Statins lower calcium-induced oxidative stress in isolated mitochondria

Abstract

Statins are widely used cholesterol-lowering agents that exert cholesterol-independent effects including antioxidative. The present study delineates the effects of statins, atorvastatin, and simvastatin on oxidative stress and functions of mitochondria that are the primary cellular sources of oxidative stress. In isolated rat liver mitochondria, both the statins prevented calcium-induced cytochrome c release, lipid peroxidation, and opening of the mitochondrial membrane permeability transition (MPT). Both the statins decreased the activity of mitochondrial nitric oxide synthase (mtNOS), lowered the intramitochondrial ionized calcium, and increased the mitochondrial transmembrane potential. Our findings suggest that statins lower intramitochondrial ionized calcium that decreases mtNOS activity, lowers oxidative stress, prevents MPT opening, and prevents the release of cytochrome c from the mitochondria. These results provide a novel framework for understanding the antioxidative properties of statins and their effects on mitochondrial functions.

Introduction

The 3-hydroxy-3-methylglutaryl-coenzyme A reductase inhibitors commonly known as statins are cholesterol-lowering drugs used in the treatment of various diseases including hypercholesterolemia. ^1,2 Statins exert multiple beneficial effects including anti-inflammatory, ^3,4 antiproliferative, and antithrombotic. ⁵ Statins also exert profound cellular antioxidant effects ^6–9 and lower oxidative stress in various tissues and organs. ⁶ However, the primary cellular targets for the antioxidative properties of statins are not fully understood.

Mitochondria are largely involved in oxidative stress and remain the primary cellular sources of reactive oxygen species (ROS). Elevated mitochondrial ROS causes oxidative injury in numerous conditions including ischemia/reperfusion, ¹⁰ aging, and neurodegenerative diseases. ¹¹ Increased mitochondrial ROS impairs mitochondrial functions and exacerbates oxidant production and mitochondrial damage. ¹² It has been shown that statins act on mitochondria and influence mitochondrial functions. ^13,14 However, the exact actions of statins on mitochondria and the underlying mechanisms are not fully understood. At low micromolar, statins improve mitochondrial functions. Proteomic studies show that statins exert protective effects on mitochondrial respiratory chain proteins. ¹⁵ In adult ventricular myocytes, cerivastatin and simvastatin protect myocardium by inhibiting β-adrenergic receptor-stimulated activation of small GTPase Rac1 that inhibits c-Jun N-terminal kinases (JNK)-dependent activation of the mitochondrial death pathway and apoptosis. ¹⁶ Simvastatin shows antihypertrophic and antioxidative effects via decreasing ROS generation. ⁹ Similarly, atorvastatin, simvastatin, and mevastatin decrease ROS in endothelial cells by S-nitrosylation of thioredoxin (TRX). ¹⁷ Increase in S-nitrosylation of TRX is accompanied by increased redox-regulatory activity of TRX and decreased cellular ROS. These studies indicate that statins exert antioxidative effects by lowering ROS and improving performance of antioxidative enzymes. Some studies have suggested that higher concentrations (>10 µM) of statins damage mitochondria and exert proapoptotic properties. For example, it has been reported that statins exert cytotoxic effects on human T cells, B cells, and myeloma tumor cells by promoting apoptosis in these cells. ¹⁸ Lovastatin has been shown to induce apoptosis in human colon cancer cells. ¹⁹ In rhabdomyosarcoma cells, simvastatin activates mitochondrial pathway of apoptosis via translocation of proapoptotic protein Bax from the cytosol to the mitochondria. ²⁰ In vitro, lovastatin induces mitochondrial permeability transition (MPT) in a dose-dependent (10–80 μM) manner. ¹³ Use of statins is associated with 0.5–1% of myopathies caused by ubiquinone depletion and block of electron transport. ²¹ Fluvastatin, lovastatin, and simvastatin induce myopathy and damage mitochondrial functions ²² particularly by lowering Q10 levels (reviewed in Ref. 23).

The present study investigated the effects of low concentrations (1–3 μM) of two widely used statins, atorvastatin and simvastatin, on oxidative stress and functions of isolated mitochondria. Our findings delineate a novel mechanism for the effects of statins in decreasing oxidative stress and improving mitochondrial functions.

Material and methods

Purification of mitochondria

Liver mitochondria were isolated from Sprague-Dawley rats by differential centrifugation and purified by Percoll purification as described. ²⁴ Purity of the isolated mitochondria was assessed by measuring cytochrome a using ∊ _{605–630 nm} 12 mM⁻¹ cm⁻¹ as described. ²⁴ Only mitochondria with less than 5% impurity were used. The respiratory control ratio was determined as described. ²⁴ Briefly, mitochondria (0.5 mg) were suspended in a thermostated chamber containing 1 ml buffer consisting of HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; (100 mM), MgCl₂ (5 mM), EDTA, ethylenediaminetetraacetic acid; (0.5 mM), and potassium phosphate (5 mM, pH 7.04), and respiration was stimulated by the addition of rotenone (5 µM), K⁺-succinate (5 mM), and adenosine diphosphate (150 µM). The ratio of state 3 to state 4 respirations was calculated and mitochondria with a ratio of ≥6 were used. Broken mitochondria were prepared by freezing the intact mitochondria in liquid nitrogen followed by thawing as described. ²⁵

Treatments

Mitochondria were treated with atorvastatin or simvastatin (1 or 3 μM; 30 min; from Sigma Aldrich, Milwaukee, WI, USA) or equivolume of vehicle dimethyl sulfoxide (DMSO) and oxidative stress was induced by loading mitochondria with Ca²⁺ (40 μM) as described ^10,26 .

Detection of cytochrome c release

Mitochondria (50 mg/ml) were incubated at 4°C in HEPES buffer (0.1 M, pH 7.1) containing protease inhibitor cocktail (leupeptin, phenylmethanesulfonyl fluoride, pepstatin A, and aprotinin; 10 µM each) as described. ²⁶ Mitochondria were energized with K⁺-succinate (800 μM) in the presence of rotenone (5 μM), and cytochrome c release was induced by loading the mitochondria with Ca²⁺ as described. ^10,26 Samples were centrifuged at 10,000g for 10 min and the supernatant was recentrifuged at 100,000g for 30 min. The supernatant of the second centrifugation was collected, and cytochrome c was detected by Western blot using monoclonal cytochrome c antibody (eBioscience) as described. ²⁷

Determination of mitochondrial LPO

2-Thiobarbituric acid (TBA) reactive substance (TBARS) was measured as the first surrogate product of lipid peroxidation (LPO). Samples were prepared as described above and in the absence of protease inhibitors. TBARSs were determined by TBA assay as described. ^26,28 Briefly, mitochondrial suspension containing Fe²⁺ was mixed with trichloroacetic acid (30%) in HCl to stop the peroxidation. TBA and butylated hydroxytoluene were added, the reaction mixture was boiled at 100°C for 20 min, cooled on ice, and centrifuged at 10,000g for 5 min. The supernatants were collected and TBARS was determined at 535 nm against the blank.

Detection of mitochondrial MPT

Mitochondria (1 mg) were suspended in a cuvette containing buffer consisting of mannitol (220 mM), sucrose (70 mM), and HEPES (10 mM, pH 7.40). Mitochondria were energized with succinate (800 μM) in the presence of rotenone (5 μM). The MPT was induced by loading mitochondria with Ca²⁺ (40 μM) and detected by following ▵OD at 540 nm. ²⁹

Determination of mtNOS activity

Mitochondrial nitric oxide synthase (mtNOS) activity was determined using oxyhemoglobin (OxyHb), citrulline, and chemiluminescence assays as described.^24,30 Briefly, for OxyHb assay, broken mitochondria (30 µg) were added to a cuvette containing HEPES buffer (100 mM; pH 7.40), l-arginine (100 µM), tetrahydrobiopterin (10 µM), and Cu/Zn-superoxide dismutase (1 KU/ml). The optical density (OD) was detected at 401–420 nm in the presence of OxyHb (4 µM). Quantitation was performed using ∊ _{401–420 nm} 100 mM⁻¹ cm⁻¹. For citrulline assay, mitochondria (1 mg) were supplemented with l-[ ³ H]arginine (30,000–50,000 cpm), and l-citrulline was measured by radioassay as described. ³⁰ For chemiluminescence assay, mitochondria (1 mg) were injected into the purge vessel containing vanadium chloride (0.8% in 1 M HCl) thermostated at 95°C, and nitric oxide (NO) was measured using Chemiluminescence Analyzer (Sievers 280i, General Electric) as described. ³⁰

Determination of intramitochondrial ionized calcium ([Ca²⁺])_m

Determination of [Ca²⁺]_m was performed using the highly sensitive dual-wavelength excitation fluorometric assay as described. ¹⁰ Briefly, isolated mitochondria were loaded with fura-2/acetoxymethylester (10 µM, 15 min) and [Ca²⁺]_m was measured by exciting the loaded mitochondria at dual-wavelengths of 352–362 nm and detecting the emission at 510 nm.

Determination of extramitochondrial calcium

Extramitochondrial calcium was determined by following the ▵OD of Arsenazo III (5 μM) at 675–685 nm. ¹⁰

Transmembrane potential (▵ψ)

The ▵ψ supported by K⁺-succinate (800 µM) in the presence of rotenone (5 μM) was determined at 511–533 nm using safranine (10 µM) as described. ²⁷

Statistical analysis

Data were analyzed using analysis of variance and differences were considered significant at p < 0.05. Barograms are mean ± standard error of mean of n ≥ 6.

Results

Cytochrome c release, LPO, and mitochondrial permeability transition

Elevation of [Ca²⁺]_m leads to cytochrome c release, increased ROS, and MPT opening. Figure 1A shows cytochrome c release upon loading mitochondria with Ca²⁺. This figure shows that Ca²⁺-induced cytochrome c release was prevented in mitochondria treated with atorvastatin or simvastatin.

Figure 1.

Statins prevent cytochrome c release, lower lipid peroxidation (LPO), and prevent membrane permeability transition (MPT) opening. (A) Release of cytochrome c from untreated mitochondria (Ctrl) or mitochondria loaded with calcium (Ca) and treated with vehicle (Veh), atorvastatin (Atr, 1 or 3 μM), or simvastatin (Sim, 1 or 3 µM) is shown. Horse heart cytochrome c (Cyto c; 25 ng) was used as positive control. Released cytochrome c was detected in the supernatant (Sup) and the remaining cytochrome c in the pellet (Pellet). (B) LPO in untreated mitochondria (Ctrl) or mitochondria loaded with calcium (Ca) and treated with vehicle, atorvastatin, or simvastatin is shown. *Significantly different from Ctrl (p < 0.05). #Significantly different from vehicle (p < 0.05). (C) Membrane permeability transition (MPT) was induced by loading mitochondria with calcium (arrow) in mitochondria treated with vehicle, simvastatin, or atorvastatin.

LPO and MPT openings are mitochondrial oxidative stress biomarkers. Figure 1B and C shows that the Ca²⁺-induced elevation of LPO and MPT openings was prevented by both statins. Ca²⁺-induced MPT opening was sensitive to MPT inhibitor cyclosporine A (not shown).

mtNOS activity

Increased [Ca²⁺]_m stimulates mtNOS activity that causes cytochrome c release and increases oxidative stress. Figure 2A–C shows that increasing [Ca²⁺]_m stimulates mtNOS activity and that both atorvastatin and simvastatin prevent Ca²⁺-induced mtNOS stimulation.

Figure 2.

Statins decrease mitochondrial nitric oxide synthase (mtNOS) activity. mtNOS activity of untreated mitochondria (Ctrl) was stimulated by loading mitochondria with calcium (Ca) in mitochondria treated with vehicle (Veh), atorvastatin (Atr, 1 or 3 μM), or simvastatin (Sim, 1 or 3 µM). mtNOS activity was determined by radioassay (panel A) and expressed as cpm l-citrulline per mg mitochondrial protein (cpm/mg) by oxyhemoglobin assay (panel B) and expressed as pmole NO per mg mitochondrial protein minute (NO pmol/µg min); and by chemiluminescence assay (panel C) and expressed as pmol NO per mg mitochondrial protein (pmol NO). *Significantly different from Ctrl (p < 0.05). #Significantly different from vehicle (p < 0.05).

Intramitochondrial and extramitochondrial calcium concentration

mtNOS is Ca²⁺ sensitive and increase in [Ca²⁺]_m stimulates mtNOS activity. Figure 3A and B shows that atorvastatin and simvastatin decreased both intra- (Figure 3A) and extramitochondrial Ca²⁺ (Figure 3B). Statins alone did not affect mitochondrial calcium flux (not shown).

Figure 3.

Statins alter mitochondrial calcium homeostasis and transmembrane potential (Δψ). Panels (A) and (B) show mitochondrial calcium homeostasis and Panel (C) shows the Δψ in mitochondria treated with vehicle (Veh) or atorvastatin or simvastatin (Atr or Sim; 1 or 3 μM). (A) Intramitochondrial ionized calcium concentration ([Ca²⁺]_m) was determined by exciting Fura-2-loaded mitochondria and detecting the fluorescence at 510 nm. At the end of the test, mitochondria were depleted of [Ca²⁺]_m by uncoupling mitochondria using carbonyl cyanide m-chlorophenylhydrazone (CCCP; 1 µM). Loading mitochondria with calcium is shown by arrow (Ca²⁺). (B) Extramitochondrial Ca²⁺ was measured at 675–685 nm using Arsenazo III. Where indicated, ▵ψ was collapsed by CCCP (1 µM) to allow [Ca²⁺]_m efflux to extramitochondria to be detected by Arsenazo III. At the end of the test, Ca²⁺ was chelated by EGTA (2 mM). (C) The Δψ was supported by K⁺-succinate (Succ) in the presence of rotenone (not shown) and measured in mitochondria loaded with Ca²⁺ and treated with vehicle, atorvastatin, or simvastatin. At the end of the test, Δψ was collapsed by uncoupling mitochondria with CCCP (CCCP; 1 µM).

Mitochondrial transmembrane potential (▵ψ)

Mitochondria retain [Ca²⁺]_m in response to ▵ψ and decreased ▵ψ decreases [Ca²⁺]_m. To test whether the effect of statins in decreasing [Ca²⁺]_m was due to decreased ▵ψ, the effect of both the statins on ▵ψ was tested. Figure 3C shows that ▵ψ of mitochondria treated with atorvastatin and simvastatin was higher than that of control mitochondria.

Discussion

Statins exert beneficial effects by improving the circulatory lipid profile. However, many beneficial effects of statins are related with their antioxidant properties.^6–9,31 While mitochondria are the primary cellular sources of ROS and involved in regulating cellular redox environment, the cellular mechanisms by which statins affect mitochondrial ROS and functions are not fully understood. We hypothesized that statins directly affect mitochondria and tested antioxidative effects of atorvastatin and simvastatin on isolated mitochondria. Our results show that both statins decrease mitochondrial oxidative stress, prevent Ca²⁺-induced cytochrome c release, and prevent MPT opening. These findings provide novel insight about beneficial properties of statins and suggest that mitochondria are important cellular targets for statins.

Increased [Ca²⁺]_m increases ROS production, causes MPT opening, ^32,33 and release of cytochrome c from mitochondria.^10,26,34,35 Statins prevent cytochrome c release from mitochondria of pancreatic islets ³⁶ and prevent apoptosis of mesenchymal stem cells. ³⁷ Mitochondria are important intracellular Ca²⁺ stores and actively participate in Ca²⁺ signaling. ³⁸ Our findings indicate that both the statins interact with mitochondria and prevent Ca²⁺-induced cytochrome c release.

Elevated mitochondrial oxidative radicals is one of the primary events underlying various pathological conditions, and lowering mitochondrial oxidative stress assists complications of those conditions. Increased LPO is one of the hallmarks of oxidative stress and apoptosis. ³⁹ We found that Ca²⁺-induced increase in LPO was significantly decreased by atorvastatin and simvastatin. Several reports have shown beneficial effects of statins in pathological conditions associated with elevated ROS and oxidative stress. Effects of statins in diminishing ROS have been shown by their antihypertrophic properties and ability to decrease kidney and liver dysfunction. ^{40 ,41} For fluvastatin, scavenging of ROS and a reduction in superoxide formation have been shown in vitro and in vivo. ⁴² Atorvastatin, pravastatin, and cerivastatin inhibit superoxide-forming capacity of the endothelial cells^31,43 Atorvastatin upregulates catalase expression both in rat vascular smooth muscle cells in vitro and in normocholesterolemic, spontaneously hypertensive rats in vivo. ³¹ Our findings suggest a novel mechanism, whereby inhibition of mitochondrial cytochrome c release and lowering mitochondrial LPO can contribute to antioxidative effects of statins.

MPT opening that leads to mitochondrial swelling and disruption of mitochondrial functions is another marker for oxidative stress. Our findings show that both atorvastatin and simvastatin inhibited Ca²⁺-induced MPT. These findings that are consistent with statins preventing cytochrome c release and LPO elevation further suggest that statins interact with mitochondria to lower the mitochondrial oxidative stress. It has been suggested that higher concentrations of lovastatin, that is, between 20 and 80 μM, induce Ca²⁺-dependent MPT opening. ¹³ Other studies have suggested that in human rhabdomyosarcoma cells, 30–100 µM simvastatin or lovastatin promote Bax translocation to mitochondria and induce apoptosis via mitochondrial pathway. ²⁰ Our findings show that both the statins at low micromolar concentrations prevent MPT opening. Thus, it appears that inducing or preventing MPT or apoptosis depends on the concentration of the statins.

Elevation of [Ca²⁺]_m stimulates mtNOS that produces NO and oxidative species in mitochondria and leads to the release of cytochrome c and oxidative stress. ^10,26,27,34 Since statins prevent the release of cytochrome c and lower oxidative stress markers, we tested whether statins decrease mtNOS activity. Our findings show that mtNOS activity stimulated by elevating [Ca²⁺]_m was inhibited by both atorvastatin and simvastatin. These novel findings suggest that inhibiting mtNOS activity serves as a mechanism underlying antioxidative properties of statins.

mtNOS is Ca²⁺ sensitive and lowering [Ca²⁺]_m inhibits mtNOS activity. ²⁵ Since both statins decreased mtNOS activity, we tested whether statins lower [Ca²⁺]_m. The [Ca²⁺]_m is regulated primarily by two principal mechanisms. First, the intramitochondrial calcium homeostasis that consists of precipitation of [Ca²⁺]_m to nonionized calcium pools, known as matrix electron-dense granules, and release of [Ca²⁺]_m from the granules. Drugs, hormones, or pathologic conditions alter this intraorganelle calcium homeostasis and increase or decrease the [Ca²⁺]_m.^10,34,44 The second mechanism for regulating the [Ca²⁺]_m is releasing the [Ca²⁺]_m to extramitochondria. This mechanism is regulated by ▵ψ that renders the inner mitochondrial membrane negatively charged. Decrease in ▵ψ leads [Ca²⁺]_m efflux from the organelles. Our findings show that both atorvastatin and simvastatin decreased [Ca²⁺]_m, which describes lower mtNOS activity of mitochondria treated with these statins. In order to study whether the decreased [Ca²⁺]_m was due to increased precipitation of [Ca²⁺]_m or increased [Ca²⁺]_m efflux, we measured extramitochondrial Ca²⁺ and ▵ψ. Our findings show that both the statins decreased the extramitochondrial Ca²⁺ and increased the ▵ψ. These findings strongly suggest that statins decrease [Ca²⁺]_m by increasing precipitation of [Ca²⁺]_m to nonionized mitochondrial calcium pools. We have shown that oxidative stress-inducing drugs ³⁴ and pathologic conditions ¹⁰ increase the [Ca²⁺]_m by increasing the release of [Ca²⁺]_m from the matrix electron-dense granules. Findings of the present study indicate that statins decrease the [Ca²⁺]_m and introduce a novel mechanism for antioxidative properties of statins reported by many investigators.

Taken together, our findings show that statins lower [Ca²⁺]_m that decreases mtNOS activity, prevents MPT opening, lowers LPO, and prevents release of cytochrome c from the mitochondria (Figure 4). These findings suggest a novel mechanism for the protective effect of statins against cellular injury caused by oxidative stress.

Figure 4.

Schematic diagram showing the suggested mechanism for lowering oxidative stress by statins. Mitochondria consist of the inner membrane (IM), outer membrane (OM), intermembrane space (IMS), and matrix. Electrons enter the respiratory chain complex I and move to the IV, also known as cytochrome c oxidase. Parallel to this intermembrane electron transport, protons (H⁺) are pumped from the matrix into the IMS. This proton gradient renders the inner face of the IM negatively charged and builds the transmembrane potential (Δψ). The Δψ is the driving force for mitochondria to uptake large amounts of Ca²⁺, however, intramitochondrial ionized calcium concentration ([Ca²⁺]_m) is maintained very low by various mechanisms including precipitating the Ca²⁺ to intraorganelle nonionized sources known as electron-dense granules. Release of Ca²⁺ from the granules stimulates the calcium-sensitive mitochondrial nitric oxide synthase (mtNOS) that converts l-arginine (l-arg) into nitric oxide (NO). NO reacts with superoxide (O²⁻) to form peroxynitrite (ONOO⁻). The mitochondrial ONOO⁻ causes a series of oxidative reactions including increasing the membrane lipid peroxidation (LPO), stimulating the mitochondrial permeability transition (MPT) opening, and causing the release of cytochrome c (cyto c) from the mitochondria. Statins lower [Ca²⁺]_m by lowering the release of Ca²⁺ from the electron-dense granules. Lowered [Ca²⁺]_m decreases mtNOS activity. Decreased mtNOS activity prevents MPT opening, lowers LPO, and prevents the release of cyto c from the mitochondria.