A monoclonal antibody to the endothelium of rat brain microvessels

A rat brain fraction enriched with microvessels was used as the immunogen to produce mouse hybridoma cell lines secreting monoclonal antibodies. One of these antibodies, selected from 156 supernatants by enzyme-linked immunosorbent and immunofluorescent assays, reacted only with the endothelium of microvessels in the brain. The endothelium-specific antibody labelled the cytoplasm of microvascular endothelial cells, their luminal membranes, and an extracellular layer, the endocapillary coat, which covered the luminal surface of these cells. In the kidney, the antibody specifically stained the brush border of the proximal tubuli, and in the liver, the antibody specifically stained bile canaliculi. This demonstrates that 3 morphological structures with important transport functions, cerebral microvascular endothelium, brush border of kidney proximal tubuli, and liver bile canaliculi, express the same epitope.     

Ultrastructure of frog cerebral and pial microvessels and their impermeability to lanthanum ions

Permeabilities of single pial microvessels in the frog are now being estimated. Fine structural information on these vessels will therefore be needed in attempts to correlate structure and permeability. Microvessels from the dorsal part of the telencephalon and from the overlying pial layer in the common frog (Rana temporaria) and their permeability to lanthanum ions were studied in the electron microscope. Extensive structural similarities between the endothelial lining of cerebral and pial microvessels were demonstrated, indicating similarity in functional characteristics. Adjacent endothelial cells in both categories of vessels were joined by elaborate tight junctions. Lanthanum ions did not permeate the endothelium. Diffusion of the ion in the interendothelial clefts was stopped by the junctions. It is concluded that the frog has an endothelial blood-brain barrier which is tight to lanthanum ions.     

Penetration of neurohypophyseal hormones from plasma into cerebrospinal fluid (CSF): Half-times of disappearance of these neuropeptides from CSF

The penetration of neurohypophyseal peptides after peripheral administration into the cerebrospinal fluid (CSF) was studied in freely moving rats. In addition, the clearance of these peptides from CSF was investigated. Increased concentrations of vasopressin (AVP) in CSF were detectable 2 min after s.c. injection of 5.0 micrograms of this peptide. Peak concentration was reached at 5 min after administration and this level declined slowly over the next hour. Administration of 5.0 micrograms oxytocin (OXT) s.c. or i.v. resulted in increased OXT levels in CSF within 10 min after application. After 60 min a significant elevation of OXT in CSF was no longer present. These data reveal that approximately 0.002% of the peripherally applied amount of AVP or OXT reached the central nervous system at 10 min after injection. AVP (2.5 ng) and OXT (5.0 ng) applied into one of the lateral brain ventricles reached the cisternal cavity within 2 min after administration. Both neuropeptides were cleared from the CSF with terminal half-times of 26 and 19 min for AVP and OXT, respectively. The present data demonstrate that neurohypophyseal hormones do cross the blood-brain barrier in amounts obviously sufficient to induce central actions.     

A mechanism for vasopressin action in the hippocampus

The action of arginine vasopressin (AVP) in the rat hippocampal slice has been extensively studied. Extracellular recording indicates that the peptide excites spontaneously active neurons in the slice, though uncertainty exists regarding the identity of this cell type. Intracellular recording from pyramidal cells also reveals an excitatory action of the peptide, but these results are confounded by simultaneous constriction of small blood vessels that surround each pyramidal cell. Here we use field potential recordings to show that AVP inhibits pyramidal cell discharge and employs a pressor-type (V₁) receptor to bring about its action. The results resolve issues regarding the identity of AVP targets in the slice. Each reported result is consistent with a model that posits direct AVP excitation of inhibitory interneurons and direct AVP constriction of slice microvessels. Inhibition of pyramidal cells recorded extracellularly and excitation of pyramidal cells recorded intracellularly are respective indirect consequences of the two direct effects.     

Vasopressin V1 receptor in rat hippocampus is regulated by adrenocortical functions

Two subtypes of arginine vasopressin (AVP) receptors (V1 and V2) have been distinguished. In this study, we examined the characteristics of AVP binding in rat hippocampus and the effects of bilateral adrenalectomy and adrenal steroids on its [³H]AVP binding. [³H]AVP binding to rat liver and the hippocampal membranes was strongly inhibited by the V1 antagonist, OPC-21268. ADX resulted in a significant decrease in theB_max of AVP binding in the hippocampus. Chronic treatment with aldosterone and corticosterone restored the ADX-induced reduction, but treatment with dexamethasone did not. These results suggest that the AVP V1 receptor in the hippocampus is regulated by adrenocortical neuroregulatory function.     

Increased collagen content of cerebral microvessels in Alzheimer's disease

Based on previous evidence suggesting abnormalities in the brain microvasculature, we examined basement membrane collagen in isolated cerebral microvessels (CMV) from subjects with Alzheimer's disease (AD) and age-matched controls. Concentrations of hydroxyproline, the principal constituent of collagen IV, were significantly increased by 55% in CMV from AD subjects compared to controls. This result was corroborated by the finding of 60% increased total collagen content in CMV as evident by the selective binding of Sirius red dye. Hydroxyproline and collagen concentrations in samples of cerebral cortex assayed in parallel were 6–20 times smaller than in CMV and were not changed between controls and AD subjects. To further differentiate AD and control samples, fractions of CMV were solubilized and the pepsin digested collagen proteins resolved by SDS-PAGE. Upon immunoblotting, AD samples with increased collagen revealed proportionally greater specific immunoreactivities detected by antibodies to collagen IV. Our observations suggest altered collagen IV content of cerebral vessels in subjects with AD that may affect brain microvascular functions.     

Vasopressin is located within lymphocytes in the rat spleen

Using dual antigen immunocytochemical staining with a specific antiserum for arginine vasopressin (AVP), we have detected AVP immunoreactivity in clusters of large immunoglobulin (Ig) G-containing cells, probably plasma cells, within the rat spleen, and in smaller cells which are IgG-negative. Vasopressin-positive cells were detected principally throughout the white pulp areas in the subcortical region of the spleen. IgG staining could only be detected within the cells and not on the cell surface, demonstrating that the antiserum is recognising genuine intracellular IgG and not cell surface antigens. Reversed-phase HPLC of spleen tissue extract revealed a single peak of AVP immunoreactivity which co-eluted with the standard. This is the first evidence that AVP is found within lymphocytes of the immune system and provides further information about the important interaction between the endocrine and immune systems.     

Differential blood-brain barrier permeabilities to [14C]sucrose and [3H]inulin after osmotic opening in the rat

The blood-brain barrier (B-BB) in 3-month-old rats was opened unilaterally by infusing 1.8 ml(+)arabinose in water into the internal carotid artery through a catheter in the external carotid. Two poorly penetrating uncharged test radiotracers of differing molecular weight and size, [¹⁴C]sucrose (340 daltons, radius 5 Å) and [³H]inulin (5500 daltons, radius 15 Å), were simultaneously injected i.v. in untreated rats, or in rats at 1, 30, or 50 min after infusion of hypertonic arabinose solution. Evans-blue solution was injected 5 min prior to osmotic treatment as a visual indicator of barrier integrity. In regions of uninfused control brains, the [¹⁴C]sucrose permeability-surface area (PA) product approximated 10^?5 s^?1, whereas PA was not measurable for [³H]inulin. In arabinose-infused animals, PA products on the ipsilateral hemisphere for both [¹⁴C]sucrose and [³H]inulin were markedly elevated 6 min after infusion, but decreased by 35 and 55 min. In nearly all regions, statistically significant differences were not found between 6-min [¹⁴C]sucrose- and [³H]inulin-PA values (P > 0.05). However, at 35 and 55 min in most regions, the PA for [³H]inulin was significantly lower (P < 0.05) than PA for [¹⁴C]sucrose. The results indicated that the B-BB closed more rapidly to larger than to smaller molecules after osmotic treatment and were consistent with a pore model for osmotic B-BB opening.     

Immunofluorescent labeling of tight junctions in the rat brain and spinal cord

Tight junctions may play an important role in maintaining the integrity of the blood-brain barrier. These junctions can be individually visualized using electron microscopy but no current technique is able to provide a more global picture of the presence and density of tight junctions in central nervous system tissue. We used an antibody that recognizes a high molecular weight protein (ZO-1) associated with tight junctions, to identify these specialized junctions within the rat brain and spinal cord. Immunofluorescent labeling showed a network of tight junctions between cells in the brain vasculature, leptomeninges and choroid plexus, and between tanycytes lining the floor of the third ventricle and the central canal of the spinal cord. Anti-ZO-1 labeled the majority of cells associated with the blood-brain barrier and may prove a useful marker, possibly in conjunction with functional dye studies, in evaluating the anatomical and functional integrity of the blood-brain barrier.     

Vasopressin depolarizes lateral horn cells of the neonatal rat spinal cord in vitro

The effects of vasopressin (VP) on lateral horn cells including a number of sympathetic preganglionic neurons contained in thin in vitro slices of neonatal rat spinal cord were investigated by means of intracellular recording techniques. Superfusion of (Arg⁸)-vasopressin (AVP, 0.01–1 μM) caused a depolarization leading, in the majority of lateral horn cells, to repetitive discharges. The AVP depolarization which could be partially reduced by low Ca/high Mg solution or tetrodotoxin, was accompanied by an increase in membrane resistance and the response was nullified near the membrane potential at which the spike afterhyperpolarization was abolished. A clear reversal of the response was not observed upon further hyperpolarization. The AVP response was blocked by the VP₁ antagonist,d-(CH₂)₅ Tyr (Me)-AVP, whereas, deamino (d-Arg⁸-vasopressin), a VP₂ agonist, at high concentrations ( 50 μM) was e ineffective or produced a small depolarization. The results indicate that AVP, acting mainly on VP₁ receptors, excited lateral horn cells by a direct depolarization and an indirect effect via the release of an excitatory transmitter(s). A reduction of a voltage-dependent K conductance may underlie the depolarizing effect of AVP.     

Nervous tissue thiamine metabolism in vivo. I. Transport of thiamine and thiamine monophosphate from plasma to different brain regions of the rat

The transport of thiamine (T) and thiamine monophosphate (TMP) across the blood-brain barrier was measured in vivo in the rat. Different doses of [¹⁴C]T (15–550 nmol) and [¹⁴]TMP (11–110 nmol) were injected into the femoral vein. The content of T and its phosphoesters in blood and brain tissue (cerebellum, pons, medulla and cerebral cortex) 20 s after the injection was determined radiometrically after electrophoretic separation. Blood flow and blood volume in the same regions of the brain was also determined. Both T and TMP entered rapidly the cerebral tissue, where they were found chemically unmodified.The cerebral tissue extracted less than 7% of plasma T. At physiological plasma T concentrations, the rate of transport ranged from 0.43 to 0.65 nmol·g⁻¹·h⁻¹ with only minor differences among the various regions. T was transported into the nervous tissue by two separate mechanisms: one saturable, that at physiological plasma T levels accounted for 95% (cerebellum) to 91% (cerebral cortex) of the total T taken up, and one non-saturable, that was most efficient in the cerebral cortex. The K_m (half-saturation constant) of the former transport mechanism ranged from 1.95 to 2.75 nmol·ml⁻¹ in the 4 areas investigated. V_max (maximal transport rate) values ranged from 6 to 9 nmol·g⁻¹·h⁻¹, the highest value being found in the cerebellum. The overall transport rate of TMP was on average 5–10 times as low as that of T and also showed a saturable and a non-saturable component. Both components were slower than those observed for T.     

Adenylate cyclase in the microvessels of the rat brain

Summary The presence of adenylate cyclase (AC) in the microvessels of the rat brain was studied by a new histochemical method for light and electron microscopy. The method is based on the precipitation of strontium and the subsequent conversion of the formed strontium salt into lead phosphate. Isoproterenol and 5-guanylylimidodiphosphate were used as enzyme activators. In the light microscope, the final reaction product was detected in the choroid plexus as well as in the walls of the microvessels in the brain parenchyma. In the electron microscope, both the luminal and abluminal endothelial membrane as well as the basal lamina of the parenchymal microvessels displayed reaction product. The observations demonstrate that isoproterenol-stimulated AC is located in the endothelium of the rat brain microvessels and suggest that the enzyme may play a role in the receptor-mediated regulation of endothelial functions.Supported by a grant from the Finnish Academy     

Stimulation of tryptophan uptake into brain microvessels byd-glutamine

The uptake of L-tryptophan into isolated porcine microvessels is increased by preincubation with L-glutamine as well as with D-glutamine. This could indicate that gamma-glutamyltranspeptidase is involved in the stimulation of uptake of large neutral amino acids into the brain observed in hyperammonemic conditions.     

A light and electron microscopic study of divalent metal transporter-1 distribution in the rat hippocampus,after kainate-induced neuronal injury

An accumulation of iron occurs in the hippocampus of rats injected with kainate over time, but thus far whether this accumulation is associated with any changes in expression of iron transporters is not known. The present study was therefore carried out using an antibody to the divalent metal transporter-1 (DMT-1) and immunoblot and immunocytochemical analyses to elucidate possible changes in expression of the transporter in the rat hippocampus after kainate injections. A significant increase in density ratios of DMT-1/beta-actin bands was observed in Western blots in the 1-week, 1-month, and 2-months post-kainate-injected hippocampus, compared to uninjected and 1-day post-kainate-injected hippocampus. The increase in DMT-1 protein was paralleled by an increase in DMT-1 immunoreactivity in astrocytes. Light staining for DMT-1 was observed in the uninjected, saline-injected, and 1-day post-kainate-injected rat hippocampus. In contrast, an upregulation of DMT-1 was observed in reactive glial cells at 1 week, 1 month, and 2 months post-kainate injection. Electron microscopy confirmed that the glial cells had morphological features of astrocytes. DMT-1 is a cellular iron transporter responsible for transport of metal ions from the plasma membrane to endosomes. The observation that DMT-1 is present on astrocytic end feet in contact with blood vessels suggests that these cells may be involved in uptake of iron from endothelial cells.     

Quinolinic acid enhances permeability of rat brain microvessels to plasma albumin

Several studies have established that increased cerebrospinal fluid (CSF) levels of quinolinic acid (QUIN), a macrophage/microglia-derived excitotoxin with N-methyl-D-aspartate (NMDA)-receptor affinity, may reflect abnormal blood-brain barrier (BBB) function in patients with acquired immunodeficiency syndrome (AIDS) dementia complex, exhibiting a relationship to their clinical and neurological status. This study was aimed to evaluate whether QUIN (250 nmol/0.25 microl/ventricle) infused into both lateral cerebral ventricles permeates adult rat brain microvessels to plasma albumin. Possible BBB dysfunction was examined 4 days after the intracerebroventricular (i.c.v.) infusion of QUIN by measuring plasma albumin extravasation using rocket immunoelectrophoresis. The i.c.v. infusion of QUIN failed to increase the extracellular tissue concentration of albumin in the entorhinal cortex, but significantly higher levels were found in the hippocampus proper (but not in the subiculum region and dentate gyrus) and in the striatum. To evaluate the possible relationship between plasma protein extravasation and QUIN-induced tissue necrosis, we quantified neuronal death in the rat hippocampal formation (subiculum, CA1/CA3 areas of the hippocampus proper, dentate gyrus). We found significantly higher tissue levels of plasma albumin in the hippocampus proper, in which the CA1 area exhibited the highest neuronal loss while the low rate of neuronal death was not accompanied by significant albumin extravasation in the dentate gyrus. However, in case of the subiculum, in which the neuronal loss reached comparable values to those in the CA1 area, we did not find significant enhancement of plasma albumin leakage into this area. The regional differences in brain microvascular permeability may depend on the density of NMDA receptors in the multicellular capillary barrier, but the differences in neuronal death may also reflect an involvement of NMDA receptors in neuronal membranes. We conclude that increased CSF QUIN levels evoke a dysfunction of the BBB that may only partially be related to sites with pronounced neuronal damage in the rat brain regions susceptible to NMDA-receptor mediated toxicity.     

Is Vasopressin Preferentially Released from Corticotropin-Releasing Factor and Vasopressin Containing Nerve Terminals in the Median Eminence of Adrenalectomized Rats?

The effects of adrenalectomy (ADX) on the amounts of immunoreactive corticotropin-releasing factor (CRF_j) and arginine vasopressin (AVP_i) that are stored in the zona externa of the median eminence (ZEME) were investigated by means of quantitative immunocytochemistry. Although ADX of male Wistar rats for 1 week or 4 weeks did not affect CRF_i in the ZEME as compared to sham-operated or intact controls, AVP_i showed a progressive accumulation. The ratio of AVP_i over CRF_i in the ZEME had already increased 1 day after ADX. However, it should be noted that the exact changes in CRF_i and AVP_j as measured by radioimmunoassay and/or quantitative immunocytochemistry were dependent on the substrain of rats used. The secretion rate of CRF_i and AVP_j was estimated in 1 week and 4 week ADX rats, by measuring the disappearance rate of CRF_i and AVP_i from the ZEME after blockade of fast axonal transport, by a low non-toxic dose of colchicine (5 μg per rat). In contrast to intact rats, where this dose of colchicine did not affect CRF_i or AVP_i in the ZEME, ADX rats showed a progressive depletion of the CRF_i and AVP_i stores as measured 2.5 and 5 h later. In 1 week ADX rats, CRF_i and AVP_i both disappeared at a rate of 7% to 8% of their stores per hour. In contrast, after 4 weeks of ADX the fractional disappearance rates of CRF_i and AVP_i were different and were 3% and 8% of the content per hour, respectively. This indicates that in long-term ADX rats the chance of AVP being released from the ZEME is more than twice that of CRF. Since most, if not all, of the CRF; containing nerve terminals in the ZEME of ADX rats costore AVP, we hypothesize that in long-term ADX rats AVP may be preferentially secreted from AVP and CRF costoring nerve terminals.     

Metabolic GHB precursor succinate binds to gamma-hydroxybutyrate receptors: characterization of human basal ganglia areas nucleus accumbens and globus pallidus

Binding of the metabolic gamma-hydroxybutyrate (GHB) precursor succinate to NCS-382-sensitive [3H]GHB-labeled sites in crude synaptosomal or purified synaptic membrane fractions prepared from the human nucleus accumbens (NA), globus pallidus (GP) and rat forebrain has been shown. This site can be characterized by binding of ethyl hemisuccinate and gap-junction blockers, including carbenoxolone hemisuccinate and beta-GRA. There was no significant binding interaction between GABAB receptor ligands (CGP 55845, (R)-baclofen) and these [3H]GHB-labeled sites. GHB, NCS-382 and succinate binding profile of [3H]GHB-labeled sites in rat forebrain, human NA or GP synaptic membranes were similar. The synaptic fraction isolated from the rat forebrain was characterized by GHB binding inhibition constants: Ki,NCS-382 = 1.2 +/- 0.2 microM, Ki,GHB = 1.6 +/- 0.3 microM and Ki,SUCCINATE = 212 +/- 66 microM. In crude membranes containing mainly extrasynaptic membranes, distinct GHB and GABAB receptor sites were found in the NA. By contrast, extrasynaptic GABAB receptor sites of rat forebrain and GP were GHB- and succinate-sensitive, respectively. The heterogeneity of GABAB sites found in native membranes indicates GABAB receptor-dependent differences in GHB action. Based on these findings, we suggest that succinate (and possibly drugs available as succinate salt derivatives) can mimic some of the actions of GHB.