Download The Blood-Brain Barrier and Its Role in Immune Privilege in the

Journal of Neuropathology and Experimental Neurology
Copyright q 2003 by the American Association of Neuropathologists
Vol. 62, No. 6
June, 2003
pp. 593 604
The Blood-Brain Barrier and Its Role in Immune Privilege in the Central Nervous System
JOEL S. PACHTER, PHD, HELGA E.
DE
VRIES, PHD,
AND
ZSUZSA FABRY, PHD
Abstract. The blood-brain barrier (BBB) provides both anatomical and physiological protection for the central nervous
system (CNS), strictly regulating the entry of many substances and blood borne cells into the nervous tissue. Increased
understanding of how the unique microenvironment in the CNS influences the BBB is crucial for developing novel therapeutic
approaches to CNS diseases. In this review, we discuss those characteristics of the BBB that play an important role in
maintaining immune privilege in the CNS, as well as factors that regulate immune cell invasion through the BBB and thereby
modulate immune responses in the nervous tissue. In general, immune cell invasion across the BBB is highly restricted and
carefully regulated. A florid invasion of activated white blood cells can create a predominantly proinflammatory local environment in the CNS, leading to immune-mediated diseases of the nervous tissue. Recent developments in cellular and molecular biological methods have allowed closer analysis of BBB function, and led to an improved understanding of the active
role of the BBB in immune-mediated diseases of the CNS.
Key Words:
Blood-brain barrier; CNS inflammation; Immune privilege.
HISTORICAL PERSPECTIVE
In 1885 Paul Ehrlich discovered that intravenous injection of acidic dyes into experimental animals caused
staining of various organs, with the notable exception of
the brain (1). Somewhat later, in 1898, Roux and Borrel
observed that tetanus toxin injected into the cerebral spinal fluid (CSF) caused marked cerebral symptoms but,
when administered intravenously, produced no discernible cerebral effect in these animals (2). These initial experiments prompted the earliest consideration of a bloodbrain barrier (BBB) that restricted the passage of
substances into and out of the brain. Subsequent experiments to test this concept were conducted by Emil Goldmann in 1908 (3). He injected trypan blue dye intravenously in large quantities and observed that the dye
stained all tissues except the brain. He also demonstrated
that under these conditions the choroid plexus stained
blue. When he injected a small amount of trypan blue
directly into the CSF, the brain was found to be colored
a deep blue. He concluded that there must be a barrier
between the blood and the brain tissue that is impermeable to trypan blue. He placed the site of this barrier at
the small, highly vascular tufts of choroid plexus epithelia within each of the 4 ventricles of the brain. During
the 1920s, numerous experiments were performed to
characterize the barrier in nervous tissue and, in 1929,
Walter and his colleagues came to the conclusion that
there must be several distinct barriers in the CNS: the
blood-CSF barrier, the brain-CSF barrier, and the BBB
(4). These observations led to our current view, which
From University of Connecticut Health Center (JSP), Farmington,
Connecticut; Department of Molecular Cell Biology (HEdV), VU Medical Center, Amsterdam, Netherlands; Department of Pathology (ZF),
University of Wisconsin-Madison, Madison, Wisconsin.
Correspondence to: Zsuzsa Fabry, PhD, Associate Professor, Department of Pathology, University of Wisconsin, 1300 University Avenue,
6130 MSC, Madison, WI 53706. E-mail: [email protected]
holds that there exists 2 principal barriers in the CNS:
the BBB that is situated along 99% of the brain’s capillary endothelium, and the blood-CSF barrier that is located at the choroid plexus epithelium of the 4 ventricles.
Anatomical Localization of Barrier Systems in the Brain
The 2 principle barriers in the CNS, the BBB and the
blood-CSF barrier, can be identified immunohistochemically in tissue sections of the CNS using different antibodies against barrier proteins. An example of such a
marker is the GLUT-1 glucose transporter protein, which
is present in virtually all brain capillary endothelium (5).
Anatomical mapping of BBB characteristics demonstrates
that there are variations in the level of BBB restriction
in different areas of the CNS. The potential importance
of such differences indicates microenvironmental regulation of brain microvessel permeability, which is exploited in neuroendocrine-feedback mechanisms in the
CNS. An example of this can be seen at the capillaries
of the hypothalamus tuber cinereum, where rich and close
connections between the endothelial cells and neurons
result in the formation of a fenestrated endothelial cell
wall that permits free diffusion of releasing and inhibitory hormones into the circulation. The absence of the
BBB at the area postrema has an opposite result, allowing
the diffusion of materials from the blood to the brain
tissue and, thus, providing an important area of the brain
with a neuroendocrine feedback mechanism. These areas,
also termed circumventricular organs, not only lack a
BBB but are also deficient in expression of neurothelin,
another BBB marker.
Brain capillaries that are approximately the diameter
of a red blood cell (7–8 mm) are the formally designated
sites of the BBB. This capillary bed has a large total
cross-sectional area where the blood flow is quite low.
Brain capillaries derive from the 2 carotid arteries, anteriorly, and the paired vertebral arteries, posteriorly.
593
594
PACHTER ET AL
Fig. 1.
Molecular organization of tight junction.
These major arteries branch into smaller arteries as they
go through the pia-arachnoid membrane and eventually
penetrate the cortex and deeper structures and branch into
capillaries. Ultrastructural tracer studies indicate that extravasation of macromolecules takes place primarily in
segments of large, penetrating cortical blood vessels in
both venules and arterioles (6). The capillaries entering
the choroid plexus are different from those of the brain
parenchyma, in that the choroidal vessels do not express
tight junctions (to be discussed in more detail in the next
section) and are porous (7). However, the choroid plexus
epithelium forms tight junctions constituting the bloodCSF barrier (7). Blood returning to the heart from the
capillary beds flows initially into the postcapillary venules then sequentially through collecting venules and
small, medium, and large veins. It was demonstrated that,
in rat brain, capillary endothelium exhibits a complex
network of continuous multistranded tight junctions.
Continuous capillary-type tight junctions extend, although in a simpler beltlike fashion, into the endothelium
of postcapillary venules, however, the endothelium of
collecting veins possess widely discontinuous single- or
double-stranded tight junctions associated with gap junctions. Arteries have endothelial tight junctions containing
focal discontinuities associated with gap junctions (6).
When in vitro BBB models are established, brain microvessels containing capillaries are trapped on small pore
size filters and the brain microvascular endothelial cells
(BMECs) are isolated and cultured.
MOLECULAR ORGANIZATION OF BRAIN
ENDOTHELIAL JUNCTIONS
The specialized brain microvessel endothelial cells not
only exhibit different metabolic characteristics compared
to endothelial cells of other organs, but also possess welldeveloped intercellular tight junctions or so-called zona
occludens (8).
Tight junctions provide a continuous seal around the
apical regions of lateral membranes of the brain endothelial cells. The presence of tight junctions determines
J Neuropathol Exp Neurol, Vol 62, June, 2003
the tightness, and thus the permeability, of the brain endothelium, resulting in a transendothelial electrical resistance (TEER) across brain endothelium that measures between 1,000 and 1,500 V/cm2 (9). In vitro models often
monitor both intercellular tightness and paracellular diffusion of marker molecules as indices of the integrity of
the endothelial monolayer. The integrity of the BBB in
vitro and in vivo is drastically compromised under inflammatory conditions (10, 11).
Cellular junctions are composed of a network of intracellular and transmembrane proteins that are specific to
each type of junction (Fig. 1). Two major transmembrane
components of the tight junctions have now been identified: occludin (12) and the claudins (13). Occludin
spans the membrane 4 times with both the amino and the
carboxy termini located intracellularly and it is one of the
primary sealing proteins in the tight junction (14). The
other transmembrane proteins of the tight junction belong
to the claudin multigene family, which is comprised of
at least 20 members. Structurally, claudins contain 2 extracellular loops and 4 transmembrane domains (review,
14). Claudins are believed to be the major transmembrane
proteins of tight junctions, as occludin knockout mice are
still capable of forming these inter-endothelial connections (15), while claudin knockout mice are nonviable
(16). With specific regard to brain endothelium, both
claudin-1 and -5 were found to be expressed (17).
The carboxy-terminal cytoplasmic tail of occludin and
claudins can interact with a number of cytoplasmic zonula occludens proteins (ZO), like ZO-1, ZO-2, and ZO3 (18, 19). These proteins belong to the membrane-associated guanylate kinase protein family that can interact
with other plaque domain molecules such as cingulin (20)
and 7H6 antigen (21). ZO proteins can also interact with
signaling molecules and cytoskeletal proteins such as cortactin and actin. Intracellular molecules like rab3, symplekin, AF-6, and 19B1 are reported to be associated with
epithelial tight junctions, but their role in BMECs is unknown (review, 17).
THE BLOOD-BRAIN BARRIER AND IMMUNITY IN THE CNS
Junctional adhesion molecules (JAMS) are transmembrane molecules that colocalize with tight junctions.
JAM-1 is a member of the immunoglobulin superfamily
(IgSF) (possessing 2 variable-region [V]-type Ig domains), and co-distributes with tight junction components
at the apical region of the junction. The carboxy-terminal
cytoplasmic tail of JAM-1 has been shown to bind guanylate kinase and/or the acidic domain of occludin, the
PSD95/dlg/ZO-1 (PDZ) domain of ZO-1, and cingulin
(22). Three different members of the JAM family have
been identified and they are heterogeneously expressed
throughout epithelial and endothelial cells (23). Recently,
it has been reported that brain endothelium expresses the
endothelial cell specific adhesion molecule ESAM (or
1G8 antigen), which is considered a structural equivalent
of JAM. ESAM also expresses a PDZ domain, although
this does not associate with ZO-1 or with ASIP/PAR3,
in contrast to JAM-1 (24).
A number of other proteins are found closely associated, or colocalized, with cellular junctions. Platelet endothelial cell adhesion molecule (PECAM-1; CD31) is
concentrated at the apical domain of the intercellular
junction, but is not structurally associated with tight junctions (25). PECAM is also a member of the IgSF and is
a single chain transmembrane glycoprotein consisting of
6 extracellular Ig-like domains and a cytoplasmic tail.
PECAM-1 is involved in cell-cell adhesion through either
homophilic interactions with other PECAM molecules or
heterophilic interactions (26) with other proteins, such as
integrin avb3 (27). Recently, altered vascular permeability
was detected in PECAM-1 deficient mice, suggesting a
role for this protein in vascular composition (28). PECAM-1 has also been demonstrated to play a role in several other functions, such as the transendothelial migration of monocytes across CNS endothelia (29). Besides
PECAM, the heavily-glycosylated molecule CD99 was
also found to be located at the intercellular borders of
human umbilical vein endothelial cells, which suggests
that CD99 may be present in endothelial junctions (30).
Whether this molecule is also present in endothelial cell
junctions along the brain microvasculature remains to be
established.
In addition to the recognized tight junction-associated
proteins, other proteins could potentially affect the function or integrity of the junctional complex. For example,
also found concentrated at tight junctions are several cytoplasmic signaling molecules (review, 31), the activation
of which may influence BBB permeability. Two types of
heterotrimeric G-proteins, Ga0 and Gai2, localized around
tight junctions were suggested to act as negative regulators for tight junction function (32). Protein kinase C has
also been implicated in the regulation of tight junctions
and is concentrated around them. Tight junction proteins
are coupled to the actin cytoskeleton and are under the
control of a number of intracellular signaling molecules.
595
The small GTPases RhoA, Rac, and Cdc42 are essential
mediators of actin reorganization (33). Recently, it was
shown that occludin is a target for this GTPase activity,
and that RhoA and its downstream kinase, p160ROCK,
are components of a signaling pathway leading to changes in tight junction permeability (34). Thus, the organization of tight junction-associated proteins may be directly or indirectly targeted by signaling molecules,
thereby providing a means for regulating the permeability
of the BBB.
THE BBB IN VITRO
Background
Nearly 3 decades have passed since the earliest descriptions of tissue culture-based paradigms of the BBB.
During this time, such in vitro systems have undergone
many reiterations and have been derived from a wide
variety of species, but are still largely comprised of enriched populations of endothelial cells derived from fragmented segments of the brain microvasculature. As the
number of reports detailing the isolation and culture of
BMECs is far too extensive to be discussed here, the
reader is referred to other excellent and up-to-date reviews that describe specifications regarding technique
(35, 36). Briefly, in most of these protocols, brain tissue
is either homogenized or pressed through filter membranes of relatively large porosity (e.g. 1,000 mm diameter) to initially fracture and disperse the parenchymal
microvessels, which are then typically separated from
contaminating cell types by density-dependent centrifugation and/or retention on filter meshes of graded pore
size (typically ranging from 20 to 70 mm in diameter).
Resulting microvessels are then treated with collagenase
to liberate BMECs. Depending on the extent of enzymatic digestion, small microvascular fragments that are
partially denuded of basement membrane, or completely
dissociated BMECs, are generated. The microvascular
fragments or BMECs are then routinely cultured on collagen substrates, mainly either collagen I, collagen IV, or
gelatin. As brain microvessels are associated with varied
types of perivascular cells in situ, such as smooth muscle
cells, pericytes, microglial cells, and astrocytes (37), it is
not uncommon for primary cultures derived from these
structures to contain components other than endothelial
cells. Hence, additional procedures have been adopted to
further purify BMECs, including fluorescence-activated
cell sorting, cell-selective lysis, differential substrate adherence, cell-selective nutrient utilization, and paramagnetic bead separation (36, 38).
These models have allowed unfettered access to
BMECs, enabling their physiologic properties and patterns of gene expression to be explored in great detail.
While growth of these cells in standard culture dishes has
permitted a plethora of data to be obtained, perhaps most
J Neuropathol Exp Neurol, Vol 62, June, 2003
596
PACHTER ET AL
significant in this area of research has been the advent of
commercially available filter inserts or permeable membrane supports (pioneered by Costar, Inc. under the brand
name Transwell), which have enabled configuration of
dual-compartment systems. In the simplest of these designs, endothelial cells are plated directly onto a filter
insert that has been coated with a collagen substrate. The
filter insert, in turn, is suspended within a microwell of
a cluster plate. When grown in this configuration,
BMECs display a physiological polarity, with the apical
culture surface reflecting the luminal microvascular surface in vivo and the basolateral culture surface representing the abluminal face of the microvasculature. The
dual-compartment system also allows for perivascular or
parenchymal cells to be grown in apposition to the
BMECs, mirroring somewhat the in vivo arrangement of
the different cell types found at or near the BBB. For
example, astrocytes, which project their foot processes
onto brain microvessels and are thought to impart at least
some of the unique properties of the BBB (39), have been
cultured on the underside of the filter insert as well as on
the floor of the microwell containing the suspended filter
(36, 39). Recently, this system has been modified so that
the BMECs are cultured atop a hydrated collagen gel
containing astrocytes (40) yielding a 3-dimensional (3D)
interactive co-culture that bears even more anatomical
similarity to the in vivo situation. The dual-compartment
arrangement, in general, is ideally suited for analyzing
transendothelial transport of soluble and cellular elements, with both upper and lower compartments being
accessible. It also represents the luminal (upper) and
abluminal (lower) aqueous environments. A more intricate extension of co-culturing BMECs and other cell
types in a 3D format has been described by Janigro and
colleagues (41), who cultured BMECs on the intraluminal
surface and astrocytes on the extraluminal surface of hollow fibers. In this system the endothelial surface can be
subject to fluid-based shear flow to the same extent encountered within microvessels in vivo, and thus offers the
opportunity to evaluate BBB properties in a near physiological microenvironment. Most recently, the same
group has advanced this technology to incorporate serotonergic neurons (42), furthering the study of neurovascular interactions at the BBB.
Features of In Vitro BBB Models
Seminal features of the BBB are variably displayed by
the myriad of in vitro models, and include tight junctions
(with corresponding reduced paracellular permeability
and high TEER value), specific transporter systems and
enzymatic activities, and low numbers of vesicles. Tight
junctions, which form the anatomical substrate of the
BBB, are features generally acknowledged to be largely
imparted by astrocyte-derived factors (for a recent review
on astrocyte influence on the BBB see [39]). Extended
J Neuropathol Exp Neurol, Vol 62, June, 2003
culturing of BMECs in the absence of such factors leads
to monolayer cultures exhibiting heightened paracellular
flux and overall attenuated barrier properties, a situation
that can partially be prevented or reversed by co-culture
of BMECs with astrocytes or astrocyte-conditioned media. While identification of these astrocyte factors remains uncertain, there is evidence that angiotensin metabolites (43) and basic fibroblast growth factor (44)
might be among them.
In addition to the anatomic restrictions imposed by
tight junctions, the presence of both specific transport
systems (many of which are carrier- or receptor-mediated
and polarized in their membrane distribution) and enzymatic activities on BMECs provide for biochemical and
metabolic qualities of the BBB, respectively. Both gamma glutamyltranspeptidase (g-GT), which catalyzes the
transfer of the g-glutamyl residue of the tripeptide glutathione to amino acids and has been postulated to function as an amino acid transport system (45), and alkaline
phosphatase have enriched expression within endothelial
cells of the brain microvasculature in situ (46). They have
also been cited as salient markers of a maintained BBB
phenotype in cultured BMECs (47). Some other transport
systems present in the BBB and detected in cultured
BMECs include those for amino acids, peptides, hexoses,
monocarboxylic acids, organic cations, nucleosides, vitamins, and various xenobiotics (review, 48). Many of
these transporters facilitate the transcellular passage of
specific solutes that are both necessary for CNS homeostasis and are unable to be synthesized de novo within
this compartment. Other transporters (e.g. P-glycoprotein
and the multidrug resistance-associated proteins) serve as
efflux proteins that preclude the entry of potentially toxic
metabolites from the blood and/or effect the export of
CNS-derived substances for targeted action in the periphery. Enzymatic activities that purportedly function in
BBB capacity include monamine oxidase and catecholO-methyl-transferase, which may act to lessen the degree
of transport of amine neurotransmitters and/or their precursor amino acids, as well as potentially toxic xenobiotics, into the CNS.
Immune Responses of In Vitro BBB Models
Like their in vivo counterparts, cultured BMECs can
be induced to express various mediators of inflammation.
For example, expression of the adhesion molecules intercellular adhesion molecule-1 (ICAM-1), vascular cell
adhesion molecule-1 (VCAM-1), and E-selectin have
been shown to be modulated by different treatments associated with leukocyte extravasation in vivo. A consistent finding by numerous independent laboratories is the
upregulated expression of these 3 adhesion molecules on
cultured BMECs treated with lipopolysaccharide (LPS)
and proinflammatory cytokines, including tumor necrosis
factor alpha (TNF-a), interleukin-1 (IL-1), and interferon
THE BLOOD-BRAIN BARRIER AND IMMUNITY IN THE CNS
gamma (IFN-g) (29, 49–52). Cytokine-stimulated ICAM-1
expression on cultured BMECs has further been shown
to display polarized membrane expression (53).
The fact that cultured BMECs display sensitivity to
immune signals has prompted their use in investigating
mechanisms regulating CNS inflammation. The dualchamber, filter format in vitro model has been most vigorously exploited in investigations of leukocyte migration
across the BBB—particularly when assaying the effects
of chemoattractants. A variety of means of quantifying
the extent of leukocyte migration across BMECs in this
configuration have been utilized, such as counting radiolabeled cells that have entered the lower compartment
(54, 55), and using conventional bright-field microscopy
to enumerate transmigrated cells that have completely traversed the filter and adhered to a coverslip (56). Most
recently, confocal microscopy combined with 3D image
reconstruction has been used to both qualitatively and
quantitatively evaluate leukocyte transendothelial migration across a 3D BBB model grown on a hydrated collagen matrix (40).
Modulation of Permeability
By far, the most extensive use of in vitro BBB models
has been to assay permeability and drug transport.
Though the vastness of this topic precludes it from being
discussed here in detail (see reference 60), suffice it to
say that efforts have largely focused on whether particular stimuli cause disruption of tight junction integrity.
For example, the proinflammatory cytokines IFN-g,
TNF-a, IL-1(a and b), and IL-6 have all been observed
to cause either increased flux of hydrophilic macromolecules or diminished TEER (57, 58) across cultured
BMECs, possibly through modulation of cyclooxygenase
and plasma membrane-associated tyrosine phosphatase activities. In contrast, the antiinflammatory cytokine IFN-b
was reported to counteract proinflammatory mediator-induced translocation of the tight junction-associated molecules ZO-1 and-2 in these cells (59), thus preserving the
integrity of the BBB in this model.
Aside from direct application of proinflammatory cytokines, hypoxic conditions and the generation of reactive
oxygen species have also been shown to cause perturbation of tight junction complexes and heightened paracellular permeability across cultured BMECs (60). On the
other hand, nitric oxide affords protection against hypoxia/reoxygenation-mediated injury of the BBB in vitro,
possibly be scavenging reactive oxygen species (61).
Proteins of human immunodeficiency virus-1 (HIV-1)
have also been shown to cause disruption of BBB integrity in vitro, thereby highlighting the means by which this
pathogen might invade the CNS and ultimately cause
HIV-1-associated dementia. Specifically, HIV-1 envelope
protein gp120 has been shown to both alter the morphology and elevate the permeability of BMEC monolayer cultures, the latter feature being antagonized by
597
spantide (a substance P antagonist) as well as anti-substance P antibody (62). The HIV Tat protein, known to
be secreted extracellularly, has been reported to stimulate
IL-8 synthesis by cultured BMECs (63), a response that
could signal loss of barrier integrity.
Caveats of In Vitro BBB Models
Notwithstanding the considerable advancements made,
one would be remiss not to view such systems with a
healthy degree of skepticism—particularly regarding
their use in investigating neuroimmune processes. In this
regard, it is important to refer back to the primary descriptions of the cellular site of the BBB and to keep in
mind the physiological context in which these were
made. Specifically, the BBB has been described as residing at the level of brain capillaries (64, 65), which demonstrate significantly reduced solute transport due to the
presence of high-resistance tight junctions and specialized transport systems. However, whether such barrier
properties are similarly manifest in postcapillary venules,
where leukocyte extravasation is primarily thought to occur, remains a matter of at least some conjecture. That
significant morphological and biochemical distinctions
have been described, even between endothelial cells from
successive segments of the vascular tree (66), legitimizes
such concern and underscores the cynicism regarding use
of other than brain microvascular-derived endothelial
cells to faithfully model the BBB in vitro (67, 68). It is
thus important to remember that BMEC cultures derived
from ‘‘microvascular fragments,’’ comprised of a collection of capillaries, arterioles, and venules, are likely to
generate, at best, mosaics of different endothelial types.
Also, cultures obtained predominantly from larger microvessels might not be able to display the barrier properties
attributed to capillaries, while cultures generated specifically from capillaries might not properly support leukocyte adhesion and/or transendothelial migration.
Aside from these matters regarding inherent endothelial differences, there is the often neglected—but nonetheless significant—issue relating hemodynamics and
blood flow to solute and cell transport across BBB models. Conventional transport assays performed with Transwell or Transwell-type, dual-chamber filter formats are
considered to be ‘‘static,’’ as they are not subject to flow
or other than gravitational forces. However, awareness
that flow-induced shear stress across the surface of monolayers of peripheral vascular endothelial cells modifies
both leukocyte diapedesis (69) and solute permeability
(70) has naturally raised speculation that flow conditions
might be similarly important in BMEC function. Indeed,
Stanness et al (41), in describing their tridimensional
BBB model cultivated on pronectin-coated hollow fibers,
argued that ‘‘endothelial cells grown under flow develop
greater differentiation than after conventional culture.’’
J Neuropathol Exp Neurol, Vol 62, June, 2003
598
PACHTER ET AL
In vitro models are clearly imperfect in reflecting the
entire complexity of the BBB. Despite this handicap,
however, their use has allowed specific BBB properties
to be evaluated that otherwise would not have been amenable to experimental scrutiny. It is expected that with
continual refinement, these model systems will come ever
closer to recapitulating the in vivo scenario.
BBB AND IMMUNE PRIVILEGE IN THE CNS
Is There Antigen Presentation at the BBB?
Antigen presentation is a crucial event for activating
naı
¨ve T cells (for more details, see review [71]).
Throughout this process, naı
¨ve T cells recognize their
specific antigen presented in the context of MHC class I
or class II molecules on antigen presenting cells (APCs).
This is considered to be the first signal for naı
¨ve T cell
activation. In order to complete the pathway of naı̈ve T
cell activation, secondary signals provided in the form of
costimulatory molecules expressed on APCs are also necessary. This activation results in changes in gene expression that drive T cell proliferation, differentiation, and
effector function in T cells. However, this activation requires a stable zone of contact formation between the T
cell and the APC. This structure is known as the immunological synapse, where there is a distinct segregation
of antigen receptors, adhesion, and signaling molecules
(72). It has been demonstrated that several hours of T
cell-APC contact were required to induce the proliferation of naı
¨ve T cells in response to antigen pulsed APC
(73). This long-term naı
¨ve T cell and APC interaction is
facilitated in specialized tissues known as the secondary
lymphoid tissues. The CNS has been considered an immunologically privileged tissue, where naı
¨ve T cell activation could not take place due to the absence of APCs
and/or insufficient expression of co-stimulatory molecules. However, in light of the potential importance of
this problem, many laboratories have revisited the question of antigen presentation at the BBB. If BMECs are
capable of presenting antigen in situ to naı
¨ve T cells, this
could potentially be important in the initiation of autoimmune diseases of the CNS. The conclusions of these
studies are quite divergent. Expression of MHC class II
molecules on BMECs in vitro, which is a prerequisite for
antigen presentation by APCs to T helper cells, has been
demonstrated by several laboratories (74). Co-stimulatory
molecule expression by in vitro cultures of BMECs has
also been demonstrated (75). The process and presentation of myelin basic protein (MBP) antigen to T cells by
in vitro cultures of murine BMECs was also reported
(76). Others have shown that both murine and guinea pig
BMECs can act as APCs for presentation of MBP to
CD41 T cells, but not for presentation of purified digested or whole ovalbumin (77). Bourdoulous et al (78)
J Neuropathol Exp Neurol, Vol 62, June, 2003
have demonstrated that IFN-g-activated BMECs, expressing MHC class II molecules, are able to stimulate proliferation of a syngeneic CD41 T cell line in the presence
of MBP. In contrast, Pryce et al (79) have demonstrated
that BMECs are poor stimulators of T cell proliferation,
despite these 2 cell types being able to engage in antigenspecific interactions in vitro.
The differential activation of T helper (Th) 1 and Th2
CD41 T cell clones was also suggested as a means by
which the BBB acts to maintain immune privilege of the
CNS (80). In these experiments, in vitro cultures of
BMECs preferred Th2 cell clones, as reflected by cell
proliferation and production of IL-4 by BMEC-activated
Th2 clones (80). As Th1 cells are the harmful effector
cells in CNS autoimmunity, the downregulating activity
of BMECs on Th1 function could potentially minimize
inflammation-associated disruption of BBB integrity.
Another point to consider is that, aside from BMECs,
other cell types that are in the vicinity of the BBB can
also present antigens to T cells. One of these cell types
might include the perivascular microglia/macrophages.
The role of these cells in shaping CNS autoimmunity has
been suggested in bone marrow chimeric animals by
Hickey and Kimura (81). These cells are capable of
phagocytosis and can constitutively express high levels
of class II antigen (82). Furthermore, using irradiated
bone marrow chimeras in CD45-congenic rats, highly purified populations of microglia and nonmicroglial, CNSassociated macrophages (CD45 highCD11b/c1) have been
isolated from the adult CNS. A minority of this CD45
high
CD11b/c1 macrophage population serves as effective
APCs, as determined by these cells participating in experimental autoimmune encephalomyelitis (EAE) and activating CD41, MBP-reactive T cells (83). Depletion of
systemic macrophages results in elimination of the perivascular microglia/macrophage population and also leads
to inhibition of the onset of EAE (84). Due both to their
close proximity to BMECs and ability to induce immune
responses, perivascular microglia/macrophages must be
considered an important, if not the most critical, APC at
the BBB.
Another mechanism by which the BBB may control
immunity in the CNS is by regulating the migration of
leukocytes across BMECs. Work from various laboratories has heightened awareness of the means by which T
cells, monocytes, and neutrophils breach the normally restrictive BBB, invade the CNS parenchyma, and ultimately cause neurologic disease.
T Cell Migration
In general, T cell migration across the BBB is a highly
regulated process involving adhesion molecules, chemokines, cytokines, and matrix metalloproteases. From a T
THE BLOOD-BRAIN BARRIER AND IMMUNITY IN THE CNS
cell point of view, the brain is considered a tertiary immune tissue that is outside the line of normal T cell circulation. Under normal conditions, T cells recirculate between the blood and secondary lymphoid tissues. This
recirculation is regulated by a group of adhesion molecules that are expressed on high endothelial venules of
the secondary lymphoid organs. Upon T cell activation,
a set of newly expressed adhesion molecules directs T
cells to inflamed tissue. The generally accepted theory is
that activated T cells, independent of their antigen specificity, can cross the BBB, but only those with specificity
for CNS antigens will be retained in the brain parenchyma (review, 85). Leukocyte transendothelial migration
across the BBB, like that which occurs in the periphery,
is a multistep process involving the following sequential
activities: 1) tethering of leukocytes to endothelial cells
by interactions between selectin and carbohydrate adhesion molecules; 2) activation of leukocytes by chemokine
stimulation of G-protein linked receptors; and 3) arrest
and transmigration of leukocytes, mediated by integrin/
immunoglobulin adhesion molecule interactions (Fig. 2).
T cell migration across nonactivated brain microvessel
endothelial monolayers has been shown to be mediated
by both LFA-1/ICAM-1 interactions and heterotrimeric
G-protein signaling pathways in BMECs (86), while
VLA-4/VCAM-1 interactions have additionally been implicated in migration across cytokine-stimulated BMECs
(87). Following transendothelial migration, there is a noticeable shift in protein expression in T cells that is characterized by downregulation of a4b1 integrins (VLA-4)
and upregulation of matrix metalloproteinases (MMP).
MMP upregulation in activated T cells enables the degradation of extracellular matrix components in order to
facilitate T cell migration into the CNS parenchyma (88).
Although several laboratories have addressed the importance of T cell migration across the BBB, the mechanism of this event is still unclear. Greater understanding
of this process would highlight potentially novel therapeutic targets to control invasion of these cells into the
CNS. While there is copious information on the importance of adhesion molecules and chemokines in the recruitment phase of CNS inflammation, the adhesion molecules and chemokines that govern the initial
antigen-specific T cell entry into the CNS have yet to be
defined. A major obstacle in studying the T cell-mediated
autoimmune process is the technical difficulty of studying
these T cells in vivo. Specifically, the frequency of antigen specific T cells represented in the total immune repertoire is below the detection limits of standard T cell
assays. However, novel technical advances, such as transgenic technology, real time PCR, and ELISPOT assay,
allow for the study of these rare T cells and their requirements for localization and function in the CNS. As an
example of applying these new technologies, T cell receptor (TCR) transgenic mice were employed (as a
599
source of monoclonal T cells with known antigen specificity to pigeon cytochrome c (PCC) residues 88–104)
to characterize adhesion molecules participating in the
early recruitment of these cells into the brain (89, 90).
By monitoring the accumulation of a small number of
antigen-specific T cells in the CSF through the use of
monoclonal antibodies to their specific T cell receptor
(Va11 or Vb3), it was demonstrated that intraventricular
delivery of PCC antigen results in a strong PECAM-1dependent antigen-specific T cell accumulation in the
CSF(89). It was also suggested that early migration of
CD41 T cells into the CNS occurs independent of the
lymphocyte integrin VLA-4 and endothelial VCAM, but
does require increased surface expression of endothelial
P selectin (91).
T cell activation and migration across the BBB is followed by recruitment of nonspecific inflammatory cells,
such as monocytes. Infiltration by monocytes significantly contributes to amplification of inflammatory reactions
in the CNS and is a major effector of tissue damage.
Monocyte Migration
Molecular mechanisms governing monocyte transmigration through brain microvascular endothelium are not
as well known as those regulating lymphocyte adhesion
and migration. The b2 integrin complement receptor-3
(CR-3 or Mac-1), which is highly expressed on monocytes, has been shown to mediate monocyte migration
across peripheral endothelial monolayers (92). In EAE,
anti-CR3 antibodies have been shown to significantly delay the onset and diminish the severity of clinical signs
of EAE, even when injections are given at the first appearance of clinical signs (93). However, in these examples, disease palliation was not accompanied by reduced cellular infiltration, implying that other
CR3-mediated cellular functions are involved in EAE,
such as myelin phagocytosis and reactive oxygen species
production (93, 94). Studies on peripheral endothelium
have suggested that the interaction of VLA-4 with
VCAM-1 is required for firm adhesion to, and subsequent
migration of, monocytes through the peripheral endothelial cell barrier (92). Similarly, an important role for the
VLA-4/VCAM-1 pathway has been demonstrated for
monocyte migration across BMECs (29). An additional
or alternative pathway, employing the integrin aDb2, has
also been implicated in the infiltration of monocytes into
the CNS. This integrin binds to VCAM-1, and treatment
with antibodies directed against aD reduced macrophage
infiltration at the lesion site of spinal cord-injured rats
(95). Whether this integrin is involved in monocyte recruitment into the CNS during EAE remains to be established. Together, these studies suggest that VCAM-1 may
be a major candidate molecule involved in the infiltration
of monocytes to the CNS. However, other IgSF members
may also participate in the transendothelial migration of
J Neuropathol Exp Neurol, Vol 62, June, 2003
600
PACHTER ET AL
Fig. 2. Molecular mechanism of leukocyte trans-endothelial migration across the BBB. The first step in leukocyte transmigration involves initial weak adhesion that is followed by reversible rolling on the BBB surface and is regulated by interaction
between selectins and their glycosylated ligands. The second step in the process of transmigration involves the stimulation or
‘‘triggering’’ of the leukocytes by chemokines and their receptors. The third step in this process involves integrin adhesion
molecule activation on leukocytes, leading to an increased adhesion of these cells to members of immunoglobulin superfamily
expressed on endothelial cells. The final step in the transmigration process is the diapedesis through the vessel wall. Molecules
currently known to be present at the junction between brain capillary endothelial cells involve CD31, JAM, and ESAM. (Please
see text for further details).
monocytes. For instance, PECAM was shown to be essential in monocyte migration across human umbilical
vein endothelium, particularly in the final step of the extravasation process (96), although the direct role of this
adhesion molecule in transendothelial migration across
BBB has not been reported. Also, other IgSF members
have a part in monocyte infiltration into the CNS. For
example, it was recently reported that the widely expressed CD47, also known as integrin-associated protein,
mediates the final postadhesion step of monocyte migration into the CNS (97). CD47 is a member of the IgSF
and contains a single extracellular Ig-like domain, 5
transmembrane segments, and a short cytoplasmic tail.
CD47 can interact with thrombospondin and also with its
monocytic ligand signal regulatory protein-a (SIRPa),
another member of the IgSF. SIRPa, also called SHPS-1
or macrophage fusion receptor, is exclusively expressed
on myeloid cells and neurons (97). SIRPa interaction
with CD47 was suggested to contribute to the recruitment
of monocytes into tissue during neuroinflammatory disease (97). Moreover, this process is mediated by CD47triggered signaling events in BMECs that in turn cause a
cytoskeletal rearrangement that facilitates monocyte migration (97).
J Neuropathol Exp Neurol, Vol 62, June, 2003
Junctional proteins like JAM also mediate monocyte
migration across brain endothelium (98). Monocyte migration into murine brain during experimental meningitis
is inhibited by using a-JAM monoclonal antibodies
(mAbs), and these antibodies have similarly been shown
to antagonize monocyte migration across cultured peripheral vascular endothelial cells (98). It is postulated that
JAM guides monocytes through endothelial cell junctions, since it is expressed at the tight junction (98). An
essential role for another junctional protein, CD99, was
recently established for monocyte migration across perivascular endothelium. In particular, CD99 appears to act
at the level of diapedesis through the tight junction (30).
It is clear that monocyte extravasation into the brain is
controlled by a number of adhesion molecule interactions
and signaling events, and that many facets of monocyte
migration across the BBB remain to be identified.
Monocytes recruited from the blood into the CNS differentiate into macrophages and contribute to neuroinflammatory processes by producing a wide range of mediators that stimulate inflammatory cascades.
Monocyte-derived inflammatory products such as proinflammatory cytokines, reactive oxygen species, or nitric
oxide can further recruit leukocytes into the CNS (99).
THE BLOOD-BRAIN BARRIER AND IMMUNITY IN THE CNS
Conversely, a neurotrophic factor like nerve growth factor may act as an anti-inflammatory agent, since it limits
the transendothelial migration of monocytes across
BMECs (100).
Current treatments of neuroinflammatory diseases, like
multiple sclerosis (MS), aim at dampening the inflammatory cascades in the CNS. For instance, IFN-b treatment leads to the reduction of new MS lesions as assessed by MRI (101). Reduced cellular infiltration may
be the result of attenuated expression of adhesion molecules on the brain endothelium, as has been reported in
EAE animals and in brain endothelial cells in vitro (29).
Cannabis, now under consideration as a potential therapeutic for MS patients, reduced spasticity and clinical
signs in EAE (102) and may influence the migration of
monocytes across the BBB. The psychoactive form of
cannabis, D or d-9-tetrahydrocannabinol, has been shown
to influence macrophage functions, including phagocytosis, antigen presentation, and migratory capacity across
endothelium (103). Lovastatin, a potent inhibitor of 3hydroxy-3-methyl glutaryl coenzyme A (HMG-CoA) reductase, a key enzyme in the cholesterol biosynthesis
pathway, may also be a promising new therapeutic agent.
Lovastatin was shown to suppress the clinical course of
EAE by inhibiting monocyte infiltration into the CNS
(104). Even sunlight influences monocyte activation
through a bioactive metabolite of vitamin D, 1,25-dihydroxyvitamin D3. Interestingly, treatment of EAE animals with this bioactive hormone was reported to decrease macrophage accumulation into the CNS (105).
Exploiting the ability of the dual-chamber model to
allow restricted access of solutes to either the apical or
basolateral surface, Andjelkovic et al (40) showed that
monocyte chemoattractant protein-1 (MCP-1) was only
able to stimulate monocyte transendothelial migration
across cultured human BMECs when exposed to the basolateral endothelial surface. This finding is consistent
with the in vivo situation wherein perivascular astrocytes
are considered major sites of MCP-1 production. Describing a similar functional polarity, Giri et al (106, 107)
reported that b-amyloid peptide (Ab) interaction, specifically with the basolateral surface of cultured human
BMECs derived from patients afflicted with Alzheimer
disease (AD), caused enhanced monocyte transendothelial migration. This has prompted these authors to postulate that perivascular accumulation of Ab in AD and
other related cerebral vascular disorders may signal the
entry of monocytes and lead to increased numbers of
potentially harmful, monocyte-derived microglia.
Neutrophil Migration
Lastly, compared to that of T cells and monocytes,
transendothelial migration of neutrophils across BBB
models has not been as intensively studied. However,
vascular endothelial growth factor (VEGF), which is
601
strongly induced during hypoxia and is accompanied during this state by an infiltration of neutrophils into the
brain, has recently been observed to stimulate neutrophil
transendothelial migration across cultured BMECs (108).
Such migration was suggested to be dependent upon
VEGF-stimulated expression of the largely neutrophiltargeting chemokine IL-8.
Full disclosure of the complex of players mediating
leukocyte migration into the CNS will be of great importance for the development of novel therapeutic strategies to effectively treat various neuroinflammatory conditions.
Conclusion
In summary, the BBB plays a very important role in
maintaining the immune-privileged status of the CNS. In
pathological conditions, the integrity of the BBB can be
breached and the inflamed brain endothelial cells can play
an important role in regulating the migration of leukocytes into the CNS. Further understanding of the unique
microenvironment that influences BBB characteristics in
the CNS will lead to novel therapies to manipulate CNS
inflammatory diseases.
Please note: Due to space considerations and the wide breadth of the
review, the list of references had to be curtailed and the authors regret
not being able to cite all.
REFERENCES
1. Davson H. History of the blood-brain barrier concept. New York:
Plenum Press, 1989
2. Roux E, Borell A. Tetanos cerebral et imunite contre le tetanos.
Ann Inst Pasteur 1898;12:225–30
3. Goldmann EE. Vitalfarbung am Zentralnervensystem. Berlin: Eimer, 1913
4. Walter FK. Die allgemeinen Grundlagen des Stoffaustusches
zwischen dem Zentralnervensystem und dem ubringen Korper.
Arch Psychit 1933;101:195–231
5. Pardridge WM, Boado RJ, Farrell CR. Brain-type glucose transporter (GLUT-1) is selectively localized to the blood-brain barrier.
Studies with quantitative western blotting and in situ hybridization. J Biol Chem 1990;265:18035–40
6. Nagy Z, Peters H, Huttner I. Fracture faces of cell junctions in
cerebral endothelium during normal and hyperosmotic conditions.
Lab Invest 1984;50:313–22
7. Brightman MW. Morphology of blood-brain interfaces. Exp Eye
Res (Suppl) 1977;25:1–25
8. Staddon JM, Rubin LL. Cell adhesion, cell junctions and the
blood-brain barrier. Curr Opin Neurobiol 1996;6:622–27
9. Butt AM, Jones HC, Abbott NJ. Electrical resistance across the
blood-brain barrier in anaesthetized rats: A developmental study.
J Physiol 1990;429:47–62
10. de Vries HE, Kuiper J, de Boer AG, Van Berkel TJ, Breimer DD.
The blood-brain barrier in neuroinflammatory diseases. Pharmacol
Rev 1997;49:143–55
11. de Vries HE, Blom-Roosemalen MC, van Oosten M, et al. The
influence of cytokines on the integrity of the blood-brain barrier
in vitro. J Neuroimmunol 1996;64:37–43
12. Furuse M, Hirase T, Itoh M, Nagafuchi A, Yonemura S, Tsukita
S. Occludin: A novel integral membrane protein localizing at tight
junctions. J Cell Biol 1993;123:1777–88
J Neuropathol Exp Neurol, Vol 62, June, 2003
602
PACHTER ET AL
13. Furuse M, Fujita K, Hiiragi T, Fujimoto K, Tsukita S. Claudin-1
and -2: Novel integral membrane proteins localizing at tight junctions with no sequence similarity to occludin. J Cell Biol 1998;
141:1539–50
14. Tsukita S, Furuse M. Occludin and claudins in tight-junction
strands: Leading or supporting players? Trends Cell Biol 1999;9:
268–73
15. Saitou M, Furuse M, Sasaki H, et al. Complex phenotype of mice
lacking occludin, a component of tight junction strands. Mol Biol
Cell 2000;11:4131–42
16. Gow A, Southwood CM, Li JS, et al. CNS myelin and sertoli cell
tight junction strands are absent in Osp/claudin-11 null mice. Cell
1999;99:649–59
17. Huber JD, Egleton RD, Davis TP. Molecular physiology and pathophysiology of tight junctions in the blood-brain barrier. Trends
Neurosci 2001;24:719–25
18. Fanning AS, Jameson BJ, Jesaitis LA, Anderson JM. The tight
junction protein ZO-1 establishes a link between the transmembrane protein occludin and the actin cytoskeleton. J Biol Chem
1998;273:29745–53
19. Haskins J, Gu L, Wittchen ES, Hibbard J, Stevenson BR. ZO-3,
a novel member of the MAGUK protein family found at the tight
junction, interacts with ZO-1 and occludin. J Cell Biol 1998;141:
199–208
20. Citi S, Sabanay H, Kendrick-Jones J, Geiger B. Cingulin: Characterization and localization. J Cell Sci 1989;93:107–22
21. Zhong Y, Saitoh T, Minase T, Sawada N, Enomoto K, Mori M.
Monoclonal antibody 7H6 reacts with a novel tight junction-associated protein distinct from ZO-1, cingulin and ZO-2. J Cell
Biol 1993;120:477–83
22. Bazzoni G, Martinez-Estrada OM, Orsenigo F, Cordenonsi M, Citi
S, Dejana E. Interaction of junctional adhesion molecule with the
tight junction components ZO-1, cingulin, and occludin. J Biol
Chem 2000;275:20520–26
23. Aurrand-Lions M, Johnson-Leger C, Wong C, Du Pasquier L, Imhof BA. Heterogeneity of endothelial junctions is reflected by differential expression and specific subcellular localization of the
three JAM family members. Blood 2001;98:3699–3707
24. Nasdala I, Wolburg-Buchholz K, Wolburg H, et al. A transmembrane tight junction protein selectively expressed on endothelial
cells and platelets. J Biol Chem 2002;277:16294–303
25. Ayalon O, Sabanai H, Lampugnani MG, Dejana E, Geiger B. Spatial and temporal relationships between cadherins and PECAM-1
in cell-cell junctions of human endothelial cells. J Cell Biol 1994;
126:247–58
26. Muller WA, Berman ME, Newman PJ, DeLisser HM, Albelda
SM. A heterophilic adhesion mechanism for platelet/endothelial
cell adhesion molecule 1 (CD31). J Exp Med 1992;175:1401–4
27. Buckley CD, Doyonnas R, Newton JP, et al. Identification of alpha
v beta 3 as a heterotypic ligand for CD31/PECAM-1. J Cell Sci
1996;109:437–45
28. Graesser D, Solowiej A, Bruckner M, et al. Altered vascular permeability and early onset of experimental autoimmune encephalomyelitis in PECAM-1-deficient mice. J Clin Invest 2002;109:383–
92
29. Floris S, Ruuls SR, Wierinckx A, et al. Interferon-beta directly
influences monocyte infiltration into the central nervous system. J
Neuroimmunol 2002;127:69–79
30. Schenkel AR, Mamdouh Z, Chen X, Liebman RM, Muller WA.
CD99 plays a major role in the migration of monocytes through
endothelial junctions. Nat Immunol 2002;3:143–50
31. Mitic LL, Anderson JM. Molecular architecture of tight junctions.
Annu Rev Physiol 1998;60:121–42
32. Denker BM, Saha C, Khawaja S, Nigam SK. Involvement of a
heterotrimeric G protein alpha subunit in tight junction biogenesis.
J Biol Chem 1996;271:25750–53
J Neuropathol Exp Neurol, Vol 62, June, 2003
33. Hall A. Rho GTPases and the actin cytoskeleton. Science 1998;
279:509–14
34. Hirase T, Kawashima S, Wong EY, et al. Regulation of tight junction permeability and occludin phosphorylation by Rhoap160ROCK-dependent and -independent mechanisms. J Biol
Chem 2001;276:10423–31
35. Gumbleton M, Audus KL. Progress and limitations in the use of
in vitro cell cultures to serve as a permeability screen for the
blood-brain barrier. J Pharm Sci 2001;90:1681–98
36. Deli MA, Joo F. Cultured vascular endothelial cells of the brain.
Keio J Med 1996;45:183–98; discussion 198–89
37. Pardridge WM. Blood-brain barrier biology and methodology. J
Neurovirol 1999;5:556–69
38. Lamszus K, Schmidt NO, Ergun S, Westphal M. Isolation and
culture of human neuromicrovascular endothelial cells for the
study of angiogenesis in vitro. J Neurosci Res 1999;55:370–81
39. Prat A, Biernacki K, Wosik K, Antel JP. Glial cell influence on
the human blood-brain barrier. Glia 2001;36:145–55
40. Andjelkovic AV, Zochowski MR, Morgan F, Pachter JS. Qualitative and quantitative analysis of monocyte transendothelial migration by confocal microscopy and three-dimensional image reconstruction. In Vitro Cell Dev Biol Anim 2001;37:111–20
41. Stanness KA, Westrum LE, Fornaciari E, et al. Morphological and
functional characterization of an in vitro blood-brain barrier model. Brain Res 1997;771:329–42
42. Stanness KA, Neumaier JF, Sexton TJ, et al. A new model of the
blood–brain barrier: Co-culture of neuronal, endothelial and glial
cells under dynamic conditions. Neuroreport 1999;10:3725–31
43. Kakinuma Y, Hama H, Sugiyama F, et al. Impaired blood-brain
barrier function in angiotensinogen-deficient mice. Nat Med 1998;
4:1078–80
44. Sobue K, Yamamoto N, Yoneda K, et al. Induction of blood-brain
barrier properties in immortalized bovine brain endothelial cells
by astrocytic factors. Neurosci Res 1999;35:155–64
45. Orlowski M, Meister A. The gamma-glutamyl cycle: A possible
transport system for amino acids. Proc Natl Acad Sci U S A 1970;
67:1248–55
46. Wolff JE, Belloni-Olivi L, Bressler JP, Goldstein GW. Gammaglutamyl transpeptidase activity in brain microvessels exhibits regional heterogeneity. J Neurochem 1992;58:909–15
47. DeBault LE, Cancilla PA. gamma-Glutamyl transpeptidase in isolated brain endothelial cells: Induction by glial cells in vitro. Science 1980;207:653–55
48. Tamai I, Tsuji A. Transporter-mediated permeation of drugs across
the blood-brain barrier. J Pharm Sci 2000;89:1371–88
49. Fabry Z, Raine CS, Hart MN. Nervous tissue as an immune compartment: The dialect of the immune response in the CNS. Immunology Today 1994;15:218–24
50. Wong D, Dorovini-Zis K. Regulation By Cytokines and lipopolysaccharide of E-selectin expression by human brain microvessel
endothelial cells in primary culture. J Neuropathol Exp Neurol
1996;55:225–35
51. Stanimirovic DB, Wong J, Shapiro A, Durkin JP. Increase in surface expression of ICAM-1, VCAM-1 and E-selectin in human
cerebromicrovascular endothelial cells subjected to ischemia-like
insults. Acta Neurochir Suppl (Wien) 1997;70:12–16
52. Gelati M, Corsini E, Dufour A, et al. High-dose methylprednisolone reduces cytokine-induced adhesion molecules on human
brain endothelium. Can J Neurol Sci 2000;27:241–44
53. Staykova M, Maxwell L, Willenborg D. Kinetics and polarization
of the membrane expression of cytokine-induced ICAM-1 on rat
brain endothelial cells. J Neuropathol Exp Neurol 2000;59:120–
28
54. Fabry Z, Topham DJ, Fee D, et al. TGF-beta 2 decreases migration
of lymphocytes in vitro and homing of cells into the central nervous system in vivo. J Immunol 1995;155:325–32
THE BLOOD-BRAIN BARRIER AND IMMUNITY IN THE CNS
55. Biegel D, Pachter JS. Growth of brain microvessel endothelial
cells on collagen gels: Applications to the study of blood-brain
barrier physiology and CNS inflammation. In Vitro Cell Dev Biol
Anim 1994;30A:581–88
56. Persidsky Y. Model systems for studies of leukocyte migration
across the blood-brain barrier. J Neurovirol 1999;5:579–90
57. Huynh HK, Dorovini-Zis K. Effects of interferon-gamma on primary cultures of human brain microvessel endothelial cells. Am
J Pathol 1993;142:1265–78
58. Gloor SM, Weber A, Adachi N, Frei K. Interleukin-1 modulates
protein tyrosine phosphatase activity and permeability of brain endothelial cells. Biochem Biophys Res Commun 1997;239:804–9
59. Kuruganti PA, Hinojoza JR, Eaton MJ, Ehmann UK, Sobel RA.
Interferon-beta counteracts inflammatory mediator-induced effects
on brain endothelial cell tight junction molecules-implications for
multiple sclerosis. J Neuropathol Exp Neurol 2002;61:710–24
60. Lagrange P, Romero IA, Minn A, Revest PA. Transendothelial
permeability changes induced by free radicals in an in vitro model
of the blood-brain barrier. Free Radic Biol Med 1999;27:667–72
61. Utepbergenov DI, Mertsch K, Sporbert A, et al. Nitric oxide protects blood-brain barrier in vitro from hypoxia/reoxygenation-mediated injury. FEBS Lett 1998;424:197–201
62. Annunziata P, Cioni C, Toneatto S, Paccagnini E. HIV-1 gp120
increases the permeability of rat brain endothelium cultures by a
mechanism involving substance P. Aids 1998;12:2377–85
63. Hofman FM, Chen P, Incardona F, Zidovetzki R, Hinton DR. HIV-1
tat protein induces the production of interleukin-8 by human brainderived endothelial cells. J Neuroimmunol 1999;94:28–39
64. Goldstein GW, Betz AL. Recent advances in understanding brain
capillary function. Ann Neurol 1983;14:389–95
65. Reese TS, Karnovsky MJ. Fine structural localization of a bloodbrain barrier to exogenous peroxidase. J Cell Biol 1967;34:207–17
66. Ghitescu L, Robert M. Diversity in unity: The biochemical composition of the endothelial cell surface varies between the vascular
beds. Microsc Res Tech 2002;57:381–89
67. MacLean AG, Orandle MS, Alvarez X, Williams KC, Lackner
AA. Rhesus macaque brain microvessel endothelial cells behave
in a manner phenotypically distinct from umbilical vein endothelial cells. J Neuroimmunol 2001;118:223–32
68. Thorin E, Shatos MA, Shreeve SM, Walters CL, Bevan JA. Human vascular endothelium heterogeneity. A comparative study of
cerebral and peripheral cultured vascular endothelial cells. Stroke
1997;28:375–81
69. Cinamon G, Grabovsky V, Winter E, et al. Novel chemokine functions in lymphocyte migration through vascular endothelium under
shear flow. J Leukoc Biol 2001;69:860–66
70. Zilberberg J, Harris NR. Role of shear and leukocyte adherence
on venular permeability in the rat mesentery. Microvasc Res 2001;
62:215–25
71. Brodsky FM, Guagliardi LE. The cell biology of antigen processing and presentation. Ann Rev Immunol 1991;9:707–44
72. Monks CR, Freiberg BA, Kupfer H, Sciaky N, Kupfer A. Threedimensional segregation of supramolecular activation clusters in
T cells. Nature 1998;395:82–86
73. Costello PS, Gallagher M, Cantrell DA. Sustained and dynamic
inositol lipid metabolism inside and outside the immunological
synapse. Nat Immunol 2002;3:1082–89
74. Male D, Pryce G. Kinetics of MHC gene expression and mRNA
synthesis in brain endothelium. Immunology 1988;63:37–42
75. Omari KI, Dorovini-Zis K. Expression and function of the costimulatory molecules B7–1 (CD80) and B7–2 (CD86) in an in vitro
model of the human blood–brain barrier. J Neuroimmunol 2001;
113:129–41
76. McCarron RM, Spatz M, Kempski O, Hogan RN, Muehl L,
McFarlin DE. Interaction between myelin basic protein-sensitized
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
603
T lymphocytes and murine cerebral vascular endothelial cells. J
Immunol 1986;137:3428–35
Rollins SA, Kennedy SP, Chodera AJ, Elliott EA, Zavoico GB,
Matis LA. Evidence that activation of human T cells by porcine
endothelium involves direct recognition of porcine SLA and costimulation by porcine ligands for LFA-1 and CD2. Transplantation 1994;57:1709–16
Bourdoulous S, Beraud E, Le Page C, et al. Anergy induction in
encephalitogenic T cells by brain microvessel endothelial cells is
inhibited by interleukin-1. European Journal of Immunology 1995;
25:1176–83
Pryce G, Male D, Sedgwick J. Antigen presentation in brain: Brain
endothelial cells are poor stimulators of T-cell proliferation. Immunology 1989;66:207–12
Fabry Z, Sandor M, Gajewski TF, et al. Differential activation of
Th1 and Th2 CD41 cells by murine brain microvessel endothelial
cells and smooth muscle/pericytes. J Immunol 1993;151:38–47
Hickey WF, Kimura H. Perivascular microglial cells of the CNS
are bone marrow-derived and present antigen in vivo. Science
1988;239:290–92
Ulvestad E, Williams K, Bo L, Trapp B, Antel J, Mork S. HLA
class II molecules (HLA-DR, -DP, -DQ) on cells in the human
CNS studied in situ and in vitro. Immunology 1994;82:535–41
Ford AL, Goodsall AL, Hickey WF, Sedgwick JD. Normal adult ramified microglia separated from other central nervous system macrophages by flow cytometric sorting. Phenotypic differences defined and
direct ex vivo antigen presentation to myelin basic protein-reactive
CD41 T cells compared. J Immunol 1995;154:4309–21
Tran EH, Hoekstra K, van Rooijen N, Dijkstra CD, Owens T.
Immune invasion of the central nervous system parenchyma and
experimental allergic encephalomyelitis, but not leukocyte extravasation from blood, are prevented in macrophage-depleted mice.
J Immunol 1998;161:3767–75
Hickey WF. Migration of hematogenous cells through the bloodbrain barrier and the initiation of CNS inflammation. Brain Pathol
1991;1:97–105
Adamson P, Wilbourn B, Etienne-Manneville S, et al. Lymphocyte
trafficking through the blood-brain barrier is dependent on endothelial cell heterotrimeric G-protein signaling. Faseb J 2002;16:
1185–94
Greenwood J, Wang Y, Calder VL. Lymphocyte adhesion and
transendothelial migration in the central nervous system: The
role of LFA-1, ICAM-1, VLA-4 and VCAM-1. Immunology
1995;86:408–15
Romanic AM, Graesser D, Baron JL, Visintin I, Janeway CA, Jr,
Madri JA. T cell adhesion to endothelial cells and extracellular
matrix is modulated upon transendothelial cell migration. Lab Invest 1997;76:11–23
Qing Z, Sandor M, Radvany Z, et al. Inhibition of antigen-specific T cell trafficking into the central nervous system via blocking PECAM1/CD31 molecule. J Neuropathol Exp Neurol 2001;
60:798–807
Qing Z, Sewell D, Sandor M, Fabry Z. Antigen-specific T cell
trafficking into the central nervous system. J Neuroimmunol 2000;
105:169–78
Carrithers MD, Visintin I, Kang SJ, Janeway CA, Jr. Differential
adhesion molecule requirements for immune surveillance and inflammatory recruitment. Brain 2000;123:1092–1101
Shang XZ, Lang BJ, Issekutz AC. Adhesion molecule mechanisms
mediating monocyte migration through synovial fibroblast and endothelium barriers: Role for CD11/CD18, very late antigen-4
(CD49d/CD29), very late antigen-5 (CD49e/CD29), and vascular
cell adhesion molecule-1 (CD106). J Immunol 1998;160:467–74
Huitinga I, Damoiseaux JG, Dopp EA, Dijkstra CD. Treatment
with anti-CR3 antibodies ED7 and ED8 suppresses experimental
J Neuropathol Exp Neurol, Vol 62, June, 2003
604
94.
95.
96.
97.
98.
99.
100.
PACHTER ET AL
allergic encephalomyelitis in Lewis rats. Eur J Immunol 1993;23:
709–15
van der Laan LJ, Ruuls SR, Weber KS, Lodder IJ, Dopp EA,
Dijkstra CD. Macrophage phagocytosis of myelin in vitro determined by flow cytometry: Phagocytosis is mediated by CR3 and
induces production of tumor necrosis factor-alpha and nitric oxide.
J Neuroimmunol 1996;70:145–52
Mabon PJ, Weaver LC, Dekaban GA. Inhibition of monocyte/
macrophage migration to a spinal cord injury site by an antibody
to the integrin alphaD: A potential new anti- inflammatory treatment. Exp Neurol 2000;166:52–64
Liao F, Ali J, Greene T, Muller WA. Soluble domain 1 of plateletendothelial cell adhesion molecule (PECAM) is sufficient to block
transendothelial migration in vitro and in vivo. J Exp Med 1997;
185:1349–57
de Vries HE, Hendriks JJ, Honing H, et al. Signal-regulatory protein alpha-CD47 interactions are required for the transmigration
of monocytes across cerebral endothelium. J Immunol 2002;168:
5832–39
Martin-Padura I, Lostaglio S, Schneemann M, et al. Junctional
adhesion molecule, a novel member of the immunoglobulin superfamily that distributes at intercellular junctions and modulates
monocyte transmigration. J Cell Biol 1998;142:117–27
Van der Goes A, Wouters D, Van Der Pol SM, et al. Reactive
oxygen species enhance the migration of monocytes across the
blood-brain barrier in vitro. Faseb J 2001;15:1852–54
Flugel A, Matsumuro K, Neumann H, et al. Anti-inflammatory
activity of nerve growth factor in experimental autoimmune encephalomyelitis: Inhibition of monocyte transendothelial migration. Eur J Immunol 2001;31:11–22
J Neuropathol Exp Neurol, Vol 62, June, 2003
101. Stone LA, Frank JA, Albert PS, et al. Characterization of MRI
response to treatment with interferon beta-1b: Contrast-enhancing
MRI lesion frequency as a primary outcome measure. Neurology
1997;49:862–69
102. Baker D, Pryce G, Croxford JL, et al. Cannabinoids control spasticity and tremor in a multiple sclerosis model. Nature 2000;404:
84–87
103. Sacerdote P, Massi P, Panerai AE, Parolaro D. In vivo and in vitro
treatment with the synthetic cannabinoid CP55, 940 decreases the
in vitro migration of macrophages in the rat: Involvement of both
CB1 and CB2 receptors. J Neuroimmunol 2000;109:155–63
104. Stanislaus R, Singh AK, Singh I. Lovastatin treatment decreases
mononuclear cell infiltration into the CNS of Lewis rats with experimental allergic encephalomyelitis. J Neurosci Res 2001;66:
155–62
105. Nashold FE, Miller DJ, Hayes CE. 1,25-dihydroxyvitamin D3
treatment decreases macrophage accumulation in the CNS of mice
with experimental autoimmune encephalomyelitis. J Neuroimmunol 2000;103:171–79
106. Giri R, Selvaraj S, Miller CA, et al. Effect of endothelial cell
polarity on beta-amyloid-induced migration of monocytes across
normal and AD endothelium. Am J Physiol Cell Physiol 2002;
283:C895–C904
107. Giri R, Shen Y, Stins M, et al. Beta-amyloid-induced migration of
monocytes across human brain endothelial cells involves RAGE
and PECAM-1. Am J Physiol Cell Physiol 2000;279:C1772–
C1781
108. Lee TH, Avraham H, Lee SH, Avraham S. Vascular endothelial
growth factor modulates neutrophil transendothelial migration via
up-regulation of interleukin-8 in human brain microvascular endothelial cells. J Biol Chem 2002;277:10445–51