PubMed Health⌕ Search

Biomedical subjects

R R Shivers

Publications and source records attributed to R R Shivers.

At least 19 recordsLinked to original sources

Structural pathways for macromolecular and cellular transport across the blood-brain barrier during inflammatory conditions. Review.

This review presents an overview of the highlights of major concepts involving the anatomical routes for the transport of macromolecules and the transmigration of cellular elements across the blood-brain barrier (BBB) during inflammation. The particular focus will include inflammatory leukocytes, neoplastic cells and pathogenic microorganisms including specific types of viruses, bacteria and yeasts. The experimental animal models presented here have been employed successfully by the authors in several independent experiments during the past twenty-five years for investigations of pathologic alterations of the BBB after a variety of experimentally induced injuries and inflammatory conditions in mammalian and non-mammalian animal species. The initial descriptions of endothelial cell (EC) vesicles or caveolae serving as mini-transporters of fluid substances essentially served as a springboard for many subsequent discoveries during the past half century related to mechanisms of uptake of materials into ECs and whether or not pinocytosis is related to the transport of these materials across EC barriers under normal physiologic conditions and after tissue injury. In the mid-1970's, the authors of this review independently applied morphologic techniques (transmission electron microscopy-TEM), in conjunction with the plant protein tracer horseradish peroxidase (HRP) to investigate macromolecular transport structures that increased after the brain and spinal cord had been subjected to a variety of injuries. Based on morphologic evidence from these studies of BBB injury, the authors elaborated a unique EC system of modified caveolae that purportedly fused together forming transendothelial cell channels, and later similar EC profiles defined as vesiculo-canalicular or vesiculo-tubular structures (VTS, Lossinsky, et al., 1999). These EC structures were observed in association with increased BBB permeability of tracers including exogenously injected HRP, normally excluded from the intercellular milieu of the CNS. Subsequent studies of non-BBB-type tumor ECs determined that the EC VTS and other vesicular structures were defined by others as vesiculo-vacuolar organelles (VVOs, Kohn et al., 1992; Dvorak et al., 1996). Collectively, these structures appear to represent a type of anatomical gateway to the CNS likely serving as conduits. However, these CNS conduits become patent only in damaged ECs for the passage of macromolecules, and purportedly for inflammatory and neoplastic cells as well (Lossinsky et al., 1999). In this review, we focus attention on the similarities and differences between caveolae, fused racemic vesicular bundles, endothelial tubules and channels (VTS and the VVOs) that are manifest in normal, non-BBB-type blood vessels, and in the BBB after injury. This review will present evidence that the previous studies by the authors and other researchers established a framework for subsequent transmission (TEM), scanning (SEM) and high-voltage electron microscopic (HVEM) investigations concerning ultrastructural, ultracytochemical and immunoultra-structural alterations of the cerebral ECs and the mechanisms of the BBB transport that occurs after CNS injury. This review is not intended to include all of the many observations that might be included in a general historical overview of the development of the EC channel hypothesis, but it will discuss several of the major contributions. We have attempted to present some of the structural evidence that supports our early contributions and those made by other investigators by highlighting major features of these EC structures that are manifest in the injured BBB. We have focused on currently established concepts and principles related to mechanisms for the transendothelial transport of macromolecules after CNS injury and also offer a critical appraisal of some of this literature. Finally, we describe more recent concepts of transBBB avenues for viruses, including HIV-1, bacterial and mycotic organisms, as well as inflammatory and neoplastic cell adhesion and migration across the injured mammalian BBB. Data from studies of EC-related adhesion molecules, both from the literature and from the author's experimental results and observations made in other laboratories, as well as from personal communications underscore the importance of the adhesion molecules in facilitating the movement of leukocytic, neoplastic cell and human pathogens across the BBB during inflammatory and neoplastic events. Exciting, ongoing clinical trials are addressing possible therapeutic intervention in neuroinflammatory diseases, including multiple sclerosis, by blocking certain glycoprotein adhesion molecules before cells have the ability to adhere to the ECs and migrate across the BBB. Approaches whereby inflammation may be reduced or arrested using anti-adhesion molecules, by restructuring EC cytoskeletal, filamentous proteins, as well as remodeling cholesterol components of the modified VTS are discussed in the context of developing future therapies for BBB injury and inflammation. Understanding new concepts about the mechanism(s) by which inflammatory cells and a variety of pathogenic microorganisms are transported across the BBB can be expected to advance our understanding of fundamental disease processes. Taken together, the literature and the author's experiences during the past quarter of a century, will hopefully provide new clues related to the mechanisms of transendothelial cell adhesion and emigration across the injured BBB, issues that have been receiving considerable attention in the clinical arena. Learning how to chemically modulate the opening and/or closure of EC VTS and VVO structural pathways, or junctional complexes prior to cellular or microorganism adhesion and breaching the BBB presents challenging new questions in modern medicine. Future studies will be critically important for the development of therapeutic intervention in several human afflictions including traumatic brain and spinal cord injuries, stroke, cancer, multiple sclerosis and conditions where the immune system may be compromised including HIV infection, infantile and adult meningitis.

Animals↗

Correlation of the presence of blood-brain barrier tight junctions and expression of zonula occludens protein ZO-1 in vitro: a freeze-fracture and immunofluorescence study.

Tight junctions are regarded as the primary anatomical structure responsible for the blood-brain barrier (BBB). The molecular components that have been defined include ZO-1, a peripheral membrane protein associated with the cytoplasmic surface of the tight junction in epithelial and endothelial cells. It has been localized to the points of membrane contact with the fibrils seen by freeze-fracture. Examination of passaged endothelial cells with freeze-fracture failed to locate the intramembrane specializations associated with tight junctions. For this reason, immunocytochemistry and freeze-fracture were used to study the correlation of ZO-1 expression with the presence of tight junctions in bovine brain and aorta endothelial cells. Indirect immunofluorescence analysis showed ZO-1 to be localized at sites of cell-cell contact. Images of freeze-fractured sites of endothelial cell-cell contacts in identical passage numbers did not display characteristic tight junctions. When bovine aorta endothelial cells were cultured in astrocyte-conditioned medium on a complete extracellular matrix, platinum replicas displayed profiles of tight junctions. The elements of tight junctions were arranged as parallel ridges which displayed free ends. The immunofluorescence staining of ZO-1 was identical to that obtained on the endothelial cells that displayed no tight junction profiles. These results suggest that ZO-1 may be present at putative junction-containing sites before the junctional structures appear in the surface membrane. Therefore, ZO-1 expression does not a priori reflect assembly of the tight junctions identified by freeze-fracture.

Animals↗

Morphology and migratory behavior of bovine pulmonary artery endothelial cells and C6 astrocytoma cells in a three-dimensional type I collagen matrix.

The purpose of this investigation was to compare the morphological characteristics and migratory behavior of bovine pulmonary artery endothelial cells (PAE) and C6 astrocytoma cells in a three-dimensional type I collagen matrix. Samples of both cell types were cultured and prepared for scanning electron microscopy and time-lapse video photomicrography using standard protocols. Observations included comparative analyses of factors such as the state of the matrix, the presence of cell processes and the distribution of cells throughout the sample. Results revealed that PAE cells formed a monolayer and were connected by many cytoplasmic extensions. They were distributed evenly throughout the collagen matrix and they migrated slowly. In contrast, C6 astrocytoma cells aggregated into clumps and digested the matrix while migrating through it quite rapidly. Photomicrography also revealed the presence of filopodia and invadopodia in the PAE samples and astrocytoma samples respectively. These cytoplasmic processes may function in either cell-cell communication or cell motility. This study demonstrates that endothelial cells and C6 astrocytoma cells behave very differently when grown in type I collagen matrices. Future studies include finding ways to slow tumor metastasis by inhibiting migration through various matrices or to slow tumor growth by preventing endothelial cell migration and angiogenesis.

Animals↗

'Rings' of F-actin form around the nucleus in cultured human MCF7 adenocarcinoma cells upon exposure to both taxol and taxotere.

The anti-cancer taxoids, Taxol (paclitaxel) and Taxotere (docetaxel), are the most promising anti-mitotic agents developed for cancer treatment in the past decade. The effectiveness of this new class of compounds lies in their unique mechanism of action on the cytoskeleton. Both taxol and taxotere bind to microtubules and shift the normal equilibrium between monomeric and polymerized tubulin to favor the polymerized form by strongly promoting tubulin assembly and inhibiting microtubule depolymerization. Although very similar in structure, these two compounds have recently demonstrated different in vitro, in vivo, and clinical activities; however, no study to date has effectively compared specific cytoskeletal alterations induced by taxol and taxotere in cultured cells. Using specific staining techniques for both F-actin and alpha-tubulin, this study provides the first detailed immunohistochemical comparison of the effects of equimolar concentrations of taxol and taxotere on both the microfilament and microtubule networks in a cultured cell line. Using human MCF7 breast adenocarcinoma cells, new observations of taxotere/taxol alterations of the cytoskeleton include: an increased abundance of parallel microtubule 'bundles' in taxotere treated cells and a definitive reorganization of the microfilament network which results in novel ring-like formations of F-actin condensed exclusively in the perinuclear zone. Reorganization of the actin cytoskeleton induced by a taxoid disruption of the microtubule equilibrium is indicative of the interdependence between microtubules and microfilaments in this transformed cell line and suggests that the indirect role of the taxoids on the microfilament network may have been overlooked in their mechanism of action as chemotherapeutic agents.

Actins↗

Ultrastructure of the blood-brain barrier in the rabbit.

The morphology of the normal blood-brain barrier in the rabbit by thin section and freeze-fracture electron microscopy is reported. Exogenous tracer horseradish peroxidase was injected to visualize the integrity of the blood-brain barrier in New Zealand White rabbits. Freeze-fracture was used to determine the intramembrane architecture of the tight junctions. Thin sections (60-100 nm) of brain capillaries from animals injected with horseradish peroxidase (HRP) possessed few pinocytotic vesicles in the cytoplasm. Junctional profiles between adjoining plasma membranes were present. Thin sections of capillaries containing electron dense HRP reaction product (HRP-RP) in the lumen revealed focal fusions of apposing plasma membranes that occluded reaction product from entering the junctional clefts. Some cytoplasmic vesicles were filled with HRP-RP; however, basal laminae and brain interstitium were free of HRP-RP in all vessel profiles examined. Freeze-fracture electron microscopy revealed tight junctions as an elaborate network of interconnecting strands of intramembrane particles appearing as ridges on the EF face and corresponding grooves on the PF face on platinum replicas. Results of this study demonstrate the architecture of rabbit brain microvessel endothelial junctions (blood-brain barrier) and provide evidence that the tight junctions prevent HRP extravasation. It is concluded that rabbit brain endothelial tight junctions (zonulae occludentes), as in other mammals, form the anatomical basis of the blood-brain barrier. Consequently, the rabbit brain microvasculature can be a useful model for establishing stereotactic radiosurgical procedures to treat brain astrocytomas (tumours).

Animals↗

Quantitative comparison of the morphology of the microvasculature of primary lung lesions and metastatic brain tumours.

Endothelial cells that make up microvessels display a differential expression of structure and function that allows them to meet the requirements of the tissues they vascularize. The growth of tumours requires neovascularization that occurs by angiogenesis, a process by which new capillaries are formed from existing vessels. It has been proposed that the local environment determines the morphological characteristics of the vessels that vascularize it. The present study is a quantitative investigation of the properties of capillaries that may contribute to their permeability and their general morphology. Tissue samples were taken from primary non-small cell lung carcinomas and from the same tumour type that had metastasized to brain. Normal samples were taken from tissue distant from the site of pathology. Using transmission electron microscopy, profiles of capillaries from the tissue samples were examined and photographed. Image-analysis was performed to measure vesicular and mitochondrial density, vessel size and vessel wall thickness. The results showed that the morphology of the capillaries that vascularize the primary lung tumours differs from the morphology of the capillaries that vascularize the metastases of these tumours in the brain. In addition to differences in the quantitative observations, qualitative differences were observed with respect to the presence of fenestrations in the vessel wall, found only in brain tumour vessels, and the presence of 'open' junctions, seen only in lung tumour vessels. These results suggest that although the environment of the tumour changed the morphology of the vessels so they were no longer normal, it did not have the same effect on the vessels that vascularize it at the two different sites. Therefore, the response of microvessels in the host tissue to factor(s) produced by tumour cells may be multifaceted and dependent upon the properties of local vessels.

Brain Neoplasms↗

Evidence that disruption of connexon particle arrangements in gap junction plaques is associated with inhibition of gap junctional communication by a glycyrrhetinic acid derivative.

Glycyrrhetinic acid exhibits many pharmacological activities, including the ability to block gap junctional communication. However, the mechanism of glycyrrhetinic action is not clear. Others have shown that glycyrrhetinic acid apparently binds to a single proteinatious binding site in the plasma membrane. We present evidence that while exposure to glycyrrhetinic acid derivatives may not affect protein synthesis or location, it does seem to alter connexon particle packing in gap junction plaques.

Animals↗

Extracellular matrix regulates microfilament and vinculin organization in C6-glioma cells.

Since the extracellular matrix (ECM) has a role in regulating cell proliferation it has been hypothesized that unregulated growth of transformed cells results from an inability of cells to interpret the constraints on growth as dictated by the ECM. The most likely candidate to communicate the restraint on growth to the nucleus is the cytoskeleton because it is the only cytoplasmic structure to physically connect the nucleus and plasma membrane. The purpose of this study was to determine whether the cytoskeleton, specifically the microfilaments (MF), of the C6-glioma cell line possesses the ability to respond to changes in the ECM. The C6-glioma cell line was chosen as a model because it exhibits the characteristics of a tumor cell, both in vivo and in vitro. In this study cells were grown on bare glass, Matrigel, laminin and type IV collagen prior to staining with phalloidin tetramethylrhodamine B isothiocyanate or antibodies specific to vinculin to visualize MF and adhesion plaques, respectively. On the basis of MF orientation, network complexity and location of adhesion plaques, this study reports that the cytoskeletal arrangements of C6 cells grown on various substrates are distinctly different. Each distinct organization pattern may reflect a change of cell behavior promoted by the different culture conditions. The significance of this study is that it demonstrates that the process of transformation in C6-glioma cells has not interfered with the cells ability to recognize, interpret and adapt to changes in the ECM.

Actin Cytoskeleton↗

Gap junctions revealed by freeze-fracture electron microscopy.

Gap junctions provide the basis for the formation of elaborate networks of communication between cells in animal tissues. Electron microscopic examination of thin sections of plastic embedded gap junctions has provided valuable information on the anatomy and function of these remarkable structures. Freeze-fracture electron microscopy, however, has made available unique vistas of gap junction-bearing intramembrane surface--surface previously inaccessible to the researcher's eyes. Data on population density, distribution, size, geometry of intramembrane particle packing, and structural responses of gap junction components to experimental manipulation are simply and easily obtained with freeze fracture. Recent developments of sophisticated protocols of immunocytochemistry as applied to freeze-fracture replicas further serve to reinforce the notion that freeze-fracture is a powerful tool for study of gap junctions. Molecular techniques of gap junction gene transfection promise to add a truly unique dimension to investigations of the broad spectrum of functional roles of gap junctions.

Animals↗

Blood-brain barrier permeability in rats is altered by exposure to magnetic fields associated with magnetic resonance imaging at 1.5 T.

We have previously reported that exposure of rats to low-field (0.15 T) magnetic resonance imaging (MRI) increases blood-brain barrier (BBB) permeability. However, a number of investigators have failed to observe this effect when high-field MRI (1.5 T or higher) is used. Therefore, we investigated whether or not we would observe changes using our technique at these higher fields. Adult male Sprague-Dawley rats were anaesthetised and then exposed to a 22.5 min imaging or sham procedure. Immediately following exposure, rats were injected with 1 MBq of 153Gd-DTPA intracardially and then immediately re-exposed for an additional 22.5 min. The rats were killed 1h following the second MRI exposure, at which time the brain was resected and 3 ml of venous blood collected. The ratio of radioactivity per gram of brain to radioactivity per milliliter of blood, known as the brain-blood partition coefficient, was determined and used as a measure of BBB permeability. Groups of animals had different exposures. Group 1 (n = 9) was exposed to a clinically relevant MRI procedure. Group 2 (n = 20) was exposed to the same procedure except the rf specific absorption rate (SAR) was reduced to 25% and the animals were positioned 15 cm from imager centre to increase the time-varying magnetic field from 0.4 to 2.8 T/s. For the sham exposures (n = 21), the animals were placed in the imager with the static field ramped down to zero and exposed to a sound recording simulating a MRI examination.(ABSTRACT TRUNCATED AT 250 WORDS)

Animals↗

Resolving the two monolayers of a lipid bilayer in giant unilamellar vesicles using deuterium nuclear magnetic resonance.

Giant unilamellar vesicles (GUVs) composed of mixtures of POPC (1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine) plus DMPG (1,2-dimyristoyl-sn-glycero-3-phosphoglycerol) and/or CHOL (cholesterol) were prepared using detergent dialysis. Vesicles containing at least 30 mol % CHOL had diameters exceeding 450 nm. POPC in such GUVs, deuterium-labeled at either the choline alpha or beta segments, yielded deuterium (2H) and phosphorus (31P) nuclear magnetic resonance (NMR) Pake pattern line shapes, quadrupole splittings and chemical shift anisotropies identical to those obtained with multilamellar vesicles (MLVs) of identical composition. Exposing exclusively the vesicle exterior to either calcium or perchlorate ions, both of which are known to influence lipid head-group conformation through surface charge effects, caused the appearance of two overlapping 2H Pake patterns of equal intensity. The quadrupole splitting of one component remained unchanged while that of the second component was altered in the manner expected for choline alpha or beta deuterons in the presence of a cationic (calcium) or anionic (perchlorate) surface charge. Freeze-thawing the GUVs to equilibrate the exterior and interior vesicular contents eliminated the initially unchanged spectral component. It was likewise possible to resolve two quadrupole splittings when Staphylococcus aureus delta-toxin, a surface-active peptide known to influence lipid head-group orientational ordering, was added to the exterior vesicular solution only. This indicates that delta-toxin upon binding remains confined to one monolayer of the lipid bilayer and does not traverse the membrane.(ABSTRACT TRUNCATED AT 250 WORDS)

Cholesterol↗

Ultrastructural analysis of gap junctions in C6 glioma cells transfected with connexin43 cDNA.

In the past, we transfected C6 glioma cells with connexin43 cDNA, resulting in a significant increase in connexin43 mRNA and protein, as well as reduced proliferation and tumorigenesis. To investigate the morphological aspects of increased connexin43 expression in these cells, we have used a combination of immunocytochemistry, cytochemistry, and electron microscopy. By confocal immunofluorescence microscopy, connexin43 protein was localized to the plasma membrane of transfected cells and extensive intracellular accumulations of connexin43 were also demonstrated. Freeze fracture preparations showed large aggregates of particles typical of mature gap junction plaques in the plasma membrane of these cells. Ultrastructural immunogold labeling with anti-connexin43 serum revealed that connexin43 protein was present in gap junctions in the plasma membrane, some of which were found in proximity to clathrin-coated pits. In addition, various intracellular membranous profiles were immunoreactive for connexin43, including annular profiles, some with fuzzy coats and some associated with lysosome-like structures. Enzyme cytochemistry revealed that these annular gap junction profiles were often associated with acid phosphatase-positive lysosomes. These studies on the intracellular localization of gap junction protein in connexin43-transfected cells are consistent with the functional expression of the transfected connexin43 cDNA and provide a useful model to study the pathways of gap junction assembly and degradation.

Acid Phosphatase↗

Heat stress affects blood-brain barrier permeability to horseradish peroxidase in mice.

The blood-brain barrier of mice subjected to hyperthermia for 0-135 min was examined using horseradish peroxidase (HRP) and Evans blue dye tracers by light, fluorescence and electron microscopy. Neither control nor heat-stressed mice exhibited extravasation of the Evans blue dye-albumin complex from the brain microvasculature. Gross examination of vibratome sections processed to reveal HRP reaction product exhibited multiple microfoci of HRP extravasation. Electron microscopic observations of the heat-stressed tissues revealed HRP reaction product within pinocytotic vesicles, tubulo-vesicular complexes, transendothelial channels and occasionally flooded within the cytoplasm of endothelial cells. HRP reaction product could clearly be seen in the basal laminae of the capillaries and in the surrounding neuropil. This study demonstrates that blood-brain barrier permeability to HRP is increased in response to heat stress and introduces a new, reproducible model of blood-brain barrier disruption.

Animals↗

A freeze-fracture study of the cuticle of adult Nippostrongylus brasiliensis (Nematoda).

The surface of the cuticle of adult Nippostrongylus brasiliensis has been studied by means of the freeze-fracture technique and by transmission electron microscopy. Some of the surface coat appears to have been shed from the surface of the cuticle of adults fixed in situ in the intestine of its host and from the surface of individuals removed from the intestine and freeze-fractured. Freeze-fracturing the cuticle of individuals removed from the host has shown that this surface coat varies in thickness from 30 to 90 nm. The epicuticle is about 20 nm thick and cleaves readily to expose E- and P-faces. The P-face of the epicuticle possesses a small number of particles, similar to intra-membranous particles, whilst the E-face possesses a few, widely scattered depressions. Despite the presence of these particles the epicuticle is not considered to be a true membrane. Freeze-fracturing the remainder of the cuticle has confirmed its structure as described by conventional transmission electron microscopy. Clusters of particles on the P-face of the outer epidermal (hypodermal) membrane and corresponding depressions on the E-face of the membrane are though to be associated with points of attachment of the cuticle to the epidermis (hypodermis). No differences in appearance of the cuticle and its surface layers were observed in individuals taken from 7-, 10-, 13- and 15-day infections.

Animals↗

Ordered distribution of membrane-associated dense plaques in intact quail gizzard smooth muscle cells revealed by freeze-fracture following treatment with cholesterol probes.

The surface distribution of membrane-associated dense plaques in intact quail gizzard smooth muscle cells was investigated by freeze-fracture. Replicas of fractured smooth muscle cell plasma membrane showed caveola-free regions with few intramembrane particles, interspersed with caveola-populated areas with a higher intramembrane particle density. Electron microscopy of thin sections of quail gizzard smooth muscle revealed the regions free of caveolae to be occupied by membrane-associated dense plaques; anchoring sites for the contractile filaments of the cell. Demarcation between the caveola-populated and caveola-free regions on the relicated intramembrane surface was not clear and thus provided little information concerning the distribution of dense plaque sites. However, treatment of the smooth muscle tissue with the cholesterol-binding agents filipin or tomatin prior to freeze-fracture allowed the dense plaque sites to be easily observed as the sites remained free of the membrane deformations characteristic of these agents. The dense plaque sites consist of caveola-free oval areas juxtaposed in regular bands that traverse the long axis of the cell. The dense plaque sites on the freeze-fracture replica were confirmed by electron microscopy of thin sections of filipin-treated quail gizzard smooth muscle cells, which showed the plasma membrane associated with the dense plaques to be unaffected by the actions of filipin, whereas that of the caveola-populated region was severely deformed. The observations presented in this study provide evidence for a highly ordered distribution of dense plaques at the cell surface of intact quail gizzard smooth muscle cells and thus corroborate existing evidence for an organized substructure of smooth muscle cells.

Animals↗

Magnetic resonance imaging increases the blood-brain barrier permeability to 153-gadolinium diethylenetriaminepentaacetic acid in rats.

In a qualitative electron microscopy study we initially reported that exposure of rats to a standard clinical magnetic resonance imaging (MRI) procedure temporarily increased the blood-brain barrier (BBB) permeability to horseradish peroxidase. In this study, we quantitatively support our initial finding. Rats were injected intracardially with radio-labelled diethylenetriaminepentaacetic acid [( 153Gd]DTPA) in the middle of two sequential 23.2 min MRI exposures. Exposed rats (n = 21) showed significantly greater (29%, P = 0.006) retention of [153 Gd]DTPA than sham-exposed rats (n = 22) 1 h after the end of the last 23.2 min exposure. These findings suggest that magnetic fields may alter BBB permeability.

Animals↗

Induction of gap junctions and brain endothelium-like tight junctions in cultured bovine endothelial cells: local control of cell specialization.

The final development of specializations by brain capillary endothelial cells, which characterize them as distinct from non-central nervous system (CNS) endothelium, is thought to be controlled by astrocyte-derived factors produced locally within the CNS. One specialization, the complex intercellular tight junction, which is unique to these cells and a major component of the blood-brain barrier, is controlled by an astrocyte-derived factor(s) and a "competent' extracellular matrix (Arthur et al., 1987). In order to test whether these factors can also trigger development of brain endothelium-like tight junctions in non-CNS microvessel endothelial cells, passaged bovine aorta and pulmonary artery endothelial cells were cultured in either 50% astrocyte-conditioned medium and 50% alpha-MEM, or in alpha-MEM alone (control). Only endothelial cells maintained in conditioned medium exhibited ultrastructural features indicative of synthesis and plasma membrane-insertion of junction components (Shivers et al., 1985). No assembled tight junctions were seen in these cells. Endothelial cells plated onto coverslips coated with ECM (Cedarlane Labs., Hornby, Ont.) and maintained in astrocyte-conditioned medium, displayed large, complex tight junctions and extraordinarily large gap junctions. Cells plated onto plastic or fibronectin-coated substrates possessed no tight or gap junctions. Results of this study show that CNS astrocytes produce a soluble factor(s) that promotes synthesis and insertion of tight junction components in non-CNS endothelial cells. Moreover, an intact, endothelial-derived extracellular matrix is required for assembly of tight junctions to complete development of this brain capillary-like specialization. This study confirms the notions that: a) the final fine-tuning of cell differentiation is under local control, and b) that endothelial cells in general do not express their final destination-specific differentiated features until those features are induced by local environment-produced conditions.

Animals↗