Impact of Oxidized Phospholipids on Outcomes from Cerebral Ischemia and Reperfusion Injury

Abstract

Background/Objectives: The mechanisms leading to oxidative stress and cellular dysfunction during stroke are not well understood. Methods: We tested if transient cerebral artery occlusion (MCAo) in mice results in the generation of oxidized phospholipids (OxPLs) that contribute to neuronal cell death and glial activation. Results: Both in vitro and in vivo cerebral ischemia and reperfusion injury (IRI) resulted in the elevation of specific OxPLs. Neuronal cell death was determined in the presence of OxPLs and the natural OxPL E06 antibody (antioxidized phospholipid antibody) protected the cells from the toxic effects. IRI in mice gave rise to increased immunoreactivity of OxPLs in the brain. E06 reduced inflammatory markers in the brain following IRI, including iba-1, GFAP and inflammatory cytokines. In addition, OxPLs gave rise to M1 and Mox microglial phenotypes which was reversed in the presence of E06 and elicited a more M2 phenotype. Nrf2-deficient mice showed increased infarct volumes and microglia from Nrf2^−/− mice showed a reduction in Mox gene expression, and E06 protects both mice and cells from the Nrf2-deficit. Finally, AB1-2 Ab which recognizes the E06 Ab, ameliorates the impact of E06 Ab on infarct volume in the mouse model. Conclusions: Taken together, the data indicate that OxPLs play an important role in inflammation and neuronal cell loss in cerebral IRI and inactivation of OxPLs may provide novel targets for potential drug targets in the treatment of stroke.

1. Introduction

Stroke is a common cause of disability and a significant contributor to death both in the US and the world as a whole [1,2]. Even though the incidence of stroke has decreased somewhat in the high-income countries, low-to-middle income countries still have significantly higher fatality rates [3,4]. Stroke care and secondary stroke prevention measures in low-to-middle income countries are not at the same level as high-income countries [5,6]. Even still, the stroke rates in the southern US are high due to genetics, diet, and lack of exercise [7]. Cerebral ischemia and reperfusion injury (IRI) is a complex pathophysiological process that involves inflammation and oxidative stress (OS) mechanisms that are difficult to control [8,9,10]. IRI results in the generation of reactive oxygen species (ROS) that can trigger mitochondrial dysfunction, lipid peroxidation, protein oxidation, oxidative DNA damage, among others that lead to microglial activation and neuronal cell loss [11,12]. The role of oxidized phospholipids has not been well-described in IRI and stroke.

Oxidized phospholipids (OxPLs) are created from (poly)unsaturated diacyl- and alk(en)ylacyl glycerophospholipids during situations of oxidative stress [13,14]. Many reaction products are identified by the amount of modifications, hydrophobicity, chemical reactivity, physical properties, and biological activity [15,16]. The specific biological behaviors of these compounds are dependent on the identification of the specific structural motifs by individual receptors and on non-specific structural and chemical properties of target molecules, such as membranes and proteins [17,18,19]. Firstly, their chemical structures are defined as their physical and (bio)chemical properties in membrane- and protein-bound form. Secondly, the biological activities of oxidized phospholipids are defined in terms of their unspecific effects on the membrane level as well as their potential interactions with specific targets (receptors) affecting a large set of (signaling) molecules. Finally, OxPLs mediate myocardial ischemia reperfusion injury. OxPL induce cell death in cardiomyocytes in vitro and neutralization of OxPL with a specific, monoclonal antibody reduced infarct size in a mouse model of ischemia reperfusion injury [20]

OxPLs are generated by the oxidation of polyunsaturated fatty acid (PUFA) residues, which are usually present in the phospholipids at the sn-2 position. Oxidation of phospholipids is instigated by lipoxygenases enzymatically or by reactive oxygen species and spreads by lipid peroxidation chain reaction [21]. The generation of OxPLs cannot be regulated by altering the amount or activity of enzymes. Therefore, the process leads to unrestrained generation of OxPLs during oxidative stress, i.e., cerebral ischemic injury [22]. Bioactive oxidized phospholipids contain fragmentation products of PUFA, such as 1-palmitoyl-2-oxovaleroyl-sn-glycero-3-phosphocholine and 9-keto-10-dodecendioic acid ester of 2-lyso-phosphatidyl choline (KOdiAPC); prostaglandins, such as 15 deoxy-delta 12, 14 prostaglandin I2 (PGI2) and 1-pelmitoyl-2-(5,6-epoxyisoprostane E2)-sn-glycero-3-phosphoryl choline (PEIPC); and levuglandins [23,24,25]. These molecules display unique biological activities. Chromatographic analysis of the products resulting from oxidation of 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphorylcholine (PAPC) allowed for the determination of 1-palmitoyl-2-(5-oxovaleroyl)-sn-glycero-3-phosphatidylcholine (POVPC), 1-palmitoyl-2-glytaroyl-sn-phosphatidylcholine (PGPC), and 1-palmitoyl-2-(5,6-epoxyisopropane E2)-sn-glycero-sn-3-phosphatidylcholine (PEIPC), among others, as potent mediators of inflammation [26]. The structural variation in the molecules determines the biological activities of the OxPLs.

In this study, we tested the impact of anti-OxPL (E06) antibodies on outcomes from cerebral ischemia and reperfusion injury. Here, we demonstrate that OxPLs generated by ischemia and reperfusion injury in the brain contribute to the pathogenesis of stroke. Treatment of animals with a monoclonal antibody specific for OxPL (E06) protected the damage, reduced inflammation, and improved behavioral outcomes. E06 treatment reduced OxPLs in the brain and attenuated the activation of microglial cells preventing neuronal cell loss. These studies suggest that E06 might be a potential therapeutic approach to treat stroke.

2. Materials and Methods

2.1. Oxidized Phospholipid Lc/Ms/Ms Analysis

OxPAPC was produced by air oxidation and analyzed by positive ion electrospray mass spectrometry (ESI-MS) as described previously [14,19,20].

OxPLs were extracted from both neuronal cells and brain tissue as previously described [20]. Cells were washed with ice-cold PBS and extracted in methanol/acetic acid (1 mL, 3% v/v) solution including 0.01% BHT and moved to a glass conical tube and capped beneath nitrogen gas. 1,2-dinonanoyl-sn-glycero-3-phosphocholine (DNPC) was added as internal. Hexane with BHT was added to the tubes (2 mL), vortexed for approximately 5 s under nitrogen gas, and then centrifuged for 5 min at 3500 rpm at 4 °C. The hexane (upper phase) was removed and discarded. The hexane/BHT treatment was repeated three times. Chloroform containing BHT (2 mL) and PBS (750 mL) were added to the tube then vortexed and centrifuged as before. The organic phase (lower) was removed and moved to a glass tube and the liquid was removed using a nitrogen evaporator, then resuspended in chloroform/methanol (300 mL, 2:1 v/v) and stored at −80 °C.

The separation of OxPLs was performed by high-performance liquid chromatography (HPLC) [20]. Brains were extracted as above, and resuspended acetonitrile:water (60:40), 10 mM ammonium formate and 0.1% formic acid. In total, 30 mL of the sample was injected onto a Cytiva C18 HPLC column (15 cm × 2.1 mm, 2.7 mm; Cytiva, Marlborough, MA, USA) with separation by a Shimadzu semi-preparative HPLC for sample fractionation and purification (UV, ELSD and RID) (Canby, OR, USA). Samples were eluted using a gradient (linear) of solvent A (acetonitrile/water, 60:40 v/v) with solvent B (isopropanol/acetonitrile, 90:10, v/v), each containing 0.1% formic acid and 10 mM ammonium formate. We used a mobile phase as follows: starting with solvent B at 32% for 4.00 min; then changed to 45% B; 5.00 min at 52% B; 8.00 min at 58% B; 11.00 min at 66% B; 14.00 min at 70% B; 18.00 min at 75% B; 21.00 min at 97% B; 25.00 min at 97% B; and 25.10 min at 32% B. We used a flow rate of 260 mL/min for analysis, and the sample tray was held at 4 °C and column oven was held at 45 °C.

The measurement of OxPLs was assessed by mass spectra (MS) using a positive polarity mode. Multiple reaction monitoring (MRM) scans were accomplished on 6 transitions operating on a product ion of 184.3 m/z, that was comparable to the cleaved phosphocholine moiety. For this study, we used six commercial standards of PONPC, POVPC, PGPC, PAzPC, KOdiAPC, and KDdiAPC that were assessed, and precise peak assignments were determined on the retention times and mass transitions. The MS settings were as previously suggested: curtain gas, 26 psi; ion spray voltage, 5500 V; temperature, 500.0 °C; collision gas, medium; ion source gas 1, 40.0 psi; ion source gas 2, 30.0 psi; entrance potential, 10 V; collision energy, 53 V; declustering potential, 125 V; collision cell exit potential, 9 V; and dwell time, 50 ms. The external mass calibration curves were determined at regular intervals. For quantitation, the MRM calibration curves were determined for each of the available OxPL standards and peaks were standardized based on their relative responses. Approximately 10 ngs of internal standards were supplemented to each sample during extraction. An Agilent LC-MS QTOF 6540 (Agilent Technologies, Santa Clara, CA, USA) mass spectrometer system with electrospray ionization sources coupled to nano-flow liquid chromatographs was used.

2.2. Mouse Primary Cerebral Cortical and Hippocampal Neurons

Cortical and hippocampal neurons were prepared from embryonic day 18 wild-type (C57) mice embryos (The Jackson Laboratory, Bar Harbor, ME, USA). Briefly, the cerebral cortexes and hippocampi were dissected and harvested under a stereomicroscope microscope. The cortices and hippocampi were sectioned into small pieces using a sterile pair of scissors followed by treatment with 10 mL of 0.25 g/mL trypsin (Invitrogen, Carlsbad, CA, USA) for 10 min at 37 °C [27,28,29]. The trypsinized tissues were then gently pipetted up and down with sterile/polished Pasteur pipette to obtain a single-neuronal cell suspension. Next, the cells were washed three times in 10 mL of Hank’s balanced salt solution (HBSS) plus 10 mg/L of gentamicin. Finally, the cells were counted and cultured in poly-D-lysine-coated 6-well culture plates at density of 1 × 10⁵ cells/cm² in neurobasal media (Invitrogen) containing 500 μM glutamine and B-27 supplement (Invitrogen). Oxidized phospholipids were purchased from Avanti Polar Lipids, Inc. (Alabaster, AL, USA). Oxidized phospholipids and standards: 1-palmitoyl-2-(5-oxovaleroyl)-sn-glycero-3-phosphocholine (POVPC), 1palmitoyl-2-glutaroyl-sn-glycero-3-phosphocholine (PGPC), 1-palmitoyl-2-azelaoyl-sn-glycero-3-phosphocholine (PAzPC), 1-palmitoyl-2-(9-oxo) nonanoyl-sn-glycero-3-phosphocholine (ALDO, PONPC), and the IgM murine NAb E06, which is LPS free, were obtained from Avanti Polar Lipids (Alabaster, AL, USA). 1-(palmitoyl)-2-(5-keto-6-octene-dioyl)-3-phosphocholine (KOdiAPC) was purchased from Cayman Chemicals (Ann Arbor, MI, USA). AB1-2 Ab was isolated and purified from HB-33 hybridoma cells (ATCC) as previously described [30]. Fresh media was applied to cells on glass coverslips and supplemented with the indicated concentrations of OxPL lipids that were sonicated into PBS, forming micelles before addition. These were physiological concentrations in the brain.

2.3. Isolation of Primary Microglial Cells from Mouse Brain

Glial cultures were prepared from cerebral cortices of 1-day-old C57BL/6 mice (Jackson Labs, Bar Harbor, ME, USA). After mechanical and chemical dissociation, cortical cells were seeded in DMEM-F12 with 10% FBS at a density of 250,000 cells/mL (~62,500 cells/cm²) and cultured at 37 °C in humidified 5% CO₂/95% air. Medium was replaced every 4–5 days and confluency was achieved after 10–12 days in vitro (DIV). Microglial cultures were prepared as follows: mouse primary mixed glial cultures were prepared on 25 cm² flasks at different seeding densities in the 60,000–200,000 cells/cm² range. The highest microglia yield was obtained when cells were plated at 100,000–120,000 cells/cm². Microglial cells were obtained by shaking the flasks overnight at 200 rpm. Floating cells were pelleted and sub-cultured at 400,000 cells/mL (~100,000 cells/cm²) on mixed glial-conditioned medium.

2.4. Animals

Adult C57BL/6 male and female mice (30–35 g body weight, 5–6 months of age) were bred as needed at the University of South Florida [31,32,33]. All animals were randomly (mice and rats were weighed, ranked in order of weight, and then assigned to groups from heaviest to lightest) assigned to the different groups, and the individuals who performed the animal studies were blinded to the different strains and groups. Both male and female mice were used in the studies. The behavioral testing experimenters were also blinded to the different groups. All the studies followed the STAIR and ARRIVE guidelines for preclinical studies [34,35]. The experimental procedures were approved by the Institutional Animal Care and Use Committee of the University of South Florida.

2.5. Ischemic Injury and Infarct Volumes

Transient (60 min) focal ischemia was induced by suture occlusion of the middle cerebral artery (MCAo), as described previously [36,37]. Mice were anesthetized with a mixture of 1.5% isoflurane, 70% N₂O, and 28.5% O₂. After the ischemia (1 h), reperfusion started and continued until the end of the study. Body temperatures were maintained at 37 °C by a water-jacketed heating pad and anal thermometer. Transcranical laser doppler was used to monitor any changes in the cerebral blood flow. Mice were included in the study when the blood flow dropped to below 20% during ischemia and returned to 90% during reperfusion of the normal value for data analysis (no mice were excluded from the study). After the indicated times of ischemia and reperfusion injury (IRI), mice were euthanized, and the brains were removed for analysis. Coronal sections at 1–2 mm intervals were prepared and stained with 2% 2,3,5-triphenyltetrazolium chloride (vital dye). Infarct volumes were calculated by summing the infarcted areas (pale) of all sections and multiplying by the thickness of the sections. For the studies involving the OxPL neutralizing antibody (E06) in the protection of the mouse brain against IRI, the E06 antibody was given i.v., via the tail vein at 20 μg/mouse or ~800 μg/kg. E06 plasma titers were determined by chemiluminescent ELISA assays for binding to phosphocholine epitopes as expressed on Cu-OxLDL, as well as to the anti-T15 idiotype antibody AB1-2 [38]. The plasma levels reached 4–5 μg/mL.

2.6. Measurement of Cerebral Blood Flow

Regional cerebral blood flow (rCBF) was analyzed by laser Doppler flowmetry every 30 min over the period 1 h before to 6 h after MCAo [39]. Mice were anesthetized with isoflurane (1.5% in 70%/28.5% NO₂/O₂) and a 2 mm hole was drilled in the skull, with the probe was positioned at 0.1 mm above the dura over the cortical surface. In the hemisphere ipsilateral to the occlusion, coordinates were as follows: point A, 1 mm posterior to the bregma and 5.4 mm lateral to the midline; point B, 1 mm posterior to the bregma and 2.1 mm lateral to the midline; point C, 1 mm anterior to the bregma and 3.4 mm lateral to the midline. The mean values of rCBF were measured before MCAO as baseline, and the data thereafter were expressed as percentages of the baseline value. The rCBF data in the present report was taken from reference point A.

2.7. Neurological Outcomes

Neurological deficits were determined following MCAo by blinded observers using a five-tiered scoring system, as previously described [40]. Briefly, neurologic scores were as follows: 0, normal motor function; 1, flexion of torso and contralateral forelimb when animal was lifted by the tail; 2, circling to the contralateral side when held by tail on flat surface, but normal posture at rest; 3, leaning to the contralateral side at rest; 4, no spontaneous motor activity. Locomotor activity was automatically quantified using the open field activity monitor (Any Maze, San Diego Instruments, San Diego, CA, USA). Mice were placed in a random corner and allowed to acclimate for 10 min prior to a 60 min testing period. External noise, lights, and other stimuli were minimized to reduce bias. A number of measures were automatically obtained during the task including total distance, number of movements, time spent at periphery, and time spent at the center of the enclosure. Activity readings acquired prior to sham procedure were used to establish baseline activities. The duration that the animal spent at the periphery vs. the center was used to assess anxiety level during the task.

2.8. RNA Isolation and Analysis

Total RNA was isolated from mouse brains using TRIzol Reagent (Invitrogen/ThermoFisher, Waltham, MA, USA) according to the manufacturer’s instructions. RNA samples (10 µg) were treated with DNase 1 (TURBO DNA-free TM kit (Invitrogen/ThermoFisher AM 1907) for 30 min at 37 °C. RNA from brain (0.5 µg) was reverse transcribed into cDNA using the Reverse Transcription System (Promega 3500, Madison, WI, USA) [29]. After 5-fold dilution, 5 µL was used as a template for real-time RT-PCR. Amplification was done for 40 cycles using SYBR^TM Green PCR master Mix (ThermoFisher, 4309155). Quantification of mRNA was performed using the ∆∆CT method and normalized to GAPDH for brain. Primer sequences are as follows: GAPDH (Accession number NM_008084) 5′-CTC ATG ACC ACA GTC CAT GCC A-3′; 5′-GGA TGA CCT TGC CCA CAG CCT T-3′; HO-1 (Accession number NM_010442) 5′- CCT TCC CGA ACA TCG GAC AGC C-3′; 5′-GCA GCT CCT CAA ACA GCT CAA-3; Txnrd1 (Accession number NM_015762.2) 5′-GGC TCA AGA GGC TGT ATG GAG-3′; 5′- TTC CAA TGG CCA AAA GAA AC-3′; srxn1 (Accession number NM_029688.5) 5′-CTA TGC CAC ACA GAG ACC ATA G-3′; 5′-GTT GAC CTG CTA ATG TGC TTT C-3′; iNOS (Accession number NM_012611) 5′-GAC CAG AAA CTG TCT CAC CTG-3; 5′-CGA ACA TCG AAC GTC TCA CA-3′; Arg-1 (Accession number NM_ 007482.3) 5′- TTG GGT GGA TGC TCA CAC TG-3′; 5′- TTG CCC ATG CAG ATT CCC-3′.

2.9. Cytokine Analysis

For quantitative analysis of cytokines, ELISA was used to measure the levels of tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), or IL-6 in brain tissue [36]. Cytokines were extracted from mouse brains as follows: frozen hemibrains were placed in tissue homogenization buffer containing protease inhibitor cocktail (Sigma, St Louis, MO, USA) 1:100 dilution immediately before use and homogenized using polytron. Tissue sample suspensions were distributed in aliquots and snap frozen in liquid nitrogen for later measurements. Invitrogen/ThermoFisher ELISA kits were then used, according to manufacturer directions (Carlsbad, CA, USA).

2.10. Immunohistochemical Analysis

Immunohistochemical staining was conducted on 8-μm paraffin sections and assessed by a blinded observer by light microscopy (Olympus BX61, Tokyo, Japan). Following antigen retrieval (IHC World, Ellicott City, MD, USA), the following primary antibodies were used: E06 (330001; Avanti Polar Lipids, Inc.). Primary antibodies were detected with ImmPress-HRP kit and NovaRed peroxidase chromagen (Vector Laboratories, Newark, CA, USA), and primary antibodies were omitted for negative controls.

2.11. In Vitro Oxygen–Glucose Deprivation Studies

OGD was performed according to a previously described method [41,42]. In brief, the culture media were removed, and cells were washed twice and incubated in DMEM without glucose. Then, the cultures were introduced into a specialized, humidified chamber filled with 95% N₂ and 5% CO₂ at 37 °C for 6 h. OGD was terminated by replacing the DMEM without glucose with Neurobasal medium supplemented with 2% B27 and the cultures were further incubated under 95% air and 5% CO₂ for 24 h at 37 °C (recovery, R). Cells in the control group were treated identically except that they were not exposed to OGD.

2.12. Isolation of Mouse Cell Mitochondria

All procedures were performed at 4 °C. Cells were placed immediately in ice-cold isolation medium containing 230 mM mannitol, 70 mM sucrose, 10 mM HEPES, and 1 mM EGTA, pH 7.4. Cells were homogenized in isolation medium using a Teflon-glass homogenizer. The homogenate was centrifuged at 2000× g for 10 min. The supernatant was then centrifuged at 12,000× g for 10 min in a Sorvall RC 5C centrifuge. The supernatant was saved for isolation of cytosol. The pellet was resuspended in isolation medium and centrifuged again at 12,000× g for 10 min. The pellet was resuspended in 15% Percoll (GE Healthcare, Chicago, IL, USA) and placed atop the discontinuous Percoll gradient consisting of a bottom layer of 40% Percoll and a top layer of 23% Percoll. The gradient was spun at 31,000× g for 30 min in an SW-Ti40 rotor in a Beckman LE80K centrifuge. The final mitochondrial pellet was resuspended in isolation medium. Protein concentration was determined with a bicinchoninic acid assay (Sigma) using bovine serum albumin as a standard [43].

2.13. Mitochondrial Permeability Transition Pore (MPTP) Activity

MPTP opening was assayed by measurements of mitochondrial swelling using a probe colorimeter (Brinkmann Instruments, Riverview, FL, USA), as described previously [44].

2.14. Statistical Analysis

Results are expressed as means ± SD. Statistical analysis was performed using one-way ANOVA. A probability value of <0.05 was considered statistically significant.

3. Results

3.1. Oxidized Phospholipids (OxPL) Are Induced in Mouse Neuronal Cells Following Ischemia and Reperfusion Injury (IRI)

Primary neuronal cultures were subjected to ischemia and reperfusion injury (IRI) and analyzed by LC/MS/MS for a number of OxPL species previously implicated in myocardial ischemia–reperfusion injury [20]. The following OxPL species were examined: POVPC (m/z 594), PGPC (m/z 610), PONPC (m/z 650), PAzPC (m/z 666), and KOdiAPC (m/z 664) (Figure 1). As seen in the neonatal cardiomyocytes (NNCM), most of the OxPL species were increased following IRI in the primary mouse neuronal cultures. Figure 1A shows the five OxPLs in both control and ischemic neuronal cell cultures. In Figure 1B, POVPC (33.8 ± 5.26 vs. 10.6 ± 3.21 ng/mg protein), PGPC (12.2 ± 3.35 vs. 3.60 ± 1.52 ng/mg protein), PONPC (73.0 ± 4.18 vs. 36.2 ± 6.61 ng/mg protein), and PAzPC (10.2 ± 1.92 vs. 3.80 ± 0.85 ng/mg protein) were significantly elevated over control cells, with POVPC and PONPC showing the greatest increases. The levels of the KOdiAPC were not significantly different in the two groups.

3.2. Ischemia and Reperfusion Injury in the Mouse Increases OxPLs in the Cortex

To determine the potential impact of OxPLs in the mouse brain (ischemic cortex) following IRI, we subjected the mice to 1 h of ischemia and 24 h of reperfusion injury. In Figure 2A, TTC staining of the control and IRI mouse brains are shown that were used for the analysis. Figure 2B quantifies the lesion volumes in the mouse brains shown in Figure 2A (n = 8/group). As seen in Figure 2C, the changes in OxPLs were similar to those seen in the neuronal cultures. All five OxPL species were significantly increased following IRI. POVPC (5.5-fold ± 0.4; p < 0.01) > PGPC (4.1-fold ± 0.3; p < 0.01) > KodiAPC (3.6-fold ± 0.3; p < 0.01) > PONPC (3.5-fold ± 0.4; p < 0.05) > PAzPC (2.6-fold ± 0.2; p < 0.01).

3.3. Neuronal Cell Death Is Mediated by OxPLs

OxPLs can induce cell death in cardiomyocytes in vitro and neutralization of OxPL with a specific, monoclonal antibody attenuated myocardial ischemia reperfusion injury in vivo [20]. We published previously that oxidized lipoproteins can elicit cell loss in isolated primary neuronal cells [27,28,29]. To determine the impact of the OxPL species identified in Figure 1 and Figure 2, mouse primary neuronal cultures were cultured for 14 days, and then exposed to the different OxPLs and evaluated the impact on cell survival (Figure 3A,B). Increasing concentrations of the OxPLs resulted in an increase in cell death as assayed using calcein-AM (for live cells, green) and ethidium homodimer-1 (for dead cells, red). There was a dose-dependent increase (1, 2, 5, and 10 μM) in cell death with all of the OxPLs. PSPC (non-oxidized PL) was used as a control and showed little if any cell death (mean ± SD; 5.15 ± 2.31%; 5.50 ± 2.92%; 9.18 ± 3.92%; and 11.5 ± 3.06%, respectively). POVPC (16.8 ± 2.66%; 19.9 ± 4.07%; 45.3 ± 10.9%; and 79.4 ± 7.07%, respectively) and PONPC (14.3 ± 2.79%; 31.1 ± 7.26%; 66.1 ± 13.1%; and 84.0 ± 6.41%, respectively) both demonstrated the highest level of cell killing, with PGPC (11.6 ± 2.32%; 28.1 ± 6.44%; 40.7 ± 11.8%; and 59.2 ± 5.56%, respectively) to a lesser degree and PAzPC (13.4 ± 2.72%; 13.9 ± 6.73%; 39.3 ± 8.46%; and 47.9 ± 7.58%, respectively) showing the least effect on the cells. All of the OxPLs showed significant cell death at most of the concentrations.

3.4. OxPLs Enhance Mitochondria Degradation in IRI

Previous studies have implicated mitochondria in the pathogenesis of IRI. Primary neuronal cultures were incubated with in vitro IRI conditions, and the mitochondria were separated from the other cellular components and the concentrations of OxPLs were determined (Figure 4). Figure 4A shows the increases in OxPLs in the neuronal cells after IRI, which shows increases in both POVPC and PONPC with only slight changes in the other OxPLs. In Figure 4B, when the mitochondrial fraction was compared to the other cellular components, the majority of the increases in OxPLs in the cells is seen in the mitochondria. Figure 4C,D show that when cells are treated with the mitochondrial membrane transition pore (mMTP) inhibitor cyclosporin A (CsA), the mitochondrial levels of OxPL are reduced and the cells are protected.

3.5. OxPL Neutralizing Antibody Prevents Neuronal Cell Death

To determine the impact of the OxPLs on neuronal cell death and if the use of neutralizing antibodies will show protection, we treated the cells with 5 μM PSPC, POVPC, and PONPC in the presence and absence of a monoclonal antibody specific to OxPL (E06, 10 mg/mL). E06 protected cardiomyocytes from cell death against OxPLs, and transgenic mice overexpressing E06 had reduced myocardial ischemia reperfusion injury as well atherosclerosis in vivo [20,45]. Here, we show that when primary neuronal cells are treated with OxPLs, they enhance cell death and when E06 was added (co-treatment) to the culture medium, protects the cells. E06 inhibited POVPC-induced cell death by 73.4% (24.8 ± 3.46% vs. 6.60 ± 1.35%) and PONPC-induced cell death by 75.9% (28.6 ± 4.22% vs. 6.90 ± 1.79%) (Figure 5A,B). The non-OxPL, PSPC, had no effect on cell death and the E06 showed no appreciable change in cell number as expected (5.30 ± 1.49% vs. 5.20 ± 1.81% with E06).

3.6. OxPL Neutralizing Antibody Protects the Mouse Brain Against IRI

To investigate the influence of OxPLs in the mouse model of MCAo (as in Figure 2), we determined the impact of the E06 antibody on infarct size and behavioral outcomes (Figure 6A–C). As seen in the figure, when the mice were subjected to IRI (1 h ischemia and 24 h reperfusion) the brains showed a significant increase in infarct volume compared to control animals (61.1 ± 2.59 mm³ vs. 0.0 ± 0.0 mm³, respectively). When mice were injected with the E06 antibody (i.v., tail vein, 20 μg/mouse or ~800 μg/kg), the plasma levels reached 4–5 μg/mL. To determine the brain concentration of E06, brains were isolated after 1 h ischemia and 15 m reperfusion. Total brain levels were 73.8 ± 0.6 ng/g wet weight. When the animals were injected immediately following the start of reperfusion, the E06-treated mice showed a significant decrease in infarct volume compared to the control antibody mice (27.3 ± 2.27 mm³; compared to 0.0 ± 0.0 mm³, in controls). When the animals were subjected to IRI (1 h ischemia, 24 h reperfusion) and examined daily for neurological severity score (NSS), the E06-treated mice showed a consistent reduction in NSS compared to the control animals, even out to 7 days (1.2 ± 0.42 vs. 2.5 ± 0.53, respectively). In addition, we determined the presence of OxPLs in the brain following IRI (Figure 6D). As seen in the figure, following IRI, there is a substantial increase in immunoreactivity of the E06 antibody staining in the IRI mice without treatment and resolution in E06-treated mice (the infarct zone in outlined in Figure 6(D2)). Plasma inflammatory markers TNF-α [26.6 ± 6.49 pg/mL (control); 108.7 ± 20.1 pg/mL (I/R); 33.0 ± 11.2 pg/mL (I/R + E06)], IL-1β [35.8 ± 9.09 pg/mL (control); 251.7 ± 29.7 pg/mL (I/R); 42.5 ± 13.4 pg/mL (I/R + E06)], and IL-6 [13.8 ± 3.70 pg/mL (control); 311.3 ± 37.8 pg/mL (I/R); 43.7 ± 11.2 pg/mL (I/R + E06)] were significantly reduced following E06 treatment in I/R injury in the mouse. There were no changes in the physiological parameters in the different conditions (

Table 1. Physiological measurements in the SAA mice. Physiological measurements were taken before, during, and post-ischemia and reperfusion injury in the mice. Mean arterial blood pressure (MABP), heart rate (HR), body temperature (Temp), and pH were determined in the study for the different mouse strains.

	Before Ischemia
	MAP	HR	Temp
Control	101.5 ± 10.2	421.3 ± 25.7	36.82 ± 0.57
I/R	100.3 ± 11.9	416.8 ± 31.2	36.73 ± 0.62
I/R + E06	100.9 ± 12.3	419.4 ± 22.8	36.52 ± 0.59
	During Ischemia
	MAP	HR	Temp
Control	97.3 ± 10.6	413.5 ± 21.3	412.7 ± 0.63
I/R	94.5 ± 11.3	421.3 ± 26.5	416.3 ± 0.67
I/R + E06	98.7 ± 9.31	417.8 ± 28.4	419.4 ± 0.58
	Post Ischemia
	MAP	HR	Temp
Control	96.9 ± 8.95	420.8 ± 27.6	420.8 ± 0.56
I/R	99.1 ± 10.5	415.4 ± 22.7	417.3 ± 0.63
I/R + E06	97.2 ± 11.1	416.2 ± 26.1	421.9 ± 0.71

)

3.7. OxPLs Induce Cytokine Expression in the Microglial Cell In Vitro

To better understand the impact of OxPLs following IRI, isolated mouse microglial cells were subjected to treatment with various OxPLs and measurement of inflammatory cytokines (Figure 7). As seen in the figure, when the primary microglial cells were treated with 10 μM control (PSPC) or OxPLs, there were significant changes in cytokine profiles. Tissue necrosis factor-α (TNF-α) was increased following treatment with POVPC (115.8 ± 13.36 pg/mL vs. 10.8 ± 1.428 pg/mL), PONPC (187.9 ± 7.801 pg/mL), and PAzPC (180.9 ± 5.076 pg/mL). PGPC and KOdiAPC showed little change (Figure 7A). Interleukin-1β (IL-1β) was elevated following treatment with POVPC (210.1 ± 5.971 pg/mL vs. 21.4 ± 1.628 pg/mL), PONPC (255.8 ± 16.72 pg/mL), and PAzPC (280.5 ± 14.08 pg/mL). Again, PGPC and KOdiAPC did not affect IL-1β levels (Figure 7B). Finally, IL-6 levels were increased with POVPC (306.8 ± 9.243 pg/mL vs. 17.6 ± 1.714 pg/mL), PONPC (343.1 ± 10.56 pg/mL) and PAzPC (391.2 ± 10.14 pg/mL). PGPC and KOdiAPC showed no significant change in IL-6 levels (Figure 7C). When E06 (10 μM) was added to the cultures with the OxPLs, the cytokine increase was eliminated.

3.8. OxPLs Induce M1 and Mox Microglia Phenotype In Vitro

To determine the impact of the OxPLs in the induction of M1 vs. M2 microglial phenotype, we treated primary mouse microglial cells with the different OxPLs (10 μM) and examined changes in inducible nitric oxide synthase (iNOS) mRNA (M1 phenotype) and arginase-1 (Arg1) mRNA (M2 phenotype). Figure 8A shows that treatment of the microglia with POVPC (199.8 ± 6.392 vs. 100.4 ± 3.537%), PONPC (447.1 ± 7.028%), and PAzPC (488.2 ± 28.29%) gave rise to the M1 phenotypic expression of iNOS. In contrast, treatment with the OxPLs reduced M2 phenotype (Arg1), POVPC (50.7 ± 4.425 vs. 102.3 ± 4.913%), PONPC (27.2 ± 2.75%), KOdiAPC (64.3 ± 3.709%), and PAzPC (23.9 ± 2.095%). PGPC showed no change from control (Figure 8B).

Recent studies have demonstrated that OxPLs induce a Mox phenotype/state in macrophages and microglial cells [29]. The implications of this phenotype are still not well known. However, Figure 8C–E show that when primary microglial cells are exposed to OxPLs (PONPC) heme oxygenase-1 (HO-1), sulfiredoxin 1 (srxn1) and thioredoxin reductase 1 (txnrd1) are all increased. PONPC increased HO-1 (5.1 ± 0.2582 vs. 1.2 ± 0.1333-fold induction), srxn1 (7.818 ± 0.3521-fold induction), and txnrd1 (6.51 ± 0.50-fold induction).

Finally, to validate the impact of the E06 Ab on attenuation of the OxPLs and reduction of infarct volume in the mouse model of ischemia and reperfusion injury, we treated mice with the AB1-2 Ab from HB-33 hybridoma cells [46]. The AB1-2 Ab recognizes the E06 Ab and can function as a blocking antibody [47]. In Figure 9 (as above), E06 Ab afforded protection against IRI in the mouse brain (IRI vs. IRI + E06 Ab; 63.5 ± 2.13 mm³ vs. 22.13 ± 1.025 mm³, respectively). When treated with the AB1-2 Ab, the infarct volume was significantly larger than the IRI treated animals alone (63.5 ± 2.13 mm³ vs. 75.75 ± 2.226 mm³, respectively). In addition, when the E06 Ab and AB1-2 Ab are added in combination, the infarct volume was similar to the IRI infarct volume alone (63.5 ± 2.13 mm³ vs. 61.75 ± 2.52 mm³, respectively).

4. Discussion

In this study, we have shown that following ischemia and reperfusion injury in the brain and OGD in primary neurons, OxPLs are generated. When primary neurons were exposed to various OxPL species, POVPC, PONPC, and PGPC resulted in neuronal cell death. Exposure of primary microglial cells to different OxPLs increased M1 and Mox phenotypes and elicited pro-inflammatory cytokine expression. Administration of E06, which is a naturally occurring Ab (NAb) that recognizes OxPLs attenuated the neuronal cell death in vitro and diminished the infarct volume in the brains of cerebral IRI in mice.

Oxidative stress results in cell death by DNA damage, lipid peroxidation, and alterations in protein structure and function. There are at least two distinct groups of DNA damage: active DNA damage and passive DNA damage, and oxidative stress primarily causes passive DNA damage [48]. DNA endonucleases, which are a part of the active DNA damage cascade, is mediated by the caspase-activated deoxynucleases, apoptosis-inducing factor (AIF), and endonuclease G (EndoG), which results in DNA double-strand breakage [49,50,51]. Passive DNA damage is caused by direct reaction with ROS or circuitously interacting with lipids or proteins altered by ROS resulting in changes of nucleotide bases [52]. The hydroxyl radical (OH−), ROS resulting from the generation of Fenton reactions, leading to lipid peroxidation [53]. OH− molecules react with unsaturated fatty acids producing alkyl radicals, which can form peroxyl radicals (ROOS) by interaction with molecular oxygen [54]. ROOS can interact with hydrogen from additional fatty acids to produce a second alkyl radical and a lipid hydroperoxide (ROOH), which creates a sequence of lipid peroxidation [55,56]. Lipid peroxidation can destroy the components of cellular membranes, leading to an increase in cell permeability, organelle dysfunction, and alterations in ion transport [57]. The consequences of lipid peroxidation contribute significantly to the role of oxidative stress injury. These products comprise 4-hydroxynonenal (HNE), malondialdehyde (MDA), and acrolein [58,59]. They can lead to the altered function of proteins by binding to thiol groups and depletion of GSH through reactions with GSH-Px and glutathione S-transferase, inducing more serious oxidative stress injury.

Oxidized phospholipids (OxPLs) can have both pro- and anti-inflammatory properties and can stimulate a slew of diverse cell types. Previous findings have revealed that OxPLs can trigger macrophages to alter their gene expression patterns and functional outcomes [60]. Macrophages can modify their metabolic status in response to changes in tissue injury or damage. Others have shown that OxPLs can reprogram the metabolic parameters of the macrophages that may contribute to response to inflammation or altered redox homeostasis [61]. OxPL-treated macrophages (Mox) are distinct from the typical M1 and M2 macrophages and are distinguished by the accrual of antioxidant metabolites that are derived from glutathione production [62]. Changes seen in the metabolic profiles of Mox macrophages appear to be coordinated with the expression of Hif1a- and Nrf2-dependent genes. Additionally, OxPLs can suppress mitochondrial respiration which is regulated by the production of ceramide through a TLR2-dependent process. Intact OxPL can induce pro-inflammatory gene expression in macrophages, without inhibiting bioenergetics, whereas truncated OxPL will suppress mitochondrial respiration and promote expression of genes controlling redox homeostasis.

Recent studies have shown that autoantibodies that recognize PC that contain oxidation-specific epitopes (OSE) are protective in various conditions such as CVD, stroke and atherosclerosis [26,63]. Low levels of IgM antibodies against PC are potential risk markers for ischemic stroke [64]. We have shown that natural antibodies against a subset of phospholipids and anti-annexin IV were capable of enhancing cerebral injury in a mouse model of ischemic stroke [65]. These studies demonstrate the presence of naturally occurring antibodies serve as not only biomarkers of disease but may provide protection as well as propagate the disease process.

5. Conclusions

In summary, we show that OxPLs play a critical role in the pathogenesis of stroke by mediating ischemic cell death and the E06 antibody reduce the damage. Further analysis showed that blocking the E06 antibody attenuated the effect. The utilization of E06 as a therapeutic may provide a unique approach to treat ischemic stroke and improve outcomes.

6. Limitations

There are limitations to the current study. In the antibody experiments, the antibody treatment was limited to the start of reperfusion, and the impact of delayed administration was not explored. In addition, the role of the OxPLs on astrocytes was not followed up and could be important to the potential outcomes. These protocols are important and will be the target of additional studies.