SEARCH

SEARCH BY CITATION

Keywords:

  • adenine nucleotide translocase;
  • Ca2+-independent phospholipase A2;
  • calcium retention capacity;
  • docosahexaenoic acid;
  • mild uncoupling;
  • reactive oxygen species

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments and conflict of interest disclosure
  7. References
Thumbnail image of graphical abstract

Ca2+-independent phospholipase A2 (iPLA2) is hypothesized to control mitochondrial reactive oxygen species (ROS) generation. Here, we modulated the influence of iPLA2-induced liberation of non-esterified free fatty acids on ROS generation associated with the electron transport chain. We demonstrate enzymatic activity of membrane-associated iPLA2 in native, energized rat brain mitochondria (RBM). Theoretically, enhanced liberation of free fatty acids by iPLA2 modulates mitochondrial ROS generation, either attenuating the reversed electron transport (RET) or deregulating the forward electron transport of electron transport chain. For mimicking such conditions, we probed the effect of docosahexaenoic acid (DHA), a major iPLA2 product on ROS generation. We demonstrate that the adenine nucleotide translocase partly mediates DHA-induced uncoupling, and that low micromolar DHA concentrations diminish RET-dependent ROS generation. Uncoupling proteins have no effect, but the adenine nucleotide translocase inhibitor carboxyatractyloside attenuates DHA-linked uncoupling effect on RET-dependent ROS generation. Under physiological conditions of forward electron transport, low micromolar DHA stimulates ROS generation. Finally, exposure of RBM to the iPLA2 inhibitor bromoenol lactone (BEL) enhanced ROS generation. BEL diminished RBM glutathione content. BEL-treated RBM exhibits reduced Ca2+ retention capacity and partial depolarization. Thus, we rebut the view that iPLA2 attenuates oxidative stress in brain mitochondria. However, the iPLA2 inhibitor BEL has detrimental activities on energy-dependent mitochondrial functions.

The Ca2+-independent phospholipase A2 (iPLA2), a FFA (free fatty acids)-generating membrane-attached mitochondrial phospholipase, is potential to regulate ROS (reactive oxygen species) generation by mitochondria. FFA can either decrease reversed electron transport (RET)-linked or enhance forward electron transport (FET)-linked ROS generation. In the physiological mode of FET, iPLA2 activity increases ROS generation. The iPLA2 inhibitor BEL exerts detrimental effects on energy-dependent mitochondrial functions.

Abbreviations used
ANT

adenine nucleotide translocase

BEL

bromoenol lactone

CAT

carboxyatractyloside

CRC

calcium retention capacity

CsA

cyclosporin A

DHA

docosahexaenoic acid

ETC

electron transport chain

FCCP

Carbonyl cyanide-p-trifluoromethoxyphenylhydrazone

FET

forward electron transport

FFA

free fatty acids

glut

glutamate

GSH

glutathione (reduced)

INAD

infantile neuroaxonal dystrophy

iPLA2

calcium-independent phospholipase A2

LysoPC

l-α-lysophosphatidylcholine

Mal

malate

MTT

3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide

O2•−

superoxide

PUFA

polyunsaturated fatty acid

RBM

rat brain mitochondria

RET

reverse electron transport

RLM

rat liver mitochondria

ROS

reactive oxygen species

RSMM

rat skeletal muscle mitochondria

Succ

succinate

UCP

mitochondrial uncoupling protein

Δψm

mitochondrial membrane potential

The Ca2+-independent phospholipase A2 (iPLA2) liberates free fatty acids (FFA) by hydrolyzing the sn-2 ester bond of membrane glycerophospholipids. iPLA2 has been found in mitochondria from various mammalian tissues [reviewed in (Six and Dennis 2000)], where iPLA2 is attached to the outer face of the inner mitochondrial membrane (IMM) (Williams and Gottlieb 2002; Kinsey et al. 2007a). iPLA2 has two major isoforms, VIA and VIB iPLA2 (also denoted iPLA2β and iPLA2γ). They are encoded by two different genes (Ackermann et al. 1994; Mancuso et al. 2000). In contrast to the secretory and the cytosolic PLA2, iPLA2 is characterized by suicide inactivation by bromoenol lactone (BEL). iPLA2 is responsible for more than 70% of the whole-brain PLA2 activity, with the highest levels found in hippocampus and striatum (Yang et al. 1999). VIB iPLA2 is the predominant phospholipase activity in mammalian mitochondria (Williams and Gottlieb 2002; Gadd et al. 2006; Kinsey et al. 2007a; Moon et al. 2012; Jaburek et al. 2013).

According to generally accepted concepts, membrane-associated iPLA2 removes oxidatively damaged fatty acids from mitochondrial membrane lipids, thereby allowing membrane repair by remodeling (Farooqui and Horrocks 2004). This process consists of deacylation of phospholipids and ATP-depending reacylation of lysophospholipids (Kudo and Murakami 2002). Consequently, in ischemia/reperfusion, iPLA2 activity is detrimental for mitochondria, because it causes loss of phospholipids because of the lack in ATP.

Mitochondria are the main oxygen consumers in cells (Murphy 2009). The electron transport chain (ETC) of mitochondria is an important source for generating the reactive oxygen species (ROS) superoxide (O2•−). O2•− dismutates mostly in the matrix to H2O2, which escapes from mitochondria or becomes converted in the matrix into H2O by the glutathione peroxidase. In addition, liberation of FFA from membrane phospholipids by iPLA2 could potentially modulate mitochondrial generation of ROS by two mechanisms. Firstly, attenuation of the reverse electron transport (RET) by mild uncoupling [reviewed in (Skulachev 1996)] and, secondly, stimulation of ROS generation by deregulation of the forward electron transport (FET) in the ETC (Schönfeld and Reiser 2006; Schönfeld and Wojtczak 2007, 2008).

The mild uncoupling mode involves well-coupled mitochondria respiring in the resting state [slow oxygen uptake, excess supply of oxidizable substrates, high mitochondrial membrane potential (∆ψm)]. They generate O2•− at high rates during the ∆ψm-driven uphill transport of electrons from the other complexes to complex I of the ETC by RET. Therefore, the amazing property of the RET-driven O2•− generation is its dramatic decrease with a slight reduction of Δψm, as evidenced by incremental additions of FCCP (Korshunov et al. 1997). This strong dependency of O2•− generation on Δψm has recently shifted the views about the physiological role of iPLA2, where iPLA2 has been claimed to attenuate the ROS generation by mitochondria of several tissues (spleen, lung) through mild uncoupling by liberated FFA (Jaburek et al. 2013).

Moreover, the uncoupling activity of FFA has been mainly attributed to several protein types of the IMM, mostly, the adenine nucleotide translocase (ANT) and the group of uncoupling proteins (UCPs). It is a well-established fact that the ANT enhances the FFA-induced uncoupling (Andreyev et al. 1989; Schönfeld 1990; Brustovetsky and Klingenberg 1994), whereas a comparable function of the UCPs (except for UCP1) is still under debate [recently reviewed (Mailloux and Harper 2011; Shabalina and Nedergaard 2011)]. It is widely accepted that FFA permeate in the protonated form, the IMM, release H+ into the matrix compartment and, thereafter, the anionic form returns to the outer leaflet of the IMM, predominantly with the assistance of ANT and the ubiquitously expressed UCPs, to pick up again H+ ions for the next cycle [fatty acid cycling model (Skulachev 1991)].

Furthermore, it is known that brain tissue is rich in polyunsaturated fatty acids (PUFA) (Cabezas et al. 2012) and, one of them, docosahexaenoic acid (DHA; 22:6n-3) is indispensable for normal brain function and becomes liberated mostly by iPLA2 activity (Strokin et al. 2003, 2007). Interestingly, PUFA are considered to target mitochondria by various activities, including diminishing of ROS generation by mild uncoupling (Rohrbach 2009).

Finally, it is important to emphasize that the mechanism of pathogenesis of some neurodegenerative disorders seems to be based on the attenuation of VIA iPLA2 activity. Two neurologic disorders, infantile neuroaxonal dystrophy (INAD) and ‘neurodegeneration with brain iron accumulation’ (NBIA), are known to be associated with mutations of the human PLA2G6 gene that encodes for VIA iPLA2 (Morgan et al. 2006; Gregory et al. 2008). Recently, we used mice hypomorph for VIA iPLA2 to study the molecular mechanisms underlying INAD. In addition, the Barth syndrome is also related to iPLA2 (Malhotra et al. 2009). We demonstrated that calcium entry into these mutant astrocytes was strongly reduced as compared to cells from wild-type mice (Strokin et al. 2012).

Indeed, several reports indicate an involvement of the iPLA2 isoforms in the protection of cells against oxidative stress (Cummings et al. 2002; Kinsey et al. 2007b, 2008). In rat insulinoma (INS-1) cells, VIA iPLA2 plays a role in protection of mitochondrial functions in oxidant-induced apoptosis (Seleznev et al. 2006).

In the first part of this study, we mimic the effect of iPLA2 activity-induced FFA enrichment on ROS generation by rat brain mitochondria (RBM). The mild uncoupling effect was simulated by low micromolar concentrations of DHA, and its action on the RET-dependent ROS generation was examined using succinate (succ) for oxidation (Loschen et al. 1971). In addition, the effect of low DHA concentrations on the FET-dependent ROS generation was studied using the complex I substrates glutamate (glut) plus malate (mal) for oxidation. In the second part, we use BEL as iPLA2 inhibitor (Ackermann et al. 1994), to study the consequences of reduced iPLA2 activity on the mitochondrial ROS generation. Finally, we propose that iPLA2 does not strengthen the anti-oxidative defense that would diminish ETC-associated O2•− generation by mild uncoupling. In contrast, iPLA2-mediated FFA release is likely to impair the FET and thereby stimulates the mitochondrial O2•− generation.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments and conflict of interest disclosure
  7. References

Materials

Adult Wistar rats were from Charles River (Sulzfeld, Germany); R-/S-BEL and BEL racemic mixture from Cayman Chemicals (Ann Arbor, MI, USA); Amplex Red and CaGreen-5N from Invitrogen (Karlsruhe, Germany); and 1-palmitoyl-2-[1-14C]-oleoyl-phosphatidyl choline (2.035 × 109 Bq/mmol) from Biotrend (Cologne, Germany); Safranine O, FCCP, 1–chloro-2,4 dinitrobenzene (CDNB), L-α-lysophosphatidylcholine mixture and other chemicals from Sigma-Aldrich (Taufkirchen, Germany).

Preparation of mitochondria

The preparation of RBM is based on a method described (Liu et al. 2002). Briefly, the brain was rapidly removed and placed in ice-cold isolation buffer (10 mM Tris-HCL pH 7.4, 320 mM sucrose, 0.5 mM EDTA, 0.5 mM EGTA, 0.1% (w/v) bovine serum albumin). All solutions used were ice-cold and all steps were carried out at 4°C. The tissue [1 : 10 tissue buffer ratio (w/v)] was minced, homogenized in the presence of protease type I (1 mg/mL), and, finally, the homogenate centrifuged (2000 g; 4 min). The supernatant was centrifuged (12 500 g; 12 min). For purification (removal of synaptosomes), the crude mitochondrial pellet was resuspended in ice-cold wash buffer (10 mM Tris-HCL pH 7.4, 320 mM sucrose, 0.02% digitonin), centrifuged (11 500 g; 12 min). Thereafter, the pellet was washed two times with digitonin-free buffer, centrifuged (11 500 g; 12 min), and resuspended in ice-cold wash buffer.

Rat liver mitochondria (RLM) were prepared according to the procedure described previously (Schönfeld et al. 1989). Rat skeletal muscle mitochondria (RSMM) were isolated from musculus gastrocnemius of adult rats, as described (Frezza et al. 2007). We use nagarse (not trypsin) for tissue digestion.

Protein concentrations in the mitochondrial stock suspension were determined by the Bradford method using bovine serum albumin as standard. Protein concentrations in the stock mitochondrial suspensions were 20–25 mg/mL.

Membrane polarization

To estimate the changes in polarization of the IMM, the cationic dye safranine O was applied (Akerman and Wikstrom 1976). Its membrane potential-sensitive uptake by energized mitochondria was measured fluorimetrically. Accumulation of safranine O inside mitochondria causes quenching of the safranine fluorescence, which has been attributed to the stacking of safranine O molecules at the inner side of the inner membrane (Zanotti and Azzone 1980). Aliquots of the RBM suspensions (0.5 mg/mL protein) were pre-incubated with 5 μM R-BEL or 5 μM S-BEL for 20 min at 25°C. The depolarization by BEL was monitored as increase in fluorescence of the dye safranine, resulting from the release of accumulated safranine O from mitochondria. Fluorescence was determined at 495 nm (excitation) and 586 nm (emission) using PerkinElmer Luminescence Spectrophotometer LS50B (PerkinElmer, Waltham, MA, USA) in 1 mL cuvette, connected to stirring devices.

Measurement of PLA2 activity with synthetic radioactive phospholipid as substrate

iPLA2 activity of RBM was assessed as described (Strokin et al. 2007). The PLA2 activity was determined as the hydrolysis of 1 μL of the synthetic substrate 1-palmitoyl-2-[1-14C]-oleoyl-phosphatidyl choline (0.925 kBq, 2.3 μM) by 50 μL of the RBM over 60 min at 37°C in reaction buffer of 100 mM Tris–HCl, pH 7.5, containing 0.1 mM ATP [for further details see (Strokin et al. 2007)]. For the estimation of iPLA2 activity, the RBM were preincubated with 5 μM BEL for 20 min at 25°C. The difference to the total PLA2 activity gave the BEL-sensitive PLA2 activity, which we denote as iPLA2 activity. Before substrate addition, we pre-incubated the RBM with different inhibitors for 20 min at 25°C. Reaction was terminated by extraction with 1-butanol. The extracts were subjected to thin layer chromatography with petroleum ether/ethyl ether/glacial acetic acid 70/30/1 v/v as mobile phase and oleic acid as a standard. Lipids were visualized by iodide vapor. The area corresponding to the oleic acid was scraped and radioactivity was analyzed by liquid scintillating counting.

Incubations

Mitochondria were suspended in standard incubation medium consisting of 10 mM KH2PO4, 0.5 mM EGTA, 60 mM TRIS, 60 mM KCl, 110 mM Mannitol pH 7.4. Substrates for the mitochondrial energization were throughout 5 mM glutamate plus 5 mM malate or 5 mM succinate. The DHA stock solutions (2 mM) were prepared freshly daily in pure ethanol. Oxygen dissolved in ethanol was removed by bubbling with N2. Further additions are described in the legends. Mitochondria were either applied in the untreated form (control, thermostated for 20 min at 25°C) or after pre-incubation with 5 μM BEL for 20 min at 25°C.

Oxygen consumption

Mitochondrial oxygen consumption was measured using an Oroboros Instruments Oxygraph® (Innsbruck, Austria) at 37°C. Respiration was stimulated by the addition of 2 mM ADP or indicated DHA concentrations.

Monitoring H2O2 release

H2O2 release by RBM was estimated fluorimetrically by recording the formation of resorufin from Amplex Red. Resorufin fluorescence was measured using excitation and emission wavelengths of 560 nm and 590 nm on a PerkinElmer Luminescence Spectrophotometer LS50B in 1 mL cuvette, connected with stirring device. RBM were used at a concentration of 0.2 mg/mL protein in standard incubation medium (substrate glut/mal) at 37°C. RBM were pre-incubated with 5 μM BEL for 20 min at 25°C. Monitoring of H2O2 release from mitochondria was initiated by fluorogenic indicator Amplex Red (5 μM) in the presence of horseradish peroxidase (2 U/mL). The fluorescence signal was calibrated by sequential additions of known H2O2 amounts (up to 750 pmol).

Estimation of the ANT content in mitochondria

The ANT content was estimated with phosphorylating mitochondria (state 3) using inhibitor-titration with carboxyatractyloside (CAT; (Schönfeld 1990). RBM, RLM, or RSMM were suspended in standard incubation medium (substrate glut/mal) supplemented with 2 mM ADP. Phosphorylating respiration was titrated by repeated additions of CAT.

Calcium retention capacity

Calcium retention capacity (CRC) was determined by repeated additions of aliquots (20 nmol Ca2+/mg) of a CaCl2 solution (2 mM) to RBM, suspended in standard incubation medium (substrate glut/mal) at 37°C. The uptake of added Ca2+ was followed as decrease of the fluorescence of the membrane-impermeant Ca2+-sensitive dye Calcium Green-5 N (100 nM). Fluorescence intensities were measured at 506-nm excitation and 532-nm emission wavelength.

Determination of mitochondrial glutathione

GSH depletion was obtained by incubation of CDNB (1-chloro-2,4 dinitrobenzene) and BEL. GSH reacts with CDNB under catalysis by GSH transferase (Jocelyn and Cronshaw 1985). Isolated RBM (5 mg/mL) were pre-incubated with different CDNB concentrations for 2 min at 25°C. In addition, RBM incubated with different concentrations of BEL for 20 min at 25°C were spun at 8000 g for 5 min and washed once in isolation buffer. Control mitochondria were treated with equal amount of solvent (ethanol). GSH amount was measured by reaction with DTNB [5,5′-Dithiobis(2-nitrobenzoic acid)] producing yellow colored TNB (5-thio-2-nitrobenzoic acid). Absorption of TNB was determined at 412 nm using LS50B PerkinElmer Luminescence Spectrophotometer. The rate of TNB production is directly proportional to GSH concentration.

Statistical analysis

For statistical analysis, relative values of mitochondrial functions were expressed as mean ± SD from at least three independent experiments with samples from individual preparations of mitochondria. Statistical significance was evaluated using the Student's t-test. A value of p < 0.01 was accepted as significant.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments and conflict of interest disclosure
  7. References

iPLA2 activity in isolated energized RBM

We investigated the total PLA2 activity with RBM suspended in standard incubation medium (substrate glut/mal) in the presence of several effectors (Fig. 1a). In addition, the data of the BEL-sensitive iPLA2 activity are given in Fig. 1b. This was obtained after subtraction of the residual activity of BEL-treated mitochondria from the total iPLA2 activity (Fig. 1a).

image

Figure 1. BEL-sensitive iPLA2 activity of isolated rat brain mitochondria (RBM). RBM (1 mg protein/mL) were suspended in standard incubation medium containing 5 mM glutamate plus 5 mM malate. The iPLA2 activity was measured, as described in Methods. (a) Total PLA2 activity in RBM. Additions: 1 μM rotenone, 10 μM H2O2, 800 nmol Ca2+/mg protein, 1 μM cyclosporine A (CsA) and 5 μM BEL. (b) BEL-sensitive iPLA2 activity, given as difference between total PLA2 activity and BEL-treated RBM. Means ± SEM from at least of three individual mitochondrial preparations. *p < 0.01, compared to control RBM (dashed line), #p < 0.01, compared to Ca2+-treated RBM, +p < 0.01, compared Ca2+- and CsA-treated RBM, $p < 0.01, compared to rotenone-treated RBM, §p < 0.01, compared to H2O2-treated RBM, **p < 0.01, compared to BEL-sensitive iPLA2 activity (control, dashed line).

Download figure to PowerPoint

The measured total PLA2 activity of RBM in the absence of effectors (resting state) was designated as control (Fig. 1a). Figure 1a shows control PLA2 activity and the activities after eliciting Ca2+-induced mitochondrial permeability transition and after exposure to oxidative stress. The basal PLA2 activity (control) amounts to about 80 pmol/min/mg of protein (Fig. 1a, first bar), demonstrating that iPLA2 is active in energized non-treated RBM. The pharmacological inhibitor of iPLA2, BEL, significantly decreased the iPLA2 activity to about 70% of that in non-treated RBM (Fig. 1a, second bar) resulting in the basal BEL-sensitive iPLA2 (Fig. 1b, first bar). Importantly, the total PLA2 activity was increased after inducing the Ca2+-induced permeability transition (about 75%; Fig. 1a, third bar). However, this activation by Ca2+ was abolished by BEL (Fig. 1a, fourth bar). That treatment also leads to an increase in iPLA2 activity (Fig. 1b, second bar). Similarly, blocking of the Ca2+-induced permeability transition by cyclosporine A (CsA), an efficient and potent inhibitor of the mitochondrial permeability transition pore, brings the total PLA2 activity as well as iPLA2 activity back to control level (Fig. 1a, fifth bar and Fig. 1b, third bar).

Moreover, oxidative stress, which was induced either by rotenone-stimulated O2•− production in complex I of the ETC or, alternatively, by exogenous addition of H2O2, enhances the total PLA2 activity by 70% (Fig. 1a, seventh bar) or 90% (Fig. 1a, ninth bar), respectively. Also, the iPLA2 activity was increased by rotenone (Fig. 1b, fourth bar) and H2O2 (Fig. 1b, fifth bar) stimulation. Taken together, these findings clearly show that BEL-sensitive iPLA2 is active in intact, energized RBM. We confirm previous work (Gadd et al. 2006; Jaburek et al. 2013), reporting that the iPLA2 activity is stimulated by oxidative stress or Ca2+-induced mitochondrial permeability transition.

ANT supports ‘mild uncoupling’ by free DHA in RBM

The enzymatic activity of iPLA2 releases several PUFA from brain phospholipids (Green and Reed 1998), especially DHA (Strokin et al. 2003). Therefore, here we added low micromolar DHA concentrations to mimic ‘mild uncoupling’ conditions in RBM. Figure 2a shows the uncoupling activity of increasing concentrations of DHA. Uncoupling was quantified by determining the increase of the resting respiration compared to control (black bars). To visualize the putative role of the ANT in the DHA-induced uncoupling, respiration was measured in the presence of the ANT inhibitor CAT. CAT is known to attenuate FFA-induced uncoupling (Andreyev et al. 1989; Schönfeld 1990). It can be seen that CAT-treated RBM (Fig. 2a, white bars) respond to DHA additions with a significantly lower increase of the respiration than untreated RBM (black bars).

image

Figure 2. Role of the adenine nucleotide translocase (ANT) in docosahexaenoic acid (DHA)-sensitive uncoupled respiration. Rat brain mitochondria (RBM), rat liver mitochondria (RLM) or rat skeletal muscle mitochondria (RSMM) (1 mg/mL protein) were suspended in 1.5 mL standard medium (substrate glut/mal) supplemented with and oligomycin (5 μM). (a) Uncoupling of RBM indicated as DHA-dependent increase of resting respiration (black and white bars). In the presence of carboxyatractyloside (CAT; white bars), DHA-dependent respiration is lower. Means ± SEM from at least four individual mitochondrial preparations. (b) CAT-sensitive rates of DHA (5 nmol DHA/mg of protein)-induced respiration (white bars) and ANT content of these mitochondria (striped bars). The ANT content was estimated from inhibitor-titration of the state 3 respiration with CAT. Data are representative for four individual mitochondria preparations. *p < 0.01, CAT untreated compared to CAT-treated RBM, #p < 0.01, compared to RLM CAT-sensitive respiration, +p < 0.01, compared to RLM ANT content, $p < 0.01, RBM compared to RSMM ANT content.

Download figure to PowerPoint

To compare brain mitochondria with mitochondria from other tissues, we measured the response of resting respiration to DHA in RBM and also in RLM and RSMM. Remarkably, poisoning of the ANT with CAT in RLM has a minimal effect on the DHA-induced increase of the respiration, whereas in RSMM the effect was slightly larger (Fig. 2b, white bars). This difference in the response of the respiration of RBM, RLM, and RSMM to DHA can be explained by different contents of ANT in the respective mitochondria (Schönfeld 1990). The ANT content was determined in RBM, RLM, and RSMM (Fig. 2b, striped bars). For this purpose, the maximal phosphorylating respiration of mitochondria was adjusted by the addition of ADP (state 3). Thereafter, state 3 respiration was progressively decreased to the resting state level by incremental additions of the ANT inhibitor CAT (not shown). This strategy is based on the finding that CAT forms a tight 1 : 1 stoichiometric complex with ANT (Weidemann et al. 1970; Klingenberg et al. 1975). Thus, the amount of CAT required to slow down the respiration gives the amount of the ANT protein.

Finally, the plot of CAT-sensitive DHA-induced respiration of RBM, RLM, and RSMM (Fig. 2b) reveals that the extent of CAT-sensitive uncoupled respiration corresponds to the ANT content. In summary, the amount of the DHA-induced CAT-sensitive respiration of RLM, RBM, and RSMM is in line with their different ANT contents.

DHA and mitochondrial ROS generation

Mitochondria can generate ROS by RET or FET in the ETC, as recently reviewed (Murphy 2009). Moreover, O2•− generation driven by the RET from complex II to complex I responds very sensitively to depolarization of the IMM (Korshunov et al. 1997).

Therefore, the influence of the DHA-induced mild uncoupling on ETC-associated O2•− generation was examined in RBM. For this purpose, RBM were energized with succinate, and the release of H2O2 was recorded with the Amplex Red/peroxidase assay. DHA caused mild uncoupling and, thereby, decreased the release of H2O2 from RBM (Fig. 3a, panel a, traces i and ii). However, this decline recovered mostly after inhibiting the ANT by CAT (Fig. 3a, trace (i), and summary results in b, second and third bar).

image

Figure 3. Effect of docosahexaenoic acid (DHA) on the H2O2 release under reversed electron transport (RET) or forward electron transport (FET), and study of mechanism of action. Rat brain mitochondria (RBM) (0.2 mg/mL) were incubated in standard incubation medium. Generation of superoxide by RET was started by the addition of 5 mM succinate. H2O2 was measured as described in Methods. (a) Typical traces of the changes in the resorufin fluorescence. Additions: DHA (10 nmol/mg protein), lysophosphatidylcholine (LysoPC) (10 nmol/mg protein), 5 μM carboxyatractyloside (CAT), 1 mM GTP or 0.25 μM FCCP. (b) Summary of RET-associated H2O2 release obtained with three individual experiments. (c) Typical traces of increase of resorufin fluorescence because of FET-associated H2O2 generation. RBM (0.4 mg/mL) were incubated in standard incubation medium. Generation of superoxide by forward FET was started either by addition of 5 mM pyruvate plus 5 mM malate (trace i) or 5 mM glutamate plus 5 mM malate (traces ii, iii). Additions: DHA (10 nmol/mg protein), 0.25 μM FCCP and 5 μM CAT. Rates of H2O2 release (pmol H2O2/min/mg protein) are given as numbers at all traces. (d) MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) reduction by DHA from four experiments. MTT reduction to blue-colored formazan dye was measured photometrically [595 nm (Liu et al. 1997)]. Mitochondria (0.1 mg protein) suspended in ice-cold incubation buffer (1 mL) supplemented with 5 mM glutamate/5 mM malate, additions DHA or rotenone (1 μM) for 5 min. Thereafter, tubes were warmed to 25°C in the presence of MTT (0.1 mg/mL). After 10-min incubation, tubes were centrifuged in 1 mL pure ethanol to measure reduced MTT. *p < 0.01, compared to control RBM, #p < 0.01, DHA-treated compared to DHA/CAT-treated RBM, ##p < 0.01, DHA/CAT-treated RBM compared to DHA/GTP-treated RBM, ###p < 0.01, DHA-treated compared to LysoPC–treated RBM, $p < 0.01 compared to untreated mitochondria (dashed line).

Download figure to PowerPoint

Moreover, it is possible that UCPs, among them most likely the ubiquitously expressed UCP2, are involved in DHA-induced uncoupling. Since guanine nucleotides are usually applied to inhibit the UCP activity (Jezek et al. 2004; Wojtczak et al. 2011; Jaburek et al. 2013), the effect of GTP on the DHA-induced depression of the H2O2 release was measured. However, the addition of GTP had no effect on DHA-induced reduction of the H2O2 release (Fig. 3a, trace (ii) and, b, fourth bar).

Furthermore, upon the hydrolysis of the sn-2 ester bond in glycerophospholipids by iPLA2, not only FFA but also lysophospholipids are formed. Therefore, lysophosphatidylcholine (LysoPC) was applied as model effector for the succinate-driven O2•− generation. In contrast to DHA, LysoPC (ester bonded fatty acids of different hydrocarbon chain-length) had no effect (Fig. 3a, trace (iv), and b, fifth bar), suggesting that LysoPC has no protonophoric activity, such as DHA. LysoPC was applied at a 2 μM concentration (10 nmol/mg protein). Such low concentrations make time-dependent actions of lysoPC because of micelle formation unlikely. Moreover, the lack in protonophoric activity of lysoPC fits the finding that only concentrations higher than 50 μM depolarize mitochondria (Rustenbeck et al. 1991). Finally, when the uncoupler FCCP was added the H2O2 release strongly declined and was not reactivated by CAT (Fig. 3a, trace (iii), and b, sixth and seventh bar).

Similarly, we studied whether DHA affects the FET-associated H2O2 release by RBM, oxidizing complex I-related substrates (substrate glu/mal). Under this condition, the rate of H2O2 release is much lower than that seen by the succinate oxidation supported RET-associated H2O2 release (compare Fig. 3a and c). Nevertheless, DHA enhanced the FET-driven H2O2 release by RBM (Fig. 3c, traces i, ii). In contrast to DHA, FCCP decreased H2O2 release probably because of oxidation of the mitochondrial NADH pool (Fig. 3c, trace iii).

DHA-enhanced FET-associated H2O2 release is consistent with previous observations obtained with oleic, arachidonic, or phytanic acid (Schönfeld and Reiser 2006; Schönfeld et al. 2011) with mitochondria from rat brain and PC12 cells. This is explained by the deregulating effect of DHA on the electron transport in the ETC. The latter is attributed to inhibitory actions of long-chain saturated and unsaturated FFA at certain sites within the ETC, partial cytochrome c depletion and increased inner membrane fluidity (Schönfeld and Wojtczak 2007). The increased fluidity likely increases the membrane ‘electron-leakage’, thereby facilitating one-electron reduction of oxygen to O2•−.

Furthermore, we also demonstrate that the reduction of the electron acceptor 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) to a formazan dye by mitochondria decreased concentration dependently with the presence of DHA (Fig. 3d). MTT reduction by mitochondria is mostly attributed to the complex I-III electron transfer to MTT (Liu et al. 1997). For comparison, rotenone, an ETC inhibitor in complex I, impairs MTT reduction (Fig. 3d, first bar). Thus, inhibition of the MTT reduction further substantiates the deregulation of the FET by DHA.

In summary, two conclusions can be drawn. Firstly, the DHA-linked decrease of the RET-driven H2O2 release from RBM is mediated only by the ANT without any indication for the involvement of UCPs in the DHA-induced uncoupling. Secondly, DHA increases the FET-associated H2O2 release by deregulation of the electron flow in the ETC.

Effect of BEL on mitochondrial ROS generation

The finding of DHA-induced effects on mitochondrial O2•− generation by RBM prompted us to study the influence of the iPLA2 activity on the mitochondrial ROS generation. For this purpose, H2O2 release was measured with untreated (control) and BEL-treated succinate-oxidizing RBM. The traces recording the resorufin fluorescence clearly demonstrate that pre-treatment of RBM with BEL for 20 min enhanced the release of H2O2 (Fig. 4a). BEL concentration dependently stimulates the H2O2 release (Fig. 4b).

image

Figure 4. Bromoenol lactone (BEL)-treated rat brain mitochondria (RBM) display enhanced reactive oxygen species (ROS) generation. (a) RBM (5 mg/mL) were pre-incubated with 5 μM BEL (grey line) or vehicle (EtOH) (black line) for 20 min at 25°C. ROS generation was followed, see legend to Fig. 3. Addition of control and pre-treated RBM (0.2 mg/mL) in standard medium (substrate succ) initiated the increase of resorufin fluorescence. (b) ROS production was significantly increased after BEL application. Means ± SEM at least of four mitochondria preparations. *p < 0.01, compared to control RBM (dashed line).

Download figure to PowerPoint

Obviously, inactivation of iPLA2 enhanced the release of H2O2 from RBM. Thus, we checked firstly the hypothesis that an increased release of H2O2 by BEL-treated mitochondria seen at RET can be explained by the decrease in iPLA2-catalyzed liberation of FFA. Since it is likely that the amount of endogenous FFA controls the polarization state of the IMM, it is suggestive to speculate that BEL-treated RBM are more polarized than untreated RBM. We assume here that hyperpolarization of mitochondria goes together with increased ROS generation. Therefore, in the next experiments we measured the effect of BEL on the polarization state of the IMM in energized RBM.

BEL treatment of RBM depolarizes the IMM, increases basal respiration, and reduces the CRC

The polarization state of the IMM was estimated in resting RBM. For detection of a possible change in the polarization state, the ∆ψm-dependent uptake of the membrane-permeable cationic dye safranine O was determined. Figure 5a shows that after the addition of RBM to the standard incubation medium (substrate glut/mal), the fluorescence signal of safranine O dramatically decreased, reflecting the dye uptake and the quenching of dye fluorescence in the matrix compartment. The strong decrease of the safranine O fluorescence indicates a high polarization state of the IMM. In contrast, in RBM, which were pre-incubated for 20 min with R-BEL or S-BEL (Fig. 5a, broken lines), the reduction of safranine O fluorescence was significantly lower, corresponding to about 85% of that obtained with untreated RBM (Fig. 5a, solid line). Finally, with the addition of FCCP, ∆ψm in RBM collapsed resulting in the maximal release of accumulated safranine O from RBM (Fig. 5a). Depolarization by BEL is also indirectly indicated by the significant, albeit slight increase of the basal respiration in BEL-treated mitochondria (Fig. 5b).

image

Figure 5. Pre-treatment of rat brain mitochondria (RBM) with bromoenol lactone (BEL) partly depolarized the inner mitochondrial membrane (IMM). RBM were pre-incubated with 5 μM R-BEL or S-BEL for 20 min in standard incubation medium (substrate glut/mal) at 25°C. The polarization state of IMM was estimated from the accumulation of the Δψm-probe safranine O (5 μM). Decrease of safranine O fluorescence following the RBM addition (0.5 mg protein/mL) indicates Δψm–driven safranine uptake combined with fluorescence quenching by energized mitochondria. Addition of 10 μM FCCP leads to decreased fluorescence indicating Δψm collapse. (a) BEL-treated RBM show lower membrane polarization (dashed lines). Traces are representative for three individual mitochondria preparations. (b) Increase of resting respiration by BEL, indicating uncoupling. (c) Acute addition of 5 μM BEL. *p < 0.01, compared to control RBM.

Download figure to PowerPoint

Furthermore, inhibition of both VIA and VIB iPLA2 isoforms, as shown by the effects of either R– or S-BEL, exerts the same depolarizing activity. To verify whether BEL acts itself directly on the membrane polarization, BEL was added acutely to untreated RBM. In this case, no change in safranine O-fluorescence could be detected (Fig. 5c). Thus, the depolarizing effect of BEL is the consequence of the pretreatment of RBM with BEL.

In addition, pre-treatment of RBM with BEL impaired energy-dependent functions, indicated by the Δψm-driven Ca2+ uptake capacity, which can be quantified by measuring the mitochondrial CRC. The CRC measurements were done by repeatedly adding Ca2+ aliquots (corresponding to 20 nmol aliquots) to energized RBM (substrate glut/mal). For the recording of the Ca2+ concentration in the incubation medium, we used Calcium Green-5 N as Ca2+-sensitive fluorochrome. CRC of untreated RBM was found to be more than 140 nmol Ca2+/mg of protein (Fig. 6a, trace i). For comparison, CRC measured with RBM treated with CsA was more than 340 nmol Ca2+/mg (trace ii). CsA is known as a potent inhibitor of the mitochondrial permeability transition pore. When RBM were pre-incubated with R-BEL (Fig. 6a, trace iii) or S-BEL (Fig. 6a, trace iv), CRC was dramatically diminished. Under this condition, CRC was only between 23 and 28% of that of control (untreated) RBM (Fig. 6b). Interestingly, BEL diminished the CRC of CsA-treated RBM (Fig. 6a, trace v).

image

Figure 6. Bromoenol lactone (BEL) treatment of rat brain mitochondria (RBM) decreases calcium retention capacity (CRC). RBM were preincubated as described in Fig. 3. Thereafter, aliquots of control (traces i, ii) and BEL-treated RBM (traces iii–v) (1 mg/mL of protein) were added to standard incubation medium (substrate glut/mal). CRC was estimated fluorimetrically by repeated additions of 20 nmol CaCl2 to the incubation mixture, as described in Methods. Cyclosporin A (CsA) enhanced CRC (trace ii). BEL partly abolished the effect of CsA (trace v). (b) Summarized CRC (traces i, iii, iv). Means ± SEM of at least three mitochondria preparations. *p < 0.01 compared to untreated RBM (dashed line).

Download figure to PowerPoint

In summary, the differences between control (untreated) and BEL-treated RBM in changes in the IMM polarization, basal respiration, and CRC clearly reveal that BEL has a deenergizing effect on RBM. Consequently, the increase in the H2O2 release observed with BEL-treated RBM (Fig. 4) cannot be attributed to a hyperpolarization of the IMM, which might result from suppression of iPLA2-mediated FFA liberation.

BEL reduces the antioxidative defense by the glutathione-peroxidase system

Alternatively, the possibility has to be considered that the increased H2O2 release by BEL-treated RBM results from partial inactivation of the glutathione/glutathione–peroxidase system in the mitochondrial matrix compartment. Such inactivation would increase the escape of H2O2 from mitochondria. Therefore, the possible effect of BEL treatment on the content of reduced glutathione (GSH) was examined. GSH is the major anti-oxidant in the brain (DeLeve and Kaplowitz 1991), and GSH plays a key role in detoxification of H2O2 in the matrix compartment of mitochondria (Chance et al. 1979). Moreover, the reduction of H2O2 by GSH to H2O is catalyzed by the combined activities of glutathione S-transferases and glutathione peroxidases (Townsend et al. 2003).

To verify our hypothesis, the depletion of GSH by 1-chloro-2,4-dinitrobenzene (CDNB) was measured colorimetrically with the Ellman's reagent, dinitrobenzoic acid. CDNB acts as a favored substrate for the matrix enzyme glutathione S-transferases, thereby forming a GSH-CDNB conjugate (Jocelyn and Cronshaw 1985). Figure 7a shows that the incubation of RBM with CDNB concentration-dependently enhanced the release of H2O2. Therefore, we studied the possible correlation with the endogenous GSH content of RBM after BEL treatment, to elucidate its relation to the antioxidant defense system.

image

Figure 7. Effects of CDNB (1-chloro-2,4 dinitrobenzene) or bromoenol lactone (BEL) on glutathione (GSH) content of rat brain mitochondrial (RBM). (a) Increase of H2O2 release from RBM initiated by glutathione depletion with indicated concentrations of CDNB. RBM (0.2 mg/mL) were suspended in standard incubation medium (substrate succ), 5 μM Amplex red and 2 U/mL horseradish peroxidase. (b) Glutathione content of RBM treated either with CDNB, BEL, or docosahexaenoic acid (DHA). RBM (5 mg/mL suspended in isolation medium) were pre-incubated with BEL, as in Fig. 4. Release of H2O2 results from lowering the matrix content of reduced glutathione by CDNB. GSH was estimated as described in Methods. Means ± SEM at least of four mitochondria preparations. *p < 0.01 compared to control RBM (dashed line).

Download figure to PowerPoint

When RBM were pre-treated with 20 μM BEL the matrix content of GSH declined by 30% compared with untreated RBM (Fig. 7b). CDNB (60 μM) had the same effect (Fig. 7b). As additional control, we also tested the influence of DHA. This was minimal.

Thus, the enhanced H2O2 release in BEL-treated RBM (Fig. 4) seems to be connected to the mitochondrial GSH depletion. Furthermore, we demonstrated that reduced GSH inside the mitochondria has quantitatively reacted with CDNB. In mitochondria permeabilized with the pore-forming agent alamethicin (He et al. 1996), to improve the access of CDNB to GSH the results were very similar to those with untreated mitochondria (data not shown).

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments and conflict of interest disclosure
  7. References

iPLA2 activity in isolated energized RBM

Brain tissue exhibits high specific oxygen uptake, has low antioxidative defense capacity, is rich in PUFA (Cabezas et al. 2012), and expresses a high level of iPLA2 (Yang et al. 1999; Balboa et al. 2002). These features prompted us to consider brain mitochondria as a suitable object for examining the physiological functions of iPLA2. In addition, previously we studied primary astrocytes from a mouse disease model of INAD, which is caused by a spontaneous VIA iPLA2 mutation. There we could show that the VIA iPLA2 plays a central role for the proper Ca2+ homeostasis (Strokin et al. 2012). These results confirm that iPLA2 isoforms are a promising pharmacological target for treatment of neurodegenerative diseases. This view seems to be substantiated by the hypothesis, assuming that iPLA2 contributes in situ to mild uncoupling of mitochondria, a process which attenuates Δψm-dependent RET-associated O2•− generation. This study addresses the role of the enzymatic activity of iPLA2 in the control of ROS generation by brain mitochondria. Oxidative stress, the bad side of the cellular ROS generation, has been discussed in numerous reports as the cause or enhancing factor for the pathological development of neurodegenerative diseases.

Here, we demonstrate firstly that iPLA2 is active in intact, energized native brain mitochondria (Fig. 1). In contrast it has been reported that intact mitochondria have little or no activity of iPLA2 (Rauckhorst et al. 2014). Nevertheless, iPLA2 exhibits activity without exposure to oxidants, in contrast to what has been claimed before (Jezek et al. 2010; Jaburek et al. 2013), and without collapse of Δψm (Gadd et al. 2006).

To explain the discrepancy between data reporting that iPLA2 is active in energized mitochondria (Kinsey et al. 2007b) and the claim that activation of iPLA2 requires mitochondrial deenergization, it is helpful to consider the methods used for assessing iPLA2 activity. The method used by the Schnellmann group (Kinsey et al. 2007b) measured iPLA2 activity in mitochondria using a radioactively labeled synthetic phospholipid substrate. However, the Pfeiffer group measured the accumulated FFA derived from hydrolysis of endogenous mitochondrial phospholipids (Gadd et al. 2006; Rauckhorst et al. 2014). Since in intact, energized mitochondria liberated FFAs are used largely for membrane remodeling, an enrichment of mitochondria with FFAs can be seen only after the interruption of the ATP supply. In conclusion, iPLA2 activity seems to be indeed present in isolated native brain mitochondria.

BEL has detrimental influence on energy-dependent functions of mitochondria indicated by decrease in CRC and content of reduced glutathione in mitochondria

We tested BEL as an inhibitor of iPLA2 (Ackermann et al. 1994), to suppress the liberation of FFA from membrane phospholipids by iPLA2 activity. If we assume that resting brain mitochondria are slightly uncoupled and that this uncoupling is because of the iPLA2 activity, it can be expected that the iPLA2 inhibitor BEL increases the RET-driven O2•− generation. Indeed, we found that BEL-treated brain mitochondria oxidizing the RET-supporting succinate show higher H2O2 release (Fig. 4). However, this slightly higher H2O2 release cannot be explained by an improved coupling, which would result from treatment of brain mitochondria with BEL. In contrast, the Δψm of the IMM is reduced compared with that in untreated brain mitochondria. Depolarization is shown by the lower uptake of safranine O as well as, indirectly, by the increased resting respiration (Fig. 5). Interestingly, BEL-induced mitochondrial depolarization was also reported for mitochondria in situ in PC12 cells (Ma et al. 2011). BEL, a lipophilic and bulky chemical compound, is likely to disturb the membrane structure after enrichment in the IMM, and consequently, membrane properties, which could lower Δψm by inducing membrane leakage. Finally, deenergizing effects of BEL on the mitochondria are indicated by the CRC decline (Fig. 6).

In addition, BEL-treated brain mitochondria enhanced ROS generation (Fig. 4). We propose that a lowered mitochondrial antioxidative defense contributes to this ROS increase. This idea is based on the observation that the BEL treatment of brain mitochondria results in partial depletion of glutathione in the mitochondrial matrix (Fig. 7), which is in the range 6–10 mM (Giulivi and Cadenas 1998). Glutathione depletion might be related to reactive elements in the BEL molecule structure. The reactive carbon–bromine bond reacts with a cysteine residue (651Cys) in the proximity of serine (Jenkins et al. 2013). The reactive lactone ring structure covalently binds serine-465Ser in the catalytic center of iPLA2, thereby inactivating the enzyme. This recently reported reactivity of BEL encourages us to speculate that BEL depletes mitochondria from glutathione by forming adducts with glutathione.

Modulation of iPLA2 is no suitable cellular tool to diminish oxidative stress

Generally, modulation of the mitochondrial O2•− generation by iPLA2-derived FFA should depend on the nature of the oxidized substrate. With FADH2-donating substrates, such as succinate, RET is important for energized mitochondria (Chance and Hollunger 1961) and the RET-associated O2•− generation is attenuated by uncoupling (Loschen et al. 1971; Korshunov et al. 1997). We used low micromolar DHA concentrations to mimic increased levels of iPLA2-liberated FFA. The release of DHA, but also of other PUFAs, is controlled by the activity of the VIB iPLA2 isoform in neuronal cells [e.g., (Strokin et al. 2007)]. DHA stimulates uncoupled respiration, as shown in Fig. 2. However, when RBM were treated with the ANT inhibitor CAT, the stimulation of respiration by DHA was significantly lower. This indicates a major contribution of the ANT in mediating FFA-dependent uncoupling. Previous work (Schönfeld 1990) shows that FFA-dependent uncoupling cannot completely be attributed to ANT. Residual uncoupling by DHA in CAT-treated mitochondria has been connected with several proteins of the inner membrane, like glutamate-aspartate carriers, phosphate carriers, and UCPs (Samartsev et al. 1997; Jezek et al. 1998; Zackova et al. 2000).

Nevertheless, we conclude that mainly the ANT mediates DHA-linked attenuation of the O2•− generation, because it becomes restored by the ANT inhibitor CAT (Fig. 3a and b). Binding of the CAT to ANT suppresses the ability of the ANT to enhance the protonophoric activity of FFA. In case of CAT application, Δψm increased and, consequently the Δψm-driven RET-based O2•− generation gets higher.

However, GTP or GDP, which inhibit UCPs (Wojtczak et al. 2011; Cardoso et al. 2013; Jaburek et al. 2013), did not influence the DHA-linked attenuation of the RET-driven O2•− generation (Fig. 3a and b). This demonstrates that the DHA-linked uncoupling is mediated mostly by the ANT and, importantly, in brain mitochondria a participation of UCPs-assisted fatty acid cycling is unlikely. Nevertheless, some reports describe that UCP2, which is expressed in the brain might be responsible for diverse pathologic changes in neurodegenerative diseases (Sullivan et al. 2003; Deierborg et al. 2008). Moreover, it has been proposed that UCP2 has a neuroprotective role in cerebral stroke (Mattiasson et al. 2003; Mehta and Li 2009). However, a recent study strongly indicates that such action of UCP2 cannot be attributed to UCP2-mediated mild uncoupling (Shabalina et al. 2014).

In addition, it is important to remember that FADH2-donating substrates (fatty acids) are poorly used as brain fuel, as discussed recently in a topical review (Schönfeld and Reiser 2013). Furthermore, it is assumed that only RET is sensitive to Δψm (Murphy 2009; Brand 2010) and a Δψm-sensitive RET-linked ROS generation is exclusively observed during mitochondrial succinate oxidation (Shabalina et al. 2014).

The strict dependence of the energy metabolism of brain tissue on glucose degradation makes it very likely that FFA modulate the O2•− generation by brain mitochondria, which oxidize NADH-donating substrates. Thus, it is important to understand, to which extent iPLA2-liberated FFA can affect FET-dependent ROS generation. We found a two-fold increase of H2O2 release, when DHA (10 nmol/mg of protein) was added to RBM oxidizing glutamate plus malate (Fig. 3c). This increase is most likely because of deregulated FET, thereby supporting the one-electron transfer to molecular O2, to form O2•−.

Two recent studies, where mitochondria isolated from heart, lung and spleen tissue were exposed to oxidants (H2O2 or tert-butylhydroperoxide), to activate the iPLA2, proposed an iPLA2-associated stimulation of mitochondrial respiration (Jezek et al. 2010; Jaburek et al. 2013). These mitochondria responded to the addition of oxidants with rapidly increased respiration, which was not found with mitochondria pre-treated with BEL. However, it should be stressed that this reported increase in resting respiration was minimal, only about 10%.

In conclusion, we demonstrate that exogenous DHA causes ANT-dependent mild uncoupling of brain mitochondria. This mild uncoupling partly abolishes RET-dependent ROS release from mitochondria. However, we find no evidence for an involvement of UCPs in the DHA-induced uncoupling. Moreover, even assuming that high succinate concentrations would be available for oxidation by mitochondria in neural cells, the decreased Δψm connected with the ATP turnover in neural cells is sufficient to abolish the succinate-driven RET-related ROS generation (Votyakova and Reynolds 2001).

Moreover, we propose that iPLA2 does not generate high FFA levels in energized mitochondria. Energized mitochondria supply ATP. ATP prevents FFA accumulation in mitochondria owing to promoting their reesterification. Thus, we rule out that mitochondrial iPLA2 has a role in the modulation of ROS production in brain mitochondria.

Nevertheless, iPLA2 has been shown to attenuate the detrimental consequences of oxidative stress, the peroxidation of lipids (Kinsey et al. 2007a, 2008). We attribute this beneficial activity to the ability of iPLA2 to hydrolyze oxidatively damaged membrane phospholipids, a precondition for the membrane repair.

Moreover, we here demonstrate for the first time that pharmacological concentrations of the iPLA2 inhibitor BEL exert detrimental influences on energy-dependent functions of mitochondria. Such detrimental activities of BEL should be taken into consideration for interpreting the effects of BEL on mitochondrial depolarization or apoptosis triggered by mitochondrial permeability transition. Finally, the present findings describing the activity of iPLA2 in maintenance of mitochondrial functions have to be connected to the role of iPLA2 in neurodegenerative diseases, such as INAD.

Acknowledgments and conflict of interest disclosure

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments and conflict of interest disclosure
  7. References

The study was supported by a grant from Deutsche Forschungsgemeinschaft (Re 563/ 22-1). We thank Petra Grüneberg and Heidelore Goldammer for their expert technical assistance. No conflict of interest.

All experiments were conducted in compliance with the ARRIVE guidelines.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments and conflict of interest disclosure
  7. References