Elucidation of Olive Oil Oxidation Mechanisms by Analysis of Triacylglycerol Hydroperoxide Isomers Using LC-MS/MS

Despite the importance of the insight about the oxidation mechanisms (i.e., radical and singlet oxygen (1O2) oxidation) in extra virgin olive oil (EVOO), the elucidation has been difficult due to its various triacylglycerol molecular species and complex matrix. This study tried to evaluate the mechanisms responsible for EVOO oxidation in our daily use by quantitative determination of triacylglycerol hydroperoxide (TGOOH) isomers using LC-MS/MS. The standards of dioleoyl-(hydroperoxy octadecadienoyl)-triacylglycerol and dioleoyl-(hydroperoxy octadecamonoenoyl)-triacylglycerol, which are the predominant TGOOHs contained in EVOO, were prepared. Subsequently, fresh, thermal-, and photo-oxidized EVOO were analyzed. The obtained results mostly agreed with the previously reported characteristics of the radical and 1O2 oxidation of linoleic acid and oleic acid. This suggests that the methods described in this paper should be valuable in understanding how different factors that determine the quality of EVOO (e.g., olive species, cultivation area, cultivation timing, and extraction methods) contribute to its oxidative stability.

Primary oxidation of lipids affords lipid hydroperoxide (LOOH) isomers, whose structures (i.e., hydroperoxyl group binding positions) depend on oxidation mechanisms (i.e., radical and 1 O 2 oxidation; Figure 1) [15,17]. In other words, triacylglycerol (TG) oxidation mechanisms can be identified by characterizing TG hydroperoxide (TGOOH) isomers. However, analyzing TGOOH isomers has been a great challenge, even using the latest analytical techniques and instruments [23][24][25][26][27]. To overcome such issues, we recently developed methods to analyze the positional isomers of various LOOHs by utilizing sodium ions during electrospray ionization (ESI)-LC-MS/MS [28,29]. Using this method, we analyzed the major TGOOH contained in canola oil (TG 18:1_18:1_18:2;OOH) and identified that canola oil was predominantly oxidized by 1 O 2 oxidation during storage [29]. The study led us to believe that our LC-MS/MS method can further be applied to elucidate the oxidation mechanisms of EVOO that possess a more complex matrix than canola oil. . Isomeric structure of TGOOH depends on oxidation mechanisms (radical and 1 O 2 oxidation). The shorthand notation of lipids was in accordance with LIPID MAPS [30].

Materials and Methods
The shorthand notations of TGOOH isomers (and other lipids) described in this study follow the LIPID MAPS nomenclature (

Oxidation of EVOO
Fresh EVOO (400 mL) was thermally oxidized (radical oxidation) in an amber 500 mL glass beaker under gentle stirring. The beaker was heated in an oil bath kept at 150 • C in the dark.
Oxidized EVOO samples were collected at 20 min intervals until 240 min (n = 3). Portions of the collected samples were diluted 10,000-fold in hexane and analyzed with LC-MS/MS (10 µL).
As described above, determining hydroperoxyl group positions is pivotal to evaluate the mechanisms responsible for EVOO oxidation. Thus, previous studies analyzed TGOOH hydroperoxyl group positions after derivatization reactions (e.g., reduction, trimethylsilylation, and methyl esterification). However, because the hydroperoxyl group is relatively unstable, artifacts can be formed during derivatization. Therefore, a direct analysis should be favored over derivatization methods. Meanwhile, most of the previous studies that directly analyzed TGOOH depended solely on molecular weight (i.e., intact TG molecular weight + 32 Da) and, hence, their isomers were not analyzed [23][24][25][26][27]. Under these circumstances, we discovered that the collision-induced dissociation (CID) of the sodium adducts of LOOH provide hydroperoxyl group position-specific product ions based on α-cleavage [28,29]. Using this method, we analyzed hydroperoxyl group positions for the main TGOOH in canola oil (TG 18:1_18:1_18:2;OOH isomers) and found that FA 18:2(9Z,12Z) in canola oil was oxidized predominantly by 1 O 2 oxidation during storage [29]. Therefore, in this study, we aimed to apply the above method to determine the oxidation mechanisms of FA 18:2(9Z,12Z) in EVOO. Additionally, to obtain further insight into the oxidation of EVOO (i.e., oxidation mechanisms of FA 18:1(9Z)), TG 18:1_18:1_18:1;OOH isomers in EVOO were also analyzed.
In summary, TG 18:1_18:1_18:2;OOH and TG 18:1_18:1_18:1;OOH isomers were directly analyzed from fresh, thermally oxidized, and photo-oxidized EVOO. To the best of our knowledge, this is the first study reporting analysis of TG 18:1_18:1_18:1;OOH isomers. The obtained results, in most cases, agreed with the previously reported characteristics of the radical and 1 O 2 oxidation of FA 18:2(9Z,12Z) and FA 18:1(9Z). Hence, the LC-MS/MS methods reported herein were advantageous in determining oxidation mechanisms of EVOO that possess a complex matrix (i.e., the presence of chlorophyll, tocopherol, various polyphenols, and carotenoids). The methods described in this paper should also be valuable in understanding how different factors that determine the quality of EVOO (e.g., olive species, cultivation area, cultivation timing, and extraction methods) contribute to its oxidative stability.