Dopaminergic Neurons Respond to Iron-Induced Oxidative Stress by Modulating Lipid Acylation and Deacylation Cycles

Metal-imbalance has been reported as a contributor factor for the degeneration of dopaminergic neurons in Parkinson Disease (PD). Specifically, iron (Fe)-overload and copper (Cu) mis-compartmentalization have been reported to be involved in the injury of dopaminergic neurons in this pathology. The aim of this work was to characterize the mechanisms of membrane repair by studying lipid acylation and deacylation reactions and their role in oxidative injury in N27 dopaminergic neurons exposed to Fe-overload and Cu-supplementation. N27 dopaminergic neurons incubated with Fe (1mM) for 24 hs displayed increased levels of reactive oxygen species (ROS), lipid peroxidation and elevated plasma membrane permeability. Cu-supplemented neurons (10, 50 μM) showed no evidence of oxidative stress markers. A different lipid acylation profile was observed in N27 neurons pre-labeled with [3H] arachidonic acid (AA) or [3H] oleic acid (OA). In Fe-exposed neurons, AA uptake was increased in triacylglycerols (TAG) whereas its incorporation into the phospholipid (PL) fraction was diminished. TAG content was 40% higher in Fe-exposed neurons than in controls. This increase was accompanied by the appearance of Nile red positive lipid bodies. Contrariwise, OA incorporation increased in the PL fractions and showed no changes in TAG. Lipid acylation profile in Cu-supplemented neurons showed AA accumulation into phosphatidylserine and no changes in TAG. The inhibition of deacylation/acylation reactions prompted an increase in oxidative stress markers and mitochondrial dysfunction in Fe-overloaded neurons. These findings provide evidence about the participation of lipid acylation mechanisms against Fe-induced oxidative injury and postulate that dopaminergic neurons cleverly preserve AA in TAG in response to oxidative stress.


Introduction
Parkinson disease (PD) is a neurodegenerative disorder associated with progressive neurodegeneration, and it is characterized by pathological α-synuclein (α-syn) aggregation, oxidative stress, elevated tissue iron, and mis-compartmentalization of copper. The ethiopathogenesis of PD is centered in the progressive loss of dopaminergic neurons, however, the mechanism underlying this process is still poorly understood [1][2][3][4].
The central role of iron (Fe) in PD pathology is supported by reports showing that the administration of Fe chelators and the overexpression of iron-binding proteins are neuroprotective in PD animal models [5][6][7]. Higher levels of divalent metal transporter DMT1 and loss of DMT1 down regulation in response to increased levels of Fe have been observed in post-mortem brains from PD patients [8].
Even though Fe-overload is a well known event in PD, a complete characterization of the state of Cu metabolism is still missing. In this respect, an important number of reports described the decrease of this metal ion in the substancia nigra neurons of PD patients [9,10]. Lack of appropriate Cu concentrations could lead to disturbances in CuZn superoxide dismutase (SOD 1) expression and, in consequence, to an increased susceptibility to α-syn oligomerization and the generation of oxidative stress [9]. Therefore, during PD, Fe-overload and Cu deficiency provokes a surrounding environment with high levels of oxidative stress for dopaminergic neurons.
One decisive event for the survival of neuronal cells under oxidative stress injury is the maintenance of cellular membrane properties. The preservation of physicochemical membrane properties (fluidity and permeability) during oxidative stress events strongly depends on phospholipid (PL) remodeling and it is therefore related to cellular mechanisms for the regulation of fatty acids (FA) availability and storage.
Phospholipids are important structural and functional components of biological membranes, which behave as reservoirs of important lipid mediators. The acyl groups of PL are highly diverse, depending on the polar head group, and are distributed in an asymmetric manner [11,12]. Rapid turnover of the sn-2 acyl moiety of phospholipids is attributed to the concerted and coordinated actions of phospholipases A 2 s (PLA 2 s) and lysophospholipid acyltransferases (LPATs). Another important storage of polyunsaturated fatty acids (PUFA), such as arachidonic acid (AA) and docosahexaenoic acid (DHA), are triacylglycerols (TAG). TAG remodeling and their fatty acid composition are also crucial for the maintenance of cellular PUFA levels, a process catalyzed by diacylglycerol acyltransferases (DGAT). Storage of TAG occurs in lipid droplets (LD) that are considered dynamic organelles regulated in response to the metabolic cellular state [13][14][15].
A link between α-syn, lipid metabolism, fatty acid oxidation, and mitochondrial damage has been established in PD. It has been proposed that α-syn can interact with FA, oxidized lipids, and LD [16][17][18]. However, little is known about the mechanisms of FA turnover and lipid remodeling that operate during oxidative stress injury in dopaminergic neurons.
In view of the above, the purpose of the present work was to determine the role of lipid acylation mechanisms during Fe-induced injury in dopaminergic neurons and to study the effect of Cu supplementation during Fe-overload. To this end, N27 dopaminergic cells were incubated under Fe-overload and Cu-supplementation conditions and the state of lipid acylation/deacylation processes and their role in the neuronal response to oxidative injury were studied.

Experimental treatments
In all the experiments carried out, cells were plated in 35 x 10 mm Cell Culture Dishes and grown to 80%-90% confluence. Fe 2+ and Cu 2+ treatments were carried out in serum-free medium. Treatments with inhibitors were performed as follows: medium was removed and replaced by serum-free media. Inhibitors were subsequently added to the desired final concentration -10 μM BEL, 50 μM ATK, 10 μM Ym 26734 and 7,5 μM CI-976-; controls received vehicle alone (DMSO 0.05%). After 30 min, 1 mM Fe 2+ , 10 or 50 μM Cu 2+ , their combination or its vehicle (ultrapure water) were added and cells were incubated under these conditions for 24 h unless otherwise specified.

Assessment of cell viability
Cell viability was assessed by MTT reduction assay. MTT is a water-soluble tetrazolium salt that is reduced by metabolically viable cells to a colored, water insoluble formazan salt. After treatments, MTT (5mg/ml) was added to the cell culture medium at a final concentration of 0.5 mg/ml. After incubating the plates for 2 h at 37°C in a 5% CO 2 atmosphere, the assay was stopped and the MTT-containing medium was replaced with solubilization buffer (20% SDS, pH 4.7). The extent of MTT reduction was measured spectrophotometrically at 570 nm [19]. Results are expressed as a percentage of the control.

Determination of lipid peroxidation
Lipid peroxidation was determined by the thiobarbituric acid reactive substances (TBARS) assay, which involves derivatization of malondialdehyde with thiobarbituric acid to produce a pink product that is quantified in a UV-VIS spectrophotometer. Briefly, after treatments, cells were scraped off into 300 μl of ice-cold water and mixed with 0.5 ml of 30% trichloroacetic acid. Then, 50 μl of 5 N HCl and 0.5 ml of 0.75% thiobarbituric acid were added. Tubes were capped, the mixtures were heated at 100°C for 30 min in a boiling water bath and the samples were centrifuged at 1000 x g for 10 min. TBARS were measured spectrophotometrically in the supernatant at 532 nm [20]. Results are expressed as a percentage of the control.

Determination of cell oxidant levels
Cell oxidative stress was evaluated using the probe 5 (or 6)-carboxy-2'7'-dichlorodihydrofluorescein diacetate (DCDCDHF). This probe can cross the membrane and after oxidation, it is converted into a fluorescent compound. After the corresponding treatments, the medium was discarded, and RPMI medium containing 10 μM DCDCDHF was added. After 30 min of incubation at 37°C, the medium was removed, cells were rinsed three times with phosphate buffer saline (PBS) and they were subsequently either imaged with an epifluorescence microscope or lysed in a buffer containing PBS and 1% NP-40. Fluorescence in the lysates (λex = 495, λem = 530) was measured in a SLM model 4800 fluorimeter (SLM Instruments, Urbana, IL) [21]. Results are expressed as a percentage of the control.

Measurement of LDH leakage
Lactate Dehydrogenase (LDH) leakage was determined as previously described [22] with slight modifications. After treatments, incubation medium was centrifuged at 1000 x g for 10 min at 4°C. The resulting supernatant was used to determine LDH activity, which was measured spectrophotometrically using LDH-P UV AA kit (Wiener lab.), following the manufacturer's instructions. Briefly, the conversion rate of reduced nicotinamide adenine dinucleotide to oxidized nicotinamide adenine dinucleotide was followed at 340 nm. Results are expressed as percentage of the control.

Evaluation of Mitochondrial Membrane Potential
To assess preservation of mitochondrial membrane potential, cultures were incubated for 20 min before fixation with the fluorescent probe MitoTracker Red (0.1 μg/mL), which labels mitochondria retaining their membrane potential with a bright red fluorescence.

Western blot analysis
For the preparation of total cell extracts, cells (10 × 10 6 cells) were rinsed with PBS, scraped and centrifuged. The pellet was rinsed with PBS and resuspended in 80 μl of a buffer containing 50 mM Tris, pH 7.5, 150 mM NaCl, 0.1% Triton X-100, 1% NP-40, 2 mM EDTA, 2 mM EGTA, 50 mM NaF, 2 mM β-glycerophosphate, 1 mM Na 3 VO 4 , 10 μg/ml leupeptin, 5 μg/ml aprotinin, 1 μg/ml pepstatin, 0.5 mM PMSF, and 0.5 mM DTT. Samples were exposed to one cycle of freezing and thawing, incubated at 4°C for 60 min and centrifuged at 10000 x g for 20 min. The supernatant was decanted and the protein concentration was measured. Cell lysates containing 25-50 μg of protein were separated by reducing 7-12.5% polyacrylamide gel electrophoresis and electroblotted to polyvinylidene difluoride membranes. Molecular weight standards (Spectra Multicolor Broad Range Protein Ladder, Thermo Scientific) were run simultaneously. Membranes were blocked with 5% nonfat dry milk in TBS-T buffer (20 mM Tris-HCl [pH 7.4], 100 mM NaCl, and 0.1% [wt/vol] Tween 20) for 1 h at room temperature and subsequently incubated with primary antibodies (anti-LPAAT-θ(T-17): sc-66789; DGAT1 (H-255): sc-32861 and Fatty Acid Synthase (G-11): sc-48357), washed three times with TBS-T and then exposed to the appropriate HRP-conjugated secondary antibody for 1 h at room temperature. Membranes were again washed three times with TBS-T and immunoreactive bands were detected by enhanced chemiluminescence (ECL; GE Healthcare Bio-Sciences, Buenos Aires, Argentina) using standard X-ray film (Kodak X-OMAT AR; GE Healthcare Bio-Sciences). Immunoreactive bands were quantified using image analysis software (Image J, a freely available application in the public domain for image analysis and processing, developed and maintained by Wayne Rasband at the Research Services Branch, National Institute of Mental Health).

Lipid phosphorus measurement
Lipid phosphorus (lipid P) was determined in total lipid extracts using the chemicals and reactions described by Rouser and collaborators [23].

[ 3 H] Arachidonic acid (AA) and [ 3 H] oleic acid (OA) incorporation into lipids
To evaluate the consequences of incubation with iron and copper on the label distribution among dopaminergic neuron lipid classes, cells were pre-labeled with [ 3 H] arachidonic acid (AA, 20:4) or [ 3 H] oleic acid (OA, 18:1). The labeled fatty acid (0.5 μCi per well) was mixed with unlabeled AA (1.5 μM) in the presence of lipid-free BSA (4 mol AA/mol BSA). Cells were pre-incubated for 5 h at 37°C with [ 3 H] AA or [ 3 H] OA supplemented medium to allow the labeled fatty acid incorporation. After this incubation, the medium was removed and replaced with serum free medium and cells were treated with 1 mM Fe 2+ , 50 μM Cu 2+ , the combination of these metals or the vehicle for 24 h.
After treatments, cells were washed 3 times with PBS, scraped off and subjected to lipid extraction following the method of Bligh and Dyer [24]. Lipid extracts were washed 2 times with Bligh and Dyer upper phase, dried under N 2 , resuspended in chloroform/methanol (2:1, vol/ vol) and spotted on silica gel plates. For the resolution of neutral lipids, silica gel G plates were used and the mobile phase consisted in hexane: eter: acetic acid (50: 50: 2.6). In order to resolve polar lipids, lipid extracts were spotted on silica gel H plates and phospholipids were separated by two dimensional thin layer chromatography exactly as described before [23]. Lipid spots were visualized by exposing the plates to iodine vapors. The regions corresponding to monoacylglycerols (MAG), diacylglycerols (DAG), free fatty acids (FFA) and triacylglycerols (TAG) were scraped off the silica gel G plates. The regions corresponding to phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylserine (PS), phospatidylinositol (PI) and phosphatidic acid (PA) were scraped off the silica gel H plates. All spots were transferred to vials where silica was deactivated by the addition of water. Five milliliters of 0.4% (wt/vol) Omnifluor in toluene/Arkopal N-100 (4:1, vol/vol) were subsequently added. Radioactivity in lipid spots from the blanks (typically less than 100 dpm) was subtracted from those of experimental samples.

Lipid droplets staining
After treatments, cells were washed 3 times with M1 buffer (150 mM NaCl, 1 mM CaCl 2 , 1 mM MgCl 2 , and 5 mM KCl in 20 mM HEPES buffer, pH 7.4) and stained for 15 min with Nile Red (Molecular Probes), 1.5 μg/ml in M1 buffer. After rinsing three times with M1 buffer, cells were observed with a Nikon Eclipse E-600 microscope (Nikon, Melville, NY, USA), using a K2E Apogee CCD camera driven by CCDOPS software (Santa Barbara Instrument Group, Santa Barbara, CA, USA) to visualize LD.

Determination of Plin-2 mRNA levels
Total RNA was isolated from dopaminergic neurons after the corresponding treatments using TRIZOL Reagent (Invitrogen) according to the manufacturer's protocol. RNA was resuspended in nuclease-free water and its concentration was assessed from A260:A280 absorbance ratio in a PicoDrop Spectrophotometer. Samples were stored at −80°C until use. Aliquots containing 2 μg total RNA were used to synthesize cDNA in reactions containing 1 μg Random Primer hexamers (Biodynamics), 1X M-MLV Reverse Transcriptase Reaction Buffer, 0.5 mM of each dNTP, 25 UI RNase inhibitor (Promega) and 200 UI M-MLV Reverse Transcriptase (Promega). Reactants were taken to a final volume of 25 μl with RNase-free water.
The cDNA resulting from RT was amplified by real-time quantitative PCR (qPCR). Gene expression levels were determined using Rotor-Gene 6000 (Corbett Research, Australia). RT-qPCR was performed in a final volume of 10 μl using Mezcla Real for Real-Time PCR (Biodynamics) and 0.2 μM of each primer. Gene-specific primers sequences designed for RT-qPCR were to Plin-2, forward: gtgacctcagtggctgtgac and reverse: tattggcgaccgcaatct and to Tbp, forward: tgggattgtaccacagctcca and reverse: ctcatgatgactgcagcaaacc and both primer pairs were purchased from Life Technologies. PCR conditions were as follows: 40 cycles of denaturation at 94°C for 20 s, annealing and extension at 58°C for 20 s and a final extension step at 72°C for 30 s. Ct values of Plin-2 mRNA obtained from 3 different experiments were normalized according to the 2 -ΔΔCt method, using TATAA-box binding protein (Tbp) as a reference gene. At the end of the amplification phase, a melting curve analysis was carried out on the product formed and agarose gel electrophoresis was carried out to confirm product size. The relative level of Plin-2 is expressed as the relative change in gene expression.

Triacylglycerol and cholesterol measurement
After separation by TLC, the spots corresponding to TAG and cholesterol were scrapped off the silica and eluted. Total TAG and cholesterol content was measured in aliquots of the lipid extracts corresponding to 10 μg lipid P, using commercial kits (TG color GPO/PAP AA and Colestat enzimático) from Wiener laboratory. Aliquots were dried under N 2 gas and resuspended in 100 μl of isopropyl alcohol and the manufacturer's instructions were subsequently followed.

Statistical analysis
Data represent the mean value ± SD of at least three independent experiments, each experiment was performed in triplicate. Statistical significance was determined by either Student's ttest or two-way ANOVA followed by a Tukey's test. p-values lower than 0.05 were considered statistically significant. For cytochemical studies, 10 fields per sample were analyzed in each case. The Western Blots shown are a representative image of samples from three independent experiments.

Effect of Fe 2+ overload and Cu 2+ supplementation on dopaminergic neurons
Our laboratory has previously characterized the effects of Fe-induced oxidative stress on cerebral cortex synaptic endings, retina and hippocampal neurons [21,22,25]. We have also described the activation of several signaling pathways related to lipid messengers during metal induced oxidative injury in the above-mentioned experimental models [19][20][21]26,27]. In the present work, we have characterized the role of PL and TAG acylation and remodeling during metal-induced toxicity in dopaminergic neurons. To this end, rat mesencephalic dopaminergic neurons, N27 cells, were exposed to Cu 2+ (10, 50 μM), Fe 2+ overload (1 mM) and the combination of both metals for 24 h. Control conditions were also assessed, replacing metals by an equal volume of vehicle (ultrapure water). To determine the extent of metal-induced effects on dopaminergic neurons, cell morphology, generation of reactive oxygen species (ROS), cell viability and lipid peroxidation levels were analyzed. Exposure to Cu 2+ ions (10 and 50 μM) did not significantly alter cell morphology. However, Fe 2+ (1 mM) caused typical morphological alterations of oxidant cellular injury, such as cytoplasmic retraction, cell body rounding and decrease in the number of cells and neuronal projections. These changes in cellular morphology were not affected when cells were incubated with the combinations Fe-Cu assayed (Fig 1A).
A highly significant increase (300% higher than the controls) in cellular ROS (measured by using the probe DCDCDHF) was observed after the treatment with 1mM Fe 2+ or with Fe-Cu combinations (1 mM-10 μM; 1 mM-50 μM). However, the presence of Cu 2+ alone did not increase ROS generation (Fig 1B). Neurons incubated in the presence of iron showed a sustained increase in ROS levels ranging from 1 to 24 h of incubation. Lipid peroxidation levels, which were analyzed by TBARS assay, were significantly increased by the presence of Fe (1mM) and with the co-incubation of both metals (approximately 100%, higher than the control; Fig 2A). Co-incubation with Cu and Fe showed no increase of TBARS levels with respect to the Fe condition. As shown in Fig 2B, experimental conditions with increased TBARS formation showed an augmentation of cell membrane permeability measured by LDH leakage. Mitochondrial membrane potential (ΔC m ) was also measured in the above mentioned experimental conditions, using the probe Mitotracker CMX-ROS (Fig 2C). In the presence of iron, ΔC m /ROS ratio strongly decreased, thus demonstrating mitochondrial depolarization. This fact agree with a decreased mitochondrial function, measured as MTT reduction (Fig 2C). TBARS and ROS generation were not increased in the presence of lower Fe 2+ concentrations than 1 mM (data not shown). The fact that Fe 2+ concentrations ranging from 10 to 200 μM produced no neuronal damage and that earlier times of incubation (1-6h) with 1 mM Fe 2+ did not produce neuronal damage were the reasons why we exposed dopaminergic neurons to 1 Cells were treated as described in A and lipid peroxidation was studied according to "Materials and Methods". Results are expressed as a percentage of the control and represent mean ± SD (n = 8). (B) LDH leakage assay was assessed in cells after their exposure to the same conditions described in A. Results are expressed as a percentage of the control and represent mean ± SD (n = 10). (C) Mitochondrial characterization. Mitochondrial membrane potential (Δᴪ) was measured using the probe Mitotracker Red CMX-ROS and mitochondrial function was evaluated by MTT reduction assay. Cells were treated as described in A, mitochondria stained with Mitotracker were visualized by confocal microscopy and MTT reduction was assessed as specified in "Materials and Methods". Δψ/ROS ratio was calculated for all the experimental conditions. Results are expressed as a percentage of the control and represent mean ± SD (n = 10). Values with different letters (a, b, c, d) are significantly different (p<0.05), one-way ANOVA and Tukey's post hoc test.
doi:10.1371/journal.pone.0130726.g002 mM Fe 2+ for 24 hs. In line with this, similar concentrations were used by the study of α-syn toxicity in a cell model of PD [28].
Cell viability, assessed by MTT reduction assay, was diminished at 50 μM Cu, 1 mM Fe and the combinations of both metals assayed (Fig 2C).
Phospholipid and triacylglycerol acylation in pre-labeled dopaminergic neurons exposed to metal-induced injury Pre-labeled dopaminergic neurons with [ 3 H] oleic acid (OA, 18:1) or [ 3 H] arachidonic acid (AA, 20:4) showed a distinctive profile of FA incorporation into PL and TAG. During Fe-induced injury, OA incorporation was increased in PS, PI and PE (Fig 3A). OA incorporation into TAG from Fe-exposed neurons showed no changes with respect to control conditions. whereas its incorporation into MAG and DAG decreased significantly (Fig 3B). Iron-exposed neurons showed decreased levels of AA uptake into PS with respect to control (Fig 4A). In contrast, AA incorporation into TAG was increased whereas decreased free AA levels were observed under the same experimental conditions (Fig 4B). Cu-incubations did not affect OA incorporation but decreased AA incorporation into TAG (Figs 3B and 4B). However, Cu-incubated neurons showed decreased OA incorporation into PE and increased AA uptake into PS (Figs 3A and 4A).

TAG levels in dopaminergic neurons exposed to metal-induced injury
The increase observed in TAG-containing AA in Fe-exposed neurons was accompanied by an increase in mRNA levels of the LD protein Perilipin-2 (PLIN-2), determined by RT-qPCR ( Fig  5A). In the presence of Fe, Nile red positive cytoplasmic bodies, were also observed ( Fig 5B). This was coincident with increased TAG levels being 40% higher in Fe-exposed neurons with respect to controls (Fig 5C). Cu did not produce any increase neither in LD appearance nor in TAG levels. Cholesterol levels were not changed in any of the experimental conditions tested ( Fig 5D). For further characterizing the mechanisms involved in AA accumulation in TAG during Fe-induced injury, the expression of DGAT 1, lysophosphatidic acid acyltransferase θ (LPAAT θ) and fatty acid synthase (FAS) proteins were analyzed by Western Blot. Intriguingly, DGAT 1 levels were found to undergo no changes in Fe-exposed neurons whereas Cu-exposed neurons were observed to decrease the levels of this enzyme (Fig 6A). FAS protein levels were diminished in Fe-exposed neurons and augmented in Cu-exposed neurons, with respect to controls and LPAAT-θ levels underwent no changes in any of the experimental conditions analyzed (Fig 6B and 6C).

Role of lipid acylation/deacylation mechanisms in dopaminergic neurons exposed to metal-induced injury
Based on the results obtained in lipid acylation process during metal-overload, and in order to further investigate the far-reaching implication of lipid remodeling during metal-induced  Western blot analyses of the enzymes involved in lipid accumulation. Cells were exposed to Cu 2+ (10, 50 μM), Fe 2+ overload (1 mM), the combination of both metals or its vehicle for 24 h. Cell lysates were prepared for Western blot as described in "Material and Methods" and (A) DGAT 1 expression levels, (B) FAS expression levels and (C) LPAAT-θ expression levels were assessed. Western blot in each case is representative of three different experiments. Bands of proteins were quantified using scanning densitometry. Data in the graphs above each blot represent the ratio between cellular protein expression and Actin, expressed as a percentage of the corresponding control condition (mean ± SD of three different experiments). *p < 0.05 for each condition with respect to the control; one-way ANOVA and Tukey's post hoc test. neuronal injury, we next investigated the involvement of deacylation and acylation processes in ROS generation and membrane permeability. To this end, PLA 2 isoform-specific inhibitors and a general inhibitor of lipid acylation (CI-976) were added to the above-mentioned assays. Dopaminergic neurons were preincubated with 50 μM ATK (cPLA 2 inhibitor), 10 μM YM 26734 (sPLA 2 inhibitor) and 10 μM BEL (iPLA 2 inhibitor) and subsequently exposed to metal incubations. cPLA 2 and sPLA 2 inhibition did not alter the increase in ROS generation observed in the presence of Fe, Cu and the combination of both (Fig 7A and 7B). Neither membrane permeability (Fig 7C and 7D) nor mitochondrial function (data not shown) were affected by the inhibition of cPLA 2 and sPLA 2 . However, whereas iPLA 2 inhibition was found to strongly increase the generation of ROS in the presence of Fe-induced injury (Fig 8A), membrane permeability and mitochondrial function were slightly affected by iPLA 2 inhibition (Fig 8B and 8C).
The inhibition of acyltransferase activities did not alter ROS levels under any of the conditions assayed (data not shown). Nevertheless, lipid peroxide generation and membrane   Role of lipid acylation mechanisms in dopaminergic neurons exposed to metal-induced injury.
(A) Determination of cellular oxidant levels. Cells were treated with Cu 2+ (50 μM), Fe 2+ overload (1 mM), the combination of both metals or its vehicle for 24 hs in the presence or absence of the acylation inhibitor (7.5 μM CI-976) and subsequently incubated in the presence of 10 μM DCDCDHF as described in "Materials and Methods". Cellular oxidant levels were quantified by spectrofluorimetry. Results are expressed as a percentage of the control and represent mean ± SD (n = 3). (B) LDH leakage assay was assessed in cells treated as described in A. Results are expressed as a percentage of the control and represent mean ± SD (n = 3-4). (C) MTT reduction assay. Cells were treated as described in A and cell viability was assessed as permeability were observed to be increased by metal-induced neuronal injury in the presence of CI-976 with respect to controls (Fig 9A and 9B). This was in accordance with a significant loss of mitochondrial function (Fig 9C).

Discussion
The mechanisms that operate during dopaminergic neuron injury are a major concern for understanding the ethiopathogenesis of PD. Excessive iron content and the resulting oxidative stress in the substancia nigra of PD patients are considered triggering factors in this pathology [2,29,30]. A less clear issue is the state of Cu metabolism [1] although recent findings have demonstrated that α-syn pathology enhances vulnerability to cell death if Cu-dependent protective mechanisms are impaired [9]. In this respect, an unexplored field is how lipid acylation/ deacylation cycles operate during dopaminergic neuron degeneration and death. Therefore, for piecing together the above-mentioned events, lipid acylation mechanisms and their role in oxidative dopaminergic injury triggered by Fe-overload were studied. Fe-exposed N27 neurons were observed to display increased levels of lipid peroxides and ROS and also augmented membrane permeability. As a result, Fe-overloaded neurons exhibited slightly diminished mitochondrial viability. Our results as well as previous evidence centering on the role of Fe in dopaminergic degeneration support the viability of the experimental model used in the present study [10,31]. In line with this, advances in non-dopaminergic treatments for PD are focused in the use of Fe chelators as therapeutic strategy [32]. The condition of Cu supplementation induced no increase in oxidative stress markers although at the highest Cu concentration tested, a reduced mitochondrial function was observed. These findings suggest that Fe-overload triggers dopaminergic neuron injury by oxidative stress whereas Cu-mediated dopaminergic neurotoxicity does not involve oxidative injury, at least at the concentrations assayed in the present study.
Neuronal phospholipid acylation and deacylation mechanisms are crucial for the maintenance of axonal and synaptic integrity [33]. Phospholipid membrane deacylation is catalyzed by different isoforms of PLA 2 that generate AA, docosahexaenoic acid and lysophospholipids. All these bioactive lipids have a variety of physiological effects and also behave as substrates for the synthesis of other potent lipid mediators, such as eicosanoids, and the lipid peroxides, 4-hydroxynonenal and malondialdehyde. Rapid clearance of peroxidized lipids is a necessary mechanism for preserving plasma membrane integrity during oxidative stress processes [34][35][36]. Our results show a different lipid acylation profile during metal-induced injury. The increase of OA acylation in PS, PI and PE and the decrease of AA incorporation into PS demonstrate that dopaminergic neurons respond to Fe-induced oxidative injury by switching PL fatty acid composition to a more saturated profile. The increased preference of OA as an acyl donor for PL acylation could be indicative of the fact that dopaminergic neurons decreased PUFA levels in their membranes in order to be less prone to peroxidation. Our results agree with previous findings according to which lipids containing arachidonate are more susceptible to oxidative degradation [37,38]. These findings are, in turn, in agreement with the membrane pacemaker hypothesis on ageing, which proposes that animals whose cellular membranes contain high amounts of PUFA have shorter life spans because of their susceptibility to oxidative stress [39]. In Cu-exposed membranes, PL acylation was opposite to the profile observed in Feexposed neurons as it increased the incorporation of AA into PS -whereas it decreased OA described in "Materials and Methods". Results are expressed as a percentage of the control and represent mean ± SD (n = 3). *Significantly different compared to the respective control group (p<0.05, one-way ANOVA test followed by Tukey post hoc test) doi:10.1371/journal.pone.0130726.g009 labeling in PE. The absence of lipid peroxides in Cu-exposed neurons justify the incorporation of AA into PS, one of the main reservoirs for this FA in the brain [40].
Differential roles for specific-PLA 2 isoforms have been described in several neurodegeneration processes. It has, in fact, been shown that recessively-inherited deficiency in the catalytic domain of iPLA 2 causes neuroaxonal dystrophy [41]. A role for cPLA 2 has particularly been observed to be involved in the generation of ROS through AA release in cerebral ischemia events and iron-induced retinal degeneration [42]. In addition, PL fatty acid composition and PLA 2 activity have been shown to be related with the progression of AD [43]. In our study, oxidative injury markers, such as ROS content, lipid peroxide levels and membrane permeability, showed to be increased by the inhibition of iPLA 2 (BEL). Furthermore, whereas PLA 2 s was found to deacylate PL in sn-2 position, AT activities were found to be necessary to reacilate this position as well as to maintain normal phospholipid composition [13]. Previous research also demonstrated that the loss of lysophosphatidylcholine acyltranferase that renders increased levels of lysophosphatodylcholine generates membrane disruption and Ca 2+ influx [44]. Loss of lysophosphatidylcholine acyltransferase 1 was found to induce photoreceptor degeneration in rd11 mice [45]. The inhibition of PL acylation processes by CI-976 was observed to yield the same effect as that of BEL on Fe-exposed neurons. These observations highlight the crucial role of lipid acylation/ deacylation processes in the response of dopaminergic neurons to oxidative injury.
The most striking finding of the present study seems to be the esterification of AA into TAG in Fe-overloaded neurons. The increase in TAG levels and the appearance of LD suggest that dopaminergic neurons respond to oxidative injury by the storage of peroxidable fatty acids in the TAG pool, thus avoiding the permanence of PUFA in PL fraction. The decreased expression of FAS also strengthens the hypothesis that under oxidative stress conditions dopaminergic neurons avoid the generation of new synthetized FA and exacerbate the mechanisms for preserving their FA from peroxidation. This strategy seemed to be successful on account of the fact that Feoverloaded neurons were found to avoid a significant death. Regarding PD experimental models, rotenone, a mitochondrial toxicant, also triggers TAG accumulation in C2C12 cells [46].
Taken together, findings from the present study provide evidence about the participation of lipid acylation mechanisms against Fe-induced oxidative stress in N27 cells and postulate that dopaminergic neurons strategically preserve AA pool in TAG in response to oxidative injury (Fig 10). Further studies in this direction will provide new knowledge on PD pathogenesis which could pave the way towards new preventive or therapeutic strategies.