Elsevier

Brain Research

Volume 768, Issues 1–2, 12 September 1997, Pages 10-18
Brain Research

Research report
F-Actin cytoskeleton and sucrose permeability of immortalised rat brain microvascular endothelial cell monolayers: effects of cyclic AMP and astrocytic factors

https://doi.org/10.1016/S0006-8993(97)00586-6Get rights and content

Abstract

The immortalised RBE4 cell line, derived from rat brain capillary endothelial cells, preserves many features of the in vivo brain endothelium, and hence is of interest as a potential in vitro model of the blood–brain barrier (BBB). This study reports the effects of elevated intracellular cAMP and factors released by astrocytes on the F-actin cytoskeleton and paracellular sucrose permeability of monolayers of RBE4 cells. RBE4 cells grown in control medium showed a marked increase in the F-actin staining at the cytoplasmic margin at confluence, which was not significantly enhanced by elevation of intracellular cAMP and/or addition of astrocyte-conditioned medium (ACM). The formation of the marginal band of F-actin was accompanied by an increase in the F-actin content of the RBE4 cells up to confluence, and a decline in F-actin content thereafter. Elevation of intracellular cAMP or co-culture above astrocytes significantly decreased the paracellular sucrose permeability of confluent RBE4 cell monolayers grown on collagen filters (P<0.01 and P<0.001, respectively). Co-culture above astrocytes together with elevated cAMP also produced a significant decrease in the sucrose permeability of the monolayer (P<0.01) but this was no greater than with astrocytes alone. These findings show that the RBE4 cell line may serve as a useful in vitro model for the study of brain endothelial cell physiology and agents which alter the permeability of the BBB.

Introduction

Primary cultures of brain capillary endothelial cells have been shown to lose many of the characteristics of the in vivo phenotype, suggesting that the presence of a factor or factors produced by one or more brain cell types is required to maintain a fully competent blood–brain barrier (BBB). It is likely that the source of these inductive factors is astrocytes which extend foot processes over the surface of the capillary endothelial cells. Arthur et al. [3], working on primary cultures, reported that only a combination of astrocyte-conditioned medium (ACM) and a specific extracellular matrix could re-induce the formation of the complex tight junctions characteristic of brain capillary endothelial cells. ACM together with elevated levels of intracellular cyclic adenosine monophosphate (cAMP) has been reported to increase the transendothelial electrical resistance of primary bovine brain capillary endothelial cell monolayers [31]. However, a study on bovine brain capillary endothelium suggests that soluble factors produced by brain cells other than glial cells may also be important in the induction of BBB characteristics [37].

It has been proposed that the actin cytoskeleton plays an important role in the structural integrity of epithelia and endothelia. In cultured intestinal epithelial cells which show a characteristic cellular polarity, there is a belt of actin around the apical part of the cell adjacent to the tight junctions 15, 21. This perijunctional actin ring has been proposed to be important in the regulation of tight junction permeability in epithelial cells under a variety of physiological and disease conditions 15, 20, 21. Electron microscopic studies have shown intimate associations between actin microfilaments and tight junctions in intestinal epithelia, which may represent the anatomical basis for cytoskeletal control of tight junction permeability [19]. Moreover, cytochalasin D has been shown to disrupt epithelial tight junctional integrity by destabilising the perijunctional actin ring 15, 20, 21.

The actin cytoskeleton may also play a key role in the maintenance and regulation of endothelial permeability [9]. In monolayers of human umbilical vein endothelial cells (HUVEC) grown on filters, there was a rapid increase in the number of F-actin microfilaments when the cells were treated with the inflammatory agent thrombin, and these changes in the cytoskeleton could be correlated with increases in endothelial monolayer permeability [36]. In cultured bovine pulmonary artery endothelial cells, bacterial lipopolysaccharide caused increases in endothelial monolayer permeability which coincided with changes in the organisation of the F-actin cytoskeleton and modulation of the F- and G-actin content of the cells [13]. Treatment of bovine pulmonary artery endothelial cells with tumour necrosis factor-α caused increases in the albumin flux across the cell monolayer, which were concomitant with qualitative and quantitative changes in the F-actin cytoskeleton [14]. Endothelial and epithelial cell tight junctions contain the protein ZO-1, which is an important site of association between the actin cytoskeleton and the tight junctions [18]. The actin cytoskeleton also has associations with actin-linking proteins, vinculin, α-actinin and catenins, which connect the actin microfilaments to intercellular adhesion molecules, the cadherins 9, 35. Cadherins are associated with adhaerens junctions of epithelial and endothelial cells, which are necessary for stabilisation of tight junctions and maintenance of barrier function 2, 35.

It has previously been reported that monolayers of primary bovine and rat brain capillary endothelial cells cultured in the presence of ACM, together with the cAMP analogue 8-(4-chlorophenylthio)-adenosine 3′,5′-cyclic monophosphate (CPt-cAMP) and the phosphodiesterase inhibitor RO-20-1724 (here abbreviated RO), show a characteristic accumulation of F-actin around the margin of the cytoplasm 1, 27, 31. This effect could not be induced in the absence of ACM or CPt-cAMP, and in the bovine cultures coincided with a marked increase in electrical resistance and a decrease in paracellular solute flux of the cell monolayer [31]. These results suggest that the actin cytoskeleton in brain endothelial cells may also play an important role in the regulation of tight junctional integrity and hence paracellular permeability.

The RBE4 cell line was derived from rat brain microvascular endothelial cells immortalised with the plasmid pE1A-neo containing the E1A region of adenovirus 2 and a neomycin resistance gene 11, 30. Before RBE4 cells can be used as an in vitro model of the BBB, it is necessary to characterise these cells in order to establish the degree to which they retain the properties of the primary rat brain microvascular endothelial cells. Previous studies have reported that RBE4 cells preserve aspects of an endothelial phenotype 1, 30, show differentiation of brain endothelial characteristics in the presence of glial factors 12, 30, show ATP receptors of similar pharmacology to those in primary cultures [23], secrete nitric oxide and endothelin [11]and preserve functional P-glycoprotein [4]. However, the epithelial characteristics of the RBE4 monolayer have so far received little attention. In this study, the organisation of the F-actin cytoskeleton and the paracellular sucrose permeability of RBE4 monolayers have been investigated and the effects of astrocytic factors and elevated intracellular cAMP on these properties have been studied.

Section snippets

Cell culture

The RBE4 cell line was immortalised from rat brain capillary endothelium by two of us (P.-O.C. and F.R.) 11, 30. Cells of passage 20–60 were grown on rat-tail collagen-coated multi-well plates or tissue culture flasks. Rat-tail collagen was prepared according to the method of Strom and Michalopoulos [34]and coating of growth surfaces was performed by application of 0.3 g/l collagen solution for 1 h at 37°C. The excess collagen solution was then removed, the collagen cross-linked in ammonia

Fluorescence microscopy

Sub-confluent RBE4 cells grown in control medium or ACM with or without CPt-cAMP+RO showed a diffuse pattern of F-actin throughout the cell cytoplasm and little concentration of staining around the cell periphery (not shown). In contrast, confluent RBE4 cells grown in either control medium (Fig. 1A) or ACM (Fig. 1B) showed a redistribution of F-actin into a dense peripheral band. RBE4 cells grown to confluence in control medium (Fig. 2A) or ACM (Fig. 2B) and exposed to CPt-cAMP+RO (1–3 days)

Discussion

Brain capillary endothelial cells have been isolated from a number of species and grown in culture in order to study aspects of BBB function [16]. Such studies have produced valuable results, but it has proved difficult to avoid contamination by other cell types, e.g. pericytes, and to maintain the brain capillary phenotype over several passages. Recently, cell immortalisation techniques have produced brain endothelial cell lines which grow vigorously in culture and have been used to study

Acknowledgements

We are grateful to Mr. R. Roberts and Mr. L. Kelberman for technical support and photographic assistance. This work was supported by the MRC and the Wellcome Trust.

References (38)

  • C Beuckmann et al.

    Induction of the blood/brain-barrier-associated enzyme alkaline phosphatase in endothelial cells from cerebral capillaries is mediated via cAMP

    Eur. J. Biochem.

    (1995)
  • E.G Chikhale et al.

    Hydrogen bonding potential as a determinant of the in vitro and in situ blood–brain barrier permeability of peptides

    Pharmacol. Res.

    (1994)
  • M.-P Dehouck et al.

    An easier, reproducible, and mass-production method to study the blood–brain barrier in vitro

    J. Neurochem.

    (1990)
  • M.-P Dehouck et al.

    Drug transfer across the blood–brain barrier: correlation between in vitro and in vivo models

    J. Neurochem.

    (1992)
  • E Dejana et al.

    Endothelial cell-to-cell junctions

    FASEB. J.

    (1995)
  • D.A DuBose et al.

    Comparisons of endothelial cell G- and F-actin distribution in situ and in vitro

    Biotech. Histochem.

    (1993)
  • O Durieu-Trautmann et al.

    Nitric oxide and endothelin secretion by brain microvessel endothelial cells: regulation by cyclic nucleotides

    J. Cell. Physiol.

    (1993)
  • B El Hafny et al.

    Synergistic stimulation of gamma-glutamyl transpeptidase and alkaline phosphatase activities by retinoic acid and astroglial factors in immortalized rat brain microvessel endothelial cells

    J. Cell. Physiol.

    (1996)
  • S.E Goldblum et al.

    Bacterial lipopolysaccharide induces actin reorganisation, intercellular gap formation, and endothelial barrier dysfunction in pulmonary vascular endothelial cells: concurrent F-actin depolymerization and new actin synthesis

    J. Cell. Physiol.

    (1993)
  • Cited by (88)

    • In Vitro CNS Models

      2017, Nanotechnology Methods for Neurological Diseases and Brain Tumors: Drug Delivery across the Blood-Brain Barrier
    • In vitro screening of nanomedicines through the blood brain barrier: A critical review

      2016, Biomaterials
      Citation Excerpt :

      Additionally, serum removal from the acceptor well better matches the BBB physiology, as the basolateral surface of the brain endothelium is exposed in vivo to serum-free interstitial fluid [221]. The monolayer tightness has also proven to be upregulated by the second messenger cAMP (cyclic adenosine monophosphate) [283] and hydrocortisone [226,284]. For this reason, culture media are often supplemented with cAMP modulators (such as 8-(4-chlorophenylthio)adenosine 3′,5′-cyclic monophosphate -CPT-cAMP-) and/or phosphodiesterase inhibitors (such as 4-(3-butoxy-4 methoxybenzyl)-2-imidazolidinone -RO-20-1724- [195]) to elevate cAMP levels; with hydrocortisone; and sometimes with puromycin [150], which boosts the response of the monolayer to glucocorticoid treatment.

    View all citing articles on Scopus
    View full text