In vitro models for the blood-brain barrier.

The aim of the present study was to identify a model for the blood-brain barrier based on the use of a continuous cell line, and to investigate the specificity of this model. A set of test compounds, reflecting different transport mechanisms and different degrees of permeability, as well as different physiochemical properties was selected. In vivo data for transport across the blood-brain barrier of this set of test compounds was generated as part of the study using two different in vivo models. A computational prediction model was also developed, based on 74 proprietary Pharmacia compounds, previously tested in one of the in vivo models. Molsurf descriptors were calculated and 21 descriptors were correlated with log(Brain(conc.)/Plasma(conc.)) using partial least squares projection to latent structures (PLS). However, the correlation between predicted and measured values was found to be rather low and differed between one and two log units for several of the compounds. The test compounds were analyzed in vitro using primary bovine and human brain endothelial cells co-cultured with astrocytes, and also using two different immortalized brain endothelial cell lines, one originating from rat and one from mouse. Cell models using cells not derived from the blood-brain barrier, ECV/C6, MDCK and Caco-2 cell lines, were also used. No linear correlation between in vivo and in vitro permeability was found for any of the in vitro models when all compounds were included in the analysis. The highest r2 values were seen in the bovine brain endothelial cells (r2=0.43) and MDCKwt (r2=0.46) cell models. Higher correlations were seen when only passively transported compounds were included in the analysis, bovine brain endothelial cells (r2=0.74), MDCKwt (r2=0.65) and Caco-2 (r2=0.86). By plotting in vivo Papp values against logDpH7.4 it was possible to classify compounds into four different classes: (1) compounds crossing the blood-brain barrier by passive diffusion, (2) compounds crossing the blood-brain barrier by blood-flow limited passive diffusion, (3) compounds crossing the blood-brain barrier by carrier mediated influx, and (4) compounds being actively excreted from the brain by active efflux. Papp and Pe values obtained using the different in vitro models were also plotted against logDpH7.4 and compared to the plot obtained when in vivo Papp values were used. Several of the in vitro models could distinguish between passively distributed compounds and efflux substrates. Of the cell lines included in the present study, the MDCKmdr-1 cell line gave the best separation of passively and effluxed compounds. Ratios between AUC in brain and AUC in blood were also calculated for six of the compounds and compared to ratios between Pe or Papp for transport in the apical to basolateral and basolateral to apical direction. Again the MDCKmdr-1 cell line gave the best correlation with only one compound (AZT) giving large discrepancy between in vitro and in vivo data. None of the in vitro models could identify compounds known to be substrates for carrier mediated influxed as such, and the results indicate that a tighter in vitro blood-brain barrier model probably is needed in order to facilitate studies on carrier mediated influx. The findings presented also indicate that identification of "batteries" of in vitro tests are likely to be necessary in order to improve in vitro-in vivo correlations and to make it possible to perform acceptable predictions of in vivo brain distributions from in vitro data.

Determination of the affinity of drugs toward serum albumin by measurement of the quenching of the intrinsic tryptophan fluorescence of the protein.

Binding of new chemical entities to serum proteins is an issue confronting pharmaceutical companies during development of potential therapeutic agents. Most drugs bind to the most abundant plasma protein, human serum albumin (HSA), at two major binding sites. Excepting fluorescence spectroscopy, existing methods for assaying drug binding to serum albumin are insensitive to higher-affinity compounds and can be labour-intensive, time-consuming, and usually require compound-specific assays. This led us to examine alternative ways to measure drug-albumin interaction. One method described here uses fluorescence quenching of the single tryptophan (Trp) residue in HSA excited at 295 nm to measure drug-binding affinity. Unfortunately, many compounds absorb, fluoresce, or both, in this UV wavelength region of the spectrum. Several types of binding phenomenon and spectral interference were identified by use of six structurally unrelated compounds and the equations necessary to make corrections mathematically were derived and applied to calculate binding constants accurately. The general cases were: direct quenching of Trp fluorescence by optically transparent ligands with low or high affinities; binding of optically transparent, non-fluorescent ligands to two specific sites where both sites or only one site result in Trp fluorescence quenching; and chromophores whose absorption either overlaps the Trp emission and quenches by energy transfer or absorbs light at the Trp fluorescence excitation wavelength producing absorptive screening as well as fluorescence quenching. Unless identification of the site specificity of drug binding to serum albumin is desired, quenching of the Trp fluorescence of albumin by titration with ligand is a rapid and facile method for determining the binding affinities of drugs for serum albumin.

Increased lipophilicity and subsequent cell partitioning decrease passive transcellular diffusion of novel, highly lipophilic antioxidants.

Oxidative stress is considered a cause or propagator of acute and chronic disorders of the central nervous system. Novel 2, 4-diamino-pyrrolo[2,3-d]pyrimidines are potent inhibitors of iron-dependent lipid peroxidation, are cytoprotective in cell culture models of oxidative injury, and are neuroprotective in brain injury and ischemia models. The selection of lead candidates from this series required that they reach target cells deep within brain tissue in efficacious amounts after oral dosing. A homologous series of 26 highly lipophilic pyrrolopyrimidines was examined using cultured cell monolayers to understand the structure-permeability relationship and to use this information to predict brain penetration and residence time. Pyrrolopyrimidines were shown to be a more permeable structural class of membrane-interactive antioxidants where transepithelial permeability was inversely related to lipophilicity or to cell partitioning. Pyrrole substitutions influence cell partitioning where bulky hydrophobic groups increased partitioning and decreased permeability and smaller hydrophobic groups and more hydrophilic groups, especially those capable of weak hydrogen bonding, decreased partitioning, and increased permeability. Transmonolayer diffusion for these membrane-interactive antioxidants was limited mostly by desorption from the receiver-side membrane into the buffer. Thus, in this case, these in vitro cell monolayer models do not adequately mimic the in vivo situation by underestimating in vivo bioavailability of highly lipophilic compounds unless acceptors, such as serum proteins, are added to the receiving buffer.

Novel, highly lipophilic antioxidants readily diffuse across the blood-brain barrier and access intracellular sites.

In an accompanying article, an in vitro assay for permeability predicts that membrane-protective, antioxidant 2,4-diamino-pyrrolo[2, 3-d]pyrimidines should have improved blood-brain barrier (BBB) permeation over previously described lipophilic antioxidants. Using a first-pass extraction method and brain/plasma quantification, we show here that two of the pyrrolopyrimidines, one of which is markedly less permeable, readily partition into rat brain. The efficiency of extraction was dependent on serum protein binding, and in situ efflux confirms the in vitro data showing that PNU-87663 is retained in brain longer than PNU-89843. By exploiting inherent fluorescence properties of PNU-87663, its distribution within brain and within cells in culture was demonstrated using confocal scanning laser microscopy. PNU-87663 rapidly partitioned into the cell membrane and equilibrates with cytoplasmic compartments via passive diffusion. Although partitioning of PNU-87663 favors intracytoplasmic lipid storage droplets, the compound was readily exchangeable as shown by efflux of compound from cells to buffer when protein was present. The results demonstrated that pyrrolopyrimidines were well suited for quickly accessing target cells within the central nervous system as well as in other target tissues.

Equilibrium distribution of HIV antiviral drugs into human peripheral blood mononuclear cells (PBMC) is controlled by free drug concentration in the extracellular medium.

Effect of protein binding on the equilibrium distribution of selected HIV antiviral drugs into isolated human peripheral blood mononuclear cells (PBMC, mainly lymphocytes) was investigated. Human PBMC from a single healthy human donor were isolated, purified, and cryopreserved. Uptake of non-peptide HIV-1 protease inhibitors PNU-96988 and PNU-103017 by these cells in vitro was evaluated as a function of increasing concentration of human serum in the cell incubation media. Both PNU-96988 and PNU-103017 were extensively bound to serum proteins. Uptake/efflux kinetics were very rapid such that accumulation by the cells was thermodynamically, not kinetically, controlled. Accumulation by human PBMCs in vitro was directly proportional to the free and not the total drug concentration in the media. For comparative purposes, the serum protein binding effect on the distribution of two HIV reverse transcriptase (RT) inhibitors, delavirdine (RESCRIPTOR) and zidovudine (AZT), was also evaluated. Like the HIV-1 protease inhibitors, delavirdine was found to be extensively associated with serum proteins and its accumulation by human PBMCs in vitro to be proportional to the free and not total drug concentration. In contrast, AZT was not bound to serum proteins to any significant extent. The uptake of this drug by human PBMCs in vitro was independent of serum concentration. However, the intrinsic cellular accumulation of PNU-96988, PNU-103017 and delavirdine were all greater than AZT. Thus, the extent to which drugs uptake by cells is affected by serum appears proportional to the binding affinity of the serum proteins for the drug.

Expression of a P-selectin ligand in zona pellucida of porcine oocytes and P-selectin on acrosomal membrane of porcine sperm cells. Potential implications for their involvement in sperm-egg interactions.

The selectin family of cell adhesion molecules mediates initial leukocyte adhesion to vascular endothelial cells at sites of inflammation. O-glycan structural similarities between oligosaccharides from human leukocyte P-selectin glycoprotein ligand-1 (PSGL-1) and from zona pellucida glycoproteins of porcine oocytes indicate the possible existence of a P-selectin ligand in the zona pellucida. Here, using biochemical as well as morphological approaches, we demonstrate that a P-selectin ligand is expressed in the porcine zona pellucida. In addition, a search for a specific receptor for this ligand leads to the identification of P-selectin on the acrosomal membrane of porcine sperm cells. In vitro binding of porcine acrosome-reacted sperm cells to oocytes was found to be Ca2+ dependent and inhibitable with either P-selectin, P-selectin receptor-globulin, or leukocyte adhesion blocking antibodies against P-selectin and PSGL-1. Moreover, porcine sperm cells were found to be capable of binding to human promyeloid cell line HL-60. Taken together, our findings implicate a potential role for the oocyte P-selectin ligand and the sperm P-selectin in porcine sperm-egg interactions.

Analytical and numerical techniques for the evaluation of free radical damage in cultured cells using scanning laser microscopy.

Scanning laser microscopy (SLM) was used to develop an assay to visualize the generation of intracellular reactive oxygen species (ROS) and to evaluate the effect of the lipophilic antioxidant U-87,663 on ROS formation. Cultured N18 neuroglioma cells were challenged by extracellular addition of cumene hydroperoxide, and subsequent intracellular generation of ROS was characterized by measuring the fluorescence intensity of the ROS indicator 2',7'-dichlorofluorescein (DCF). The kinetics of the reaction between ROS and the indicator DCF, or the antioxidant U-87,663, can be most accurately assessed if results from individual cell clusters are analyzed independently. It is possible and necessary to account for the these experimental and analytical properties in order to characterize the properties of the antioxidant activity precisely. We determined that the temporal increase in DCF fluorescence was consistent with the reaction of DCF with free radicals generated from cumene hydroperoxide, as was the loss of fluorescence from U-87,663. The rate constants for the free radical reactions revealed that ROS reaction with DCF is 10 times faster than with U-87,663. These differences in reaction rates combined with differences in the cellular distribution of the ROS indicator DCF, the antioxidant U-87,663, and the bulk of the ROS prevented detection of any protection of U-87,663 may offer.

Signal transduction and glial cell modulation of cultured brain microvessel endothelial cell tight junctions.

Noncontact coculture of postconfluent bovine brain microvessel endothelial cell (BMEC) monolayers with rat C6 glioma cells results in markedly decreased transmonolayer permeability measured by transendothelial electrical resistance (TER) and solute flux. An elevation in adenosine 3',5'-cyclic monophosphate (cAMP) in response to forskolin correlates with an increase in TER through a threshold event; however, unlike forskolin, the severalfold increase in TER induced by C6 cells is cAMP independent. Activation of protein kinase C (PKC) enhances the C6 cell-induced increase in TER, and PKC inhibition blocks C6 cell induction. Treatment of control or C6 cell-induced BMEC monolayers with pertussis toxin immediately and irreversibly obliterates TER, without an apparent change in guanosine 3',5'-cyclic monophosphate levels or viability, indicating the importance of a G protein-mediated event. A similar effect is observed with transforming growth factor-beta 1, and both responses are polarized, since the loss in TER is significantly faster in BMEC monolayers exposed basolaterally. These results suggest that astroglia modulate the blood-brain barrier through a cAMP-independent PKC-dependent pathway maintained by a G protein-coupled mechanism.

Neuroprotective efficacy of microvascularly-localized versus brain-penetrating antioxidants.

The 21-aminosteroid (lazaroid) tirilazad mesylate has been demonstrated to be a potent inhibitor of lipid peroxidation and to reduce traumatic and ischemic damage in a number of experimental models. Currently, tirilazad is being actively investigated in phase III clinical trials in head and spinal cord injury, ischemic stroke and subarachnoid hemorrhage. This compound acts in large part to protect the microvascular endothelium and consequently to maintain normal blood-brain barrier (BBB) permeability and cerebral blood flow autoregulatory mechanisms. However, due to its limited penetration into brain parenchyma, tirilazad has generally failed to affect delayed neuronal damage to the selectively vulnerable hippocampal CA1 and striatal regions. Recently, we have discovered a new group of antioxidant compounds, the pyrrolopyrimidines, which possess significantly improved ability to penetrate the BBB and gain direct access to neural tissue. Several compounds in the series, such as U-101033E, have demonstrated greater ability to protect the CA1 region in the gerbil transient forebrain ischemia model with a post-ischemic therapeutic window of at least four hours. In addition, U-101033E has been found to reduce infarct size in the mouse permanent middle cerebral artery occlusion model in contrast to tirilazad which is minimally effective. These results suggest that antioxidant compounds with improved brain parenchymal penetration are better able to limit certain types of ischemic brain damage compared to those which are localized in the cerebral microvasculature. On the other hand, microvascularly-localized agents like tirilazad appear to have better ability to limit BBB damage.

Passive diffusion of weak organic electrolytes across Caco-2 cell monolayers: uncoupling the contributions of hydrodynamic, transcellular, and paracellular barriers.

A systematic approach was used to demonstrate the quantitative interplay of pH, pKa, lipophilicity, charged and uncharged molecular species, molecular size, aqueous diffusivity, and stirring in passive transport across the aqueous boundary layer, microporous filter support, and transcellular and paracellular barriers in Caco-2 cell monolayers. The relationship of permeability of the aqueous boundary layer and hydrodynamic stirring was elucidated from transmonolayer fluxes of testosterone. Adrenergic receptor antagonists including propranolol (PPL), alprenolol (APL), pindolol (PDL), and atenolol (ATL) represented the model series of structurally similar weak bases with pKa values between 8.8 and 9.65. Although intrinsically lipophilic, their apparent log PC (n-octanol/water) at pH 7.4 and 6.5 ranged from -2.6 to 1.3. Effective permeability coefficients (Pe) correlated with log PC at both pH 7.4 and 6.5 showing a single sigmoidal-like curve: PPL > APL > PDL > or = ATL. The Pe approached a minimum plateau value established by the protonated ATL for the paracellular route (pore radius of 12 A) by molecular size-restricted diffusion within a negative electrostatic field of force. The Pe of the weak bases was delineated into component permeability coefficients of the aqueous boundary layer and porous filter support, the intrinsic permeabilities of charged and uncharged species for the transcellular and paracellular routes, and the extent to which the routes were utilized at each pH. This study emphasized a generally applicable approach to quantitatively analyze passive transport data on weak organic electrolytes and neutral molecules generated using cell culture monolayers.

A general, wide-rage spectrofluorometric method for measuring the site-specific affinities of drugs toward human serum albumin.

Binding of drugs to serum albumin is one of the most important pharmacokinetic determinants and the design of drugs should take advantage of this property. In the present work, the fluorescent ligands Warfarin and dansylsulfonamide were used as probes of IIA site of human albumin and dansylsarcosine as the probe of the IIIA site. From the changes in fluorescence upon binding at 37 degrees C, pH 7.4, the following dissociation constants were determined: Warfarin, 3.43 +/- 0.69 microM; dansylsulfonamide, 7.57 +/- 0.88 microM; and dansylsarcosine, 6.06 +/- 1.09 microM. Nonfluorescent ligands displace these probes competitively and the type of probe displaced identifies the site specificity of the ligands. Nonlinear least-squares analysis of the decrease in fluorescence accompanying the displacement yields the stoichiometry and the dissociation constant may also be estimated rapidly from displacement at a single competitor concentration. The method yields reliable Kd values for at least the range of 0.2 to 100 microM. Representative dissociation constants for the IIA site-specific ligands are as follows: phenylbutazone, 1.9 +/- 0.3 microM; U-99,499, 1.8 +/- 0.2 microM; U-96,988, 5.3 +/- 1.5 microM; and U-105,665, 42 +/- 7 microM. For the IIIA site we find the following Kd values: oxazepam, 27.7 +/- 2.1 microM; diazepam, 7.7 +/- 1.0 microM; and ibuprofen, 2.7 +/- 1.2 microM. The method is eminently suitable for large-scale screening.

Quantitative approaches to delineate paracellular diffusion in cultured epithelial cell monolayers.

When using cultured cell monolayers to determine the mechanism of transcellular diffusion of molecules, it may be important to identify the fraction that moves through the paracellular route or passively diffuses through tight junctions. We characterized the apparent diameter of the junctional pore in a variety of epithelial cell monolayers (Caco-2, MDCK, alveolar). Using hydrophilic extracellular permeants varying in molecular radii and charge (neutral, anionic, cationic, zwitterionic), rate-determining steps and factors of the paracellular route were quantitatively delineated by the model for molecular size-restricted diffusion within a negative electrostatic field of force. Protonated amines permeated the pores faster than their neutral images while organic anions were slower. With increasing molecular size the influence of charge diminished. This approach was used to quantify the relationship between permeant radius and transepithelial electrical resistance and to analyze changes in junctional pore size as a function of pharmacological perturbation, such as in the use of absorption promoters or adjuvants.

Expression of a widely distributed, novel sulfated membrane glycoprotein by bovine endothelia and certain transporting epithelia.

We generated monoclonal antibodies (MAbs) against cultured bovine brain microvessel endothelial cells (BMEC) for use as probes to study membrane protein traffic and polarity. One MAb recognized a heterogeneous family of acidic sulfoglycoproteins called gp4A4 with molecular weights of 50-65 KD and 85 KD. Gp4A4 is a long-lived integral membrane protein which resides mostly at the plasma membrane, and a portion appears to be in equilibrium with an intracellular pool via endocytosis. Gp4A4 is expressed by many endothelial cells, except for fenestrated capillaries in choroid plexus, and specific epithelial cells in bile duct, kidney, and choroid plexus. A comparison of indirect immunoperoxidase and immunofluorescence detection using semi-thin cryosections gave contrasting results on the apparent distribution of gp4A4 on the apical and basolateral membranes of cerebral endothelia and choroid plexus epithelia. Immunogold labeling of ultra-thin cryosections showed that gp4A4 was expressed by the apical and basolateral membrane domains of BMEC and choroid plexus epithelia. This was consistent with the results using indirect immunofluorescence microscopy. On an average, gp4A4 expression by cerebral endothelia was not asymmetric and was considerably variable between capillaries. These results emphasize the need to compare several different techniques in assessing polarized expression of cell surface antigens in vivo.

Transcellular permeability of chlorpromazine demonstrating the roles of protein binding and membrane partitioning.

Transcellular permeability of the neuroleptic-anesthetic chlorpromazine (CPZ) was examined using a cell type (MDCK) that forms a confluent monolayer of polarized cells resulting in distinct apical (AP) and basolateral (BL) membrane domains separated by tight junctions. Because CPZ is membrane interactive, transmonolayer flux was analyzed as two kinetic events: cell uptake from the AP donor solution and efflux into the BL side receiver. Using the rate of cell uptake in the presence of different concentrations of BSA, an intrinsic cell partition coefficient of 3700 +/- 130 and an operational dissociation binding constant of 0.4 +/- 0.05 mM were calculated. In contrast to uptake, efflux of CPZ from either the AP or the BL side of the cell monolayer was approximately 10(4)-fold slower and was dependent upon the avidity of CPZ for the protein acceptor in the receiver solution. These results emphasized the importance of simultaneously measuring disappearance of a lipophilic molecule from the donor solution and its appearance in the receiver and demonstrated how interactions with proteins on either side of the cellular barrier influence permeability. Appearance kinetics showed that the composition of the receiving environment is critical to model a particular in vivo situation and implied that the intrinsic permeability of membrane-interactive molecules in vitro does not necessarily predict penetration beyond the initial cellular barrier in vivo.

The Alzheimer beta-amyloid protein precursor/protease nexin-II is cleaved by secretase in a trans-Golgi secretory compartment in human neuroglioma cells.

Alzheimer beta-amyloid protein precursor (beta APP) is expressed endogenously and abundantly by human neuroglioma (H4) cells. Its secretory processing has been shown to involve discrete proteolysis within the beta A4 region, thus preventing beta-amyloid formation, by an enzyme which has been referred to as 'beta APP secretase'. This cleavage results in secretion of a soluble N-terminal 135 kDa protein and retention of an integral membrane C-terminal fragment within the cell. The membrane-associated C-terminal fragment is sorted to lysosomes where it undergoes limited degradation. We show here that most newly synthesized beta APP is degraded via a non-lysosomal pathway before maturation in H4 cells, and most mature beta APP is processed predominantly by the so-called secretase. The rapid kinetics of appearance/disappearance of a cleaved 135 kDa protein within a microsomal fraction and the slow accumulation of this form in the extracellular medium indicated that secretase cleaves beta APP in an intracellular compartment. Low-temperature block (20 degrees C) was used to demonstrate that beta APP is cleaved within a late Golgi compartment after sulphation which occurs in the trans-Golgi network (TGN). This is consistent with (1) the immunolocalization of most of the beta APP within a Golgi compartment that reacts with wheat germ agglutinin, (2) the fact that less than 1.5% of the total mature full-length beta APP is present at the plasma membrane and (3) subcellular fractionation studies which showed that the mature full-length and intracellular cleaved beta APPs co-sediment with a membrane fraction that is slightly more dense than the plasma membrane. This study provides evidence that most of the beta APP secretase in H4 cells is intracellular, and confirms that the resulting C-terminal fragment is delivered to lysosomes immediately after cleavage. These results are discussed with regard to the possibility that mature full-length beta APP escapes secretase cleavage and is delivered directly from the TGN to the lysosome without passing through the plasma membrane. Either pathway will result in the generation of amyloidogenic fragments.

Use of a biophysical-kinetic model to understand the roles of protein binding and membrane partitioning on passive diffusion of highly lipophilic molecules across cellular barriers.

The novel antioxidants U-78517F and U-74006F, or lazaroids, are highly lipophilic organic molecules with poor brain uptake. To understand this paradoxical behavior better, continuous monolayers of Madin-Darby canine kidney (MDCK) epithelial cells with distinct apical (AP) and basolateral (BL) plasma membrane domains grown on polycarbonate membrane filters and plastic were used to examine the mechanism of transcellular diffusion. Independent kinetic experiments were used to quantify AP to BL flux, efflux from the AP and BL membranes and AP membrane partitioning as functions of bovine serum albumin (BSA) concentration. Fluxes were appropriately reduced to permeability coefficients (Pe) for the membrane, aqueous boundary layer (ABL) and filter, BSA-drug binding constants, and effective (Ke) and intrinsic (Kintr) membrane partition coefficients in the absence of metabolism. Both Pe and Ke decreased exponentially with increased BSA concentration and a concomitant decrease in free drug concentration. Uptake was ABL-controlled under the conditions used and its Pe was 1,000-fold faster than that for efflux due to a large Kintr. Therefore, diffusion across the cellular barrier was limited kinetically by the equilibrium between protein-bound drug and free drug partitioned into the cell membrane and the rate-limiting desorption of drug from the cell membrane into the aqueous receiver. This suggests that brain uptake of these lipophilic antioxidants is limited by interactions with plasma proteins and, possibly, by unfavorable partitioning from the endothelium into the underlying tissue. The present biophysical kinetic model is proposed as generally useful in studying the penetrative ability of other membrane interacting molecules.

The activity of Golgi transport vesicles depends on the presence of the N-ethylmaleimide-sensitive factor (NSF) and a soluble NSF attachment protein (alpha SNAP) during vesicle formation.

An assay designed to measure the formation of functional transport vesicles was constructed by modifying a cell-free assay for protein transport between compartments of the Golgi (Balch, W. E., W. G. Dunphy, W. A. Braell, and J. E. Rothman. 1984. Cell. 39:405-416). A 35-kD cytosolic protein that is immunologically and functionally indistinguishable from alpha SNAP (soluble NSF attachment protein) was found to be required during vesicle formation. SNAP, together with the N-ethylmaleimide-sensitive factor (NSF) have previously been implicated in the attachment and/or fusion of vesicles with their target membrane. We show that NSF is also required during the formation of functional vesicles. Strikingly, we found that after vesicle formation, the NEM-sensitive function of NSF was no longer required for transport to proceed through the ensuing steps of vesicle attachment and fusion. In contrast to these functional tests of vesicle formation, SNAP was not required for the morphological appearance of vesicular structures on the Golgi membranes. If SNAP and NSF have a direct role in transport vesicle attachment and/or fusion, as previously suggested, these results indicate that these proteins become incorporated into the vesicle membranes during vesicle formation and are brought to the fusion site on the transport vesicles.

Permeability of bovine brain microvessel endothelial cells in vitro: barrier tightening by a factor released from astroglioma cells.

It has been shown both in vivo and in culture that astrocytes communicate with brain microvessel endothelial cells (BMECs) to induce many of the blood-brain barrier characteristics attributed to these unique cells. However, the results using cultured cells are conflicting as to whether this communication is dependent upon cell-cell contact. In this study we used primary cultures of bovine BMECs grown as monolayers on polycarbonate filters to study the formation of the barrier in vitro and examine its modulation by rat C6 glioma cells. Effects were examined by treating postconfluent BMEC monolayers with medium conditioned continually by C6 cells from the basolateral side to mimic the in vivo orientation. Cell monolayer integrity was assessed using electrical resistance and by measuring diffusion of uncharged molecules. BMEC monolayers form a functionally polarized and leaky barrier, with maximal resistance of 160 omega . cm2 and significant flux of molecules of molecular weight less than 350 Da. Treatment with rat or human astroglioma cells rather than pericytoma cells or transformed fibroblasts results in a concentration-dependent 200-440% increase in electrical resistance and a coincident 50% decrease in permeability to sucrose and dextran (70 kDa). The decrease in passive diffusion is most likely due to a change in tight junctions and not to transcellular vesicular traffic. The findings support that astroglioma cells release one or more signals that are required for cultured BMECs to express a "differentiated" phenotype associated with a tighter barrier, increased gamma-glutamyl transpeptidase activity, and decreased pinocytic activity. The relative ease and quickness of this culture system makes it amenable to studies on cell-cell interaction and regulation of barrier maintenance.