• arachidonic acid metabolites
• micro-extraction by packed sorbent
• plasma
• stroke disease
• UHPLC-ESI-MS/MS analysis

Oxidative stress is defined as an imbalance between the production of reactive oxygen species (ROS) and antioxidant defenses. Oxidative stress is involved in several chronic diseases such as cardiovascular diseases, cancer, diabetes, neurodegenerative diseases, stroke condition and chronic inflammatory diseases.
ROS and free radicals can induce lipid peroxidation of polyunsaturated fatty acids (PUFA), e.g. arachidonic acid, causing an alteration of the biological properties of phospholipidic membranes, such as increase of membrane’ permeability and inactivation of enzymes. In specific conditions, arachidonic acid can undergo alteration either by enzymatic peroxidation to form eicosanoids (i.e. prostaglandins, leukotrienes, and thromboxanes) via cyclooxygenase, lipoxygenase or cytochrome P-450 metabolic pathway, or by non-enzymatic peroxidation to form isoprostanes, malondialdehyde, 4-hydroxynonenal and other products.
Biomarkers such as prostaglandins, isoprostanes and thromboxanes have been suggested to assess levels of oxidative stress and lipid peroxidation both in vitro and in vivo. The evaluation of oxidative damage is usually performed in serum, plasma and urine samples. The complexity of such biological matrices, deriving from the presence of interfering compounds (e.g. proteins and lipoproteins) and the tendency of auto-oxidation of lipids, makes the determination of oxylipins in such fluids difficult.
The aim of this thesis was to develop and validate an analytical procedure for the simultaneous determination of different arachidonic acid metabolites in plasma. The method involved the use of an innovative technique based on the automated micro-extraction by packed sorbent (MEPS) coupled to an ultra-high performance liquid chromatography-electrospray ionization tandem mass spectrometry (UHPLC-ESI-MS/MS) analysis. Fifteen compounds were selected as potential biomarkers of oxidative stress and determined: 8-iso prostaglandin F2α (8-isoPGF2α), prostaglandin E2 (PGE2), 8-iso prostaglandin E2 (8-isoPGE2), Prostaglandin D2 (PGD2), thromboxane B2 (TX-B2), leukotriene B4 (LEU-B4), 15-deoxy-Δ12,14-Prostaglandin J2 (15-deoxy-PJ2), 20-hydroxy-5Z,8Z,11Z,14Z-eicosatetraenoic acid (20-HETE), (±) 15-hydroxy-5Z,8Z,11Z,13E-eicosatetraenoic acid (15-HETE), (±) 12-hydroxy-5Z,8Z,10Z,14Z-eicosatetraenoic acid (12-HETE), (±) 5-hydroxy-6E,8Z,11Z,14Z-eicosatetraenoic acid (5-HETE), (±)14(15)-epoxy-5Z,8Z,11Z-eicosatrienoic acid (±14,15-EET), (±)11(12)-epoxy-5Z,8Z,14Z-eicosatrienoic acid (11,12-EET), (±)8(9)-epoxy-5Z,11Z,14Z-eicosatrienoic acid (8,9 EET), (±) 13-hydroxy-9Z,11E-octadecadienoic acid (13-HODE).
After the optimization of instrumental parameters, stability assays performed at 4 °C for one month and at room temperature for 24 h showed that all analytes were stable in water solution for at least 1 month at 4 °C and for 24 h at room temperature except PGD2 and isoprostanes that were stable respectively for 24 h and 14 days at 4 °C and PGD2 that resulted stable until 6 hours at room temperature. Protein precipitation, involving the use of acetonitrile and saline solution of copper sulphate pentahydrate (10% w/v) and sodium tungstate dehydrate (12% w/v), and ultrafiltration (with a 3kDa cut-off) approaches were tested in order to find a reliable and effective sample clean-up method. The use of copper sulphate (sol.A) and sodium tungstate solutions (sol.B) (plasma:solA:solB, 2:1:1 v/v/v) and the subsequent addition of acetonitrile (plasma:acetonitrile, 1:1 v/v) produced clearer extracts without significantly impacts on the recovery of analytes. The extract was subjected to MEPS procedure to further remove interferences and to pre-concentrate the target analytes. Regardless of the analytes concentration, the procedure showed an almost quantitative recovery for most compounds (8-iso-PGF2α, 8-iso-PGE2 90-120 %; PGE2 85-100%; TX-B2 85-100%; Leu 80-105%; 13-HODE 98-130%; 15-deoxy-PJ2 90-105%; 12-HETE 80-108%; 14,15-EET, 11,12-EET, 8,9-EET 60-104%; 20-HETE, 15-HETE, 5-HETE 60-110%), and a low intra- and inter-day variability (relative standard deviation close to 15%). Matrix effect was evaluated by comparing the slopes between calibration curves obtained with working solution and in matrix (at a confidence level of 95%). Linear calibration ranges (R2 >0.995) over three orders of magnitude were observed for all the analytes. The variability induced by the sample preparation, the extraction and instrumentation was minimized by the addition of a set of deuterated internal standard (PGE2-d4, 20-HETE-d6, 14(15)-EET-d11, 8-iso-PGE2-d4, 8-iso-PGF2-d4), thus improving the analytical performances.
Future application will be devoted to assess the levels of these arachidonic acid metabolites in patients affected by ischemic and hemorrhagic stroke and healthy patients in order to evaluate potential relationship between the selected compounds and type of stroke diseases and to evaluate the possibility to use such analytes to predict the onset of such diseases.