Csongor S. Ábrahám

Permeability studies on in vitro blood-brain barrier models: Physiology, pathology, and pharmacology

Summary1. The specifically regulated restrictive permeability barrier to cells and molecules is the most important feature of the blood–brain barrier (BBB). The aim of this review was to summarize permeability data obtained on in vitro BBB models by measurement of transendothelial electrical resistance and by calculation of permeability coefficients for paracellular or transendothelial tracers.2. Results from primary cultures of cerebral microvascular endothelial cells or immortalized cell lines from bovine, human, porcine, and rodent origin are presented. Effects of coculture with astroglia, neurons, mesenchymal cells, blood cells, and conditioned media, as well as physiological influence of serum components, hormones, growth factors, lipids, and lipoproteins on the barrier function are discussed.3. BBB permeability results gained on in vitro models of pathological conditions including hypoxia and reoxygenation, neurodegenerative diseases, or bacterial and viral infections have been reviewed. Effects of cytokines, vasoactive mediators, and other pathogenic factors on barrier integrity are also detailed.4. Pharmacological treatments modulating intracellular cyclic nucleotide or calcium levels, and activity of protein kinases, protein tyrosine phosphatases, phospholipases, cyclooxygenases, or lipoxygenases able to change BBB integrity are outlined. Barrier regulation by drugs involved in the metabolism of nitric oxide and reactive oxygen species, as well as influence of miscellaneous treatments are also listed and evaluated.5. Though recent advances resulted in development of improved in vitro BBB model systems to investigate disease modeling, drug screening, and testing vectors targeting the brain, there is a need for checking validity of permeability models and cautious interpretation of data.

Recombinant human tumor necrosis factor α constricts pial arterioles and increases blood-brain barrier permeability in newborn piglets

Tumor necrosis factor alpha (TNF alpha) plays a significant role in the pathogenesis of central nervous system infections. We investigated the effect of intracisternal injection of recombinant human TNF alpha (50-50,000 IU) on pial vasoreactivity and blood-brain barrier permeability in newborn piglets. The cytokine administration resulted in arterial vasoconstrictions, blood-brain barrier opening for Na-fluorescein (mol. wt. 376 Da) and increased Na-fluorescein uptake in brain regions examined (parietal and occipital cortex, cerebellum, pons/medulla, periventricular white matter) in a dose-dependent manner. TNF alpha may be involved in the pathophysiology of neonatal brain injuries.

Intracarotid tumor necrosis factor-α administration increases the blood-brain barrier permeability in cerebral cortex of the newborn pig: quantitative aspects of double-labelling studies and confocal laser scanning analysis

Tumor necrosis factor-alpha (TNF-alpha) plays a crucial role in the pathogenesis of the central nervous system infections. The aim of the present study was to analyze quantitatively the changes in the blood-brain barrier (BBB) permeability after the intracarotid injection of TNF-alpha. Recombinant human TNF-alpha was injected into the left internal carotid artery of anesthetized newborn pigs (n = 48) in the doses of 0, 1000, 10 000 and 100 000 IU, respectively. Before, as well as 1, 2, 4, 8, and 16 h after the challenge, the extravasation of a small (sodium fluorescein (SF), mw 376), and a large (Evans blue-albumin (EBA), mw 67 000) tracer was determined concomitantly by spectrophotometry in the cerebral cortex of the animals. There was a time- and dose-dependent increase in BBB permeability both for SF and EBA; however, significant (P < 0.05) BBB opening for albumin only developed 2 h after the challenge. In the morphological study the same excitable tracers, identical experimental protocol and groups were used. Cryostat sections of brain tissue were viewed for optical sectioning with a confocal laser scanning microscope equipped with an argon/krypton ion laser. A diffuse BBB opening for SF and a moderate perivascular extravasation for EBA were found in the cortices of TNF-alpha-treated animals. We conclude that significant increases in intravascular TNF-alpha-concentration during neonatal infections may result in vasogenic brain edema formation.

Adrenomedullin regulates blood-brain barrier functions in vitro.

Adrenomedullin (AM) is an important vasodilator in cerebral circulation, and cerebral endothelial cells are a major source of AM. This in vitro study aimed to determine the AM-induced changes in blood–brain barrier (BBB) functions. AM administration increased, whereas AM antisense oligonucleotide treatment decreased transendothelial electrical resistance. AM incubation decreased BBB permeability for sodium fluorescein (mol. wt 376 Da) but not for Evans blue albumin (mol. wt 67 kDa), and it also attenuated fluid-phase endocytosis. AM treatment resulted in functional activation of P-glycoprotein efflux pump in vitro. Our results indicate that AM as an autocrine mediator plays an important role in the regulation of BBB properties of the cerebral endothelial cells.

Pentosan polysulfate protects brain endothelial cells against bacterial lipopolysaccharide-induced damages

Peripheral inflammation can aggravate local brain inflammation and neuronal death. The blood-brain barrier (BBB) is a key player in the event. On a relevant in vitro model of primary rat brain endothelial cells co-cultured with primary rat astroglia cells lipopolysaccharide (LPS)-induced changes in several BBB functions have been investigated. LPS-treatment resulted in a dose- and time-dependent decrease in the integrity of endothelial monolayers: transendothelial electrical resistance dropped, while flux of permeability markers fluorescein and albumin significantly increased. Immunostaining for junctional proteins ZO-1, claudin-5 and beta-catenin was significantly weaker in LPS-treated endothelial cells than in control monolayers. LPS also reduced the intensity and changed the pattern of ZO-1 immunostaining in freshly isolated rat brain microvessels. The activity of P-glycoprotein, an important efflux pump at the BBB, was also inhibited by LPS. At the same time production of reactive oxygen species and nitric oxide was increased in brain endothelial cells treated with LPS. Pentosan polysulfate, a polyanionic polysaccharide could reduce the deleterious effects of LPS on BBB permeability, and P-glycoprotein activity. LPS-stimulated increase in the production of reactive oxygen species and nitric oxide was also decreased by pentosan treatment. The protective effect of pentosan for brain endothelium can be of therapeutical significance in bacterial infections affecting the BBB.

Cerebral endothelial cells are a major source of adrenomedullin.

Adrenomedullin is a peptide hormone with multifunctional biological properties. Its most characteristic effects are the regulation of circulation and the control of fluid and electrolyte homeostasis through peripheral and central nervous system actions. Although adrenomedullin is a vasodilator of cerebral vasculature, and it may be implicated in the pathomechanism of cerebrovascular diseases, the source of adrenomedullin in the cerebral circulation has not been investigated thus far. We measured the secretion of adrenomedullin by radioimmunoassay and detected adrenomedullin mRNA expression by Northern blot analysis in primary cultures of rat cerebral endothelial cells (RCECs), pericytes and astrocytes. We also investigated the expression of specific adrenomedullin receptor components by reverse transcriptase‐polymerase chain reaction and intracellular cAMP concentrations in RCECs and pericytes. RCECs had approximately one magnitude higher adrenomedullin production (135 ± 13 fmol/105 cells per 12 h; mean ± SD, n = 10) compared to that previously reported for other cell types. RCECs secreted adrenomedullin mostly at their luminal cell membrane. Adrenomedullin production was not increased by thrombin, lipopolysaccharide or cytokines, which are known inducers of adrenomedullin release in peripheral endothelial cells, although it was stimulated by astrocyte‐derived factors. Pericytes had moderate, while astrocytes had very low basal adrenomedullin secretion. In vivo experiments showed that adrenomedullin plasma concentration in the jugular vein of rats was approximately 50% higher than that in the carotid artery or in the vena cava. Both RCECs and pericytes, which are potential targets of adrenomedullin in cerebral microcirculation, expressed adrenomedullin receptor components, and exhibited a dose‐dependent increase in intracellular cAMP concentrations after exogenous adrenomedullin administration. Antisense oligonucleotide treatment significantly reduced adrenomedullin production by RCECs and tended to decrease intraendothelial cAMP concentrations. These findings may suggest an important autocrine and paracrine role for adrenomedullin in the regulation of cerebral circulation and blood–brain barrier functions. Cerebral endothelial cells are a potential source of adrenomedullin in the central nervous system, where adrenomedullin can also be involved in the regulation of neuroendocrine functions.

Transient Forebrain Ischemia Increases the Blood-Brain Barrier Permeability for Albumin in Stroke-Prone Spontaneously Hypertensive Rats

Abstract1. The aim of the present study was to reveal the effect of transient forebrain ischemia on the regional and temporal changes in the permeability of the blood-brain barrier (BBB) permeability for sodium fluorescein (MW: 376 Da) and Evans blue-labeled albumin (MW: 67 kDa) in stroke-prone spontaneously hypertensive rats (SHRSP).2. BBB permeability was significantly higher in the brain regions of 16-week-old control SHRSP than those in age-matched normotensive Wistar-Kyoto rats.3. Transient forebrain ischemia evoked by 10-min bilateral carotid occlusion increased the permeability of the BBB for albumin, but not for sodium fluorescein, after 6 and 24 h of reperfusion in brain regions of SHRSP.4. Extravasation of serum macromolecules may contribute to neuronal loss and development of hypertensive encephalopathy in SHRSP.

Human serum amyloid P component attenuates the bacterial lipopolysaccharide-induced increase in blood-brain barrier permeability in mice.

Endotoxin challenge leads to septic shock, multi-organ failure and death in mice. Permeability of the blood-brain barrier (BBB) is increased by endotoxemia. Serum amyloid P component (SAP) is a lipopolysaccharide (LPS)-binding protein that can modulate the host reactions during infections. It is controversial whether SAP can protect from LPS toxicity in vivo or not. We have tested the effect of human SAP on BBB permeability of Salmonella typhimurium LPS-injected mice. The animals showed signs of sickness behaviour including immobility, anorexia, and diarrhoea. Intraperitoneally administered LPS increased the BBB permeability for sodium fluorescein for about 4-fold, and for albumin for more than 2-fold in brain cortex. SAP, given intravenously, had no effect on basal BBB permeability for albumin, although it decreased sodium fluorescein extravasation to brain tissue. In LPS-treated mice, SAP administration alleviated the symptoms of septic shock, and significantly inhibited the enhanced BBB permeability for both tracers. Our data indicate that human SAP may counteract the toxic effects of LPS during septic shock.

PrP fragment 106-126 is toxic to cerebral endothelial cells expressing PrP(C).

A hydrophobic, fibrillogenic peptide fragment of human prion protein (PrP 106–126) had in vitro toxicity to neurons expressing cellular prion protein (PrPC). In this study, we proved that primary cultures of mouse cerebral endothelial cells (MCEC) express PrPC. Incubation of MCEC with PrP 106–126 (25–200 μM) caused a dose-dependent toxicity assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay, lactate dehydrogenase release, bis-benzimide staining for nuclear morphology, and trypan blue exclusion test. Pentosan polysulphate (50–100 μg/ml), a drug effective in scrapie prophylaxis, dose-dependently attenuated the injury. MCEC cultures from mice homogenous for the disrupted PrP gene were resistant to the toxicity of PrP 106–126. In conclusion, cerebral endothelium expressing PrPC may be directly damaged during spongiform encephalopathies.

Escherichia coli 0111 B4 Lipopolysaccharide Given Intracisternally Induces Blood-Brain Barrier Opening during Experimental Neonatal Meningitis in Piglets

ABSTRACT: Neonatal bacterial meningitis remains a life-threatening infection, and severe neurologic sequelae may be left in survivors as well. The goal of the study was to develop and characterize a porcine model of the disease with intravital observation of the permeability changes in cerebral microvessels. Eighteen newborn piglets were given doses of 0 ng (group 1), 20 ng (group 2), and 200 ng (group 3) of Escherichia coli 0111 B4 endotoxin (LPS) intracisternally (n = 6 in each group). Cardiovascular parameters were without changes, but a compensated metabolic acidosis occurred in group 3 4 h after LPS injection. Using the open cranial window technique combined with fluorescence excitation, there was no blood-brain barrier leakage in pial-arachnoid microvessels for sodium fluorescein during the 4 h of experiments in group 1 piglets, whereas spotty extravasations occurred in group 2 and in group 3 after the LPS injections (70.5 ± 10.5 and 55.2 ± 4.1 min, respectively, mean ± SEM). A dose-dependent increase in sodium fluorescein uptake in brain regions examined (parietal and occipital cortex, cerebellum, and periventricular white matter) was also found by fluorescence spectropho-tometry. LPS-treated piglets had developed pleocytosis. Four h after the challenge, the white blood cell counts in cerebrospinal fluid were (mean ± SD): group 1, 8.2 ± 7.6 μL-1; group 2, 453 ± 703 μL-1; and group 3, 1027 ± 620 μL-1; respectively, whereas there was no change in white blood cell count of peripheral blood samples. Sixty min after the intracisternal injection, LPS content measured by Limulus amebocytes lysate assay was mildly elevated in cerebrospinal fluid in group 2 and group 3, whereas it was unchanged in sera in either group. We conclude that our model may be appropriate for further in vivo examination of the pathogenesis and treatment of neonatal bacterial meningitis.