The Proteome of Extracellular Vesicles Released from Pulmonary Microvascular Endothelium Reveals Impact of Oxygen Conditions on Biotrauma.

The lung can experience different oxygen concentrations, low as in hypoxia, high as under supplemental oxygen therapy, or oscillating during intermittent hypoxia as in obstructive sleep apnea or intermittent hypoxia/hyperoxia due to cyclic atelectasis in the ventilated patient. This study aimed to characterize the oxygen-condition-specific protein composition of extracellular vesicles (EVs) released from human pulmonary microvascular endothelial cells in vitro to decipher their potential role in biotrauma using quantitative proteomics with bioinformatic evaluation, transmission electron microscopy, flow cytometry, and non-activated thromboelastometry (NATEM). The release of vesicles enriched in markers CD9/CD63/CD81 was enhanced under intermittent hypoxia, strong hyperoxia and intermittent hypoxia/hyperoxia. Particles with exposed phosphatidylserine were increased under intermittent hypoxia. A small portion of vesicles were tissue factor-positive, which was enhanced under intermittent hypoxia and intermittent hypoxia/hyperoxia. EVs from treatment with intermittent hypoxia induced a significant reduction of Clotting Time in NATEM analysis compared to EVs isolated after normoxic exposure, while after intermittent hypoxia/hyperoxia, tissue factor in EVs seems to be inactive. Gene set enrichment analysis of differentially expressed genes revealed that EVs from individual oxygen conditions potentially induce different biological processes such as an inflammatory response under strong hyperoxia and intermittent hypoxia/hyperoxia and enhancement of tumor invasiveness under intermittent hypoxia.

Effects of Hyperoxia and Hyperoxic Oscillations on the Proteome of Murine Lung Microvascular Endothelium.

Patients presenting with insufficient tissue oxygenation and impaired lung function as in acute respiratory distress syndrome (ARDS) frequently require mechanical ventilation with supplemental oxygen. Despite the lung being used to experiencing the highest partial pressure of oxygen during healthy breathing, the organ is susceptible to oxygen-induced injury at supraphysiological concentrations. Hyperoxia-induced lung injury (HALI) has been regarded as a second hit to pre-existing lung injury and ventilator-induced lung injury (VILI) attributed to oxidative stress. The injured lung has a tendency to form atelectasis, a cyclic collapse and reopening of alveoli. The affected lung areas experience oxygen conditions that oscillate between hyperoxia and hypoxia rather than remaining in a constant hyperoxic state. Mechanisms of HALI have been investigated in many animal models previously. These studies provided insights into the effects of hyperoxia on the whole organism. However, cell type-specific responses have not been dissected in detail, but are necessary for a complete mechanistic understanding of ongoing pathological processes. In our study, we investigated the effects of constant and intermittent hyperoxia on the lung endothelium from a mouse by an in vitro proteomic approach. We demonstrate that these oxygen conditions have characteristic effects on the pulmonary endothelial proteome that underlie the physiological (patho)mechanisms.

Oxygen-Dependent Changes in the N-Glycome of Murine Pulmonary Endothelial Cells.

Supplemental oxygen is frequently used together with mechanical ventilation to achieve sufficient blood oxygenation. Despite the undoubted benefits, it is vigorously debated whether too much oxygen can also have unpredicted side-effects. Uncertainty is also due to the fact that the molecular mechanisms are still insufficiently understood. The lung endothelium is covered with an exceptionally broad glycocalyx, carrying N- and O-glycans, proteoglycans, glycolipids and glycosaminoglycans. Glycan structures are not genetically determined but depend on the metabolic state and the expression level and activity of biosynthetic and glycan remodeling enzymes, which can be influenced by oxygen and the redox status of the cell. Altered glycan structures can affect cell interactions and signaling. In this study, we investigated the effect of different oxygen conditions on aspects of the glycobiology of the pulmonary endothelium with an emphasis on N-glycans and terminal sialylation using an in vitro cell culture system. We combined a proteomic approach with N-glycan structure analysis by LC-MS, qRT-PCR, sialic acid analysis and lectin binding to show that constant and intermittent hyperoxia induced time dependent changes in global and surface glycosylation. An siRNA approach identified St6gal1 as being primarily responsible for the early transient increase of α2-6 sialylated structures in response to hyperoxia.

Oxygen conditions oscillating between hypoxia and hyperoxia induce different effects in the pulmonary endothelium compared to constant oxygen conditions.

The pulmonary endothelium is an immediate recipient of high oxygen concentrations upon oxygen therapy and mediates down-stream responses. Cyclic collapse and reopening of atelectatic lung areas during mechanical ventilation with high fractions of inspired oxygen result in the propagation of oxygen oscillations in the hypoxic/hyperoxic range. We used primary murine lung endothelial cell cultures to investigate cell responses to constant and oscillating oxygen conditions in the hypoxic to hyperoxic range. Severe constant hyperoxia had pro-inflammatory and cytotoxic effects including an increase in expression of ICAM1, E-selectin, and RAGE at 24 hr exposure. The coagulative/fibrinolytic system responded by upregulation of uPA, tPA, and vWF and PAI1 under constant severe hyperoxia. Among antioxidant enzymes, the upregulation of SOD2, TXN1, TXNRD3, GPX1, and Gstp1 at 24 hr, but downregulation of SOD3 at 72 hr constant hyperoxia was evident. Hypoxic/hyperoxic oscillating oxygen conditions induced pro-inflammatory cytokine release to a lesser extent and later than constant hyperoxia. Gene expression analyses showed upregulation of NFKB p65 mRNA at 72 hr. More evident was a biphasic response of NOS3 and ACE1 gene expression (downregulation until 24 hr and upregulation at 72 hr). ACE2 mRNA was upregulated until 72 hr, but shedding of the mature protein from the cell surface favored ACE1. Oscillations resulted in severe production of peroxynitrite, but apart from upregulation of Gstp1 at 24 hr responses of antioxidative proteins were less pronounced than under constant hyperoxia. Oscillating oxygen in the hypoxic/hyperoxic range has a characteristical impact on vasoactive mediators like NOS3 and on the activation of the renin-angiotensin system in the lung endothelium.

CA1 Hippocampal Pyramidal Cells in Rats, Resuscitated From 8 Minutes of Ventricular Fibrillation Cardiac Arrest, Recover After 20 Weeks of Survival: A Retrospective Pilot Study.

PURPOSE: The cornu ammonis 1 (CA1) region of the hippocampus is specifically vulnerable to global ischemia. We hypothesized that histopathological outcome in a ventricular fibrillation cardiac arrest (VFCA) rat model depends on the time point of the examination. METHODS: Male Sprague-Dawley rats were put into VFCA for 8 min, received chest compressions for 2 min, and were defibrillated to achieve return of spontaneous circulation. Animals surviving for 80 min, 14 days and 140 days were compared with controls. Viable neurons were counted in a 500 µm sector of the CA1 region and layer thickness measured. Microglia cells and astrocytes were counted in a 250×300 µm aspect. RESULTS: Control and 80 min surviving animals had similar numbers of pyramidal neurons in the CA1 region. In 14 days and 140 days survivors neuron numbers and layer thickness were severely diminished compared with controls (P < 0.001). Two-thirds of the 140 days survivors showed significantly more viable neurons than the last third. Microglia was increased in 14 days survivors compared with controls and 140 days survivors, while astrocytes increased in 14 days and 140 days survivors compared with controls (P < 0.001). 140 days survivors had significantly higher astrocyte counts compared with 14 days survivors. CONCLUSIONS: The amount and type of brain lesions present after global ischemia depend on the survival time. A consistent reduction in pyramidal cells in the CA1 region was present in all animals 14 days after VFCA, but in two-thirds of animals a repopulation of pyramidal cells seems to have taken place after 140 days.

Hyperoxia Induces Inflammation and Cytotoxicity in Human Adult Cardiac Myocytes.

Supplemental oxygen (O2) is used as adjunct therapy in anesthesia, emergency, and intensive care medicine. We hypothesized that excessive O2 levels (hyperoxia) can directly injure human adult cardiac myocytes (HACMs). HACMs obtained from the explanted hearts of transplantation patients were exposed to constant hyperoxia (95% O2), intermittent hyperoxia (alternating 10 min exposures to 5% and 95% O2), constant normoxia (21% O2), or constant mild hypoxia (5% O2) using a bioreactor. Changes in cell morphology, viability as assessed by lactate dehydrogenase (LDH) release and trypan blue (TB) staining, and secretion of vascular endothelial growth factor (VEGF), macrophage migration inhibitory factor (MIF), and various pro-inflammatory cytokines (interleukin, IL; chemokine C-X-C motif ligand, CXC; granulocyte-colony stimulating factor, G-CSF; intercellular adhesion molecule, ICAM; chemokine C-C motif ligand, CCL) were compared among treatment groups at baseline (0 h) and after 8, 24, and 72 h of treatment. Changes in HACM protein expression were determined by quantitative proteomic analysis after 48 h of exposure. Compared with constant normoxia and mild hypoxia, constant hyperoxia resulted in a higher TB-positive cell count, greater release of LDH, and elevated secretion of VEGF, MIF, IL-1ß, IL-6, IL-8, CXCL-1, CXCL-10, G-CSF, ICAM-1, CCL-3, and CCL-5. Cellular inflammation and cytotoxicity gradually increased and was highest after 72 h of constant and intermittent hyperoxia. Quantitative proteomic analysis revealed that hypoxic and hyperoxic O2 exposure differently altered the expression levels of proteins involved in cell-cycle regulation, energy metabolism, and cell signaling. In conclusion, constant and intermittent hyperoxia induced inflammation and cytotoxicity in HACMs. Cell injury occurred earliest and was greatest after constant hyperoxia, but even relatively brief repeating hyperoxic episodes induced a substantial inflammatory response.

Reduced long-term memory in a rat model of 8 minutes ventricular fibrillation cardiac arrest: a pilot trial.

BACKGROUND: Evaluating beneficial effects of potential protective therapies following cardiac arrest in rodent models could be enhanced by exploring behavior and cognitive functions. The Morris Water Maze is a well-known cognitive paradigm to test spatial learning and memory. RESULTS: Behavioral testing with the Morris Water Maze in Sprague-Dawley rats (300 ± 25 g) resuscitated after 8 min of ventricular fibrillation cardiac arrest was carried out 5 and 12 weeks after cardiac arrest (CA) and compared to results of naïve rats (CONTROL). At 5 weeks, within each group latency time to reach the hidden platform (reflecting spatial learning) decreased equally from day 1 to 4 (CA: 105.6 ± 8.2 vs. 8.9 ± 1.2 s, p < 0.001; CONTROL: 75.5 ± 13.2 vs. 17.1 ± 4.5, p < 0.001) with no differences between groups (p = 0.138). In the probe trial 24 h after the last trial, time spent in the target sector (reflecting memory recall) within each group was significantly longer (CA: 25 ± 1.3; CONTROL: 24.7 ± 2.5 s) than in each of the three other sectors (CA: 7.7 ± 0.7, 14.3 ± 2.5, 8.4 ± 0.8 and CONTROL: 7.8 ± 1.2, 11.7 ± 1.5, 10.3 ± 1.6 s) but with no significantly differences between groups. Seven days later (reflecting memory retention), control group animals remained significantly longer in the target sector compared to every other sector, whereas the cardiac arrest group animals did not. Even 12 weeks after cardiac arrest, the single p values showed that the control animals displayed a trend to perform better than the resuscitated animals. CONCLUSIONS: Memory recall was impaired early after 8 min of ventricular fibrillation cardiac arrest and might be a more valuable tool for cognitive testing than learning recall after global ischemia due to cardiac arrest.

The molecular background of the differential UV absorbance of the human lens in the 240-400 nm range.

The ultraviolet (UV) absorption of various sections of the human lens was studied and compared with protein expression paralleling differential UV absorbance in anterior and posterior lenticular tissue. The UV absorbance of serial lens cryostat sections (60 µm) and that of lens capsules was determined using a Shimadzu scanning spectrophotometer, and the absorption coefficients were calculated. Two-dimensional gel electrophoresis was performed using two pooled lenticular protein extracts (anterior and posterior sections). Protein spots were quantified and significantly different spots were identified by mass spectrometry following in-gel digestion with trypsin and chymotrypsin. The UV-C and UV-B absorption of the human lens increased toward the posterior parts of the lens. The anterior and posterior lens capsules also effectively absorbed UV radiation. Levels of molecular chaperone proteins Beta-crystallin B2 (UniProtKB ID:P43320), A3 (UniProtKB ID:P05813) and of glyceraldehyde 3-phosphate dehydrogenase (UniProtKB ID:P04406) were significantly higher in the anterior part of the lens, whereas lens proteins Beta-crystallin B1 (UniProtKB ID:P53674) and Alpha-crystallin A chain (UniProtKB ID:P02489) were higher in the posterior sections. These results provide evidence that differential UV absorption in the anterior and posterior lens is accompanied by differential protein expression.

Hippocampal polyamine levels and transglutaminase activity are paralleling spatial memory retrieval in the C57BL/6J mouse.

Although a potential role for polyamines and transglutaminases (TGs) in memory mechanisms have been proposed, hippocampal spermine (SPM) and spermidine (SPD) levels as well as transamidating activity of TG in spatial memory have not been addressed yet. It was therefore the aim of the study to assess hippocampal polyamines and TG activity at the probe trial in a spatial memory paradigm. C57BL/6J mice (20 animals per group) were used for the experiments and divided into a trained and a yoked (untrained) group. The Morris water maze (MWM) was selected as the memory test, animals were sacrificed within 5 min following the probe trial and hippocampi were taken for biochemical analysis. SPD and SPM levels were assessed by an analytical procedure according to Gismondi et al. Transamidating activity of TG was determined following the method described by Chung and Folk using [14C] methylamine as substrate. γ-(Glutamyl)-polyamine levels were evaluated by ion exchange chromatography according to Folk et al. Animals learned the task in the MWM as latencies and pathlengths were significantly reduced. At the probe trial mice showed significantly higher preference for the target quadrant. Free SPD and SPM levels were manifold decreased in the trained as compared to the yoked group. Transamidating activity of TG was fourfold increased in trained as compared to yoked controls. γ-(Glutamyl)-SPD was comparable while γ-(glutamyl)-SPM was significantly higher in the trained group. The findings show a potential role for polyamines, their derivative γ-(glutamyl)-SPM and transamidating activity of TG at memory retrieval or formation. Results from this study are extending and knowledge on polyamines and report for the first time involvement of γ-(glutamyl)-SPM and transamidating activity of TG that may form the basis for future neurochemical and pharmacological studies and indeed, modulation of polyamine and TG activity has been already proposed as a tentative therapeutical concept.

Hippocampal lipids linked to spatial memory in the C57BL/6J mouse.

Although the role of individual brain lipids for learning and memory has been reported, no systematic approach associating brain lipids with spatial memory has been carried out. It was therefore the aim of the study to determine brain lipids in hippocampus of mice forming and yoked controls that did not form spatial memory using the probe trial as the endpoint. 10 animals were trained in the Morris water maze (MWM) and 10 mice were serving as yoked controls i.e. no platform was used during the whole experiment. Hippocampal tissue lipids were extracted and data were acquired with Fourier transformation ion cyclotron resonance mass spectrometry (LTQ-FT) coupled to HPLC. Glycerophosphatidylethanolamines (18:0/22:6, 18:0/20:4 and 18:1/18:1), plasmalogens (16:0-10/22:6 and 18:0-10/22:6) and ceramides (18:0) showed higher levels in the trained group, while glycerolysophosphatidylcholines (16:0, 18:1, 18:0, 20:4), sphingomyelins (16:0, 24:1), ether linked glycerophosphatidylcholines (16:0-10/18:0), glycerophosphatidylcholines (16:0/18:1, 16:0/18:0, 18:0/18:1, 38:7, 18:1/20:1, 20:4/20:4, 22:1/18:1, 22:0/18:1, 20:4/22:6, 22:6/22:6), glucosylceramide (24:1) and plasmalogen (18:0-10/20:1) revealed lower levels in the trained group. Decreased levels of certain species of lysophosphatidylcholine, sphingomyelin, plasmenylphosphatidylcholine, phosphatidylcholine, glycosylceramide and plasmalogen at the probe trial for spatial memory may indicate catabolism in terms of consumption during this process. Increased hippocampal levels of long chain highly unsaturated phosphatidylethanolamines, plasmalogens and ceramides may reflect increased synthesis or decreased degradation at the endpoint of memory testing, probably representing interactions in the brain lipid pathways. The study shows pathways involved in spatial memory, may propose the use of individual brain lipids as probable cognitive enhancers and forms the basis for further studies on the role of brain lipids per se.