Emerging epidemic viral encephalitides with a special focus on henipaviruses

In the last few decades, there is an increasing emergence and re-emergence of viruses, such as West Nile virus, Enterovirus 71 and henipaviruses that cause epidemic viral encephalitis and other central nervous system (CNS) manifestations. The mortality and morbidity associated with these outbreaks are significant and frequently severe. While aspects of epidemiology, basic virology, etc., may be known, the pathology and pathogenesis are often less so, partly due to a lack of interest among pathologists or because many of these infections are considered “third world” diseases. In the study of epidemic viral encephalitis, the pathologist’s role in unravelling the pathology and pathogenesis is critical. The novel henipavirus infection is a good example. The newly created genus Henipavirus within the family Paramyxoviridae consists of two viruses, viz., Hendra virus and Nipah virus. These two viruses emerged in Australia and Asia, respectively, to cause severe encephalitides in humans and animals. Studies show that the pathological features of the acute encephalitis caused by henipaviruses are similar and a unique dual pathogenetic mechanism of vasculitis-induced microinfarction and parenchymal cell infection in the CNS (mainly neurons) and other organs causes severe tissue damage. Both viruses can cause relapsing encephalitis months and years after the acute infection due to a true recurrent infection as evidenced by the presence of virus in infected cells. Future emerging viral encephalitides will no doubt continue to pose considerable challenges to the neuropathologist, and as the West Nile virus outbreak demonstrates, even economically advanced nations are not spared.     

Relapsing herpes simplex encephalitis: pathological confirmation of viral reactivation

This case is reported to raise awareness of herpes simplex encephalitis as a persisting brain disorder. A 66 year old immunocompetent man developed status epilepticus and died of pneumonia in the course of progressive hemiparesis, cognitive decline, and atrophy of the brain over a five year period after herpes simplex encephalitis. In addition to a completely destroyed left temporal lobe, necropsy revealed active encephalitis consisting of necrosis and lymphocyte infiltration with a large number of intranuclear inclusions in the neurones and glial cells in the markedly oedematous parenchyma of the right frontal and parietal lobes. Herpes simplex virus type 1 (HSV-1) antigen was detected by immunohistochemistry, HSV-1 DNA by in situ hybridisation, and herpes simplex virus nucleocapsids by electronmicroscopy. These clinical and pathological findings suggest that direct viral reactivation might result in a relapse of herpes simplex encephalitis, causing progressive clinical deterioration associated with the persistence of HSV-1 in the brain. This is the first case report demonstrating HSV-1 antigen, HSV-1 DNA, and herpes simplex virus nucleocapsids in a case of relapsing herpes simplex encephalitis.     

A CLINICO–PATHOLOGICAL STUDY OF HERPES SIMPLEX ENCEPHALITIS

A retrospective clinical and pathological analysis has been performed of 24 cases of herpes simplex virus encephalitis (HSE) seen at the Institute of Neurological Sciences, Glasgow, between 1972 and 1985. All patients had been diagnosed on the basis of isolation of herpes simplex virus (HSV) from, and/or the demonstration of characteristic histological changes of acute necrotizing encephalitis (ANE) in brain biopsy and/or autopsy tissue. Clinical presentation on admission included a prodromal influenza-like illness (46%), sudden onset of headache and confusion (54%), meningism (38%), deep coma (42%), aphasia (54%) and focal neurological signs (79%). Seizures occurred in 46% of cases during the course of the illness. Of the 24 cases, 14 (58%) died and 10 (42%) survived. Intravenous acyclovir treatment was associated with the best prognosis. Cerebral biopsy of one temporal lobe was performed in 22 cases and in 19 of these a positive histological diagnosis of HSE could be made. HSV was isolated from 15 of the 19 (79%) biopsied cases in whom virus isolation was attempted. Only seven out of the 15 cases (47%) in which immunofluorescence assays for HSV antigens were performed were unequivocally positive. Herpes simplex virus was isolated in culture from all cases which were negative by immunofluorescence. Immunocytochemical analysis on tissue sections of five representative brain biopsies demonstrated the presence of HSV antigens in some astrocytes, neurons and macrophages especially within areas of inflammatory infiltration. In situ hybridization experiments with a cloned HSV DNA probe demonstrated viral RNA in astrocytes, neurons and macrophages in two human biopsies and mouse brains in areas broadly corresponding to the distribution of viral antigen labelling. The combined immunocytochemical and in situ hybridization procedure showed that many but not all of the cells containing viral RNA also contained HSV antigens, indicating a productive infection in these double-labelled cells.     

Seizures and encephalitis: Clinical features, management, and potential pathophysiologic mechanisms

Encephalitis is an inflammation and swelling of the brain, which is often caused by a viral infection; it is an important cause of acute symptomatic seizures as well as subsequent epilepsy. Herein we describe the definition, epidemiology, and etiology of encephalitis as a cause of seizures. We then focus on encephalitis due to herpes simplex virus (the most common sporadic viral cause of encephalitis) and Japanese encephalitis virus (the most common epidemic viral cause). We also discuss the evidence for seizures occurring in the context of antibody-associated encephalitis, an increasingly important condition. Finally, we describe the acute and longer-term management of encephalitis-related seizures and their potential pathophysiologic mechanisms, concluding with the emerging etiologic role of human herpesvirus 6.     

Temporal analyses of the neuropathogenesis of a West Nile virus infection in mice

A West Nile virus (WNV) infection in humans can produce neurological symptoms including acute flaccid paralysis, encephalitis, meningitis and myelitis. To investigate the pathogenesis of WNV in the peripheral and the central nervous system (PNS and CNS), the authors used a murine footpad inoculation model of WNV infection. Survival curves of virus-infected animals of ages 4- to 6-weeks-old demonstrated age-dependent mortality where older animals (6-weeks-old) had a higher mortality rate compared to younger animals (4- and 5-weeks-old). The mice that survived the virus infection formed WNV-reactive antibodies, confirming viral infection and clearance. The localization of viral RNA (vRNA) and antigen in infected murine tissues was investigated using TaqMan and immunohistochemistry (IHC) respectively. During a nine day infection, vRNA levels in the spinal cord and brainstem fluctuated, suggesting early viral clearance from these tissues by days 3-4 p.i. with later re-introduction. Viral antigens detected using IHC were primarily observed in three main regions of the brain: cortex, hippocampus and brainstem. Additionally, the dorsal root ganglion neurons of the PNS stained positive for viral antigens. These data are consistent with multiple routes of neuroinvasion following a peripheral inoculation of virus and do not preclude the previous observation that virus-infected peripheral neurons can introduce virus into the CNS by a retrograde transport mechanism.     

ICAM-1 is crucial for protection from TMEV-induced neuronal damage but not demyelination

Previous work has suggested that the factors protecting mice from Theiler's murine encephalomyelitis virus (TMEV)-induced spinal cord demyelination are distinct from those involved in protection of the brain during the acute encephalitic phase. In this study, we examined the requirement for intercellular adhesion molecule-1 (ICAM-1) in both of these processes. During the acute phase of infection (days 7 to 10 after intracerebral infection with TMEV), no differences in brain or spinal cord pathology or virus burdens were observed between ICAM-1-knockout mice and the infected immunocompetent control mice of a similar background. Examination of brain pathology later in infection (that is, day 45 post infection [p.i.]) revealed that ICAM-1-deficient mice experienced increased levels of pathology in gray matter regions of the brain. We observed an increase in striatal damage and meningeal inflammation in the brains of TMEV-infected ICAM-1-knockout mice compared to C57BL/6J mice. Despite the increase in brain pathology, no immunoreactivity to viral antigens was detected, suggesting that the virus had been cleared by this time. Resistance to demyelination was similar in both groups, indicating that the resulting immune response was sufficient for protection of the spinal cord white matter.     

Enterovirus 71 encephalomyelitis and Japanese encephalitis can be distinguished by topographic distribution of inflammation and specific intraneuronal detection of viral antigen and RNA

K. T. Wong, K. Y. Ng, K. C. Ong, W. F. Ng, S. K. Shankar, A. Mahadevan, B. Radotra, I. J. Su, G. Lau, A. E. Ling, K. P. Chan, P. Macorelles, S. Vallet, M. J. Cardosa, A. Desai, V. Ravi, N. Nagata, H. Shimizu and T. Takasaki (2012) Neuropathology and Applied Neurobiology 38, 443–453 Enterovirus 71 encephalomyelitis and Japanese encephalitis can be distinguished by topographic distribution of inflammation and specific intraneuronal detection of viral antigen and RNA Aims: To investigate if two important epidemic viral encephalitis in children, Enterovirus 71 (EV71) encephalomyelitis and Japanese encephalitis (JE) whose clinical and pathological features may be nonspecific and overlapping, could be distinguished. Methods: Tissue sections from the central nervous system of infected cases were examined by light microscopy, immunohistochemistry and in situ hybridization. Results: All 13 cases of EV71 encephalomyelitis collected from Asia and France invariably showed stereotyped distribution of inflammation in the spinal cord, brainstem, hypothalamus, cerebellar dentate nucleus and, to a lesser extent, cerebral cortex and meninges. Anterior pons, corpus striatum, thalamus, temporal lobe, hippocampus and cerebellar cortex were always uninflamed. In contrast, the eight JE cases studied showed inflammation involving most neuronal areas of the central nervous system, including the areas that were uninflamed in EV71 encephalomyelitis. Lesions in both infections were nonspecific, consisting of perivascular and parenchymal infiltration by inflammatory cells, oedematous/necrolytic areas, microglial nodules and neuronophagia. Viral inclusions were absent. Conclusions: Immunohistochemistry and in situ hybridization assays were useful to identify the causative virus, localizing viral antigens and RNA, respectively, almost exclusively to neurones. The stereotyped distribution of inflammatory lesions in EV71 encephalomyelitis appears to be very useful to help distinguish it from JE.     

The immune response to picornavirus infection and the effect of immune manipulation on acute seizures

Viral infection of the central nervous system can result in encephalitis. About 20% of individuals who develop viral encephalitis go on to develop epilepsy. We have established an experimental model where virus infection of mice with Theiler’s murine encephalomyelitis virus (TMEV) leads to acute seizures, followed by a latent period (no seizures/epileptogenesis phase) and then spontaneous recurrent seizures—epilepsy. Infiltrating macrophages (CD11b⁺CD45^hi) present in the brain at day 3 post-infection are an important source of interleukin-6, which contributes to the development of acute seizures in the TMEV-induced seizure model. Time course analysis of viral infection and inflammatory [CD11b⁺CD45^hiLy-6C^hi] and patrolling [CD11b⁺CD45^hiLy-6C^low] monocyte and T cell infiltration into the brains of TMEV-infected C57BL/6J mice over the entire course of the acute viral infection was performed to elucidate the role of virus and the immune response to virus in seizures and viral clearance. The infiltrating inflammatory macrophages were present early following infection but declined over the course of acute viral infection, supporting a role in seizure development, while the lymphocyte infiltration increased rapidly and plateaued, advocating that they play a role in viral clearance. In addition, we showed for the first time that, while TMEV infection of RAG1^?/? mice did not alter the number of mice experiencing acute seizures, TMEV infection of C57BL/6J mice depleted of macrophages resulted in a significant decrease in the number of mice experiencing seizures, again supporting a role for infiltrating macrophages in the development of acute seizures in the TMEV-induced seizure model.     

Relapsed and late-onset Nipah encephalitis

An outbreak of infection with the Nipah virus, a novel paramyxovirus, occurred among pig farmers between September 1998 and June 1999 in Malaysia, involving 265 patients with 105 fatalities. This is a follow-up study 24 months after the outbreak. Twelve survivors (7.5%) of acute encephalitis had recurrent neurological disease (relapsed encephalitis). Of those who initially had acute nonencephalitic or asymptomatic infection, 10 patients (3.4%) had late-onset encephalitis. The mean interval between the first neurological episode and the time of initial infection was 8.4 months. Three patients had a second neurological episode. The onset of the relapsed or late-onset encephalitis was usually acute. Common clinical features were fever, headache, seizures, and focal neurological signs. Four of the 22 relapsed and late-onset encephalitis patients (18%) died. Magnetic resonance imaging typically showed patchy areas of confluent cortical lesions. Serial single-photon emission computed tomography showed the evolution of focal hyperperfusion to hypoperfusion in the corresponding areas. Necropsy of 2 patients showed changes of focal encephalitis with positive immunolocalization for Nipah virus antigens but no evidence of perivenous demyelination. We concluded that a unique relapsing and remitting encephalitis or late-onset encephalitis may result as a complication of persistent Nipah virus infection in the central nervous system.     

Simultaneous in situ detection of viral antigens and RNA in brain tissues from a patient with herpes simplex encephalitis by immunocytochemistry and in situ hybridization

In order to elucidate the correlation between herpes simplex virus (HSV-1) and the central nervous system tissue, we performed the simultaneous detection of viral antigens and RNA in the brain tissue sections from a patient with herpes simplex virus (HSV) encephalitis using immunocytochemistry and in situ hybridization. In the present study the hybridization protocol reported by Brahic M et al. in 1984 were applied for the simultaneous detection of viral RNA and antigens with a few modification. The sections were first immunocytochemically stained to detect HSV-1 antigens by ABC method, and then hybridized with 3H-labelled HSV-1 cDNA probe for the detection of RNA after the acetylation of slides for the prevention of nonspecific bindings of isotope to slides. In the present study, viral antigens were immunocytochemically stained to brown-colored deposits located in the cytoplasm and nucleus whereas viral RNA were detected as the accumulation of many silver grains over the nuclei or cytoplasm. In this case the light microscopic findings in a part of temporal lobe showed multiple areas of necrosis mainly involving the gray matter and a few inflammatory changes such as perivascular cell cuffings. HSV-1 infected Vero cells as positive control demonstrated both antigens and RNA as shown in Fig.1 a. However, no hybridization signals and color deposits were observed in uninfected Vero cells.(ABSTRACT TRUNCATED AT 250 WORDS)     

疱疹病毒性脑炎并发视力下降二例

目的探讨疱疹病毒性脑炎和视力下降的关系。方法对2例疱疹病毒性脑炎患者腰椎穿刺,行脑脊液测压、常规、生化、细胞学及病原学检查及头颅磁共振(MRI)检查;例1患者行眼科超声检查。结果2例患者均在入院治疗过程中出现视力下降。例1为单纯疱疹病毒感染,例2为巨细胞病毒感染。2例患者头颅MRI扫描均有脑实质炎性改变,均涉及岛叶。例1眼科超声提示视网膜坏死及视网膜脱离。结论疱疹病毒感染与急性视网膜坏死综合征及视神经炎有关。对于病毒性脑炎患者应给与足够的抗病毒治疗,期间若出现视力下降,警惕合并急性视网膜坏死综合征及视神经炎。     

Expression of caspase-3 in brains from paediatric patients with HIV-1 encephalitis

Apoptosis of neurones, macrophages, and microglia occurs in the brains of paediatric patients with human immunodeficiency virus (HIV) type 1 encephalitis, which is often associated with pre-mortem neurological disease (progressive encephalopathy). We have previously reported that TUNEL-positive neurones in brain tissue from paediatric patients with HIV type 1 encephalitis and progressive encephalopathy are strikingly devoid of the pro-apoptotic gene product Bax, in marked contrast to brain-resident macrophages and microglia. Using immunocytochemical methods, the present study demonstrate that neurones in patients with HIV type 1 encephalitis and progressive encephalopathy, as well as macrophages and microglia, but not astrocytes, overexpress caspase-3, a pro-apoptotic enzyme that is proteolytically activated downstream of Bax-Bcl-2 dysregulation. Co-localization of neuronal cytoplasmic caspase-3 and nuclear TUNEL staining, a marker for fragmented DNA, was also infrequently observed in brain tissue from patients with HIV type 1 encephalitis and progressive encephalopathy. These findings suggest that vulnerable neurones in brain tissue from patients with HIV virus type 1 encephalitis and progressive encephalopathy undergo apoptosis by a mechanism that involves upregulation of caspase-3 in a pathway that is independent of Bax-Bcl-2 dysregulation. Furthermore, caspase-3 upregulation in apoptotic neurones likely occurs prior to DNA fragmentation.     

Virulent and attenuated canine distemper virus infects multiple dog brain cell types in vitro.

Canine Distemper Virus (CDV) produces an encephalitis in dogs that varies with viral strain. We have studied the cell tropisms of two virulent strains (CDV-SH and CDV A75-17) and an attenuated strain, Rockborn (CDV-RO), in cultured canine brain cells. Infected cell types were identified by double immunofluorescent labeling of specific cell markers and viral antigens. All viral strains studied produced infection in astrocytes, fibroblasts, and macrophages. Neurons were not infected by CDV A75-17 but were rapidly infected by CDV-SH and CDV-RO. Multipolar oligodendrocytes were very rarely infected by any of the virus strains. In contrast, a morphologically distinct subset of bipolar oligodendrocytes were commonly infected by CDV-SH and CDV-RO. The kinetics of infection in the astrocytes, oligodendrocytes, neurons, and macrophages varied between strains. Both CDV-SH and CDV-RO rapidly infected bipolar oligodendrocytes, astrocytes, neurons, and macrophages by 14 days post infection while infection by CDV A75-17 was delayed until after 28-35 days post infection. The differences in the growth kinetics and cell tropisms for some brain cells, exhibited by the three viral strains examined in this in vitro study, may relate to the different CNS symptoms that these strains produce in vivo.     

A comparative study of acute and chronic diseases induced by two subgroups of Theiler’s murine encephalomyelitis virus

Theiler’s murine encephalomyelitis viruses (TMEV) are divided into two subgroups on the basis of their different biological activities. The GDVII strain produces acute polioencephalomyelitis in mice, whereas the DA strain produces demyelination with virus persistence in the spinal cord. A comparative study of GDVII and DA strains suggested that low host immune responses are responsible for the development of acute GDVII infection and that the persistence of infected macrophages plays a crucial role in the development of chronic white matter lesions in DA infection. All 78 mice infected with GDVII died or became moribund by day 13, while none of 54 mice infected with DA died. In the acute stage, the distribution of viral antigens in the central nervous system (CNS) tissue was similar in both GDVII and DA infections, although the virus titer was higher in GDVII infection. In DA infection, a substantial number of T cells were recruited to the CNS on day 6 when they were virtually absent in GDVII infection. The titer of neutralizing antibody was already high on day 6 in DA infection but was negligible in GDVII infection. Development of chronic paralytic disease from day 35 of the DA infection was accompanied by focal accumulation of viral antigen-positive macrophages in the spinal white matter. In addition, white matter lesions comparable to those in chronic DA infection were induced in the spinal cord within 7 days after intracerebral injection of DA-infected murine macrophages. Received: 26 June 1995 / Revised, accepted: 27 December 1995     

T-cells in human encephalitis

Encephalitis literally means inflammation of the brain. In general, this inflammation can result from a viral or bacterial infection in the brain itself or alternatively from a secondary autoimmune reaction against an infection or a tumor in the rest of the body. Besides this, encephalitis is present in (believed autoimmune) diseases with unknown etiology, such as multiple sclerosis or Rasmussen encephalitis (RE). This article summarizes the existing data on the role of T-cells in the pathogenesis of three types of human encephalitis: RE, paraneoplastic encephalomyelitis, and virus encephalitis. In all of them, T-cells play a major role in disease pathogenesis, mainly mediated by major histocompatiblity complex class I-restricted CD8⁺ T-lymphocytes.     

Upregulation of COX-2 and CGRP expression in resident cells of the Borna disease virus-infected brain is dependent upon inflammation

Infection of immunocompetent adult rats with Borna disease virus (BDV) causes severe encephalitis and neural dysfunction. The expression of COX-2 and CGRP, genes previously shown to be implicated in CNS disease and peripheral inflammation, was dramatically upregulated in the cortical neurons of acutely BDV-infected rats. Neuronal COX-2 and CGRP upregulation was predominantly seen in brain areas where ED1-positive macrophages/microglia accumulated. In addition, COX-2 expression was strongly induced in brain endothelial cells and the number of COX-2 immunoreactive microglial cells was increased. In contrast, despite increased expression of viral antigens, neither COX-2 nor CGRP expression was altered in the CNS of BDV-infected rats treated with dexamethasone, or tolerant to BDV. Thus, increased CGRP and COX-2 expression in the BDV-infected brain is the result of the inflammatory response and likely to be involved in the pathogenesis of virus-induced encephalitis.     

Herpes simplex encephalitis: an autopsy case with isolation of type 1 herpes simplex virus

Summary An adult case of herpes simplex encephalitis was studied after autopsy. Postmortem examination revealed necrotizing encephalitis associated with Cowdry type A intranuclear inclusion bodies in glial cells. Herper simplex virus type 1 was isolated from the removed brain. Herpes simplex virus antigens were detected diffusely in wide areas of the brain by immunofluorescent test and viral particles characteristic to herpes simplex virus were demonstrated by electron microscopy. There was an apparent discrepancy between severity of histological changes and distribution of virus antigen.     

Emerging and re-emerging epidemic encephalitis: a tale of two viruses

Two major epidemics of viral encephalitis occurred in Asia in 1997 and 1998. The first was a re-emergence of neurovirulent strains of enterovirus 71, which caused severe encephalomyelitis in children in Malaysia, Taiwan and Japan, on a background of hand, foot and mouth disease. Necropsy studies of patients who died of enterovirus 71 infection showed severe perivascular cuffing, parenchymal inflammation and neuronophagia in the spinal cord, brainstem and diencephalon, and in focal areas in the cerebellum and cerebrum. Although no viral inclusions were detected, immunohistochemistry showed viral antigen in the neuronal cytoplasm. Inflammation was often more extensive than neuronal infection, suggesting that other factors, in addition to direct viral cytolysis, may be involved in tissue damage. The second epidemic of viral encephalitis was the result of a novel paramyxovirus called Nipah, which mainly involved pig handlers in Malaysia and Singapore. Pathological evidence suggested that the endothelium of small blood vessels in the central nervous system was particularly susceptible to infection. This led to disseminated endothelial damage and syncytium formation, vasculitis, thrombosis, ischaemia and microinfarction. However, there was also evidence of neuronal infection by the virus and this may also have contributed to the neurological dysfunction in Nipah encephalitis. Some patients who seemed to recover from the acute symptoms have been re-admitted with clinical findings suggestive of relapsing encephalitis. As these two epidemics indicate, the emergence and re-emergence of viral encephalitides continue to pose considerable challenges to the neuropathologist, in establishing the diagnosis and unravelling the pathogenesis of the neurological disease.     

Impact of direct virus-induced neuronal dysfunction and immunological damage on the progression of flavivirus (Modoc) encephalitis in a murine model

Flavivirus encephalitis is believed to be the result of two main mechanisms: (i) direct damage to and dysfunction of neurons as a result of viral replication and (ii) destruction of the brain tissue by an inflammatory response. The differential impact of both mechanisms on the progression of flavivirus encephalitis has not been clearly determined. We have now studied the encephalitis caused by Modoc virus (MODV) infection in (i) severe combined immunodeficiency (SCID) mice, (ii) immunocompetent NMRI mice, and (iii) NMRI mice under varying immunosuppressive treatment regimens. In SCID mice, Modoc virus infection proved to be uniformly lethal (100%). The virus replicated extensively in neurons and no signs of inflammation of the brain were observed. In immunocompetent NMRI mice, intranasal (but not intraperitoneal) inoculation with MODV caused severe encephalitis accompanied by a fulminate inflammatory response. When NMRI mice, infected with MODV via the intraperitoneal route, were submitted to a brief immunosuppressive treatment, they also developed encephalitis with an obvious inflammatory component. These animals died significantly earlier than NMRI mice, which received immunosuppressive treatment for a longer period of time. In the latter group, no signs of inflammation of the brain were noted. These models thus allow us to distinguish between the impact of direct viral replication and that of immunological factors in the development of MODV encephalitis, and let us to conclude that (i) replication of the virus in neurons is sufficient to cause fatal encephalitis and (ii) immunological factors contribute significantly to disease progression.     

Cerebrospinal fluid is an efficient route for establishing brain infection with feline immunodeficiency virus and transfering infectious virus to the periphery

Like human immunodeficiency virus (HIV), feline immunodeficiency virus (FIV) invades and infects the central nervous system (CNS) soon after peripheral infection. The appearance of viral RNA is particularly prominent in the cerebrospinal fluid (CSF), suggesting an efficient route of virus transfer across the blood-CSF barrier. This raises the concern whether this route can establish a stable viral reservoir and also be a source of virus capable of reseeding peripheral systems. To examine this possibility, 200 mul of cell-free NCSU1 FIV or FIV-infected choroid plexus macrophages (ChP-Mac) was directly injected into the right lateral ventricle of the brain. Negative controls were sham inoculated with uninfected ChP-Mac or virus-free culture supernatant and positive controls were infected systemically by intraperitoneal (i.p.) injection. Intracerebroventricular (i.c.v.) inoculation with cell-free FIV resulted in high levels of plasma FIV RNA detected as early as 1 to 2 weeks post inoculation in all cats. In each case, the plasma viremia preceded the detection of CSF viral RNA. Compared to i.p. cats, i.c.v. cats had 32-fold higher CSF viral loads, 8-fold higher ratios of CSF to plasma viral load, and a 23-fold greater content of FIV proviral DNA in the brain. No FIV RNA was detected in plasma or CSF from the cats inoculated with FIV-infected ChP-Mac but an acute inflammatory response and a slight suppression of the CD4+:CD8+ ratio were observed. These results indicate that free FIV circulating in the CSF promotes infection of the CNS and provides a highly efficient pathway for the transfer of infectious virus to the periphery.