In studying a model inflammatory lesion in the nervous system, such as that of EAE (experimental autoimmune encephalomyelitis), or a naturally occurring lesion in man, such as the plaque of MS (multiple sclerosis), we attempt to establish the presence and significance of one or more standard immunopathologic mechanisms (Boese, ed., 1980; Waksman, 1981). These were first distinguished over a half century ago, but the detailed cellular and molecular events underlying each are still in the process of elucidation.
The term BBB is used to describe the restricted movement of various metabolically important micro- and macro-molecules essentially soluble(solutes), between blood and brain. The function of the BBB rests at the level of endothelial cells (ECs) and results from their special ultrastructural and biochemical properties (6,8,9). Although this barrier is comprised of brain microvessels, choroid plexus epithelium and arachnoid membrane, our main attention will be focused on microvessels (capillaries, precapillary arterioles and postcapillary venules) where extensive processes of exchange of various metabolites between blood and brain parenchyma take place.
The blood-brain barrier (BBB) is selectively permeable and shows a wide range of permeability to various substances in health and becomes generally permeable to all small molecules in almost all disease states. The capillary bed in brain (the BBB) is uniquely impermeable to many polar substances. The mechanism by which the capillary bed is maintained in this state is unknown, but may be an astrocytic function. In most major neuropathology, the BBB is lost because these unique structural features creating the BBB come to resemble the permeable capillaries in other tissues. Early isotopic brain scans using polar isotopes such as 99m-TcO4 depend upon this increased permeability to demonstrate focal lesions. Positron emission tomography has made it possible to label interesting biochemical substrates allowing the tracing of certain aspects of regional brain metabolic activity. The equipment required is very expensive, but single photon tomography using lipophilic tracers is much less expensive and provides tomographic representation of regional blood flow. To penetrate a healthy BBB requires that radiotracers either be lipid soluble or have an affinity for one of the selective BBB transport systems.
Experimental allergic encephalomyelitis (EAE) is an autoimmune disease resulting from injection of homogenates of central nervous system tissue. Acute variants of this disease have been extensively studied during the past decades and from the view of pathohistology can be regarded as suitable models for acute inflammatory demyelinating diseases in humans (acute perivenous leucoencephalomyelitis and acute hemorrhagic leucoencephalomyelitis; Alvord 1970, Levine 1974). More recently reproducible models of chronic progressive or chronic relapsing EAE have been introduced using various animal species (Stone and Lerner 1965, Wisniewski and Keith 1977, Massanari 1980, Lublin et al.1981) which pathohistologically differ from acute models by the large extent of demyelination in the lesions (Raine et al.1974, Wisniewski and Keith 1977, Lassmann and Wisniewski 1979, Lassmann 1983). A more extensive comparative neuro- pathological study of cr-EAE and MS lesions revealed a close similarity in the essential features of the demyelinated plaques (inflammation, demyelination and sclerosis) as well as in more variable aspects like lesional topography, patterns of plaque growth, oligodendroglia involvement and remyelination, vascular pathology, axonal and neuronal alterations and others (Lassmann 1983). Differences in pathohistology of the two diseases are rather found in quatitative than in qualitative aspects (e.g. in the size of the lesions, or in the degree of oligodendroglia involvement and remyelinating).
In recent years, considerable interest has focused on the possibility that in experimental allergic encephalomyelitis (EAE), antigen other than myelin basic protein (MBP) may be required for the initiation of demyelination and for the development of exacerbating-remitting disease. Previous results from these laboratories have implicated a role for antibodies against galactocerebroside (GC) in initiating demyelination of central nervous system (CNS) tissue in vitro (8) and in vivo (9). We have now used the rabbit eye model to dissect further the role of antibodies in causing CNS demyelination. The results show: that in animals directly sensitized against GC, no spontaneous CNS lesion develops but primary demyelination is observed if a mononuclear inflammatory reaction is superimposed; that in rabbits sensitized against MBP, antiserum against GC causes enhanced demyelination; and that in normal animals, anti-GC serum initiates primary demyelination only when an inflammatory reaction is induced by supernatants of activated lymphocytes. Injection of anti-GC serum alone has no pathologic effect. These results suggest that antibodies against lipid haptens are capable of causing primary demyelination in the CNS in vivo but that effector cells provided by an inflammatory response are required. Thus, the development of the fully demyelinating lesion probably depends on both cellular and humoral mechanisms.
Demyelinating disease is characterized by breakdown of myelin sheaths based upon allergic processes. Detailed mechanism involved in the processes is not clearly explored yet. However, demyelinating agents have been suggested to be present in the circulating fluid, lymphocytes and brain tissue. The agent was first described by Bornstein and Appel (1) in 1961. They applied sera from EAE rabbits to the myelinated cerebellar cultures and found the sera produced myelin breakdown without involvement of axons. The demyelinating agent was found to be a complement dependent antibody against myelin sheaths (2). In 1968, similar demyelinating agent was reported by Yonezawa et al (3) in EAN rabbit sera. In 1969, Arnason, Winkler & Hadler (4) reported a cellular factor of demyelinating agent. Lymph node cells obtained from EAN rabbits were applied to ganglion cultures and they found the node cells produced delayed type of demyelination. Identical study was performed by us with EAN node cells (5). On the studies of node cells from EAE animals were also reported by us (6) and Bornstein and Iwanami (7).
The changes of the blood-brain barrier (BBB) play undoubtedly a significant role in pathophysiology of cerebrovascular disorders. First of all, breakdown of the BBB associated with extravasation of serum proteins leads to development of vasogenic brain edema in the brain tissue. Furthermore, there may occur, as it was observed in cerebral ischemia, an abnormal passage from the blood of various pharmacologically active substances, such as biogenic amines, entry of which is otherwise restricted by a normally functioning barrier. Finally, the BBB dysfunction, associated with faulty out-transport of metabolites and removal of toxic waste products, may contribute to tissue damage.
It is generally accepted that the blood brain barrier (BBB) is very important in the regulation of the internal environment of the brain. Because of its very large surface area relative to the choroid plexus ( 5000:1) and its intimate contact with the neuropil it must be viewed as a major component in such a regulatory process. Despite the potential importance of the BBB in the regulation of brain fluid and electrolyte balance as well as substrate delivery, its role has been viewed as largely static (24) in contrast to the choroid plexus, where dynamic regulation of fluid and electrolyte exchange is well recognized although incompletely characterized. Thus, the BBB is often viewed as a membrane system that once developed, very effectively excludes a wide variety of substances from the central nervous system (e.g., various ions, proteins), while through the mechanism of facilitated diffusion, admits and discharges a limited number of essential compounds (e.g., glucose, specific amino acids).
Probably, the most important evidence concerning the breakdown of the BBB is a large inflow of hematogenous fluid into the extravascular spaces. Thus all parenchymal cell elements represent freely floating cells in this fluid medium called the edema fluid. These essential morphological alterations, such as extremely expanded extracellular space and freely floating cells within the fluid, were also observed in the developing normal fetal brain. Many neuroblasts were vigorously migrating in the fetal brain, as were the macrophages and reactive astrocytes in the edema fluid. Obviously, hematogenous cells and reactive astrocytes in the edematous lesion take part in its repair. Many astrocytes, GFAP positive, in the 3 or 4 day-old lesions, revealed mitosis. And in vitro, when we immersed these astrocytes in the protein and glucose rich medium, they demonstrated a remarkably changed morphology and were moved into the M and G1 phases, thus gaining the ability of cell motility. This was also true in the edema fluid. Brain edema is definitely a serious "pathological" condition. But it is also conceivable from a different biological aspect that as a result of BBB "opening", free extracellular space essential for cell motility and a source of their energy is ingeniously provided for these cells, and thus the lesion can be effectively repaired. The biological significance of the edema fluid was emphasized and stereotaxic morphology and cinematography, supporting the above evidence, have also been presented.
Experimental tumors and abscesses were produced by intrahemispheric inoculation of a blastomatous glial cell clone and of staphylococcus aureus, respectively. In both models severe vasogenic brain edema developed. The site of the barrier lesion was identified by systemic application of Evans blue or peroxidase, and the spread of edema by immunoautoradiographic localisation of extravasated serum proteins. In both experimental conditions, serum proteins accumulated diffusely in the white matter of the ipsilateral hemisphere, although the barrier lesion was strictly confined to the pathological focus. Water content of the edematous white matter in the vicinity of tumors and abscesses increased from 69.1 to 80.6 and 82.3 ml/100g w.w., respectively. This increase was associated with a volume-dependent decrease of flow, a parallel increase of sodium and an increase of extravasated serum proteins. The latter was determined by a newly developed immunochemical approach with appropriate corrections for the intravascular fraction of total serum protein content. The calculated concentration of sodium in edema fluid of tumors and abscesses amounted to 132 and 129 ueq/ml, respectively. The concentration of serum proteins was 8.7 and 6.4 mg/ml, respectively. Protein content of edema fluid, in consequence was less than 10% of blood serum. This suggests that fluid accumulation in vasogenic edema cannot be explained by the oncotic properties of extravasated proteins alone.
Considering the current evidence available on pathophysiological mechanisms of blood-brain barrier disturbances (cf. above contributions) it appears questionable, whether or not methods of treatment today in use influence the specific processes involved. Nevertheless, this paper is an attempt, first, to analyse the influence of conventional forms of treatment of acute brain damage on barrier mechanisms, then, to comment on drugs which experimentally were found to protect the blood-brain barrier, and finally, to discuss the therapeutic potential of induced opening of the blood-brain barrier. Obviously, a most pertinent result of barrier damage occurring under clinical circumstances is vasogenic brain edema, acutely developing in head injury, focal ischemia, subacutely or chronically in brain tumours, brain abscess, inflammation, etc.
Compared with other organs of the body the brain and spinal cord show unique behaviour as far as the exchange of metabolites to and from the blood is concerned. This peculiarity has led to the concept of the so-called "blood-brain barrier" (BBB). The concept is useful not only when dealing with questions of metabolic exchange in general but more specifically for understanding the mechanism of the production of and the changes in the cerebral spinal fluid; the permeability of the brain for toxins, viruses and antibodies; and, finally, the capability of various drugs to reach the brain. The pathologist is often confronted with blood-brain barrier function since blood pigments appear in the cerebrospinal fluid in certain diseases. For example, in neonatal jaundice the yellow pigment reaches the orain because of an insufficiently developed blood-brain barrier; the phenomenon cannot be found in adults.
Homeostasis of the neuronal microenvironment is brought about by the selective permeability of the cerebral endothelium. The endothelium is able, in part, to transport certain substrates while excluding other molecules because it comprises a continuous barrier that cannot be bypassed extra-cellularly. Each endothelial cell is joined to its neighbor by belts of tight junctions which occlude the intercellular clefts between them. In this way, the extracellular fluid (ECF) of the central nervous system (CNS) is isolated from the blood (33,4).
Hydrophilic substrates necessary for brain function cross the capillary by facilitated diffusion. The facilitation has many features in common with enzyme-catalyzed reactions and is probably subserved by protein entities in the endothelial wall. The proteins act as receptors, recognizing substrate molecules, and as translocators, giving the molecules access to an aqueous path through the endothelium. These receptor-translocators can be saturated, and the transport is subject to competitive inhibition by substrate analogs. Thus, amino acids inhibit the transport of each other, and galactose can inhibit glucose transport in suckling rats. The proteins can be induced, as in the case of ketone transport in starvation, and repressed, as in the case of glucose transport in hyperglycemia. In rats with hyperglycemia for three weeks, the maximum glucose transport capacity of the blood-brain barrier decreased from 400 to 290 mumol/hg/min. An important result of the description is the understanding that rigid distinctions between the function of receptors, translocators, and enzymes is impossible. Understanding of the biochemical properties of facilitated diffusion may help explain a variety of symptoms in many 'inborn errors of metabolism'. This understanding has followed greater, recent insights into the general properties of the blood-brain barrier (45,46,47).
The present report, compares two murine models of virus induced chronic relapsing demyelination. MHV-induced demyelination in the BALB/c mouse results from the direct virus mediated cytolysis of oligodendrocytes. Extensive remyelination by oligodendrocytes is noted. Recurrent demyelination occurs in small areas. Infectious virus persists and viral antigens are localized within oligodendrocytes and their processes. TMEV-induced demyelination in SJL/J mice is associated with perivascular inflammatory infiltrates and is diminished by immunosuppressive measures. Remyelination by oligodendrocytes is delayed and incomplete. Chronic demyelination is widespread and associated with perivascular inflammatory infiltrates. The virus persists and viral antigen is localized within oligodendrocytes. The findings indicate virus persistence in oligodendrocytes in both models. Demyelination follows the disintegration of infected oligodendrocytes. Virus replication in oligodendrocytes is responsible for cell lysis in the MHV model whereas immune mediated injury of infected oligodendrocytes is considered to play a role in the pathogenesis of demyelination in the TMEV model.
