Indomethacin and cerebrovascular permeability to albumin in acute hypertension and cerebral embolism in the rat

Summary The experiments were performed to determine if indomethacin, a prostaglandin synthesis inhibitor, could reduce albumin extravasation and brain edema in some models of blood-brain barrier dysfunction.The blood pressure was increased by i.v. adrenaline or bicuculline in conscious rats with indwelling catheters in the aorta and jugular vein. ¹²⁵I-labeled serum albumin and Evans blue-albumin were used as tracers of the blood-brain barrier function. Pretreatment with indomethacin significantly reduced albumin extravasation after the administration of adrenaline but not after bicuculline, i.e. when acute hypertension was combined with a metabolically mediated cerebral vasodilatation. It is argued that the protective effect of indomethacin in adrenaline-induced hypertension probably is related to the vasoconstrictory effect of the drug.Five l of air or Lipiodol were injected into the right internal carotid artery in rats anesthetized with pentobarbitone. The albumin content in the injected hemisphere was seven to nine times higher after fat than after air embolism. No significant reduction of tracer extravasation was obtained in rats treated with indomethacin. Rats subjected to fat embolism had a significant homolateral cerebral edema (i.e. increased water content) which was not reduced by pretreatment with indomethacin. By contrast, the water content was significantly increased also in the non-injected side in rats given indomethacin indicating a larger spread of edema fluid in these animals.     

Capsaicin and blood-brain barrier permeability

The influence of perivascular sensory afferent nerve fibres upon the permeability of the blood-brain barrier was examined in halothane-anaesthetised rats. Blood-brain barrier permeability was assessed by the rate of transfer of the small neutral amino acid, [14C]alpha-aminoisobutyric acid, from blood to brain and by gross extravasation of Evans blue albumin. In every brain area examined, the acute administration of capsaicin (0.15-5 mumol/kg, i.v.) failed to alter significantly the passage into the brain of the small neutral amino acid tracer or to effect dye extravasation. Capsaicin, at concentrations which cause the release of vasoactive neuropeptides from nerve endings and increase vascular permeability in peripheral tissues, does not increase the permeability of the blood-brain barrier.     

Intracarotid hypothermic saline infusion: a new method for reversible blood-brain barrier disruption in anesthetized rats

An animal model for reversible blood-brain barrier disruption has been developed. Retrograde infusion of hypothermic saline solution (8 ± 1°C) into the left external carotid artery of normothermic, Wistar rats reversibly increases cerebrovascular permeability to Evans blue albumin in the left cerebral hemisphere. Isotonic saline solutions at 37°C for Group I and at 8 ± 1°C for Group II were infused for 30 s at a constant rate of 0.12 ml/s into the left external carotid artery. Evans blue, the barrier tracer, was administered intravenously either prior to or at intervals 5, 30, 180, 360 min after the hypothermic saline infusion under pentobarbital anesthesia. All animals receiving hypothermic saline perfusion had disturbed blood-brain barrier permeability. Based on visual inspection, disruption grade in the left hemispheres of 10 of 16 animals was 3+. Mean values for Evans blue dye were found to be 0.32 ± 0.08 mg% in left the hemisphere after normothermic saline infusion (Group 1), and 2.9 ± 0.4 mg% in the same hemisphere after hypothermic saline infusion (Group II). The difference was found to be significant between Group I and Group II (P < 0.001). The increase in cerebrovascular permeability was temporary, however, although Evans blue albumin extravasion remained slightly elevated 3 h after infusion, it was normal 6 h after infusion.     

Vasodilator Phosphostimulated Protein (VASP) Protects Endothelial Barrier Function During Hypoxia

The endothelial barrier controls the passage of solutes from the vascular space. This is achieved through active reorganization of the actin cytoskeleton. A central cytoskeletal protein involved into this is vasodilator-stimulated phosphoprotein (VASP). However, the functional role of endothelial VASP during hypoxia has not been thoroughly elucidated. We determined endothelial VASP expression through real-time PCR (Rt-PCR), immunhistochemistry, and Western blot analysis during hypoxia. VASP promoter studies were performed using a PGL3 firefly luciferase containing plasmid. Following approval by the local authorities, VASP ^−/− mice and littermate controls were subjected to normobaric hypoxia (8% O₂, 92% N₂) after intravenous injection of Evans blue dye. In in vitro studies, we found significant VASP repression in human microvascular and human umbilical vein endothelial cells through Rt-PCR, immunhistochemistry, and Western blot analysis. The VASP promoter construct demonstrated significant repression in response to hypoxia, which was abolished when the binding of hypoxia-inducible factor 1 alpha was excluded. Exposure of wild-type (WT) and VASP ^−/− animals to normobaric hypoxia for 4 h resulted in an increase in Evans blue tissue extravasation that was significantly increased in VASP ^−/− animals compared to WT controls. In summary, we demonstrate here that endothelial VASP holds significant importance for endothelial barrier properties during hypoxia.     

Effects of increasing arterial pressure on cerebral blood flow in the baboon: Influence of the sympathetic nervous system

The influence of stimulation of the cervical sympathetic chain on the response of cerebral blood flow to hypertension induced by the intravenous infusion of angiotensin was studied in anaesthetised baboons. Cerebral blood flow was measured by the intracarotid¹³³Xenon injection technique. Possible lesions of the blood-brain barrier were studied by injecting Evans blue towards the end of the experiment and ischaemic brain damage was assessed following perfusion fixation.In a control group of five baboons blood flow increased by 53±9% (mean ±S.E.) from the base line values in the arterial pressure range 130–159 mm Hg.In four baboons subjected to unilateral sympathetic stimulation flow increased by 16±4% in the same pressure range.In three babbons subjected to bilateral sympathetic stimulation there were not significant increases in flow until the arterial pressure had increased above 159 mm Hg.Disruption of the blood-brain barrier in the parietooccipital regions was only seen in the control animals but not in the stimulated baboons. Ischaemic brain damage was not observed with the exception of one small lesion in a single stimulated baboon.These findings provide strong support for the observations of Bill and Linder (1976) that activation of the cervical sympathetic can modify the level at which breakthrough of cerebral blood flow occurs in association with systemic hypertension.These investigations were supported by the Medical Research Council and Tenovus (Scotland)     

Local versus regional cerebral blood flow in the rat at high (hypoxia) and low (phenobarbital anesthesia) flow rates

Local cerebral blood flow (CBF) was measured in rats, using an autoradiographic technique with ¹⁴C-iodoantipyrine as diffusible tracer, in situations with low, normal and high flow rates (phenobarbital anesthesia, analgesia with 75% N₂O, and hypoxia, respectively). A comparison of the results with previous data obtained in conscious rats (Sakurada et al. 1978) demonstrates that 75 % N₂O moderately reduces local CBF in some, but not all, cortical and subcortical areas, that phenobarbital anesthesia reduces local CBF to between 30 and 65 % of (conscious) control, and that pronounced hypoxia (arterial Po₂ about 25 mmHg) increases local CBF 3- to 4-fold. A comparison of the values obtained for cortical structures with those previously measured with a technique based on the Fick principle shows that the autoradiographic technique gives similar values at low and normal flow rates but that it moderately underestimates CBF at high flow rates, probably due to diffusion limitation.     

Cerebral circulation in acute arterial hypertension—protective effects of sympathetic nervous activity

The cervical sympathetic chain was stimulated electrically at 6 or 3 Hz on one side in anesthetized cats. Acute arterial hypertension was induced by ligation of the aorta. Evans blue was given as tracer for protein leakage. The regional blood flow in the brain was determined by using labelled microspheres. At high blood pressures there was a multifocal breakdown of the blood-brain barrier. The regions with breakdown had 10–20 times the normal flow rates. With a maintained hypertension regions which were overperfused at 5 min were still overperfused at 10 min, but there was little addition of new overperfused areas. Normalization of the pressure resulted in almost twice the normal flow rates in previously overperfused regions. The breakdown of the blood-brain barrier was restricted to the non-stimulated side, or more marked on that side. The protective effect of the sympathetic stimulation lasted more than 10 min. The results indicate that acute arterial hypertension tends to cause forced and long-lasting vasodilation in some areas in the brain but regions which are resistant to the acute rise have an increase in the vascular tone. Sympathetic activity helps in developing this tone. Normalization of the blood pressure results in partial recovery of the vascular tone in previously overperfused regions and normalization in other areas.     

Influence of intravenously administered catecholamines on cerebral oxygen consumption and blood flow in the rat

In order to study effects of catecholamines on cerebral oxygen consumption (CMRo₂) and blood flow (CBF), rats maintained on 75 % N₂O and 25 % O₂, were infused i.v. with noradrenaline (2, 5, or 8 μpg. kg^-1. min^-1) or adrenaline (2 or 8, μg. kg^-1.min^-1) for 10 min before CBF and CMR_oz were measured. In about 50% of animals infused with 2–8, μg. kg^-1 min^-1 of noradrenaline, CMRo_z (and CBF) rose. However, there was no dose-dependent response, and CMRo₂, did not exceed 150% of control. The effects of noradrenaline in a dose of 5 μg. kg^-l. min^-1 on CMRo₂, and CBF were blocked by propranolol (2.5μg.kg^-1). In animals infused with adrenaline (8 μg.kg^-1.min^-1) CMRo₂, was doubled and, in many, CBF rose 4- to 6-fold. It is concluded that, when given in sufficient amounts, catecholamines have pronounced effects on cerebral metabolism and blood flow, the effects of adrenaline on CMRo₂, and CBF resembling those observed in status epilepticus.     

The effect of indomethacin on cerebral blood flow and oxygen consumption in the rat at normal and increased carbon dioxide tensions

The effect of the fatty acid cyclo-oxygenase inhibitor indomethacin on cerebral blood flow (CBF) and the metabolic rate for oxygen (CMRO₂) was studied in paralyzed and artificially ventilated rats. In normocapnic animals, the drug (10 mg·kg^-1i. v.) reduced CBF to 50% of control without a measurable effect on CMRO₂. During hypercapnia (Pa_CO2 70–80 mmHg) the increase in CBF was reduced by about 80% but CMRO₂ remained unchanged. Autoradiographic evaluation of local CBF in 20 brain structures indicated that the reduction in CBF was relatively uniform throughout the brain. Dose response curves showed that an effect on CBF was evident already at an indomethacin dose of 1 mg·kg^-1 and maximal effects were obtained with 3–5 mg·kg^-1. Following i. v. injection of the drug reduction in CBF was observed already after 10 s and the full response occurred after 1–2 min. It is concluded that metabolites of arachidonic acid, possibly mainly prostacyclin, are powerful modulators of normal cerebrovascular tone, and help to mediate the CBF response to increased CO₂ tensions. However, since indomethacin does not modify the circulatory response in other conditions with increased CBF these substances do not qualify as general coupling factors controlling CBF in physiological or pathological states.     

Kainic acid neurotoxicity: Characterization of blood-brain barrier damage

The alterations in the water, sodium (Na) and potassium (K) contents of the frontoparietal cortex, hippocampus and thalamus as well as the protein permeability of the blood-brain barrier were investigated in rats 4 h after systemic kainic acid administration. Increases in the water and Na contents and a decrease in the K content were observed together with Evans blue extravasation in the thalamus area indicating the development of vasogenic brain edema. Changes observed in the ion contents of the frontoparietal cortex and hippocampus may be due to a general cell membrane permeability damage but are not caused by a primary disturbance of the blood-brain barrier.     

Effects of sympathetic stimulation on cerebral and ocular blood flow Modification by hypertension,hypercapnia, acetazolamide,PGI2 and papaverine

The effect of unilateral, electrical stimulatio of the cervical sympathetic chain in rabbits anesthetized with pentobarbital sodium and vasodilated by hypercapnia, acetazolamide, papaverine or PGI₂ was investigated to determine to what extent the sympathetic nerves to the brain and the eye cause vasoconstriction and prevent overperfusion in previously vasodilated animals. Evans blue was given as a tracer for protein leakage. Blood flow determinations were made with the labelled microsphere method during normotension and acute arterial hypertension. Hypertension was induced by ligation of the thoracic aorta and in some animals metaraminol or angiotensin was also used. Acetazolamide caused a two to threefold increase in cerebral blood flow (CBF) and hypercapnia resulted in a fivefold increase. CBF was not markedly affected by papaverine or PGI₂. In the choroid plexus, the ciliary body and choroid, papaverine and hypercapnia caused significant blood flow increases on the control side. Sympathetic stimulation induced a 12 % blood flow reduction in the brain in normotensive, hypercapnic animals. Marked effects of sympathetic stimulation at normotension were obtained under all conditions in the eye. In the hypertensive state the CBF reduction during sympathetic stimulation was moderate, but highly significant in hypercapnic or papaverine-treated animals as well as in controls. Leakage of Evans blue was more frequently seen on the nonstimulated side of the brain. In the eye there was leakage only on the control side except in PGI₂-treated animals where 2 rabbits had bilateral leakage. The effect of sympathetic stimulation on the blood flow in the cerebrum and cerebellum in vasodilated animals seems to be small or absent if the blood pressure is normal. In the eye pronounced vasoconstriction occurs under these conditions. In acute arterial hypertension sympathetic stimulation protects both the cerebral and ocular barriers even under conditions of marked vasodilation.     

Changes in the permeability of the blood-brain barrier following sodium dodecyl sulphate administration in the rat

 The blood-brain barrier (BBB) arises from epithelial-like tight junctions that virtually cement adjoining capillary endothelium together in the brain microvascolature. Several experimental manipulations have been shown able to increase the permeability of brain capillaries, by altering endothelial cell membrane integrity or activating specific biochemical pathways involved in regulation of BBB functionality. Because of its amphiphilic nature, sodium dodecyl sulphate (an anionic surfactant widely used as solubilizer or stabilizer in several pharmaceutical preparations; SDS) may enter into interactions with the major membrane components, which are lipids and proteins. The aim of the present study was to determine the effect of an intracarotid infusion of SDS (25, 50 and 100 μg/kg; infusion rate: 3 ml/min for 30 s) on the functionality of the BBB in the rat. An extensive, dose-dependent Evans blue extravasation was observed, in the ipsilateral brain hemisphere, 15 min following SDS infusion. These results were confirmed by the significant increase in [¹⁴C]α-aminoisobutyric acid ([¹⁴C]AIB) transport (evaluated by calculating a unidirectional transfer constant, K _i, for the tracer from blood to brain) measured in several ipsilateral brain regions 2 min after SDS infusion; this SDS-elicited BBB opening to [¹⁴C]AIB proved to be reversible. Since the BBB is created by the plasma membrane and tight junctions of the endothelial cells, the change in BBB permeability caused by SDS might be explained as a nonspecific surfactant-membrane interaction. Furthermore, SDS might affect the functional characteristics of brain vascular endothelial cells by an interaction with specific BBB proteins and/or biochemical pathways. In conclusion, one can suggest that intracarotid infusion of SDS might provide a useful clinical approach for the intentional introduction of different substances into the brain. On the other hand, these findings should call attention to possible dangerous consequences of using SDS as solubilizer in drug excipients. Received: 14 June 1996 / Accepted: 3 February 1997     

Cerebrovascular sympathetic denervation and blood-brain barrier function in conscious rats

Electrical stimulation of the sympathetic nerves to the cerebrovascular bed enables the resistance vessels to better withstand a high blood pressure in terms of blood-brain barrier integrity. Sympathetic denervation could hence be expected to lead to a decrease in cerebrovascular tone and increased vulnerability of the blood-brain barrier. In the present study acute hypertension was induced in conscious unrestrained rats by administration of angiotensin or bicuculline. The albumin leakage into the brain, as studied by Evans blue-albumin and ¹²⁵I labelled human serum albumin. was not enhanced in acutely or chronically sympathectomized rats compared to controls.     

Effects of lactoferrin on rat dermal interstitial fluid pressure (Pif) and in vitro endothelial barrier function

We recently demonstrated that intravenous (i.v.) injection of the iron‐binding protein lactoferrin (Lf) followed by antilactoferrin (aLf) antibodies or iron‐saturated Lf alone increased albumin extravasation in vivo in several tissues including skin. Increased driving pressure for blood‐tissue exchange or direct effects of Lf on the endothelial barrier are possible mechanisms. We therefore, firstly, measured interstitial fluid pressure (P_if) in dermis of rats given 1 mg Lf i.v. followed 30 min later by aLf or saline and circulatory arrest 1 or 5 min thereafter and compared with controls. Secondly, transmonolayer passage of Evans blue labelled albumin (EB‐albumin) was evaluated in porcine pulmonary artery endothelial cells exposed to iron‐free or iron‐saturated Lf (both 100 μg mL^–1) in the absence and presence of 0.5 mM hydrogen peroxide. P_if increased significantly at 11–30 min following Lf to +2.1 ± 0.3 and +1.7 ± 0.2 mmHg at 11–20 and 21–30 min, respectively, compared with +0.1 ± 0.2 mmHg before Lf (P < 0.05, n=25). Endothelial transmonolayer passage of EB‐albumin during 3 h was not affected by iron‐free or iron‐saturated Lf neither in the absence nor presence of hydrogen peroxide that increased passage 3.5 times compared with controls. In conclusion, Lf‐induced increase in albumin extravasation in rat skin is not explained by changes in P_if (because Lf raised P_if significantly) or direct effects of Lf on the endothelial barrier.     

Computerised image analysis in conjunction with fluorescence microscopy for the study of blood-brain barrier permeability in vivo

The present paper describes a new method using computerised image analysis techniques for quantification of tracer extravasation over the blood-brain barrier as studied by intravital fluorescence microscopy. Cats were equipped with an open cranial window and continuously infused with fluorescein isothiocyanate-labelled dextran (FITC-dextran, mol. wt. 70 000) to maintain a steady plasma concentration. Several cortical fields were recorded in each experiment and the images stored on video tape for off-line analysis. This procedure, which largely eliminates the superficial pial vasculature and allows extraction of the extravasation areas, consists of the following steps: (1) averaging of images, (2) software shading correction based on the original images for compensation of optical non-uniformity, (3) correction of displacement artefacts, (4) intensity adjustment, (5) generation of subtraction images by subtracting the first image of a series from the subsequent ones, (6) median filtering and thresholding, (7) a length recognition algorithm, and (8) elimination of small areas. Compared to the previously described method, step (2) has been newly developed and steps (4) and (8) added to enhance sensitivity for detecting tracer extravasation. The degree of extravasation in a cortical field at a given time point [E(f) value] was calculated as the mean intensity of the remaining pixels. TheE(f) is a quantitative value computed by a fully automatised procedure which takes into account the number, as well as the size and intensity, of extravasation areas in a given cortical field. TheE(f) values obtained at different times in a series of experiments were averaged to give theE(I) value. TheE(I) value did not alter when hypercapnia was employed to induce pure vasodilatation. On the other hand it increased dramatically, indicating tracer extravasation, during topical application of high concentrations of adenosine (10^–5–10^–3 M). The new computerised image analysis procedure may therefore be suitable for measuring quantitatively tracer extravasation over the blood-brain barrier in vivo under different experimental conditions. It may also be applicable to study changes of vascular permeability in peripheral vascular beds.     

Repetitive measurements of intracranial pressure in awake rabbits

Previous measurements of the intracranial pressure in experimental animals suffer from acute and subacute effects of cannulation. In order to obtain reliable, repetitive or continuous values, we measured the intracranial pressures in awake rabbits with a new permanent adjustable ventricular cannula that included a separate entrance to the subarachnoid space. The mean intraventricular pressure ten days after the operation was 5.2 +/- 1.1 mmHg (SD) (70 animals). Manipulation of the cannula system and infusion of artificial cerebrospinal fluid did not damage the blood-brain barrier (indicated by extravasation of Evans Blue). The intracranial pressure was constant for as long as 6 months and as many as 22 separate measurements and infusions. The cerebrospinal fluid cells and protein content did not change in animals with permanent cannulae and in animals perfused with 2-4 ml artificial cerebrospinal fluid. In 30 animals the ventricular cannula functioned for 10-180 (median 65) days and the subarachnoid entrance for 11-23 (median 16) days.     

Sympathetic Control of Cerebral Blood Flow in Acute Arterial Hypertension

The cervical sympathetic chain on one side was stimulated electrically at 10–20 Hz and an acute rise in arterial blood pressure was produced by: intravenous injection of angiotensin, ligation of the thoracic aorta, or ligation of the aorta combined with injection of metaraminol. The blood flow through the cerebrum and the cerebellum was determined by using labelled microspheres. At high blood pressures there was multifocal breakdown of the blood-brain barrier in the cerebrum as indicated by leakage of Evans blue. The breakdown was restricted to the control side or much more marked on that side than on the stimulated side. Sympathetic stimulation prevented also breakdown of the blood-aqueous barrier. The blood flow through the cerebrum on the control side was higher than that on the stimulated side in all experiments. Regions with breakdown of the blood-brain barrier had flow rates which were about 10 times normal values. Cerebellar blood flow was less affected by the hypertension and did not react significantly to sympathetic stimulation. The results indicate that stimulation of the sympathetic nerves to the brain tends to prevent forced dilatation of the arterioles with a resulting regional overperfusion with blood and breakdown of the blood-brain barrier. It is concluded that one role of the sympathetic nerves supplying the brain is to extend the pressure region with autoregulation in its upper part under conditions of a general increase in sympathetic vasomotor activity.     

A Novel Method for Efficient Drug Delivery

Local delivery of anti-thrombotic and anti-restenotic drugs is desired to achieve high concentrations of agents which may be rapidly degraded systemically or which exhibit very short half-lives in vivo. In this article, the operating characteristics of a novel local drug delivery method are described and its effectiveness demonstrated computationally and experimentally. Computational models used a finite volume method to determine the concentration field. Optical dye density measurements of Evans blue in saline were performed in an in vitro steady flow system. Modeling parameters were kept in the physiologic range. Experimental flow visualization studies demonstrated high concentrations of infusate near the vessel wall. Computational studies predicted high, clinically significant drug concentrations along the wall downstream of the infusion device. When the radial infusion velocity is large (infusion flow rate, Q_inf > 0.5% of the main flow rate, Q), the wall concentration of the infused drug remains high, e.g., levels are greater than 80% of the infusate concentration 5 cm downstream of the infusion device. At lower infusion rates (Q_inf < 0.001Q), the drug concentration at the wall decreases exponentially with axial distance to less than 25% of the infusate concentration 5 cm downstream of the infusion device, although therapeutic drug levels are still readily maintained. The near wall drug concentration is a function of flow conditions, infusion rate, and the drug diffusivity. Good agreement was obtained between computational and experimental concentration measurements. Flow simulation and experimental results indicate that the technique can effectively sustain high local drug concentrations for inhibition of thrombosis and vascular lesion formation.     

Sympathetic control of cerebral blood flow in acute arterial hypertension.

The cervical sympathetic chain on one side was stimulated electrically at 10-20 Hz and an acute rise in arterial blood pressure was produced by: intravenous injection of angiotensin, ligation of the thoracic aorta, or ligation of the aorta combined with injection of metaraminol. The blood flow through the cerebrum and the cerebellum was determined by using labelled microspheres. At high blood pressures there was multifocal breakdown of the blood-brain barrier in the cerebrum as indicated by leakage of Evans blue. The breakdown was restricted to the control side or much more marked on that side than on the stimulated side. Sympathetic stimulation prevented also breakdown of the blood-aqueous barrier. The blood flow through the cerebrum on the control side was higher than that on the stimulated side in all experiments. Regions with breakdown of the blood-brain barrier had flow rates which were about 10 times normal values. Cerebellar blood flow was less affected by the hypertension and did not react significantly to sympathetic stimulation. The results indicate that stimulation of the sympathetic nerves to the brain tends to prevent forced dilatation of the arterioles with a resulting regional overperfusion with blood and breakdown of the blood-brain barrier. It is concluded that one role of the sympathetic nerves supplying the brain is to extend the pressure region with autoregulation in its upper part under conditions of a general increase in sympathetic vasomotor activity.     

Regional brain uptake of noradrenaline following mechanical or osmotic opening of the blood-brain barrier

The passage of noradrenaline from the cerebrovascular circulation into the brain (vessel walls and parenchyma) was studied quantitatively and regionally by a modification of Oldendorf's technique for determination of brain uptake index. Using 14C-ethanol as the highly diffusible internal standard, the index for noradrenaline varied between 2.7 and 4.5% in different regions, confirming the poor penetration of this neurotransmitter. The barrier was impaired transiently as evidenced by Evans blue extravasation, through osmotic opening by internal carotid injection of a hyperosmolar urea solution or mechanical disruption by a short-lasting elevation of the intracarotid hydrostatic pressure. This resulted in a 3-4-fold increase in the passage of noradrenaline from the circulation into the brain.