key: cord-022395-rk31pwoa
authors: Schuller-Levis, Georgia; Kozlowski, Piotr B.; Kascsak, Richard J.
title: Central Nervous System: Viral Infection and Immune-Mediated Inflammation
date: 2012-12-02
journal: Xenobiotics and Inflammation
DOI: 10.1016/b978-0-12-628930-5.50019-9
sha: 
doc_id: 22395
cord_uid: rk31pwoa

nan

locally disabled for long periods of time (Reese and Karnovsky, 1967; Hirano etal, 1970) . A peculiarity of the BBB is that it excludes macromolecules from entering the CNS but allows megastructures such as entire hematogenous inflammatory cells to cross the barrier without opening the tight junctions (Hirano et al, 1970; Azarelli et al, 1984; Lossinsky et al, 1989; Raine et al, 1990; Powell, 1991) . Nonetheless, the entry of hematogenous inflammatory cells into the CNS appears to be strictly controlled by the ECs.

An initial event in the inflammatory reaction in the CNS is the egress of hematogenous cells into the extravascular milieu of the CNS parenchyma ( Figure 1 ). Only the previously sensitized inflammatory cells, i.e., lympho cytes, can invade the CNS (Raine et al, 1990; Powell, 1991) . The prerequisite for transendothelial passage is the attachment of sensitized cells to the luminal surface of endothelial cells. A growing body of evidence suggests that lymphocytes and monocytes (but not neutrophils) attach to parajunctional endothelial cell receptors and subsequently insert pseudopodial pro jections into tubular, channel-like openings in the endothelial cells (Lossin sky et al, 1989; Powell et al, 1991) . These channels serve as eventual conduits for the transendothelial cell passage into the extravascular space. Numerous studies have shown a great variety of specific membrane recep tors on endothelial cells that are critical for the binding of hematogenous cells. These highly specific receptors include several distinct classes of adhesion molecules. These receptors include the immunoglobulin super family (ICAM-1 and -2, LFA-3, VCAM-1, ENDOCAM-1, and NCAM-1); the integrins (LFA-1 or C D l l a / 1 8 , MAC-1, LPAM-1, 0 1 , β2, β3 families); the LEC-CAM family for lectins ; Cadherins (Ρ, Ν, E-Cadherins), and the Hermes group of antigens (CD-44) (Harlan, 1985; Dustin and Springer, 1988; Lassmann et al, 1991 , Staunton et al, 1990 Yednock, 1985) . The expression of these receptors, also called addressins, on the endothelial surface is under the control of several cytokines, some of which are present only in local inflammatory response and absent under normal conditions. Neutrophils, which also require addressins for adherence, are able to enter the CNS by opening and passing through a tight junction between adjacent endothelial cells (Broadwell, 1989) .

Second, there are structural cells of the CNS that, in addition to their "main" roles, may, under certain conditions, play a role in antigen presenta tion and/or processing. These cells include astrocytes, endothelial cells, and microglial cells (Fontana et al, 1984) . Under normal conditions, only a small portion of meningeal, perivascular dendritic cells and resident mi croglial cells may express low levels of Class II major histocompatibility complex (MHC) antigens. The expression of these antigens on various cell types seems to be related to the severity of the inflammatory response. In mild inflammation, Class II MHC antigen expression is upregulated mostly on microglia cells; only in severe inflammation can some expression of these antigens be seen on astrocytes. Gamma interferon (IFN-γ) is known to stimulate in vitro the expression of Class II MHC antigens on endothelial cells, microglial cells, and astrocytes (Wong et al., 1984; Hirsch et al., 1983; Fierz et al., 1985) . The nonhematogenous cells that are native to the CNS have a very limited role in CNS inflammation. Neurons, oligodendroglia cells, and ependymal cells, if not affected directly by an infectious agent, appear to be inert bystanders, even in the midst of severe inflammation.

Astrocytes perform a variety of functions within the CNS. Astrocytic processes are a vital part of the BBB, and the cells themselves maintain the homeostasis of the neuropil (Figure 2 ). Astrocytic cells are the main cells involved in scar formation following CNS injury. They are now considered an accessory cell to the immune system of the CNS (Hertz etal., 1990; Yong et al., 1991; Rodriguez et al., 1987) . They express MHC antigens, may act as antigen-presenting cells (APCs), and secrete cytokines. In inflammation, astrocytic cells (especially type 1) may be stimulated by LAF-and ICAM-1, by cytokines from Τ cells, macrophages, microglial cells, and/or other astrocytes. IL-2 stimulates proliferation of astroglia in experimental brain injury. In vitro, IL-1, tumor necrosis factor (TNF), and IFN-γ stimulate proliferation of astrocytes. IFN primes astrocytic cells to the effect of IL-2 by secretion of TNF, IL-1, IL-3, IL-6, and lymphotoxin.

The endothelial cells are not only an anatomical and functional barrier between the blood and the brain; they are also pivotal cells in inflammation. ECs not only regulate the influx of cells and solutes, but they may also act as APCs, and the luminal surface of ECs may be the site of antigen presenta tion and recognition. Class I MHC molecules necessary for antigen recogni tion can be found on all cells constituting the BBB, whereas Class II MHC can be induced by IFN-γ stimulation. Class II MHC antigens expressed on ECs increase with the increased severity of inflammation, suggesting that the MHC expression is progressively turned on and may, under certain conditions, lead to amplification of local inflammation. ECs activation in chronic (and not necessarily severe) inflammatory processes may not be a secondary consequence of a local process but may, in fact, initiate and perpetuate the inflammation (Hertz et al., 1990; Gimborne and Beviaqua, 1988; Montovani and Dejana, 1989; Cotran, 1987) .

It also appears that the endothelial cell may be a pivotal cell in autoim mune inflammatory reaction (with local intrathecal antibody formation) without the involvement of extraneuronal antigens or infectious agents. ECs may present antigens that are specific for the CNS and attract sensitized Τ cells. Τ cells may attach to CNS antigens and perpetuate inflammation by secreting lymphokines (IFN, IL-2) and attract even more lymphocytes. This critical mass of lymphocytes and lymphokines may open the BBB, which allows Τ cells to enter the CNS parenchyma to recruit Β cells, which then locally produce antibodies against the CNS antigens presented by endothelial cells.

Microglia cells are considered immune-competent cells that are native to the CNS but whose origin is still unclear (Graeber and Streit, 1990) . They are found throughout the CNS and constitute from 5 to 20% of all CNS cells (Figure 3 ). Microglia cells share common surface antigens with mono cytes and macrophages but not with neuroectodermal cells. The functions of microglial cells include phagocytosis, presentation of antigens, produc tion of IL-1, and stripping of synaptic buttons (deafferentation) from dis-

The antigen is also present on brain resident microglial cells. Note the cell surface staining and presence of the dichotomically branching processes. Cells are evenly spaced, and do not form a connection with each other. Anti-LN-1 monoclonal antibody, Avidin-Biotin Technique, 70-/Ltm-thick vibratome section; nuclei were counterstained with hematoxylin, x 470.

tressed neurons. There is a low level of expression of Class II MHC antigens on microglial cells under normal conditions, and this expression can be dramatically increased with infusion of IFN-γ (Graeber and Streit, 1990; Tribolet de et al, 1984; Hayes et al, 1987; Hayes et al, 1988; Delisle et al, 1986; Hickey and Kimura, 1988; McGeer et al, 1987; McGeer et al, 1988) . Microglia cells also express the CD4 molecule, which is considered an entry point for the human immunodeficiency virus (HIV) (Watkins et al, 1990; Bourdial et al, 1991; Levy, 1988; Michaels et al, 1988 ). Yet another unique feature of the CNS is the presence of CNS-specific antigens such as myelin basic protein (MBP), glial fibrillary acidic protein (GFAP), or neuron-specific enolase (NSE). Such proteins can be the focus of autoimmune responses within the CNS. This autoimmune response can be a major contributing factor to pathology and clinical disease (see later discussions on viral-induced autoimmune response and experimental aller gic encephalomyelitis).

A strong body of evidence now suggests that some diseases of the CNS have an immunologic pathogenesis (Paterson, 1977 (Paterson, , 1978 (Paterson, , 1979 (Paterson, , 1982 . Ex amples include experimental allergic encephalomyelitis (EAE), acute dis seminated encephalomyelitis (ADEM), and multiple sclerosis (MS). Acute and chronic relapsing EAE can be induced in laboratory animals by an injection of CNS tissue, CNS myelin, myelin basic protein, or more recently, T-cell lines specific for nervous system antigens. ADEM occurs in humans and animals after an antecedent virus infection (see later discussion on viral induced autoimmunity) or vaccination with a living virus or a vaccine containing nervous tissue. MS is an acute or chronic progressive remitting disorder of humans characterized by inflammatory cell infiltration and demyelination.

EAE is a useful autoimmune disease model for the study of mechanisms of cellular immune reactions in the CNS and has been considered an experi mental model of MS since the earliest descriptions by Rivers et al (1935) . In recent years, reproducible small animal models of chronic EAE have become available (Stone and Lerner, 1965; Wisniewski and Keith, 1977; Massanari, 1980; Lubin et al, 1981; Brown et al, 1982) . These small animal models cover the full spectrum of MS pathology (Wisniewski et al, 1982; Lassmann and Wisniewski, 1979) and allow the study of the pathogenesis of chronic inflammatory demyelinating plaque formation. Progress on the pathogenesis and treatment of EAE has been advanced by the selection and maintenance of permanent autoantigen MBP-specific T-cell lines. Τ cells can now be isolated from animals with EAE that will mediate EAE in the recipient animal. This isolation is possible as a result of progress in the characterization of immune cells subsets, using monoclonal antibodies, and the definition of their growth requirements (i.e., IL-2) (Ben-Nun et al, 1981) . It is not clear how the sensitized Τ cells present in the circulation of EAE animals, cross the BBB and recognize an antigen or antigens that are located in the major dense lines of the myelin sheath. One possible explanation of this phenomenon is that the low rate of physiological ex change of lymphocytes that appears to exist between the CNS and the circulation allows the entry of the sensitized Τ cells (Lassmann et al, 1991; Wekerle et al, 1986; Wekerle et al, 1991) .

Interaction between inflammatory cells and the endothelial cells lining the BBB is an important initial event during inflammation of the CNS. The expression of la and of the endothelial leukocyte adhesion molecule (ELAM-1) on endothelial cells in the CNS are some of the earliest changes detected to date in brains of animals with EAE (Rose et al, 1991) .

Although the etiology of MS is still unknown, the similarity of MS to remitting-relapsing EAE suggests that MS, like EAE, might be the conse quence of either a direct or indirect autosensitization to myelin antigens (i.e., MBP and proteolipid apoprotein) (Rose et al, 1991) .

As a result of the application of advances in molecular biology, such as the polymerase chain reaction, and developments in cellular immunology, such as the ability to grow T-cell clones, great progress has been made in the pathogenesis and treatment of MS (Steinman, 1991) . Recent reports that most encephalitogenic (BP-specific) T-cell clones derived from mice or rats use common variable region genes of the rearranged T-cell receptor (TCR) have intensified the search for analogous associations in humans (Burns et al, 1989; Urban et al, 1988) . Steinman (1991) has recently reviewed the strong parallels between T-cell receptor (TCR) usage in the pathogenesis of EAE and TCR usage in MBP, and specific Τ cells in MS peripheral blood and Τ cells in demyelinated plaques in MS brains. The development of strategies for selective immunotherapy in EAE was based on these similarit ies. This therapy uses monoclonal antibodies on either Class II molecules of the MHC or TCR-variable regions or peptides that compete with HLA Class II molecules or vaccination against TCR-V regions. For example, vaccination of mice with appropriate TCR peptides is highly effective at inducing both cellular and humoral response to TCR and at inhibiting EAE (Howell et al, 1989; Vanderbark et al, 1989) . Monoclonal antibodies that bind only the complex of BP and I-As inhibited EAE in H-2s mice when injected within -2 to + 1 0 days of disease induction (Aharoni et al, 1991) . Sriram and Carroll (Sriram and Carroll, 1991) have shown that in vivo treatment with I-A antibodies at the earliest sign of EAE results in decreased homing of radiolabeled cells to the brain. These data suggest that Class II antigens play an important role in cellular migration across the BBB.

Other potential therapies for EAE include the use of cytokines (see cyto kine section) and suppressor Τ cells. Although mechanisms involved in EAE remission are unclear, suppressor Τ cells have been postulated to play a role in the prevention of autoimmunity (Ofosu-Appiah and Mokhtarian, 1991) . An adaptively transferred suppressor T-cell line, obtained from spleens of mice that recovered from EAE, was able to downgrade EAE in mice subsequently challenged with MBP-activated Τ cells.

Immunologically important cytokines are produced chiefly by lympho cytes and macrophages. Inflammation in the nervous system is likely to be the same as in other organ systems, with a few modifications. Because of the BBB and the macromolecular nature of cytokines, some of these important inflammatory molecules are likely to be CNS resident derived. Recent studies have also indicated astrocytes, microglia, and endothelial cells as possible additional sources of CNS-derived cytokines (Fontana et al, 1982; Frei et al, 1985; Giulian et al, 1985; Frei et al, 1989; Sawada et al, 1989) . Immunocytochemical techniques have demonstrated the presence of a variety of cytokines (see earlier discussion) (i.e., MIF, IFN-γ, TNFa, IL-1,2,3) during chronic relapsing EAE . IL-1, IL-2, TNFa, and IL-6 have also been shown in the serum and/or CSF of EAE animals and/or MS patients (Merril et al, 1989; Gallo et al, 1988; Gijbels et al, 1990) .

A. TGF-/31 and IL-1 Transforming growth factor β (TGF-β) has pleotrophic effects. Evidence to date includes the possible role of down-regulation of the IFN-γ induction of Class II MHC expression on the inductive phase of the immune response, as well as inhibition of cytotoxic T-cell development and antagonism of TNF at the effector end of the immune response (Racke et al, 1991; Wahl et al, 1988; Rook et al, 1986; Ranges et al, 1987; Czarniecki et al, 1988; Schluesener, 1990; Shalaby and Ammann, 1988) . When administered dur ing EAE induction, TGF-/31 slightly delays the onset of disease (Kuruvilla et al, 1991) . However, when given during remission, TGF-/31 prevents the occurrence of relapses in relapsing EAE, suggesting an anti-inflammatory effect of TGF-/31. When TGF-j81 was preincubated in vitro, activation and proliferation of myelin basic protein-specific lymph node cells in vitro were decreased and severity of the clinical course was reduced (Racke et al, 1991) .

Recent data indicate that IL-1 may promote CNS inflammation, and that blocking IL-1 activity may prove beneficial in the treatment of CNS inflammatory disease. For example, EAE in the Lewis rat has been shown to be exacerbated by IL-Ια. In addition, the disease was significantly delayed and attenuated by IL-1 receptor antagonist (IL-lra) (Jacobs et al, 1991) .

IL-1 receptor antagonists have been used to ameliorate other inflamma tory diseases such as rheumatoid arthritis and septic shock, and colitis in a rabbit model (Cominelli et al, 1990; Ohlsson et al, 1990; Wakabayashi et al, 1991; Arend and Dayer, 1990; Higgins and Postlethwaite, 1991) . Several structural variants of IL-lra have been described (Arend, 1991) that might be of interest in exploring the therapeutic potential of IL-lra. Administration of IL-lra into the lateral ventricle of rabbit brains has been reported to block IL-l-induced non-rapid-eye movement sleep as well as IL-l-induced fever (Opp and Krueger, 1991) .

The similarities of macrophages and microglia are striking (Merz et al, 1987) . For example, IL-1, IL-6, and TNFa are known to be produced by monocytes and macrophages. In the CNS, microglia have been implicated as a major contributor to the production of IL-1, IL-6, and TNF (Woodroofe et al., 1991) . Using a continuous microdialysis probe, Woodroofe et al. (1991) showed a 15-fold increase in IL-1 over a 24 to 48 hr period and a slight increase in IL-6 at day 1, as a result of mechanical trauma to the brain. Another potential cellular source of IL-1 is the astrocyte. However, the authors have immunocytochemical evidence that in this model the astrocyte response appears much later-at day 7. An increase in peripheral benzodi azepine receptors after local injection of IL-1 and TNFa has been demon strated on glial cells, but not on neurons (Bourdial et al, 1991) , raising the possibility of a sequential mechanism involving the activation of microglia, the release of IL-1 and TNFa, and the promotion by these cytokines of the astroglial reaction.

Administration of human serum amyloid A to mice inhibited fever in duced by rIL-1/3 or rTNFa in vivo, whereas the addition of human serum amyloid A to murine hypothalamic slices inhibited IL-1/3-or TNFa-induced prostaglandin E2 production. These data suggest a possible feedback rela tionship between serum amyloid A and cytokines (Shainkin-Kestenbaum et al, 1991) .

Injection of IL-1 results in an increased release of corticotropin-releasing factor (Ohgo et al, 1992; Besedovsky et al, 1986; Sapolsky et al, 1987; Uehara et al, 1987) . In addition, IL-Ια injected into the third ventricle of castrated rats inhibited the pulsatile release of luteinizing hormone (Rettori et al, 1991) . In a recent report, Palazzolo et al (1990) suggest that the central and neuroendocrine effects of IL-1 are most likely produced through changes in neurotransmitter metabolism in the brain. In another report, Mohankumai et al. (1991) provide evidence that IL-1 stimulates the release of catecholamines (e.g., dopamine and its metabolite, dihydroxyphenylacetic acid (DOPAC) from discrete hypothalamic nuclei of conscious, freely mov ing rats. The precise role that IL-1 may play in neuroimmunomodulation is still not clear. The important known role of the hypothalamicpituitary-adrenocortical (ΗΡΑ) axis in regulating inflammation mandates further studies of the role of IL-1 as a neuroimmunomodulator.

Another potential interaction of the nervous system and the immune system is the involvement of cytokines in neurogenic inflammation (Kim ball, 1991) . In the Kimball model, substance P, a neuropeptide, is at the center, interacting with fibroblasts, mast cells, Β cells, and macrophages. In addition to substance P, other neuropeptides, neurotransmitters, and other vasoactive amines act in concert with cytokines to affect immunologic mechanisms.

B. IL-2, IL-6, TNFa, and IFN-γ

The roles of established monokines (IL-6 and TNFa) and T-cell products (IFN-γ and IL-2) on CNS pathology and physiology are stimulating areas of ongoing research. TNF appears to be a major cytokine involved in cellular injury. Axon and myelin abnormalities have been demonstrated in vitro by TNF (Selmaj and Raine, 1988) . After IL-2 perfusion in rats, TNF was noted to coincide with myelin damage. Interestingly, IL-2 perfusion has been shown to compromise the BBB (Ellison and Merchant, 1991) . Several struc tural variants of IL-lra have been described (Arend, 1991) that might be of interest in exploring the therapeutic potential of IL-lra. Several phosphodi esterase inhibitors (e.g., theophylline, pentoxifylline, and 3-isobutyl-lmethylxanthine) have been shown to suppress TNFa synthesis (Endres et al., 1991) . High TNF levels have been reported to be associated with several manifestations of malaria (Kremsner et al, 1991; Shaffer et aL, 1991) , and decreasing TNF, levels with recovery. In a murine model of cerebral malaria, pentoxifylline, which reduces TNF, given for 10 days after infection, pro tected the mice from development of cerebral malaria (Kremsner et aL, 1991) .

One of the pleotrophic effects of IL-6 is a neurotrophic effect that has been described in viral diseases (Frei, 1989) and improved survival of mes encephalic catachoaminergic and septal cholinergic neurons (Hama et al., 1991) . Mice infected with lethal lymphocytic choriomeningitis (LCM) virus as well as LCM carrier mice have been shown to be correlated with high levels of secretion of IL-6 (Maskophidis et al., 1991) . Of interest is the possible role of IL-6 in the formation of Alzheimer amyloid plaque (Baurer and Strauss, 1991) . Bauer et al. have shown a potent human proteinase inhibitor, a-2 macroglobulin (a2M), after stimulation with IL-6 (Ganter et al, 1991) . Subsequently, they examined whether a 2 M and IL-6 could be detected in Alzheimer disease (AD) brain. They report that AD cortical senile plaques display strong a2M and IL-6 immunoreactivity, with no such immunoreactivity found in age-matched control brains. The data indicate that neuronal cells are the site of a2M synthesis in AD brains.

Among the best known pleotrophic effects of IFN-γ is its ability to induce the expression of Class II MHC molecules on several cell types. Class II MHC antigens are known to be essential in antigen presentation. Studies have shown that IFN-γ can induce expression of Class II MHC molecules on astroglia (Sedgwick et al., 1991) . However, Sedgewick et al. (1991) were unable to show that astroglial cells act as stimulators of C D 4 + Τ cells and therefore postulated another cell type as the major antigen-presenting cell in CNS inflammation. Infection with mouse hepatitis virus (MHV), has been shown to block expression of MHC molecules on murine cerebral endothelial cells (see later discussion) (Joseph et al., 1991) . Joseph et al. postulate possible release of cytokines by endothelial cells as a mechanism for blocking IFN-induced Class II antigens.

Several additional cytokines have been reported that may play a role in CNS pathology and physiology. Growth factors and, to some extent, IL-1/3 and TNFa appear to increase nerve growth factor synthesis and secretion by astrocytes (Yoshida and Gage, 1992) . Cultured astrocytes have been reported to express macrophage colony-stimulating factor activity at a high level, which may be important in the replication of microglial precursors (Alliot et al, 1991) .

Growth differentiation factor-1, a recently described member of the trans forming growth factor β superfamily, has been reported to be restricted almost exclusively to the CNS (Lee, 1991) . This extracellular signaling factor is postulated to play a general role in nervous system maintenance and function. Of interest is one of two platelet-activating factor (PAF) antago nists, which has been shown to also inhibit IL-l/3-induced ACTH secretion (Rougeot et al, 1991) . This suggests that IL-1 ΗΡΑ secretion may be medi ated, at least in part, by the production of PAF.

Virus infection of the CNS can result in a wide range of pathogenic outcomes: a rapid, severely acute form; a chronic, progressive form; a reversible, nonproductive, latent type; or various intermediate stages. Dis ease is a direct result of interaction between host and infecting virus. Host parameters include age, genetic makeup, and immune competency, whereas viral factors include class and strain of agent, target specificity, and genomic variation. The phenotypic expression of disease is a complex multifactorial interaction among these parameters.

The first step in the disease process involves entry of the virus into the CNS. Various host protective barriers prevent direct viral entry into the CNS and force indirect routes of infection. Agents that cause viremia can enter by replication in the endothelial cells that line the blood vessels. Viruses known to enter by this route include the toga viruses, enteroviruses, and certain retroviruses. Viruses can also invade from the bloodstream by entering the stroma and the choroid plexus through the fenestrated capillary and endothelium and either infecting or being transported across the cho roid plexus into the CSF. Recently, it has been demonstrated that fibroblastlike cells isolated from the human choroid plexus can be infected with HIV (Harouse et al, 1989) . Once in the CSF, the virus can infect ependymal cells lining the ventricles and invade the underlying CNS tissue. The virus can also enter through "Trojan horse" mechanisms, being transported within circulating leukocytes such as lymphocytes or macrophages/mono cytes, which can contain such viruses as measles, mumps, canine distem per, lentiviruses, or togaviruses. Viruses can also infect peripheral neurons at such sites as the neuromuscular junction and can be axonally transported toward the CNS. Rabies and herpes simplex may utilize such a route (John son, 1984) . Herpes simplex may also gain entrance to the CNS by means of the sensory neurons of the olfactory system.

Once within the CNS, viruses encounter a differentiated cell population with complex, functionally integrated cell-to-cell interactions. The highly specialized cytoplasmic membranes of this cell population allow for great variation in virual receptor sites and in the abilities of cells to support viral replication. Viral tropism to cells within the CNS involves both cell and viral receptor molecules . Polioviruses display a particular affinity for anterior horn motor neurons, whereas the rabies virus normally prefers neurons of the limbic system (Johnson, 1980) . Infections with orthomyxoviruses or paramyxoviruses usually involve the selective infection of ependymal cells (Wolinsky et al, 1976) . The polyomavirus, which is the causative agent of progressive multifocal leukoencephalopathy (see later discussion), produces a lytic infection of oligodendrocytes and a nonpermissive infection of astrocytes (Richardson, 1961) . The la antigen receptors present on a variety of cell types have been implicated as viral receptors (Inada and Mims, 1990) . Cellular CD4 protein interacts with HIV gp 120 and is the receptor for this virus on lymphocytes and monocytes (Wigdahl and Kunsch, 1990) . Several studies suggest the presence of the CD4 receptor on cells within the CNS (Dewhurst et al, 1987) . The specific affinity of rabies virus to acetylcholine receptors serves to facilitate uptake and transfer of virus to the CNS and to determine neuronal specificity (Lentz and Burrage, 1982) .

As stated earlier, one potential outcome of CNS viral infection is rapid acute disease, usually including encephalitis and/or meningoencephalitis. This can be accompanied by perivascular inflammation involving polymor phonuclear cells followed later by macrophages, lymphocytes, and microg lia. During infection by certain viruses, i.e., poliovirus, the inflammatory response consists of a meningeal reaction, perivascular cuffing, and paren chymal infiltration. Polymorphonuclear leukocytes are seen early, with a later shift to predominantly mononuclear cells. Microglial cell infiltration and neuronophagia are also seen (Johnson, 1982) .

Pathogenesis is normally a two-step process consisting of viral-mediated cell destruction as well as immune-mediated, viral-induced cellular destruc tion. Viral antigens are normally good immunogens, and host immune surveillance directs a T-cell-dependent immune response (Burns, 1975) . Such a response is directed primarily against infected cells that express viral or altered cell surface proteins. Antibodies can function as intermediaries in the destruction of infected cells by means of complement-mediated cytotox icity or antibody-dependent cellular cytotoxicity or by serving as opsinizing factors. Cytotoxic Τ cells can destroy such cells directly or release lympho kines to recruit, activate, and/or sensitize other lymphoreticular elements. Cytolytic Τ lymphocytes (CTLs) are an important effector in antiviral immu nity (Zinkernagel and Doherty, 1986) . Such cells can recognize foreign viral antigens on cell surfaces only in the context of self MHC structures. Until recently, the HLA Class I molecules were thought to be the primary, if not the only, HLA recognition structure for CTLs. Studies of measles and Epstein-Barr virus (EBV) infection suggest that HLA Class II molecules can also serve as recognition sites, further expanding the potential action of CTLs. Viral antigens recognized by CTLs have also been expanded beyond the traditional cell surface molecules (Braciale and Braciale, 1986) . Prelimi nary data indicate that recognition of internal viral polypeptides and per haps even nonstructural gene products may be a common feature of antivi ral CTLs. Coronavirus MHV-4 can selectively block gamma-interferoninduced Class II antigen expression on cerebral endothelial cells (Joseph etal, 1991) . Studies by other investigators have shown Class I modulation on mouse glial cells by other related coronaviruses (Suzumura etal., 1986) . In such a manner, viruses may be able to evade immune-mediated events that occur at the level of the BBB. The virus is able to avoid host immune surveillance and to enter the CNS. Such events probably also prevent late immune-mediated demyelinating disease, which is absent in MHV-4 virus infection (see later discussion).

Immunodeficiency states, such as AIDS or iatrogenic immunosuppres sion in transplant recipients, may sometimes modify the inflammatory response of the CNS to pathogens. In AIDS, especially at the terminal stage of disease, the picture of CNS inflammation, i.e., in toxoplasma gondii encephalitis or in progressive multifocal leukoencephalopathy caused by JC virus, can differ from that seen in nonimmuno-suppressed individuals. The inflammation seems muted, and the destructive component of the process (fulminant necrosis) may dominate. This may serve as another example that the inflammatory response within the CNS is partly depen dent on the peripheral immune system.

As discussed earlier, immune surveillance in response to viral infections can lead to virus-induced immune response to normal host components. During acute infections, this response can lead to increased tissue damage. During chronic or persistent infections, such responses can perpetuate or be the primary cause of the disease process. Involvement can take the form of humoral response (autoantibodies) as well as cell-mediated autoimmune (CMAI) response (Ter Meulen, 1989) . Such autoimmunity may involve a phenomenon known as molecular mimicry (Srinvasappa et al, 1986) , an immune process in which molecules coded for by dissimilar genes share similar structures. In such a manner, antiviral antibodies may bind to host antigens as well as to antigens of the virus. Many examples of such mimicry have now been identified: measles virus phosphoprotein and keratin ; vaccinia virus hemagglutin and vimentin (Dales et al., 1983) ; fusion protein of measles and heat shock protein (Shesheberadarin and Norby, 1984) ; HIV gpl20 and neuroleukin (Mizarachi, 1989) . In an analysis of monoclonal antibodies to 11 different viruses, 4% of such anti bodies cross-reacted with host cell determinants (Srinvasappa et al., 1986) . Generation of cytotoxic cross-reactive effector lymphocytes or antibodies would recognize "self proteins" located at target cells. The virus need not be present for such events to take place. Myelin basic protein exhibits a significant degree of homology with several viral proteins, including hepati tis Β virus polymerase. Inoculation of hepatitis Β virus polymerase peptide into rabbits caused perivascular infiltration localized to the CNS, reminis cent of EAE (Fujinami and Oldstone, 1985) .

Several other mechanisms are potentially operational in virus-induced autoimmune disease. Viruses may disrupt normal immune regulation by direct interaction with cells of the reticuloendothelial system (RES). As discussed earlier, many viruses can replicate in these cell types, leading to destruction of lymphocyte subpopulations or stimulation of autoreactive clones. Viruses may incorporate host molecules into their envelope or coat, or they may insert, modify, or expose other cellular components on the cell surface. Immune pathological events may also be triggered by induction of Class II MHC antigens on the surface of infected CNS cells. MHC antigens are expressed only at low levels or not at all on the majority of CNS cells (Hart and Fabre, 1981 ). Astrocytes exposed in vitro to measles or corona virus JHM begin to express MHC Class II antigens, which are further enhanced in the presence of TNF (Massa et al, 1987 ). An analogous situation in vivo would certainly facilitate CTL-mediated autoimmune response.

Postinfectious encephalomyelitis has been associated with a wide range of virus infections including measles, mumps, rubella, vaccinia, and herpes zoster and certain respiratory viral infections (Johnson and Griffin, 1986) . Neuropathological changes resemble those seen in EAE and appear to be a direct result of an MBP-directed immune response (Ter Meulen, 1989) . Coronavirus JHM causes acute demyelinating encephalitis by selective in fection of oligodendrocytes. Lymphocytes from sick animals that are pas sively transferred to syngeneic animals produce lesions in the recipients that resemble those in individuals with allergic encephalomyelitis (Ter Meulen, 1989) . Viral infection of oligodendrocytes appears capable of inducing an autoimmune response against the myelin proteins produced by these cells. In measles-induced encephalomyelitis, there is little evidence that virus invades the CNS. Deregulation of autoreactive cells may occur secondarily to viral infection of lymphoid cells. Demyelinating disease caused by Theiler's murine encephalomyelitis virus (TMEV), a nonbudding enterovirus, may result from a virus-specific, delayed-type hypersensitivity (DTH) (Clatch et al., 1986) . Lymphokines produced by MHC Class II-restricted, TMEV-specific, DTH Τ cells primed by interaction with TMEV-infected macrophages would recruit and activate additional macrophages, leading to nonspecific macrophage-induced damage.

In contrast to the autoimmune mechanisms described above, persistent viral infections can be a consequence of the ability of the virus to avoid immune surveillance. In the face of excessive viral antigen, both virusspecific antibodies and CTLs are suppressed or rendered ineffective. Vi ruses may be able to down regulate appropriate recognition molecules on the surface of immune cells. Expression of virus surface proteins involved in immune recognition may also be reduced. Reduced expression of glyco protein has been observed during persistent infections with arenaviruses, paramyxoviruses, retroviruses, and rhabdoviruses (Oldstone, 1989) . Antibody-induced antigenic modulation can lead to measles virus persistent infection in the CNS (Fujinami and Oldstone, 1984) . Presumably, this mod-ulation can occur in body compartments that are devoid of the effector molecules of complement such as the CNS. Subacute sclerosing panenceph alitis (SSPE) results from a persistent CNS infection with defective measles virus (Wechsler and Meissner, 1982) . A defect in viral envelope genes prevents expression of viral antigen on cell surfaces. Infection is propagated by cell-to-cell transmission of virus, allowing the agent to avoid immune surveillance. Viruses can also replicate in the cellular constituents of the immune system and in such a way disable specific immune responsiveness. Viruses thus persist within cells that are ordinarily utilized to provide clearance. Viruses such as measles and HIV can infect lymphocytes and/ or macrophages, resulting in generalized immunosuppression. Such immu nosuppression leaves the host susceptible to infection by a wide range of other microorganisms, leading to the opportunistic infections often associ ated with HIV and other such viruses.

Viruses can also avoid immune surveillance through latency, a state of reversible, nonproductive infection. Viral genetic information can remain within the host in either an integrated or an episomal state. Retroviruses such as HIV use reverse transcriptase to insert their DNA proviral form into host-cell DNA; this integration step is a prerequisite for the replication of these RNA viruses (Mahry, 1985) . Herpes simplex virus (HSV), which resides in sensory ganglion cells, integrates its DNA directly into host nuclear DNA. The DNA of other DNA viruses such as EBV or JC agent, the polyomavirus that causes progressive multifocal leukoencephalopathy, remains latent in a nonintegrated episomal form. Latent infections estab lished by these viruses may result from a lack of host factors that are critical for the expression of viral early gene products (Garcia-Blanco and Cullen, 1991) . The subsequent activation of specific cellular transcription factors in response to extracellular stimuli can induce the expression of virus and lead to CNS disease. Herpesviruses such as HSV-1, HSV-2, or VZV cause initial acute peripheral infection followed by a latent infection of neurons. In this environment, sheltered from immune surveillance, the virus can remain for the life of the host. HSV-1 can be activated by a number of seemingly unrelated stimuli such as physical or emotional stress or damage to adjacent tissue. Reactivation probably results from a signal transduction event that directly or indirectly induces HSV-1 gene transcription (Leib et ah, 1989) . Latently EBV-infected Β cells can also persist for the life of the individual. If immediate early viral gene expression does not occur, EBV may establish a latent infection (Miller, 1985) . Both activated Τ lymphocytes and nondividing macrophages are able to support the synthesis and integra tion of HIV-1 proviruses (Fauci, 1988) . Resting Τ cells are nonpermissive and do not support replication or integration. Resultant unintegrated viral intermediates may persist in a viable yet transcriptionally inert form for extended periods of time. In quiescent macrophages or in Τ cells in a resting state, cellular factors critical for proviral transcription may not be active. Postintegration latency in HIV-1 may result from inefficient proviral tran scription and from a suboptimal amount of REV, a viral-coded transactivation regulatory protein (Pomerantz et al, 1990) . Stimulation of these cells in some manner could induce cellular transcription factors increasing REV which leads to productive virus replication. The rare demyelinating disor der, progressive multifocal leukoencephalopathy (PML) (see earlier discus sion), is caused by the reactivation of latent polyomavirus ( J C V ) in the oligodendroglia cells of affected individuals (Mazlo and Tariska, 1980) . This disease is usually associated with an underlying disorder of the RES system in which immune responsiveness is impaired. Loss of immune surveillance allows reexpression of virus and cytocidal effects on oligodendrocytes, lead ing to demyelination.

The final type of interaction of virus within the CNS is a chronic persistent infection that does not elicit an immune response. Wild mouse ecotropic immune leukemia virus (MuLv) induces a progressive form of hind limb paralysis in a natural population of mice as well as in laboratory animals (Gardner et al., 1973) . Pathologically, the disease is characterized by the absence of host inflammatory response at the site of tissue destruction and by a picture of neuronal degeneration, spongiform gray matter lesions, and gliosis. Paralysis is a direct result of viral replication, most notably in anterior horn motor neurons, but virus can also infect endothelial and glial cells. There is no virus-mediated CTL or antibody response. The CNS is devoid of inflammatory infiltrates or deposits of IgG or complement. In genetically susceptible laboratory strains of mice, ability to induce disease is age depen dent and involves tolerance to the virus. Passive immunization with anti bodies to MuLv virus can prevent paralysis. Neuronal destruction and spongiosis may be a direct result of the interaction of viral gene expression in the neuron. Recent evidence suggests that a posttranscriptional step in virus envelope protein synthesis is impaired and that neurological disease may be a consequence of abortive replication of virus within neurons (Sharpe et al, 1990) . Such abortive infections can make the presence of virus difficult to detect. Analogies to other retrovirus CNS infections, notably HIV and HTLV-1, suggest that similar mechanisms may be operational in CNS disease caused by these agents.

Another group of infectious agents that establish CNS disease in the absence of inflammatory response are the agents of the transmissible spon giform encephalopathies (TSEs) or prion diseases (Carp et al, 1989) . Unlike the retroviral disease described above, in which immune response to virus is age dependent and the disease process may include tolerance, there is no evidence for the ability of the host to generate any immune response to the agents of TSE. These agents cause Creutzfeldt-Jakob disease (CJD), Gerstmann-Straussler-Scheinker syndrome (GSS), and kuru in humans and are associated with scrapie, bovine spongiform encephalopathy (BSE), chronic wasting disease, and transmissible mink encephalopathy (TME) in animals. The agents that cause these diseases are poorly characterized, but it is certain that they do not exhibit properties of known viruses or virus like agents. These agents replicate peripherally in the RES system, notably in spleen and lymph nodes. Agent crosses into the CNS, leading to spon giosis, gliosis, and neuronal destruction in the absence of any inflammatory response, similar to that described above for the retroviruses. Modulation of the immune system only affects the agent given peripherally. Immuno logical compounds, given close to the time of peripheral infection, that stimulate the immune response shorten incubation periods, whereas com pounds that suppress the immune response extend the incubation period. Treatment with dextran sulfate (Farkuhar and Dickinson, 1986 ) extends disease and may mediate its effect through interaction with macrophages. Recent studies suggest the importance of follicular dendritic cells (Kitamoto et al., 1991) in agent peripheral replication. Similar dendritic cells are in volved in antigen presentation and virus replication in infections caused by other viruses. Similar to evidence described for the neurotropic retroviruses, pathology within the CNS may include abnormal protein processing, in this case, the production of an abnormal host-coded protein termed PrP (Prusiner, 1991) . Both agent replication and spongiform change are associ ated with the presence of this protein within the CNS.

In conclusion, the inflammatory response within the CNS appears to have several aspects. First, the CNS itself is unique from other organs in the presence of BBB with specialized endothelial cells, the presence of CNSspecific cells (e.g., microglia or astrocytes), and the presence of CNS-specific antigens. Second, infectious agents such as viruses comprise a wide spec trum of agents with varied neural tissue virulence and varied specificity of agents to certain CNS cell types. Some of these agents may produce fulmi nant disease with severe, if not lethal, CNS damage, whereas others may persist for decades, resulting in chronic or latent effects. The interactions among the cells and environment of the CNS, infectious agents, and the immune response are quite complex. Studying the unique aspects of the CNS-immune response provides us the window of opportunity to view fascinating and diverse aspects of these responses that otherwise would not be available.

Immunomodulation of experimen tal allergic encephalomyelitis by antibodies to the antigen-la complex

Microglial progenitors with a high proliferative potential in the embryonic and adult mouse brain

Interleukin-1 receptor antagonist

Cytokines and cytokine inhibitors or antagonists in rheumatoid arthritis

The leptomeningeal vein. A site for re-entry of leukemic cells into the systemic circulation

Cytokines in the central nervous system of mice during chronic relapsing experimental allergic encephalomyelitis

Interleukin-6 and alpha 2-macrogroleutin indicate an acute phase state in Alzheimer's disease cortices

The rapid isolation of clonable antigen specific Τ lymphocyte lines capable of mediating autoimmune encephalomyelitis

Immunoregulatory feedback between interleukin-1 and glucocorticoid hormones

Increase in W3 (peripheral type benzodiazepine) binding sites in the rat cortex and striatum after local injection of interleukin-1, tumor necrosis factor-alpha and lipopolysaccharide

CTL recognition of transfected H-2 gene and viral gene products

The Concept of a Blood-Brain Barrier

Hypoxia and vascular disorders at the central nervous system

Implications of the Blood-Brain Barrier and Its Manipulation

Transcytosis of macromolecules through the blood-brain barrier: A cell biological perspective and critical appraisal

Chronologic neuropathology of relapsing experimental allergic encephalomyelitis in the mouse

Viral antigens

Both rat and mouse T-cell receptors specific for encephalitogenic determinant of myelin basic protein use similar V-alpha and V-beta-chain genes even though the major histocompatibility complex and encephalitogenic determinants being recognized are different

The nature of the unconventional slow infection agents remains a puzzle

Characterization of Theiler's murine encephalomyelitis virus (TMEV) specific delayed type hypersensitivity responses in TMEV induced demyelinating disease: Correlation with clinical signs

Interleukin 1 (IL-1) gene expression, synthesis, and effect of specific IL-1 receptor blockade in rabbit immune complex colitis

New roles for the endothelium in inflammation and immunity

Transforming growth factor betal modulates the expression of class II histocompatibility antigens on human cells

Serologic relatedness between Thy-1.2 and actin revealed by monoclonal antibody

What's new in cerebral pathology in acquired immune deficiencies?

Expression of the T4 molecule (AIDS virus receptor) by human brain-derived cells

Lymphocyte-function-associated antigen-1 (LFA-1) interaction with intercellular adhesion molecule-1 (ICAM-1) is one of at least three mechanisms for lymphocyte adhesion to cultured endothelial cells

Appearance of cytokine-assodated central nervous system myelin damage coincides temporarily with serum tumor necrosis factor induction after recombinant interleukin-2 infusion in rats

Cyclic nucleotides differentially regulate the synthesis of tumor necrosis factor alpha-and interleukin-lbeta by human mononuclear cells

Prolongation of scrapie incubation period by an injection of dextran sulfate 500 within the month before or after injection

The human immunodeficiency virus: Infectivity and mechanisms of patho genesis

Astrocytes as antigen presenting cells: Induction of la antigen expression on astrocytes by Τ cells via immune interferon and its effects on antigen presentation

Production of prostaglandin Ε and interleukin-1 like factor by cultured astrocytes and C6 glioma cells

Astrocytes present myelin basic protein to encephalitogenic T-cell lines

Astrocytes of the brain synthesize interleukin 3-like factor

On the cellular source and function of IL-6 produced in the central nervous system in viral disease

Antibody initiates virus persistence: Immune modulation and measles virus infection

Amino acid homology and immune responses between the encephalitogenic site of myelin basic protein and virus: A mechanism for autoimmunity

Molecular rnimicry in viral infection: Cross-reaction of measles phosphoprotein or of herpes simplex virus protein with human intermediate filaments

Interleukin-2 levels in serum and cerebrospinal fluid of multiple sclerosis patients

Alpha 2-macroglobulin synthesis in interlukin-6 stimulated human neuronal (SH-SY5Y neuroblastoma) cells: Potential significance for the processing of Alzhei mer beta-amyloid precursor protein

Molecular basis of latency in pathogenic human viruses

A spontaneous lower motor neuron disease appar ently caused by indigenous type C RNA virus in wild mice

Interleukin-6 production in the central nervous system during experimental autoimmune encephalomy elitis

Vascular endothelium. Functional modula tion at the blood interface

Interleukin-1 of the central nervous system is produced by ameboid microglia

Microglia: Immune network in the CNS

Interleukin-6 improves the survival of mesencephalic catecholaminergic and septal cholin ergic neurons from postnatal two-week old rats in cultures

Leukocyte-endothelial interactions

Human choroid plexus cells can be latently infected with human immunodeficiency virus

Demonstration and characterization of IA positive dendritic cells in the interstitial tissues of rat heart and other tissues but not brain

Microglia are the major cell type expressing MHC class II in human white matter

Characterisation of microglia isolated from adult human and rat brain

Astrocytes: Auxiliary cells for immune responses in the central nervous system?

Privascular microglial cells of the CNS are bone marrowderived and present antigen in vivo

Interleukin-1 inhibitors and their significance in rheumatoid arthritis

Infection and Inflammation of CNS Fine structural alterations of the blood-brain barrier in EAE

Expression of la antigens by cultured astrocytes treated with gamma interferon

Vaccination against experimental allergic encephalomyelitis with Τ cell receptor peptides

Ig antigen and Fc receptors of mouse peritoneal macro phages as determinants of susceptibility of lactic dehydrogenase virus

Experimental autoimmune encephalomyelitis is exacerbated by IL-lalpha and suppressed by soluble IL-1 receptor

The pathogenesis of viral infections of the nervous system

Selective vulnerability of neural cells to viral infections

Viral Infections of the Nervous System

Virus induced autoimmune demyelinating disease of the CNS

Mouse hepatitis virus (MHV-4, JHM) blocks gamma interferon-induced major histocompatibility complex class II antigen expression on murine cerebral endothelial cells

Pathogenesis of virus-induced and autoimmune nervous system injuries

Involvement of cytokines in neurogenic inflammation

Abnormal isoform of prion protein accumulates in follicular dendrite cells in mice with Creutzfeldt Jacob disease

Pentoxifylline prevents murine cerebral malaria

Protective effect of transforming growth factor betal on experimental autoim mune diseases in mice

Chronic relapsing experimental allergic encepha lomyelitis. Clinicopathological comparison with multiple sclerosis

Expression of adhesion molecules and histocompatibility antigens at the blood-brain barrier

Expression of growth/differentiation factor 1 in the nervous system: Conserva tion of bierstronic structure

Immediate early regulatory gene mutants define different stages in establishment and reactivation of herpex simplex virus latency

Is the acetylcholine receptor a rabies virus receptor?

The human immunodeficiency virus and its pathogenesis

Sites of egress of inflammatory cells and horseradish peroxidase transport across the bloodbrain barrier in a murine model of chronic relapsing experimental allergic encephalomyeli tis

Delayed, relapsing experi mental allergic encephalomyelitis in mice

Strategies of virus persistence

Production of random classes of immunoglobulins in brain tissue during persistent viral infection paral leled by secretion of interleukin-6 (IL-6) but not IL-4, IL-5 and gamma interferon

Tumor necrosis factor amplifies measles virus-mediated la induction on astrocytes

A latent-relapsing neuroautoimmune disease in Syrian hamsters

Morphological demonstration of the first phase of polyomavirus replication in oligodendroglia cells of human brain in progressive multifocal leukoencephalopathy (PML)

Reactive microglia in patients with sende dementia of the Alzheimer type are positive for the histocompatibility glycopro tein HLA-DR

Expression of the histocompatibdity glycoprotein HLA-DR in neurological disease

In vitro study of mediators of inflammation in multiple sclerosis

Isolation and characterization of macrophages from scrapie-infected mouse brain

Microglia in the giant cell encephalitis of acquired immunodeficiency syndrome:proliferation, infection and fusion

Epstein Barr virus

Neurotropic activity of monomelic glucophosphoisomerase was blocked by human immunodeficiency virus and peptides from HIV-1 envelope glycoproteins

Interleukin-1 stimulates the release of dopamine and dihydroxyphonyacetic acid from hypothalamus in vivo

Cytokines as communication signals between leukocytes and endothelial cells

Characterization of a T-suppressor ceU line that downgrades experimental aUergic encephalomyelitis in mice

Stimulation by interleukin-1 (IL-1) of the release of rat corticotropin-releasing factor (CRF), which is independent of the cholinergic mechanism, from supervised rat hypothalomaneurohypophysial coplexes

Interleukin-1 receptor antagonist reduces mortality from endotoxin shock

Viral persistence

Interleukin-1 receptor antagonist blocks interleukin-1 induced sleep and fever

Interleukin-1 stimulates catecholamine release from the hypothalamus

Molecular and cellular determinants of neuroimmunologic inflammatory disease

Autoimmune neurological disease: Experimental animal systems and implications for multiple sclerosis

The demyelinating diseases: Clinical and experimental studies in animals and man

Neuroimmunologic diseases of animals and humans

Cells nonproductively infected with HIV-1 exhibit an aberrant pattern of viral RNA expression: A molecular model of latency

Response of the axon and barrier endothelium to ΕΑΝ induced by autoreactive Τ cell lines

Molecular biology of prion diseases

Prevention and treatment of chronic relapsing experimental allergic encephalomyeli tis by transforming growth factor -betal

Homing to central nervous system vasculature by antigen-specific lymphocytes. II Lymphocyte/endothelial cell adhesion during the initial stages of autoimmune demyelination

Inhibition of cytotoxic Τ cell development by transforming growth factor-beta and reversal by recombinant tumor necrosis factor-A

Fine structural localization of a blood-brain barrier to exogenous peroxidase

Interleukin-1 alpha inhibits prostaglandin E-2 release to suppress pulsatile release of lutein izing hormone but not follicle-stimulating hormone

Progressive multifocal leukoencephalopathy

Encephalomyelitis accompanied by myelin destruction experimentally produced in monkeys

Immune response gene products (la antigens) on glial and endothelial cells in virus-induced demyelination

Effects of transforming growth factor-beta on the functions of natural killer cells: Depressed cytolytic activity and blunting of interferon responsiveness

Short analytical review: Remitting-relapsing EAE in nonhuman primates: A valid model of multiple sclerosis

Basal and PAF-, interleukin 1-, ether stress-induced hypothalamic pituitary adrenal secretion of conscious rat: Modulation by PAF antagonists

Interleukin-1 stimu lates the secretion of hypothalamic corticotropin-releasing factor

Production of tumor necrosis factor-alpha by microglia and astrocytes in culture

Transforming growth factors type Bl and B2 suppress rat astrocyte autoantigen presentation and antagonize hyperinduction of class II major histocompatibil ity complex antigen expression by interferon-gamma and tumor necrosis factor-a

Major histocompati bility complex-expressing nonhematoporetic astroglial cells prime only CD8 + Τ lympho cytes: Astroglial cells as perpetrators but not initiators of CD4 + Τ cell responses in the central nervous system

Tumor necrosis factor mediates myelin and oligodendro cyte damage in vitro

Tumor necrosis factor and severe malaria

Acute phase protein, serum amyloid A, inhibits IL-1 and TNF-induced fever and hypothalamic PGE2 in mice

Suppression of immune cell function in vitro by recombinant human transforming growth factor-beta

Role of abortive retroviral infection of neurons in spongiform CNS degeneration

Three monoclonal antibodies against measles F protein cross-react with the cellular stress proteins

Molecular mimicry: Frequency of reactivity of monoclonal antibodies with normal tissues

Haplotype-specific inhibition of homing of radiolabeled lymphocytes in experimental allergic encephalomyelitis following treatment with anti-la antibodies

The arrangement of the immunoglobulin-like domains of ICAM-1 and the binding sites of LFA-1 and rhinovirus

The development of rational strategies for selective immunotherapy against autoimmune demyelinating disease

Chronic disseminated allergic encephalomyelitis in guinea pigs

Coronavirus infection induces H-2 antigen expression on oligodendrocytes and astrocytes

Virus-induced cell mediated immunity

Demonstration of HLA-DR in normal human brain

Interleukin-1 stimulates ACTH release by an indirect action which requires endogenous corticotropin releasing factor

Restricted use of T-cell receptor V genes in murine autoimmune encephalomyelitis raises possibilities for antibody therapy

Immunization with a synthetic T-cell receptor V-region peptide protects against experimental allergic encephalomyelitis

Transforming growth factor-beta is a potent immunosuppressive agent that inhibits IL-1-dependent lymphocyte proliferation

A specific receptor antagonist for interleukin-1 prevents Escherichia coli-induced shock in rabbits

Specific tropism of HTV-1 for microglial cells in primary human brain cultures

Measles and SSPE: Similarities and differences

Cellular immune reactivity within the CNS

Interaction of Tlymphocytes with cerebral endothelial cells in vitro

Persistence of neuroadapted mumps virus in brains of newborn hamsters after intraperitoneal inoculation

Human immunodeficiency virus infection and neurologic dysfunction

Chronic relapsing EAE an experimental model of multiple sclerosis

Multiple sclerosis: Immunological and experimental aspects

Inducible expression of H-2 and la antigens on brain cells

Detection of interleukin-1 and interleukin-6 in adult rat brain, following mechanical injury, by in vivo microanalysis: Evidence of a role for microglia in cytokine production

Lymphocyte homing receptor: Relationship between the Mel-14 antigen and cell surface lectin

Expression and modulation of HLA-DR on cultured human adult astrocytes

Fibroblast growth factors stimulate nerve growth factor synthesis and secretion by astrocytes

MHC restricted cytotoxic Τ cells: Studies on the biological role of polymorphic major transplantation antigens determining T-cell restriction specificity, function and responsiveness