Effects of PDE4 pathway inhibition in rat experimental stroke.

PURPOSE: The first genomewide association study indicated that variations in the phosphodiesterase 4D (PDE4D) gene confer risk for ischemic stroke. However, inconsistencies among the studies designed to replicate the findings indicated the need for further investigation to elucidate the role of the PDE4 pathway in stroke pathogenesis. Hence, we studied the effect of global inhibition of the PDE4 pathway in two rat experimental stroke models, using the PDE4 inhibitor rolipram. Further, the specific role of the PDE4D isoform in ischemic stroke pathogenesis was studied using PDE4D knockout rats in experimental stroke. METHODS: Rats were subjected to either the ligation or embolic stroke model and treated with rolipram (3mg/kg; i.p.) prior to the ischemic insult. Similarly, the PDE4D knockout rats were subjected to experimental stroke using the embolic model. RESULTS: Global inhibition of the PDE4 pathway using rolipram produced infarcts that were 225% (p<0.01) and 138% (p<0.05) of control in the ligation and embolic models, respectively. PDE4D knockout rats subjected to embolic stroke showed no change in infarct size compared to wild-type control. CONCLUSIONS: Despite increase in infarct size after global inhibition of the PDE4 pathway with rolipram, specific inhibition of the PDE4D isoform had no effect on experimental stroke. These findings support a role for the PDE4 pathway, independent of the PDE4D isoform, in ischemic stroke pathogenesis.

Rapid endocytosis of interleukin-15 by cerebral endothelia.

Interleukin (IL)-15 receptors are present in the cerebral endothelia composing the blood-brain barrier where they show robust up-regulation by neuroinflammation. To determine how IL15 receptor subunits participate in the endocytosis and intracellular trafficking of IL15, we performed confocal microscopic imaging and radioactive tracer uptake assays in primary brain microvessel endothelial cells and related cell lines transfected with modulatory molecules. By immunostaining and co-localization studies with organelle markers, we showed that IL15 was rapidly endocytosed via lipid rafts and was directed to diverse intracellular pathways. During the course of intracellular trafficking, Alexa dye-conjugated IL15 was partially co-localized with both the specific receptor IL15Rα and the co-receptor IL2Rγ. However, deletion of one of the receptor subunits had only a minor effect in slowing IL15 uptake when primary brain microvessel endothelial cells from the receptor knockout mice were compared with those from wildtype mice. IL15 was trafficked to early, recycling, and late endosomes, to the Golgi, and to lysosomes. The diffuse distribution suggests that IL15 activates multiple endothelial signaling events.

Tissue plasminogen activator expression and barrier properties of human brain microvascular endothelial cells.

BACKGROUND: Tissue plasminogen activator (tPA) regulates fibrinolysis and is routinely used as ischemic stroke pharmacotherapy. We hypothesized that brain microvascular tPA expression and barrier properties of endothelial cells are substantially related. METHODS: Human brain microvascular endothelial cells were incubated with two agents known to modify cAMP pathways: forskolin and rolipram. We analyzed development of endothelial barrier properties, i.e., trans-endothelial electrical resistance (TEER), permeability of endothelial cell monolayer, expression of influx transporter glut-1 and endothelial tight junction molecules occludin and claudin-5, tPA antigen release, and levels of endothelial tPA mRNA. RESULTS: Forskolin plus rolipram-treated endothelial cells showed increased TEER compared to controls (174±20% of control at day six, p<0.01), while permeability to albumin and 70kDa dextran was reduced (21±6.8% of control and 3.8±0.3% of control, respectively, p<0.001). In addition, occludin and claudin-5 protein were up-regulated, occludin mRNA was increased to 206±60% of control (p<0.05), glut-1 mRNA was increased to 196±68% of control (p<0.05), levels of tPA protein were reduced to 35±7.0% of control (p<0.001) after six days, and tPA mRNA was reduced to 32±7.7% of control (p<0.01). TPA and occludin mRNA levels were inversely associated (r=-0.68, p<0.05). CONCLUSIONS: In this in vitro model, barrier properties were strongly linked (by inverse association) with tPA expression of brain microvascular endothelial cells.

The role of cerebral vascular NFkappaB in LPS-induced inflammation: differential regulation of efflux transporter and transporting cytokine receptors.

BACKGROUND/AIMS: The transcription factor NFkappaB is a major mediator of lipopolysaccharide (LPS) signaling. We determined the role of NFkappaB activation in regulatory changes of the P-glycoprotein (Pgp) drug efflux transporter at the blood-brain barrier (BBB) and proinflammatory cytokine receptors. METHODS: We treated NFkappaB knockout and wildtype mice with LPS or vehicle, obtained enriched cerebral microvessels, and determined target mRNA by qPCR for MDR1a/b, IL15Ralpha, IL2 Ralpha, IL2Rgamma, LIFR, gp130, and TNFR1/2, and protein expression by western blotting for P-gp, IL15Ralpha, IL2Rgamma, LIFR, and gp130. RESULTS: The effects of LPS on the transporters and cytokine receptors showed differences between wildtype and NFkappaB knockout mice, and between mRNA and protein changes. NFkappaB not only mediated the LPS-induced increase of MDR1b, IL2Rgamma, and TNFR2 mRNA in the wildtype mice, but it showed opposite effects by elevating IL15Ralpha and TNFR1 mRNA and decreasing IL2Ralpha in the knockout mice. Although basal vinblastine uptake was unchanged in the NFkappaB knockout mice, LPS induced an increase of the uptake (depressed efflux transport) greater than that seen in the wildtype mice, indicating that NFkappaB helps to maintain Pgp efflux transporter function. CONCLUSION: The results show differential involvement of NFkappaB signaling in response to LPS at the BBB.

Cerebral microvascular IL15 is a novel mediator of TNF action.

The blood-brain barrier is a gatekeeper and modulatory interface for the CNS. Cerebral endothelial cells are the major component of the blood-brain barrier, and they modify inflammatory signals from the circulation to the CNS by production and secretion of secondary substances. The inflammatory mediators induced by tumor necrosis factor alpha (TNF) were determined by microarray analysis of RBE4 cerebral endothelial cells, at 0, 6, 12, or 24 h after TNF treatment. Interleukin (IL)-15 and its receptors were among the most robustly up-regulated genes. This was confirmed by real-time RT-PCR and western blotting. The three subunits of the IL15 receptor complex (IL15Ralpha, IL2Rbeta, and IL2Rgamma) showed differential regulation by TNF in their time course and amplitude of increased expression. Consistent with increased expression of the specific high affinity receptor IL15Ralpha, TNF increased cellular uptake of (125)I-IL15 and enhanced the fluorescent intensity of Alexa568-IL15 in RBE4 cells. TNF treatment in mice also increased the level of expression of IL15 receptors in enriched cerebral microvessels. We conclude that the cerebral microvascular IL15 system is a novel inflammatory mediator that transduces the actions of TNF.

Astrocyte leptin receptor (ObR) and leptin transport in adult-onset obese mice.

The agouti viable yellow (A vy) spontaneous mutation generates an unusual mouse phenotype of agouti-colored coat and adult-onset obesity with metabolic syndrome. Persistent production of agouti signaling protein in A vy mice antagonizes melanocortin receptors in the hypothalamus. To determine how this disruption of neuroendocrine circuits affects leptin transport across the blood-brain barrier (BBB), we measured leptin influx in A vy and B6 control mice after the development of obesity, hyperleptinemia, and increased adiposity. After iv bolus injection, (125)I-leptin crossed the BBB significantly faster in young (2 month old) B6 mice than in young A vy mice or in older (8 month old) mice of either strain. This difference was not observed by in situ brain perfusion studies, indicating the cause being circulating factors, such as elevated leptin levels or soluble receptors. Thus, A vy mice showed peripheral leptin resistance. ObRa, the main transporting receptor for leptin at the BBB, showed no change in mRNA expression in the cerebral microvessels between the age-matched (2 month old) A vy and B6 mice. Higher ObRb mRNA was seen in the A vy microvasculature with unknown significance. Immunofluorescent staining unexpectedly revealed that many of the ObR(+) cells were astrocytes and that the A vy mice showed significantly more ObR(+) astrocytes in the hypothalamus than the B6 mice. Although leptin permeation from the circulation was slower in the A vy mice, the increased ObR expression in astrocytes and increased ObRb mRNA in microvessels suggest the possibility of heightened central nervous system sensitivity to circulating leptin.

Permeation of blood-borne IL15 across the blood-brain barrier and the effect of LPS.

Interleukin15 (IL 15) is a proinflammatory cytokine with elevated concentrations in autoimmune diseases involving the periphery (e.g. rheumatoid arthritis) and CNS (e.g. multiple sclerosis). Its interactions with the blood-brain barrier (BBB) were studied in normal and lipopolysaccharide (LPS)-treated mice. (125)I-IL15 remained intact for at least 10 min after i.v. injection and reached CNS parenchyma with regional differences between brain and spinal cord. Both in vivo and in situ brain perfusion of (125)I-IL15 showed that its permeation of the BBB was non-saturable. LPS induced a significant increase of IL15 uptake by the brain and spinal cord, partly related to a higher general permeability of the BBB. The results suggest that the BBB is an interface for blood-borne IL15 to interact with the CNS in the basal state and during inflammation.

Neuroinflammation activates Mdr1b efflux transport through NFkappaB: promoter analysis in BBB endothelia.

BACKGROUND/AIMS: Although it is known that drug delivery across the blood-brain barrier (BBB) may be hampered by efflux transport activity of the multidrug resistance (mdr) gene product P-glycoprotein, it is not clear how inflammation regulates efflux transporters. In rat brain endothelial (RBE4) cells of BBB origin, the proinflammatory cytokine TNF mainly induced transcriptional upregulation of mdr1b, and to a lesser extent mdr1a, resulting in greater efflux of the substrates. This study further determines the mechanisms by which TNF activates mdr1b promoter activity. METHODS/RESULTS: Luciferase reporter assays and DNA binding studies show that (1) maximal basal promoter activity was conferred by a 476 bp sequence upstream to the mdr1b transcriptional initiation site; (2) TNF induced upregulation of promoter activity by NFkappaB nuclear translocation; and (3) the NFkappaB binding site of the mdr1b promoter was solely responsible for basal and TNF-activated gene transcription, whereas the p53 binding site was not involved. Binding of the p65 subunit of NFkappaB to nuclear DNA from RBE4 cells was shown by electrophoretic mobility shift assay and chromatin immunoprecipitation assays. CONCLUSION: NFkappaB mediates TNF-induced upregulation of mdr1b promoter activity, illustrating how inflammation activates BBB efflux transport.

Design of a Conceptual Bumper Energy Absorber Coupling Pedestrian Safety and Low-Speed Impact Requirements.

The car front bumper system needs to meet the requirements of both pedestrian safety and low-speed impact which are somewhat contradicting. This study aims to design a new kind of modular self-adaptive energy absorber of the front bumper system which can balance the two performances. The X-shaped energy-absorbing structure was proposed which can enhance the energy absorption capacity during impact by changing its deformation mode based on the amount of external collision energy. Then, finite element simulations with a realistic vehicle bumper system are performed to demonstrate its crashworthiness in comparison with the traditional foam energy absorber, which presents a significant improvement of the two performances. Furthermore, the structural parameters of the X-shaped energy-absorbing structure including thickness (tu), side arc radius (R), and clamping boost beam thickness (tb) are analyzed using a full factorial method, and a multiobjective optimization is implemented regarding evaluation indexes of both pedestrian safety and low-speed impact. The optimal parameters are then verified, and the feasibility of the optimal results is confirmed. In conclusion, the new X-shaped energy absorber can meet both pedestrian safety and low-speed impact requirements well by altering the main deformation modes according to different impact energy levels.

TNFalpha trafficking in cerebral vascular endothelial cells.

Using small tags, we tracked the pathway of tumor necrosis factor (TNF)alpha across cerebral vascular endothelial cells. In cerebral microvessel derived RBE4 cells, (125)I-TNFalpha had rapid endocytosis within the first 20 min and showed substantial exocytosis in the intact form. Biotinylated TNFalpha was detected at different time points after endocytosis by streptavidin-Quantum dots which showed its time-dependent colocalization with intracellular organelles. In mice, electron microscopic autoradiography after intravenous injection of (125)I-TNFalpha showed its transcytosis, as signals emerged on the abluminal side of the endothelial cells and reached brain parenchyma. The vesicular trafficking of TNFalpha reflects the immunomodulatory potential of peripheral cytokines for the CNS.

Opposing effects of proteasomes and lysosomes on LIFR: modulation by TNF.

The blood-brain barrier (BBB), a communicating interface for inflammation, transports cytokines through its endothelial cells. This study shows how tumor necrosis factor alpha (TNF) regulates the expression of the leukemia inhibitor factor receptor (LIFR) gp190 in RBE4 cells. The high expression of LIFR was rapidly downregulated by the proinflammatory agents lipopolysaccharide, TNF, and LIF. Downregulation by TNF affected LIFR endocytosis and lysosomal degradation, preceding decreased LIFR mRNA. Lysosomal inhibitors reversed the rapid disappearance of LIFR, whereas inhibition of the ubiquitin-proteasome pathway did not. Rather, blockade of proteasome activity, as well as inhibition of NFkappaB activation, reduced the basal expression of LIFR. Thus, NFkappaB activity and proteasome degradation of IkappaB stabilized LIFR and prevented its rapid lysosomal degradation. By a non-NFkappaB-mediated mechanism, TNF facilitated LIFR degradation and reduced LIFR activation indicated by pStat3. The novel opposite effects of proteasomes and lysosomes in controlling receptor expression shows the functional implications and interactions of circulating inflammatory cytokines in acutely modulating BBB activity.

Diet-induced obesity suppresses expression of many proteins at the blood-brain barrier.

The blood-brain barrier (BBB) is a regulatory interface between the central nervous system and the rest of the body. However, BBB changes in obesity and metabolic syndrome have not been fully elucidated. We hypothesized that obesity reduces energy metabolism in the cerebral microvessels composing the BBB, reflected by downregulation of protein expression and function. We performed comparative proteomic analyses in enriched microvessels from the cerebral cortex of mice 2 months after ingestion of a high-fat diet or regular rodent chow. In mice with diet-induced obesity (DIO), there was downregulation of 47 proteins in the cerebral microvessels, including cytoskeletal proteins, chaperons, enzymes, transport-related proteins, and regulators for transcriptional and translational activities. Only two proteins, involved in messenger RNA (mRNA) transport and processing, were upregulated. The changes of these proteins were further validated by quantitative polymerase chain reaction (qPCR), western blotting, and immunofluorescent staining of freshly isolated microvessels, in samples obtained from different batches of mice. The predominant downregulation suggests that DIO suppresses metabolic activity of BBB microvessels. The finding of a hypometabolic state of the BBB in mice at the chronic stage of DIO is unexpected and unprecedented; it may provide novel mechanistic insight into how obesity influences CNS function via regulatory changes of the BBB.

The role of pericytes in blood-brain barrier function and stroke.

Central nervous system pericytes have critical and complex inductive, structural, and regulatory roles interacting with other cell types of the neurovascular unit, especially endothelial cells and astrocytes. Pericyte-endothelial interactions are particularly prominent for blood-brain barrier (BBB) maintenance, with profound effects on basement membrane and endothelial tight junction structure and function. Under experimental conditions of hypoxia-ischemia mimicking stroke, pericytes migrate from their usual microvascular location and influence, directly or indirectly, BBB permeability. The contractile properties of pericytes provide the capacity to regulate capillary blood flow, but this may have detrimental effects on ischemic injury. Stem cell characteristics of pericytes imply an important regenerative role following stroke. Pericytes thus appear to orchestrate multiple critical functions in stroke, involving blood flow, permeability, and repair of the neurovascular unit.

Phosphodiesterase inhibitor modulation of brain microvascular endothelial cell barrier properties.

BACKGROUND AND PURPOSE: Brain microvascular disorders, including cerebral microscopic hemorrhage, have high prevalence but few treatment options. To develop new strategies for these disorders, we analyzed the effects of several phosphodiesterase (PDE) inhibitors on human brain microvascular endothelial cells (HBECs). METHODS: We modified barrier properties and response to histamine of HBECs using cilostazol (PDE3 inhibitor), rolipram (PDE4 inhibitor), and dipyridamole (non-specific PDE inhibitor). RESULTS: Cilostazol and dipyridamole altered the distribution of endothelial F-actin. Cilostazol increased expression of tight junction protein claudin-5 by 118% compared to control (p<.001). Permeability to albumin was decreased by cilostazol (21% vs control, p<.05), and permeability to dextran (70Kd) was decreased by both cilostazol (37% vs control, p<.001) and dipyridamole (44% vs control, p<.0001). Cilostazol increased trans-endothelial electrical resistance (TEER) after 12h by 111% compared to control (p<.0001). Protein kinase A (PKA) inhibitors H89 and KT5720 attenuated the TEER increase by cilostazol. Transient increased permeability in response to histamine was significantly mitigated by cilostazol, but not by other PDE inhibitors. CONCLUSIONS: These findings demonstrate distinctive effects of cilostazol and other PDE inhibitors on HBECs, including enhanced barrier characteristics and mitigation of response to histamine. PKA-mediated effects of cilostazol were prominent in this model. These in vitro findings are consistent with therapeutic potential of PDE inhibitors in human brain microvascular disorders.

PDE4 regulates tissue plasminogen activator expression of human brain microvascular endothelial cells.

INTRODUCTION: Factors regulating brain tissue plasminogen activator (tPA) are pertinent for stroke. Recent observations have suggested a role for the phosphodiesterase-4 (PDE4) pathway in stroke pathogenesis, via an uncertain mechanism. We studied PDE4 regulation of tPA expression by human brain microvascular endothelial cells in a variety of conditions, including an in vitro model of ischemia. MATERIALS AND METHODS: We analyzed tPA antigen and mRNA of human brain microvascular endothelial cells (HBECs) during normoxia and oxygen-glucose deprivation (OGD) following inhibition of PDE4 and PDE4D, using HBEC monocultures and co-cultures with astrocytes and pericytes, and analyzed relevant signal transduction pathways. RESULTS: PDE4 inhibitor rolipram enhanced OGD effects on endothelial tPA release in endothelial monocultures and co-cultures with astrocytes; there was a 54±10% (p<0.001) reduction of tPA release in astrocyte-endothelial co-cultures under OGD. PDE4D siRNA reduced endothelial tPA mRNA to 40-55% of control (p<0.05). Use of Epac inducer mimicked, while use of Epac siRNA inhibited, these effects. CONCLUSIONS: Inhibition of PDE4 and PDE4D reduced expression of tPA by HBEC via Epac pathway.

Brain Activation by Peptide Pro-Leu-Gly-NH(2) (MIF-1).

MIF-1 (Pro-Leu-Gly-NH(2)) is a tripeptide for which the therapeutic potential in Parkinson's disease and depression has been indicated by many studies. However, the cellular mechanisms of action of MIF-1 are not yet clear. Here, we show the specific brain regions responsive to MIF-1 treatment by c-Fos mapping, and determine the kinetics of cellular signaling by western blotting of pERK, pSTAT3, and c-Fos in cultured neurons. The immunoreactivity of c-Fos was increased 4 hours after MIF-1 treatment in brain regions critically involved in the regulation of mood, anxiety, depression, and memory. The number of cells activated was greater after peripheral treatment (intravenous delivery) than after intracerebroventricular injection. In cultured SH-SY5Y neuronal cells, c-Fos was induced time- and dose-dependently. The activation of cellular c-Fos was preceded by a transient increase of mitogen-activated protein kinase pERK but a reduction of phosphorylated Signal Transducer and Activator of Transcription (pSTAT3) initially. We conclude that MIF-1 can modulate multiple cellular signals including pERK, and pSTAT3 to activate c-Fos. The cellular activation in specific brain regions illustrates the biochemical and neuroanatomical basis underlying the therapeutic effect of MIF-1 in Parkinson's disease and depression.

Mass spectrometric quantification of MIF-1 in mouse brain by multiple reaction monitoring.

MIF-1 (Pro-Leu-Gly-NH(2)) has potent therapeutic effects in depression and Parkinson's disease, but its CNS sites of production are not yet clear. In this study, the concentration of MIF-1 in different brain regions was measured by the multiple reaction monitoring technique on a 4000 QTRAP mass spectrometer. The limit of quantification was 300 fg of MIF-1, and limit of detection was 60 fg. The low molecular weight fractions of tissue homogenates from different regions of mouse brain were analyzed. The concentration of MIF-1 ranged from 22+/-3 fg/microg protein in cerebral cortex to 930+/-60 fg/microg protein in the hypothalamus. Moderate concentrations were also detected in all other regions tested, including the striatum, thalamus, and hippocampus. By incubation of stable isotope-labeled oxytocin with tissue preparations, it was also confirmed that oxytocin at least partially contributed to the production of MIF-1 in the hypothalamus by action of peptidases. Regional differences were also found. The results are the first to show the ultrasensitive quantification of MIF-1 in different brain regions, and support the neuromodulatory actions of MIF-1 in the striatum.

Neuroinflammation facilitates LIF entry into brain: role of TNF.

Leukemia inhibitory factor (LIF) is a proinflammatory cytokine mediating a variety of central nervous system (CNS) responses to inflammatory stimuli. During lipopolysaccharide (LPS)-induced inflammation, blood concentrations of LIF increase, correlating with lethality of sepsis. Circulating LIF crosses the blood-brain barrier (BBB) by a saturable transport system. Here we determine how this transport system is regulated in neuroinflammation. Using transport assays that quantify the influx rate and volume of distribution of LIF in mice, we show that LPS facilitated the permeation of LIF from the blood to the brain without compromising the paracellular permeability of the BBB as determined by coadministration of fluorescein. Concurrently, gp130 (shared by the interleukin-6 family of cytokines), but not gp190 (the specific receptor for LIF) or cilliary neutrophic factor (CNTF-Ralpha, a unique receptor for cilliary neurotrophic factor that also uses gp130 and gp190), showed increased levels of mRNA and protein expression in cerebral microvessels from the LPS-treated mice. The upregulation of gp130 by LPS was at least partially mediated by vascular tumor necrosis factor receptor (TNFR)1 and TNFR2. This was shown by elevated TNFR1 and TNFR2 mRNA and protein in cerebral microvessels after LPS and by the absence of the LPS effect on gp130 in knockout mice lacking these receptors. The results show that neuroinflammation by LPS induces endothelial signaling and enhances cytokine transport across the BBB.

TNF reduces LIF endocytosis despite increasing NFkappaB-mediated gp130 expression.

To examine how the proinflammatory cytokine tumor necrosis factor alpha (TNF) modulates the response of cerebral microvessels to other cytokines, we used rat cerebral microvessel endothelial RBE4 cells to simulate the in vitro blood-brain barrier (BBB). The gp130 receptor, which is shared by the interleukin (IL)-6 family of cytokines, showed specific upregulation by TNF. TNF treatment (5 ng/ml for 30 min to 24 h) increased gp130 at both the levels of transcription and protein expression. The stability of gp130 protein was mediated by NFkappaB activity, as the inhibitors quinazoline and MG132 not only blocked the increase induced by 6 h of TNF treatment, but also reduced its basal level of expression. By contrast, the lysosomal inhibitor chloroquine and the extracellular regulated kinase inhibitor U0126 showed no effect. Despite the increase of gp130, TNF caused a significant reduction in the cell binding and endocytosis of leukemia inhibitory factor (LIF), another proinflammatory cytokine that binds to the gp130 co-receptor and its unique gp190 receptor. This is consistent with our previous findings that TNF reduces gp190 expression and Stat3 activation. Thus, TNF stimulation results in decreased responsiveness of RBE4 cells to LIF, indicating complex regulatory interactions of cytokines at the BBB.

Gamma glutamyl transpeptidase is a dynamic indicator of endothelial response to stroke.

Gamma glutamyl transpeptidase (gammaGT) is enriched at the apical surface of the cerebral capillaries that constitute the blood-brain barrier (BBB). This study tested the effects of hypoxia and inflammation on gammaGT activity in mice after stroke induced by transient cerebral artery occlusion (tMCAO) and in cultured cerebral microvessel endothelial cells. In microvessel-enriched preparations from mice after tMCAO, gammaGT activity was higher than in the sham controls in both ipsilateral and contralateral hemispheres from 12 h to 5 days after stroke, but lower at later time points (10-15 days). To identify the roles of different cytotoxic and stimulatory signals in this event, we further studied the dynamic changes of gammaGT activity in rat brain endothelial (RBE4) cells. Tumor necrosis factor alpha and lipopolyssachride significantly increased gammaGT activity in a time-dependent manner, an effect not seen after re-oxygenation. Such endothelial activation correlated with reduced total cellular ATP production. Thus, hypoxia and inflammatory stimulation appeared to have opposite effects on endothelial function. With the co-existence of inflammation and hypoxia in the brain after ischemic stroke, dynamic changes of gammaGT activity reflect evolving changes of endothelial function.