ABSTRACT
Lipid peroxidation (LPO) is reported to be involved in the pathogenesis of several oxidative diseases, and several therapeutic approaches using antioxidants have been proposed. LPO is thought to progress via a complicated series of multistep reactions suggesting that the activity of each antioxidant may be different, and depends on the reacting molecules. Hence, in this study, we evaluated the inhibitory mechanisms of several antioxidants toward arachidonic acid (AA) peroxidation induced by the azo initiator 2,2’ azobis(2 amidinopropane) dihydrochloride (AAPH) or a lipid hydroperoxide, hydroperoxyoctadecadienoic acid (HpODE)/hemin. Edaravone, ferrostain1, TEMPO and trolox effectively inhibited the production of malondialdehyde (MDA) and several oxidized AAs generated in the AAPH induced LPO because of their scavenging ability toward lipid peroxyl radicals. In contrast, ebselen and ferrostatin1 showed strong antioxidative activity in the HpODE/hemin induced peroxidation. Under this condition, ebselen and ferrostatin1 were thought to reduce HpODE and its derived alkoxyl radicals to the corresponding lipid alcohols. In conclusion, we found that each antioxidant had different antioxidative activities that prevented the progression of LPO. We expect that these findings will contribute to the design of novel therapeutic strategies using an appropriate antioxidant targeted to each step of the development of oxidative stress diseases.
KEYWORDS: lipid peroxidation, antioxidant, lipid radical, lipid hydroperoxide.
INTRODUCTION
Lipid peroxidation (LPO) is recognized as an important contributor to the pathological events of oxidative diseases [1]. A polyunsaturated fatty acid (PUFA), which contains several carbon carbon double bonds, is easily oxidized by reactive oxygen species (ROS), causing chain peroxidation [2]. The primary step of this chain reaction is hydrogen atom abstraction from the bis allylic methylene groups of the PUFA and addition of molecular oxygen, resulting in the formation of lipid peroxyl radicals [3]. Then, these radicals promote the excessive generation of lipid hydroperoxides or electrophiles, such as malondialdehyde (MDA) and 4 hydroxynonenal (4 HNE) [4]. Additionally, lipid hydroperoxides easily decompose to highly reactive lipid alkoxyl radicals via the Haber Weiss reaction in the presence of metal ions, and further accelerate the propagation step in the chain reaction [5]. Each oxidized lipid and electrophile contributes to the onset or progression of several oxidative diseases. For example, accumulation of lipid hydroperoxide leads to lipid peroxidation dependent cell death or ferroptosis [6,7]. Lipid derived electrophiles cause the chemical modification of proteins [8,9] and induce cardiovascular and neurodegenerative diseases [10,11].
Hence, antioxidants that specifically inhibit lipid peroxidation may be effective therapeutic agents [12]. To date, there have been many studies on the protective effects of antioxidants toward several diseases. “ Tocopherol, which is a lipophilic antioxidant, has been reported to improve the pathological condition of non alcoholic steatohepatitis [13]. Edaravone has been clinically applied for the treatment of acute cerebral infarction and amyotrophic lateral sclerosis [14,15]. Additionally, ferrostatin1 and liproxstatin1, which are potential inhibitors of ferroptosis, reduced the pathogenesis in various disease model animals, including those for acute renal failure, myocardial disease and ischemic stroke [1618]. Thus, LPO is a promising therapeutic target in oxidative stress diseases, and is the subject of novel antioxidant therapeutic strategies.
However, it is unclear as to which peroxidation step each antioxidant targets. As described above, LPO involves a complex multistep reaction and a number of oxidized lipids and oxidative factors are involved in the reaction processes [2,5]. Thus, the types and amounts of oxidized lipids vary greatly depending on the reaction conditions [19]. Therefore, it is also assumed that antioxidants show different inhibition effects toward LPO depending on where in the multistep reaction process they act. However, commonly used antioxidant assays [20], such as oxygen radical absorbance capacity and ferric reducing antioxidant power (FRAP) [21] [22], only enable a limited evaluation of the capabilities of antioxidants, such as their ROS scavenging and reducing capacities. Accordingly, if each reaction target of an antioxidant could be comprehensively investigated and clarified, it will contribute to the design of appropriate antioxidants for each LPO condition.
In this study, we investigated the inhibitory mechanisms of several antioxidants toward LPO. For this purpose, we comprehensively analyzed the oxidized lipid species generated in the presence of antioxidants using the thiobarbituric acid reactive substances (TBARS) assay and LC/MS/MS measurements. As a result, the employed antioxidants showed different removal activities toward oxidized lipids, such as lipid radicals and lipid hydroperoxides.
MATERIALS AND METHODS
Reagents
Arachidonic acid (AA, from porcine liver, ≥99%), linoleic acid (LA, ≥99%) and soybean lipoxidase from Glycine max (s15LOX) were purchased from Sigma Aldrich (St. Louis, MO), 2,2’ azobis (2 amidinopropane) dihydrochloride (AAPH) was obtained from Wako Pure Chemical Industries, Ltd. (Osaka, Japan), and hemin was from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). The antioxidants edaravone and ebselen were purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan), trolox was from Merck Millipore (Burlington, MA) and ferrostatin1 was from Focus Biomolecules (Plymouth Meeting, PA).
4 (4 Nitrobenzo[1,2,5] oxadiazol 7 ylamino) 6 pentyl 2,2,6 Immune clusters trimethylpiperidine1 oxyl (NBD Pen) was synthesized according to a previously reported method [23]. For high performance liquid chromatography (HPLC), acetonitrile (LCMS grade, ≥99.9%) was purchased from Nacalai Tesque (Kyoto, Japan), and ammonium acetate (Wako 1st grade) was from Wako Pure Chemical Industries, Ltd.
Preparation of hydroperoxyoctadecadienoic acid
LA (500 µM, 0.5% ethanol) and s15LOX (10 µg/mL) were mixed in PBS (pH 7.4) and incubated 15 min at 37 °C. After the incubation, the reaction mixture was extracted by the Bligh and Dyer method [24]. Functionally graded bio-composite The organic layer was dried with nitrogen gas and dissolved in 100 µL of ethanol. The solution of hydroperoxyoctadecadienoic acid (HpODE) was stored at 80 °C until use.
AA peroxidation systems
AA (500 µM) and each antioxidant (ebselen, edaravone, ferrostatin1, TEMPO and trolox; 10 or 100 µM) were mixed in PBS containing 0.5% ethanol (pH 7.4). Radical initiators (system1; 50 mM AAPH, system 2; 50 µM HpODE and 5.0 µM hemin) were added to the reaction mixture. After incubation for 15 min at 37 °C, the reaction mixture was extracted by the Bligh and Dyer method
[24]. The organic layer was dried with nitrogen gas and dissolved in 200 µL of methanol. The solution was stored at 80 °C until use.
Thiobarbituric acid reactive substances assay
A mixture of 10% sodium dodecyl sulfate, 20% acetic acid and 1.3% thiobarbituric acid were added to oxidized AA solutions and incubated at 60 °C for 1 h. Samples were cooled and an n butanol/ pyridine (15:1) solution was added. After centrifugation at 720 g for 10 min, samples were plated onto a 96 well microplate and the fluorescence was measured (λex 512 nm, λem 553 nm) using an EnSpire Multimode Plate Reader (PerkinElmer Japan Co., Ltd., Kanagawa, Japan).
Measurement of oxidized arachidonic acids by liquid chromatography tandem mass spectrometry
Liquid chromatography–tandem mass spectrometry (LC/MS/MS) was carried out using a LCMS 8060 (Shimadzu Co., Kyoto, Japan). The LC/MS/MS analyses were carried out using LabSolutions version 5.80 (Shimadzu Co.). The mass spectrometer was equipped with an electron spray ionization source. The extracted solution was measured by LC/MS/MS with multiple reaction monitoring (MRM) and a LC/MS/MS Method Package for Lipid Mediators (Shimadzu Co.).
Detection of lipid derived radicals
AA (500 μM) and antioxidants at the indicated concentration (50, 100, 200 and 400 μM) were mixed in PBS (pH 7.4). A radical initiator (system1; 50 mM AAPH, system 2; 50 µM HpODE and 5.0 µM hemin) was added to the reaction mixture and incubated at 37 °C for 60 min. Then NBD Pen (10 μM) was added to the mixture, and further incubated at 37 °C for 15 min. After the reaction, the fluorescence intensity (λex 470 nm, λem 530 nm) of the reaction solution was measured using an Enspire Multimode Plate Reader.
Statistical analysis
Statistical analyses were carried out using StatView 5.0 (SAS Institute Inc.), and the data were analyzed by the Tukey Kramer test. The results are expressed as mean ± standard deviation, and P<0.05 was considered statistically significant. RESULTS In this study, LPO was induced using two radical initiators (Fig. 1a). In system1, the azo initiator AAPH was used, which pyrolytically decomposes to radical species and promotes the production of peroxyl radicals [25]. In system 2, a combination of HpODE and hemin was applied, which mimics LPO induced by the decomposition of lipid hydroperoxide [5]. The chemical structures of the antioxidants used in this study, ebselen, edaravone, ferrostatin1, TEMPO and trolox, are shown in Fig. 1b. Lipid peroxide inhibitory effects of antioxidants in an AAPH induced lipid peroxidation system. First, we examined whether antioxidants inhibit MDA production using system1 (Fig. 2). After adding AAPH to the AA solution, most of the tested antioxidants inhibited MDA production except for ebselen. Next, we investigated the effect of the antioxidants on the production of AA derived oxidants, including hydroperoxyeicosatetraenoic acids (HpETEs; 15 HpETE, 12 HpETE and 5 HpETE; Fig. 3a c), hydroxyeicosatetraenoic acid (HETEs; 15 HETE, 12 HETE and 5 HETE; Fig. 3d f) and prostaglandins (PGs; 8 iso PGE2, PGE2 and PGD2; Fig. 3g i) using LC/MS/MS. The formation of oxidized AAs was enhanced by addition of AAPH, and decreased by addition of all the tested antioxidants except ebselen. Interestingly, ebselen inhibited HpETEs formation, but not the production of HETEs and PGs. In addition, ebselen enhanced the formation of 15 HETE, 5 HETE and 8 iso PGE2. The formation of MDA and PGs are reported to occur via the cyclization of lipid peroxyl radicals through a reaction process that does not involve Lirametostat supplier the formation of lipid hydroperoxides [26,27] (Fig. 4a). Therefore, we hypothesized that the small inhibitory effect observed for ebselen toward MDA and PGs formation is attributed to its low reactivity toward lipid derived radicals. To investigate whether ebselen could react with the AA derived radicals generated in system1, we used NBD Pen, which is a probe developed in house for the detection of lipid radicals [23]. As expected, ebselen did not inhibit the enhanced fluorescence intensity of NBD Pen, but TEMPO suppressed it in a concentration dependent manner (Fig. 4b). These results clearly suggest that most of the antioxidants used in this study inhibited the generation of AA derived radicals, while ebselen only suppressed the accumulation of lipid hydroperoxide but not AA derived radicals and their secondary products, such as MDA and PGs.
LPO inhibitory effects of antioxidants in an HpODE/hemin induced lipid peroxidation system.
Next, we examined the inhibitory effect of antioxidants on MDA production in a HpODE/hemin induced peroxidation system (system 2) (Fig. 5). In system 2, enhanced MDA production was inhibited by all tested antioxidants, even ebselen although it did not suppress the MDA level in system1. Next, the amounts of oxidized lipids derived from AA and 13 HpODE were measured (Fig. 6). 13 HpODE was completely consumed by adding hemin (Fig. 6a). AA derived oxidized lipids increased in the presence of the HpODE/hemin peroxidation system and decreased in the presence of antioxidants (Fig. 6b j). Interestingly, ebselen and ferrostatin1 increased the level of 13 hydroxyoctadecadienoic acid (13 HODE) (Fig. 6k). Ebselen is known to reduce lipid hydroperoxides to lipid alcohols [28]. Hence, ferrostatin1 was also assumed to react with 13 HpODE derived active species to prevent the propagation of LPO.
Reducing capacities of ferrostain 1 toward HpODE derived alkoxyl radicals
In the presence of metal ions, lipid hydroperoxides immediately decompose to lipid derived alkoxyl radicals via the Haber Weiss reaction. Lipid alkoxyl radicals are usually reduced to lipid alcohols by a one electron reduction. In comparison, lipid alkoxyl radicals may also be converted to epoxyaryl radicals via a radical rearrangement [29], which are highly reactive carbon centered radicals that act to accelerate LPO as chain carriers (Fig. 7a) [30]. Hence, we investigated whether ferrostatin1 eliminates lipid alkoxyl radicals by promoting the one electron reduction to produce 13 HODE. As a result, 13 HODE increased with the addition of ferrostatin1 in a concentration dependent manner in the presence of hemin, but not in the absence of hemin (Fig. 7b). Furthermore, under this condition, lipid derived radical generation was suppressed by ferrostatin1 (Fig. 7c). These results suggest that ferrostatin1 reduced 13 HpODE derived radicals, and then effectively inhibited the subsequent LPO process.
DISCUSSION
As shown in this study, the antioxidants exhibited different inhibitory effects toward LPO and regulated the production of oxidized lipids depending on their mechanism of action. Scheme 1 demonstrates the reaction points of the antioxidants using in this study. Edaravone, TEMPO and trolox inhibited the generation of a wide variety of oxidized lipids using both the system1 and system 2 peroxidation mixtures (Fig. 2, 3, 5, 6). The inhibition was attributed to the scavenging effect of the above antioxidants toward the lipid derived radicals, which are the upstream active species in LPO. Moreover, TEMPO showed much higher antioxidative activity than edaravone or trolox. TEMPO has been reported to catalytically inactivate lipid derived radicals via the electron transfer or radical radical coupling reaction [31,32]. Indeed, TEMPO had a higher anti oxidative effect even at a lower concentration (10 µM) (Fig. 2, 5). In contrast, chromanol derivatives, such as trolox, react with peroxyl radicals to produce unstable semiquinone radical intermediates resulting in loss of their antioxidative capacities [33]. Edaravone forms an enolate anion in aqueous solution, and is inactivated after reaction with the radical species [34]. For these reasons, despite the fact that the above antioxidants react with the same reaction target (lipid derived radicals), TEMPO in particular showed strong antioxidative effects because of its catalytic scavenging capacity against radical species.
Although ebselen completely inhibited the production of lipid hydroperoxides (Fig. 3a c), it did not suppress the generation of MDA, PGs and lipid derived radicals (Fig. 2, 3g i, 4b). Organic selenium is known to reduce hydroperoxides to their corresponding alcohols [28]. Indeed ebselen, a synthetic organoselenium compound, exhibited a large cytoprotective effect against ferroptosis characterized by the accumulation of lipid hydroperoxide [6]. In contrast, there are few studies that show ebselen has radical scavenging ability. In agreement with these observations, ebselen sufficiently inhibited LPO induced by lipid hydroperoxides (system 2), but not by a radical reaction (system1). On the other hand, ebselen enhanced the production of some HETEs and PGs, including 15 HETE, 5 HETE and 8 iso PGE2 in system1 (Fig. 3). These oxidized AAs were thought to generate through the reduction of hydroperoxides (Scheme S1), and ebselen can promote this reaction [35]. Therefore, ebselen could increase the formation of 15 HETE, 5 HETE and 8 iso PGE2 in AAPH induced LPO system.
The ferroptosis inhibitor ferrostatin1 showed excellent antioxidative effects in both systems (Fig. 2, 5). It is thought that ferrostatin1 exhibits scavenging activity against not only lipid peroxyl radicals, but also lipid alkoxyl radicals. In fact, ferroptosis requires both the accumulation of lipid hydroperoxide and iron [6,7], which may indicate that a large amount of lipid alkoxyl radicals are generated from this mechanism of cell death. These observations are consistent with the result described herein that ferrostatin1 showed a high scavenging ability toward lipid alkoxyl radicals (Fig. 7). In addition, other anitoxidants, including edaravone, TEMPO, and trolox, also enhanced the production of 13 HODE from HpODEs in the presence of hemin, but these capabilities were weaker than that of ferrostatin1 (Fig. S1).
Furthermore, the strong LPO inhibitory effect of ferrostatin1 was observed even at a lower concentration (10 µM) in both peroxidation systems (Fig. 2, 5). In recent years, aniline type antioxidants including ferrostatin1 have been reported to exhibit excellent protective effects against oxidative stress because of their catalytic antioxidative activities [36,37]. Additionally, the high hydrophobicity of ferrostatin1 allows its successful accumulation in lipid vesicles [38]. Taken together, these properties of ferrostatin1 support why it efficiently inhibits LPO in both systems.
Since all the experiments in this study were performed in test tubes, it is unclear whether these antioxidants show similar activities in vivo. Therefore, future work should include investigation of the antioxidative activity and the reaction target of each antioxidant in biological systems using disease model animals.
To date, many researchers have worked on treatments using antioxidants, but their effects have been controversial [39]. In this study, we demonstrated that each tested antioxidant has a different reaction target in LPO, including lipid peroxyl radical, lipid hydroperoxide and lipid alkoxyl radical. These observations suggest that antioxidants may not be broadly applicable to conventional treatment methods because of a lack of understanding of their reaction points in LPO. Based on these findings, we expect that novel antioxidant therapeutic strategies must be developed with knowledge of the specific LPO reaction targets responsible for the pathology of each oxidative stress disease.