential (E_hGSSG) was −264 mV, which was based upon the value of −240 mV for pH 7.0 [26].

Cardiolipin Quantification

Cardiolipin was quantified using previously published methods with normal phase liquid chromatography coupled to electrospray ionization mass spectrometry in an API 4000 mass spectrometer (Sciex, Framingham, MA) [27]. Lung tissue was homogenized in PBS, and lipids were extracted using a modified Bligh Dyer method according to previously published methods with 1000 nmoles tetramyristal cardiolipin as an internal standard (Avanti Polar Lipids, Alabaster, AL, US) [27, 28]. Three species of cardiolipin were quantified, (1) tetralineoyl cardiolipin (L4) containing 4 linoleic acid side chains with a mass to charge ratio (m/z) of 1448, (2) L4 cardiolipin with a single hydroxyl group (m/z 1464) and (3) L4 cardiolipin with a single peroxyl group (m/z 1480). Retention times for peaks were approximately 7.5 minutes for L4, and 9 minutes for oxidized and peroxidized CL.

High-Resolution Respirometry

Oxidative respiration of lung homogenates was measured by high-resolution respirometry (Oxygraph 2000, Oroboros Instruments, Innsbruck, Austria) using a stepwise substrate-uncoupler-inhibitor titration (SUIT) protocol to evaluate various components of the electron transport system [29–31]. Fresh unlavaged flushed lungs were homogenized at a concentration of 200 mg of tissue per 1 mL KME buffer (20 mM MOPS, 120 mM KCl, and 1 mM EGTA, pH 7.4) at speed 4 with an IKA^® T25 digital ULTRA-TURRAX^® homogenizer (Staufen, Germany) and 25 μl added to 2 ml of solution in each chamber of the Oxygraph 2000 apparatus at 37°C containing Mir05 respiration medium (0.5 mM EGTA, 3 mM MgCl₂.6H₂O, 20 mM Taurine, 10 mM KH₂PO₄, 20 mM HEPES, 110 mM Sucrose, 1 gram per liter essentially fatty acid free bovine serum albumin, and 60mM L-lactobionic acid at a pH of 7.1) and standard protocols were followed for calibration of the instrument (Veksler, 1987; Letellier, 1992; Gnaiger, 2000). Basal respiration was recorded and followed by stepwise addition of 0.2 mM Palmitoylcarnitine and 2 mM Malate, 4 mM ADP, 0.4 mM Octanoylcarnitine, 5 mM Pyruvate, 10 mM Glutamate, 10 mM Succinate, 1 μM Carbonyl cyanide p-(trifluoro-methoxy) phenyl-hydrazone, 1 μM Rotanone, and 1 μM Antimycin A. Oxygen flux was normalized to protein by conducting a protein assay on lung homogenates (Quick Start Bradford Protein Assay, Bio-Rad, Hercules, California).

Western Blot Analysis

Tissue was homogenized in KME buffer (20 mM MOPS, 120 mM KCl, and 1 mM EGTA at a concentration of 200mg of tissue/ml). Protein concentration was determined using the Pierce Rapid Gold BCA kit (Thermo Fisher Scientific). Protein (30 μg) was separated by gel electrophoresis using Criterion XT 4–12% Bis-Tris Precast Gel (Bio-Rad, Hercules, CA) with XT MES SDS running buffer (Bio-Rad). Proteins were transferred to poly-vinylidene fluoride membranes (Bio-Rad). Membranes were activated in methanol and blocked in 5% nonfat dry milk in Tris buffered containing 0.05% Tween 20 (TBST) for 1 hour. Membranes were incubated in the following primary antibodies: rabbit monoclonal to citrate synthase (D7V8B; Cell Signaling) at 1:1,000, anti-SOD1 (Cu,Zn SOD) antibody (ab13498, Abcam) at 1:1000, SOD2 (MnSOD) (D9V9C) rabbit mAb (13194, Cell Signaling) at 1:1000, and rabbit monoclonal to Vinculin (E1E9V; Cell Signaling) at 1:10,000 in 5% milk in TBST at 4°C, overnight. The appropriate horseradish peroxidase conjugated goat anti-rabbit (Cell Signaling) was applied at 1:5,000 in TBST for 1 hour at room temperature. Detection was accomplished using SuperSignal Femto Chemiluminescent substrates (Thermo Scientific). Bands were quantified by a Bio-Rad chemicon transiluminator.

Statistical Analysis

Data were analyzed using Prism (GraphPad Software, La Jolla, CA, USA) by unpaired t-test or two-way analysis of variance (ANOVA). Post-hoc analysis was performed using Tukey’s test when significant differences were found between groups. Data are expressed as mean ± SEM. Significance was defined as p < 0.05.

RESULTS

Superoxide levels were lower in plasma and BALF cells in R213G mice 7 days post bleomycin.

We previously demonstrated that plasma and BALF EC-SOD levels were higher in the R213G mice compared to the WT mice [1, 4]. To determine if the elevated EC-SOD levels in these extracellular fluids influenced the local superoxide levels, we measured superoxide in plasma and BALF cells at 7 days post Bleo by EPR. Plasma superoxide levels were consistently lower at baseline in R213G mice compared to WT by the overall 2-way ANOVA, though there were no statistical differences between specific groups by post-hoc analysis. Plasma superoxide levels also modestly increased in response to Bleo by the overall 2-way ANOVA (p=0.047) with no differences between specific groups by post-hoc analysis (Fig 1A). In BALF cells, superoxide production significantly increased in the Bleo-treated WT mice compared to PBS-treated WT mice while BALF cell superoxide production did not increase in R213G mice post Bleo. BALF cell superoxide production was significantly lower in the Bleo-treated R213G mice compared to Bleo-treated WT mice (Fig 1B).

Fig 1. Lower superoxide was observed in R213G mice in plasma and BALF cells while superoxide in lung increased similarly in both strains 7 days post bleomycin.

Mice were treated with a single dose of intratracheal bleomycin (Bleo) (5U/Kg) or PBS vehicle. At 7 days, mice were anesthetized; CPH spin probe was administered in vivo through a single intraperitoneal injection (IP) followed by a single subcutaneous injection. After 1h circulation mice euthanized (A and C). (A) To detect extracellular superoxide in plasma, blood was collected via right ventricular puncture into a syringe coated with 1000 USP/mL. Freshly drawn blood was spun down, and plasma was collected. 50 μL of plasma was loaded in a capillary tube and EPR measurements done at RT (B). BALF cells were pelleted by centrifugation, resuspended in KHB contains 100 μM DTPA and incubated with CMH for 1h at 37 °C. The samples were loaded into PTFE tubing and measured at 77K. (C). To detect superoxide in lung tissue, lungs were flushed with cold PBS and 10 mg of lung tissue was used and EPR measurements were done at room temperature (RT) using the cell tissue accessory. Nitroxide radical concentration was obtained by SpinFit for RT measurements or double integration for 77 K followed by the Spin Count module (Bruker). Data was analyzed by two-way ANOVA with Tukey’s multiple comparisons post-hoc analysis . Data expressed as mean ± SEM; ^# p<0.05 for treatment by two-way ANOVA with no difference between specific groups on post-hoc ^$ p<0.05 for genotype by two-way ANOVA with no difference between specific groups on post-hoc analysis due to treatment, *p<0.05,***p< 0.001 differences between groups,

Lung cellular superoxide increased similarly in both mouse strains 7 days post bleomycin.

We previously demonstrated that the R213G mice showed improved resolution of inflammation post-bleomycin despite low lung EC-SOD content [4, 13, 14]. To further understand how the R213G variant impacted lung redox status in the setting of low lung EC-SOD levels, we next examined lung cellular and mitochondrial superoxide production and lung glutathione redox status. Lung cellular superoxide production increased significantly 7 days post intratracheal Bleo in WT mice. We previously demonstrated that lung EC-SOD levels were significantly reduced in the R213G mice, with release of active EC-SOD into plasma and alveolar fluid, and these mice were protected against lung inflammation and edema at day 7 post Bleo. Cellular superoxide production was similar at baseline and increased with bleomycin similarly in the R213G and WT mice, despite the established lower EC-SOD content in R213G mice [1, 4](Fig 1C). The decrease in lung EC-SOD in the R213G mice did not result in a compensatory increase in Cu,Zn SOD (SOD1) at baseline or with bleomycin, shown by western blot analysis (Supplement 2A,C).

Lung GSH redox potential was preserved in the R213G mice 7 days post Bleo.

The antioxidant tripeptide glutathione protects cells against oxidative stress and regulates numerous redox sensitive signaling pathways. We next examined the thiol redox state of the primary thiol redox GSH/GSSG in the lung at baseline and 7 days post- bleomycin by HPLC. Lung E_hGSSG was significantly more oxidized in WT compared to R213G mice (Fig 2A) due to an increase in GSSG concentration in WT mice lung (Supplement S1B). The concentration of GSH and GSSG used to calculate the redox potentials along with total glutathione are shown in supplement S1.

Fig 2. Lung GSH redox potential was preserved in the R213G mice and lower mitochondrial superoxide was observed 7 days post bleomycin compared to WT mice.

Mice were treated with a single dose of intratracheal bleomycin (Bleo) (5U/Kg) or PBS vehicle. At 7 days, lung tissue was immediately placed in PCA solution then homogenized for HPLC and redox potential was calculated from concentrations of GSH, and GSSG using the Nernst equation. E_hGSSG in lung (A). (B) Fresh left lung tissue was weighed and homogenized immediately in 300 μl sucrose in Tris-EDTA buffer using the BEAD RUPTOR and placed on ice. 30 μl homogenate was added to 166 μl KHB + 100 μM DTPA, treated with 4 μl of mito-TEMPO-H (9.5 mM stock) and incubated for 1h at 37 °C. After 1h, 150 μl was loaded into PTFE tubing, flash frozen in liquid nitrogen and measured at 77K., Data were analyzed by two-way ANOVA with Tukey’s multiple comparisons post-hoc analysis. Data expressed as mean ± SEM; *p<0.05.

Lung mitochondrial superoxide was lower in R213G mice compared to WT mice 7 days post bleomycin.

We next measured lung mitochondrial superoxide production in WT and R213G mice 7 days post-bleomycin. At the 7 day timepoint, bleomycin increased lung mitochondrial superoxide only in WT mice and mitochondrial superoxide post bleomycin was lower in the R213G mice compared to WT mice (Fig 2B). Furthermore, the mitochondrial isoform of SOD, Mn SOD or SOD2, did not differ between WT and R213G mice at baseline or following bleomycin, indicating that there was not a compensatory increase in SOD2 to explain the lower mitochondrial superoxide production after bleomycin in R213G mice (Supplement 2B,C).

Oxidation of mitochondrial phospholipid, cardiolipin is blocked in R213G

Since the R213G variant altered mitochondrial superoxide, we next tested measures of mitochondrial oxidation and function. Cardiolipin is a specialized mitochondrial phospholipid containing four acyl groups and is required for normal mitochondrial function [27, 28]. Decreases in the percent of tetralineoyl (L4) cardiolipin impair mitochondrial respiration and reduce ATP production [32], and bleomycin has been reported to decrease cardiolipin levels in lung [12]. We therefore evaluated L4 cardiolipin content and oxidation in the two mouse strains in response to Bleo. L4 cardiolipin content declined in the lung 7 days post-bleomycin in both strains (Fig 3A). Generally, cardiolipin abnormality could be attributed to the lower content or the oxidation. The oxidation of L4 cardiolipin, depicted by the ratio of peroxidized L4 cardiolipin to L4 cardiolipin, increased in the WT mice post Bleo but was unchanged from baseline in R213G mice, (Fig 3B).

Fig 3. Oxidation of mitochondrial phospholipid, cardiolipin is blocked in R213G:

Mice were treated with a single dose of intratracheal bleomycin (Bleo) (5U/Kg) or PBS vehicle. At 7 days, mice were anesthetized and euthanized, and the lungs were flushed with cold PBS to remove blood. Cardiolipin was quantified with normal phase liquid chromatography coupled to electrospray ionization mass spectrometry. Lung tissue was homogenized in KME buffer and lipids were extracted using a modified Bligh Dyer method with 1000 nmoles tetramyristal cardiolipin as an internal standard. (A) L4 content to cardiolipin and (B) peroxidized L4 to non-oxidized L4 ratio. (n=4) Data analyzed by two-way ANOVA with Tukey’s multiple comparisons test post-hoc analysis. Data expressed as mean ± SEM; *p<0.05.

Mitochondrial respiration was suppressed at baseline in R213G mice but only decreased in response to bleomycin in WT mice.

We next assessed the impact of the R213G variant on mitochondrial function. Oxygen consumption of lung homogenate was measured under basal conditions as well as following the addition of substrates in the SUIT protocol [29–31]. Respiration was suppressed in WT mice 7 days after bleomycin. Respiration following addition of Palmitoyl carnitine + Malate and Pyruvate, indicating the ability to utilize long chain fatty acids and carbohydrates respectively in the presence of ADP, was suppressed in bleomycin treated WT compared to PBS WT mice. While R213G lungs showed lower respiration under control conditions following ADP and Pyruvate, there was no further impairment in bleomycin treated R213G mice (Fig. 4A and 4B). Suppressed respiration after the addition of Glutamate indicated impairment of Complex I function, whereas lower respiration after the addition of Succinate and Rotenone indicates impairment of Complex 2 function (Fig 4C–4F). Altogether, β-fatty acid metabolism and electron transport chain function were both impaired after bleomycin treatment in WT mice. Mitochondrial function was impaired at baseline in R213G mice but, in contrast to the WT mice, these mice were resistant to further suppression of mitochondrial function by bleomycin.

Fig 4. Mitochondrial respiration was suppressed significantly post bleomycin in WT mice but not R213G mice compared to each strain at baseline :

Mice were treated with a single dose of intratracheal bleomycin (Bleo) (5U/Kg) or PBS vehicle. At 7 days, mice were anesthetized and euthanized, and the lungs were flushed with cold PBS to remove blood. Lung tissue was homogenized in KME buffer. Oxygen consumption measured in O2 Flux/mg of Protein of WT and SOD3 R213G lung; (A) following the addition of 0.2 mM Palmitoylcarnitine, 2 mM Malate and 4 mM ADP, (B) Conditions in B plus 5 mM Pyruvate, (C) Conditions in B plus 10 mM Glutamate, (D) Conditions in C plus 10 mM Succinate, (E) Conditions in D plus1μM Carbonyl cyanide p-(trifluoro-methoxy) phenyl-hydrazone (FCCP), and (F) Conditions in E plus 1μM Rotenone. Data were analyzed by two-way ANOVA with Tukey’s multiple comparisons post-hoc analysis. Data expressed as mean ± SEM; *p<0.05.

Mitochondrial content was not impacted by genotype or treatment 7 days post bleo.

To determine whether the suppression in respiration could be due to lower mitochondrial content, citrate synthase levels were assessed with western blot. Protein expression levels were not significantly different at baseline and post-bleomycin between the strains, suggesting that impairment in mitochondrial function rather than mitochondrial content is responsible for the observed suppression of respiration (Fig 5).

Fig 5. Citrate synthase protein expression was not impacted by genotype or Bleomycin treatment:

Mice were treated with a single dose of intratracheal bleomycin (Bleo) (5U/Kg) or PBS vehicle. At 7 days, mice were anesthetized and euthanized, and the lungs were flushed with cold PBS to remove blood. Lung tissue was homogenized in KME buffer. (A) Western blotting of Citrate synthase (CS) expression in lung homogenate with antibodies for CS and vinculin. Data normalized to WT PBS and expressed as mean ± SEM

DISCUSSION

We recently reported that knock-in of the human R213G variant of EC-SOD in mice enhanced the resolution of bleomycin-induced inflammation and protected against fibrosis [4]. To further understand how the redistribution of EC-SOD due to this SNP could mediate the resolution of inflammation, we tested the hypothesis that the change in the local EC-SOD content would influence not only the extracellular redox microenvironment where EC-SOD is localized but also protect the intracellular redox status of the lung. Studies were conducted at 7 days post bleomycin, coinciding with peak inflammation and a timepoint in which we previously demonstrated that R213G mice begin to resolve inflammation, unlike WT counterparts [4]. We observed that the R213G variant had lowered superoxide levels in both blood and bronchoalveolar cells after bleomycin compared to WT mice. In contrast, despite the low lung EC-SOD content in the R213G mice, lung superoxide production was not increased further in the R213G strain compared to the WT mice, consistent with the previously observed protection against inflammation in R213G mice. The R213G variant also protected the lung intracellular redox environment, specifically preserving lung GSH level in the lung, reduced lung mitochondrial oxidative stress and prevented bleomycin-induced mitochondrial dysfunction.

Our first key observation was that the superoxide level in both plasma and BALF cells was lower in the mice expressing the R213G variant compared to WT. It is established that the R213G EC-SOD variant lowers matrix binding affinity, altering the localization of the enzyme without impacting its activity [1]. We previously reported that plasma EC-SOD levels are elevated in the R213G mice, and BALF EC-SOD levels increase 7 days post bleomycin [4], indicating that superoxide levels in these compartments directly correlated with the local concentration of EC-SOD. We also previously observed that the total cell count in BALF was similar between WT and R213G mice at 7 days post bleo, suggesting that the lower BALF cell superoxide levels in the R213G mice was attributable to increased BALF EC-SOD activity rather than a decrease in the number of inflammatory cells. In contrast to the extracellular fluids, the lung superoxide levels did not correlate with EC-SOD content, but instead, with inflammation and lung injury post-bleomycin. The lack of elevated lung superoxide levels in the R213G mice is consistent with our prior report that the R213G strain showed resolution of inflammation and protection against fibrosis compared to WT mice. The protection in the R213G mice was not explained by a compensatory increase in SOD1 or SOD2 protein, consistent with our prior published data in both adult and neonatal mice [4, 38]. The lung superoxide measurements using the CPH spin probe reflect both extracellular and intracellular levels; thus it is possible that there could be differences specifically in the extracellular component with the redistribution of EC-SOD. However, technical barriers using the cell impermeable spin probes in vivo have limited our ability to test extracellular superoxide at this time. Also, future studies will be necessary to elucidate the specific lung cells responsible for ROS production. Overall, the results indicate that the level of ROS production is location-specific in the setting of disease and the extracellular redox environment in blood and BALF is influenced by the local EC-SOD content [33]. However, the lung protection is more complicated, indicating that the lung ROS production correlated with inflammation and not lung EC-SOD content. These findings supported further studies focused on the impact of the R213G variant on lung intracellular redox state.

The next key observation was that the R213G variant of EC-SOD modulated lung intracellular redox environment, showing protection against both GSH oxidation and mitochondrial ROS production in response to bleomycin. These experiments were performed because of the strong evidence by our lab and others that the extracellular redox environment can regulate the intracellular redox environment and redox sensitive intracellular molecules [7] [34–37]. We focused specifically on the intracellular thiol, glutathione, and mitochondrial ROS since they are both critical redox-sensitive intracellular targets implicated in the pathogenesis of bleomycin-induced injury [9, 12] The preservation of GSH, measured by HPLC, was consistent with studies in bleomycin-treated rats in which bleomycin lowered GSH level lung as well as studies from our lab showing the R213G mice had preserved GSH/GSSG levels at the later fibrotic timepoints [4],[38]. Mitochondrial ROS production have been implicated in several lung diseases [39–41]. Mitochondrial reactive oxygen species (ROS) have also been shown to increase in response to an oxidized extracellular E_hCySS in the setting of cardiovascular disease [7, 45]. The lower mitochondrial superoxide in the lungs of R213G mice following bleomycin compared to the WT mice is consistent with the observed protection in the lungs of these mice rather than a compensatory change in SOD2.

After testing the impact of the R213G variant on mitochondrial superoxide production, the final series of experiments demonstrated that the EC-SOD R213G variant also protected against mitochondrial lipid oxidation and prevented the bleomycin induced reduction in mitochondrial function. Cardiolipin, a key energy-transducing membrane phospholipid localized to the inner mitochondrial membrane (IMM), is necessary for mitochondrial respiration and energy production [46–48]. Furthermore, cardiolipin is involved in mitochondrial cristae morphology and stability [46, 49], and in mitochondrial quality control including biogenesis, mitophagy and dynamics [49–51]. Cardiolipin is highly susceptible to oxidation and peroxidation due its location in the IMM near to electron transport complexes, the main sites of reactive oxygen species (ROS) production [52, 53]. Oxidized cardiolipin accumulates in the outer mitochondrial membrane (OMM) and serves as a vital signaling stage during cell death as it results in the opening of the mitochondrial permeability transition pore (mPTP) and in the release of cytochrome c (cyt c) to the cytosol [54–56]. Also, oxidation of cardiolipin would negatively impact mitochondrial function and dynamics [49]. These events may play a causative role in the etiology and progression of bleomycin- induced lung cell damage and fibrosis. The oxidation of CL is one of the important measures of mitochondrial damage and is critical to the proper function of the mitochondrial ETC complexes. Therefore, after establishing that CL oxidation was blocked in R213G mice, we next assessed mitochondrial function by evaluating lung mitochondrial respiration.

High-resolution respirometry revealed that bleomycin caused a dysfunction of both fatty acid and citric acid cycle derived electron transport chain function, and mice harboring the R213G variant were protected against the bleomycin induced suppression of baseline respiration. Differences between PBS and bleomycin treated WT were evident in response to every substrate profile analyzed, suggesting robust mitochondrial impairment associated with bleomycin treatment. This is consistent with previous reports showing mitochondrial dysfunction in a bleomycin model of fibrosis [57]. Our results also show a protective effect of R213G demonstrated by the lack of a bleomycin-induced decreases in respiration in R213G mice relative to WT. The impairment in mitochondrial function at baseline in R213G mice and the resistance to the effects of bleomycin is parallel to our previous observation that R213G mice exhibit pulmonary hypertension (PH) at baseline but are protected against further bleomycin-induced PH [4]. We find the mitochondrial respiration data to be the most interesting observation in this study as they overall provide compelling evidence of a link between the extracellular redox environment regulated by EC-SOD and mitochondrial function. Currently, there are several studies supporting the concept that alterations in EC-SOD can impact mitochondrial function. Kusuyama et al. showed that inducible loss of EC-SOD in the placenta prevented exercise-induced AMPK activation and downstream glucose metabolism in the fetal hepatocyte [42]. Another study showed that EC-SOD overexpression promoted the survival of SH-SY5Y cells by maintaining mitochondrial homeostasis [43]. Further studies will determine the underlying cause of the impairment in mitochondrial function at baseline in the R213G and elucidate the exact mechanism of the link between R213G variant, the extracellular redox environment and the mitochondria, which may involve the activation of specific redox sensitive signaling molecules, energy demand and metabolism regulation sensors such as AMPK [41].

In summary, the data indicate that the R213G variant preserves intracellular redox state and protects against bleomycin-induced mitochondrial dysfunction. Importantly, these data confirm the protection of the lung due to the increase of EC-SOD in plasma and BALF and provide a novel link between the extracellular redox environment regulated by EC-SOD and mitochondrial dysfunction observed in bleomycin-induced lung injury and inflammation.

Highlights.

• The R213G variant of extracellular superoxide dismutase, which lowers matrix binding affinity and releases EC-SOD into extracellular fluids, lowers local superoxide levels in plasma and bronchoalveolar lavage fluid following bleomycin

• The R213G EC-SOD variant protects the lung intracellular redox state, exemplified by less oxidized E_h GSSG, lower mitochondrial ROS level, and less mitochondrial cardiolipin oxidation

• The R213G EC-SOD variant protects mitochondrial function in the setting of bleomycin-induced inflammation

Abbreviations

DTPA

Diethylenetriaminepentaacetic acid

CPH

1-Hydroxy-3-carboxy-2,2,5,5-tetramethylpyrrolidine. HCl

CMH

1-Hydroxy-3-methoxycarbonyl-2,2,5,5-tetramethylpyrrolidine. HCl

Mito-TEMPO-H

1-Hydroxy-4-[2-triphenylphosphonio)-acetamido]-2,2,6,6-tetramethylpiperidine,1-Hydroxy- 2,2,6,6-tetramethyl-4-[2-(triphenylphosphonio)acetamido]piperidinium dichloride

PTFE

Polytetrafluoroethylene

KHB

Krebs-Henseleit buffer

EhGSSG

redox potential of GSH/GSSG

GSH

Glutathione

GSSG

glutathione disulphide

bleomycin

a medication used to treat cancer

EC-SOD

Extracellular superoxide dismutase

R213G

substitution of arginine to glycine at amino acid 213

PBS

Phosphate Buffered Saline

WT

Wild type

HPLC

High-performance liquid chromatography

SNP

Single-nucleotide polymorphism

mtDNA

Mitochondrial DNA