1995 Abstracts

AUTHORS: Jay D. Amsterdam, M.D., Greg Maislin, M.S., M.A.
AFFILIATION: Depression Research Unit, University of Pennsylvania Medical Center, 3600 Market Street, 8th Floor, Philadelphia, PA (USA)
TEXT:In vitro studies have shown an inhibitory effect of lithium salts on herpes simplex virus (HSV) replication by mechanisms that appear to interfere with viral DNA synthesis. Several clinical observations indicate that lithium may have an anti-viral activity in recurrent HSV infections.

We conducted a randomized, double-blind, placebo-controlled, two-year trial of oral lithium carbonate prophylaxis in 11 subjects with recurrent HSV (type-II) infections.

We observed a modest reduction in 4 out of 5 illness severity measures during lithium therapy. In contrast, placebo resulted in an increase in 3 out of 5 outcome measures of illness severity. Although the comparisons between groups failed to achieve statistical significance due to the limited sample sizes, these data support prior observations of an anti-viral activity of lithium, and suggest the possibility of a prophylactic, anti-HSV activity with oral lithium carbonate in patients with recurrent HSV infections.

Acknowledgements: This project was partially supported by B.R.S. grant So7-RR-05415-28 awarded by the Biomedical Research Support Grant Program, Division of Research Resources at the National Institutes of Health.

AUTHORS: S.E.Bachus*, C. Ojeh1, M.M. Herman*, E.F. Torrey*, J. Cervenak*, J.E. Kleinman* and R.H. Yolken1
AFFILIATION: *Clinical Brain Disorders Branch, NIMH Neuroscience Center at St. Elizabeths, Washington, DC 20032; 1The Stanley Neurovirology Laboratory, Johns Hopkins University, 600 N. Wolfe Street, Blalock 1111, Baltimore, MD 21287
TEXT:In light of the suggestion that there may be a viral etiology in schizophrenia, we have been developing a cat model in which to study schizophrenia, by exposing kittens postnatally to the neurotropic pestivirus, Bovine Viral Diarrhea Virus (BVDV). Pestiviruses can infect humans, and BVDV may cause neurological disorders. We report here the results of our preliminary attempt to localize mRNA from BVDV in the brain of cats exposed postnatally to either BVDV, control saline treatment, normal human brain, or schizophrenic brain.

At 1 week of age, kittens, reared in a pathogen-free environment, received intracerebral injections of one of the following: saline, BVDV, pooled normal human brain or pooled schizophrenic brain. No kitten displayed signs of illness or behavioral alternations. At 7 days post-injection, cats were euthanized and their brains fixed in formalin. Subsequently, 5 micron thick paraffin-embedded sections were mounted on gelatin-coated slides. In situ hybridization was performed on slides with cerebral cortex or cerebellum, using 35-S labelled oligonucleotide probes directed against 4 different regions of the BVDV mRNA. Images from films were quantified, using a 14C standard curve and the program IMAGE (Rasband, NIH), and statistics were done by 3-way ANOVA (group X region X probe).

No signal was seen in any of the samples from saline control-treated cats. For all 4 probes, labelling was greater in both cortex and cerebellum of kittens treated with BVDV, normal human brain or schizophrenic brain than it was in saline controls (p=.0007). These results demonstrate that exposure of kittens to BVDV results in the presence of mRNA for BVDV in cerebral cortex and cerebellum of these animals. Moreover, detection of BVDV mRNA in the brain of kittens exposed to human brain suggests that BVDV is present in the brain of some humans. We are continuing to explore whether BVDV or other pestiviruses may play a role in the etiology of schizophrenia.

AUTHORS: LD Bobo, NL Johnston, NG Novak, EF Torrey, RH Yolken
AFFILIATION: Johns Hopkins Univ. School of Medicine, Stanley Neurovirology Laboratory, 600 N. Wolfe St., Blalock 1111, Baltimore, MD 21287
TEXT:Cytokines are known mediators of both immune and central nervous system function and homeostasis that can be produced in response to infection and can cause neurodegeneration. Schizophrenia may be viewed as a heterogenous disorder resulting from the possible interactions of exogenous (infectious), autoimmune or hormonal elements with key susceptible gene(s) affecting cognition and behavior. Studies employing cytokine protein measurements on peripheral fluids, such as plasma and CSF, have yielded conflicting results. The purpose of this study is to compare the local cytokine milieu in persons with schizophrenia to controls. Using both a quantitative reverse transcriptase PCR-EIA and a fluorescence-tagged primer PCR, we are first comparing types and quantities of interleukins-1b , 2, 3, 4, 6, 8, 10, and 12, intercellular adhesion molecule-1, tumor necrosis factor-% and interferon-g mRNAs from visual, prefrontal, frontal and parietal cortex. We are examining autopsy samples from persons who had schizophrenia, bipolar disorder and controls with no psychiatric history. One person from the schizophrenia group differentially expressed a retroviral gene transcript. Samples from the bipolar group are included as a control for the effect of medication. Later studies will use in situ methods on select brain areas to determine the cellular localization of relevant transcripts. We will also determine if MHC markers are present as another function of immune activation. It is hoped that this avenue of research may offer an alternative intervention for treatment of central disorders.

AUTHORS: LV Jones Brando, LE Holland, KM Carbone, SA Rubin, RH Yolken, EF Torrey
AFFILIATION: Johns Hopkins University School of Medicine, Stanley Neurovirology Laboratory, 600 N. Wolfe St., Blalock 1111, Baltimore, MD 21287
TEXT:Current therapy for schizophrenia relies on the use of typical antipsychotic medications, which specifically inhibit binding of ligand at the D2 dopamine receptor, and atypical medications which display little activity for this receptor interaction. While atypical antipsychotic agents have been shown to variably inhibit other neuroreceptor-ligand interactions, the exact mechanisms for the therapeutic efficacy of these medications have not been completely defined. Clozapine, an atypical antipsychotic, and nine of its metabolites were studied in vitro for possible antiviral activity against models of human neurotropic viruses, including human immunodeficiency virus (HIV) type 1. In an assay for inhibition of HIV-induced cytopathic effect (CPE) four chemically similar metabolites demonstrated antiviral activity (ID5037 . 150 m g/ml while other metabolites demonstrated antiviral medications had limited or no effect. Two of the metabolites tested thus far, 8-hydroxy-8-deschloro-clozapine and 8-hydroxy-desmethyl-clozapine inhibited the production of p24, the major internal antigen of HIV (ID50 9.2 – 10.7 m g/ml as measured by ELISA. These data suggest that the therapeutic efficacy of some antipsychotics may be due in part to an ability to inhibit viral replication. Antiviral agents may prove to be effective adjuncts in the treatment of schizophrenia.

AUTHORS: James S. Brown, Jr., M.D.
AFFILIATION: Keller Army Community Hospital, U.S. Military Academy, West Point, NY, 10996
TEXT:Schizophrenia prevalence in the United States is highest in urbanized northeastern, northwestern and Great Lakes states. The viral theory of schizophrenia attributes this distribution to enhanced susceptibility to viral infections in crowded, urban areas. Examination of the geographical distribution of infectious diseases in the United States reveals that the spreading foci of Lyme disease and its primary vectors, Ixodes ticks, correlate significantly with high schizophrenia areas. Ixodes ticks are vectors in North America and throughout the world of over 40 viruses including tick-borne encephalitis. The international distribution of tick-borne encephalitis is shown to be concentrated in countries where the highest rates of schizophrenia are found: Croatis, Norway, Finland, Germany, Ireland, and others. The geographical specificity of this correlation and the plausibility of an Ixodes tick virus theory of schizophrenia is discussed. More than half of the known Ixodes tick viruses were first isolated from migratory seabirds. The possibility that viruses from migratory seabirds contribute to the global distribution of schizophrenia is presented.

Alan S. Brown, M.D.; Ezra S. Susser, M.D., Dr. P.H.; Catherine Schaefer, Ph.D.; Richard J. Wyatt, M.D.
Columbia University College of Physicians and Surgeons, New York State Psychiatric Institute
The majority of epidemiologic studies indicate a potential role of prenatal influenza exposure in schizophrenia, although there are notable negative studies. These conflicting findings may be due to several limitations in study design, including the use of ecologic data, rather than defining exposure in individual pregnancies, diagnoses based upon hospital registry data, and lack of control for confounding variables. In order to address these limitations, it is critical to understand these research design issues. We shall therefore begin by presenting a critical review of prior epidemiologic studies of prenatal influenza and schizophrenia. This review will lead to a discussion of a seroepidemiologic study initiated by our group, which is designed to overcome each of these limitations. We have begun a schizophrenia follow-up study of a large birth cohort, the California Child Health and Development Study (CHDS). Mothers of cohort members were studied in the course of their prenatal care between 1959 and 1966, and the serum samples were drawn at prenatal and postpartum visits; these samples have been stored and are being obtained for analysis. Thus, it will be possible for the first time to link serologic data, indicating influenza infection during pregnancy, with schizophrenia in the offspring from these pregnancies. This should permit a unique and potentially definitive test of putative associations between prenatal exposure to influenza and other viral agents in the etiopathogenesis of schizophrenia. 

AUTHORS: K.M. Carbone*, P.M. Silvas*, S.A. Rubin*, M. Vogel^, T.H. Moran#, and G. Schwartz#
AFFILIATION: *Viral Immunopathogenesis Laboratory, Departments of Medicine and #Psychiatry, The Johns Hopkins University School of Medicine; ^Maryland Psychiatric Research Center, University of Maryland School of Medicine.
TEXT:Neonatal infection with Borna disease virus (BDV) damages cerebellar development, and causes hippocampal degeneration and physiological and behavioral abnormalities in the rat1, 2, 3, 4. BDV is a candidate infectious etiology for schizophrenia, based on several lines of evidence: 1) Antibodies to BDV have been identified in patients with multiple psychiatric diseases5. 2) Recent development of a Western blot for BDV allowed the first publication of an association between BDV antibodies in humans and schizophrenia6. 3) The development of an RT-PCR assay for BDV and the discovery of BDV in peripheral blood mononuclear cells from infected rats7, 8 permitted the detection of BDV-like RNA sequences in peripheral blood cells of patients with mixed psychiatric disorders9. 4) The hippocampus is a target for BDV infection, degenerates over time in neonatally BDV infected rats, is a site for schizophrenia-associated pathology, and is the center for spatial learning and memory4.

We measured the neuronal loss from the dentate gyrus and spatial learning and memory in neonatally BDV infected Lewis rats between 23 to 73 days p.i. 1) Neuronal dropout in the dentate gyrus, measured as neuron density, was strongly correlated with decreasing performance in the Morris Water Maze test (R=0.81). 2) Spatial learning and memory are indicated by the swim time required for the rats to navigate by spatial clues to a submerged escape platform. At every timepoint the BDV-infected rats had slower mean swim times as compared to the uninfected, control rats (p<0.05). 3) Relative performance improvement, indicative of spatial learning ability, was significantly worse in BDV infected rats than control rats (p<0.05). 4) The number of times the rat crosses the target area indicates spatial memory. Again, the BDV infected rats performed significantly worse than the control rats (p<0.05). These results demonstrate a clear and quantitative correlation between the increasing damage to the hippocampus in neonatally BDV infected rats and progressively poor performance in spatial learning and memory tasks.

1Dittrich W et. al. Biol. Psych., 26:818-828, 1989.

2Bautista JR et. al. In Press, Brain Res., 1995.

3Bautista JR et. al. Brain Res. Bull., 34(1):31-40, 1994.

4Carbone KM et. al. J. Virol., 65(11):6154-64, 1991.

5Bode, L. In Borna Disease, p. 103-130, Eds. Koprowski & Lipkin. Springer Verlag, NY, 1995.

6Waltrip, RW II et. al. Psych. Res. 56 33-44, 1995.

7Sierra-Honigmann AM et. al. Brit. J. Psych. 166:55-60, 1995.

8Rubin SA et. al. Blood 85: 2762-2769, 1995.

9Bode L et. al. Nature Medicine, 1(3):232-236, 1995.

AUTHORS: William T. Carpenter, Jr., M.D.
AFFILIATION: Professor of Psychiatry, Univ. of Md. School of Medicine; Director, Maryland Psychiatric Research Center.
TEXT:Schizophrenia is a clinically heterogenous syndrome. This results in a substantial impediment to research even when the dependent variables are conceptualized as a pathophysiological final common pathway, treatment response, or the consequence of illness. If this clinical heterogeneity derives from etiological heterogeneity, studies of viral or immune causal mechanisms in schizophrenia face a profound challenge. Type II error and failure in independent replication will be common.

Heterogeneity in schizophrenia can be significantly reduced by the application of a Domains of Psychopathology approach resulting in subgroups which differ in age and nature of onset, course and treatment response, and a range of structural and functional neurological variables. The Domains of Psychopathology consist of (1) hallucinations and delusions; 2) dissociative thought; and 3) deficit negative symptoms. distinct neural circuits are implicated in each domain and the specific pathology, rather than the schizophrenia syndrome, is the disease target. Therefore, any hypothesized immune or viral etiology with a distinctive neuroanatomic or neurobehavioral correlate will find a more robust study design using a Domains of Psychopathology approach.

One of the Domains of schizophrenia (deficit psychopathology) will be used to illustrate the enhanced power of study designs which reduce heterogeneity. In another presentation, application of this design in the study of Borna disease virus serology in schizophrenia will be presented (Waltrip et al., pending acceptance).

S. Chatterjee, S. Dey, M.M. Herman, J.E. Kleinman, T.M. Hyde, R.H. Yolken
Stanley Laboratory for the Study of Schizophrenia and Bipolar Disease, Department of Pediatrics, Johns Hopkins University, Baltimore, MD 21287-4933
Glycosphingolipids (GSL) are major components of brain cell membranes and are principal determinants of intracellular signaling. The GSL components of the brain frontal lobes of individuals with schizophrenia has not been previously determined. We extracted neutral GSLs from the frontal lobes of brains from 29 individuals. These included individuals with schizophrenia (n=7), normal controls (n=7), suicides (n=9) and non-schizophrenic individuals who had received neuroleptic medication (n=6). The individual neutral GSL components were resolved by high performance thin layer chromatography and quantified by densitometry. 

The levels of ceramide, monoglycosyl ceramide, lactosyl ceramide, and sulfatide did not differ among the study groups. However, the mean level of the A lipid that co-migrated with globotriosyl ceramide (GbOse3Cer) in the brain frontal lobe of individuals with schizophrenia (74.1 ug/gm of tissue) was significantly lower (p<.01) than the mean levels measured in the brains of normal controls (194.5 ug/gm) and in the brains of non-schizophrenic individuals who committed suicide (171.7 ug/gm) and who were taking neuroleptics (130.4 ug/gm).

This study indicates that the levels of this uncharacterized lipid are decreased in the brains of individuals with schizophrenia. Glycosphingolipids are of biological importance since they serve as receptor for infectious agents and for inflammatory mediators such as a 2-interferon. Altered levels of GSL may modulate susceptibility to infectious agents and contribute to the pathophysiology of schizophrenia.

AUTHORS: Indra Dé, Lorraine Brando, Robert Yolken, EF Torrey
AFFILIATION: Johns Hopkins Univ. School of Medicine, Stanley Neurovirology Laboratory, 600 N. Wolfe St., Blalock 1111, Baltimore, MD 21287-4933
TEXT:There are several aspects of the epidemiology and pathophysiology of schizophrenia which suggests a major role for viral infections. However, previous studies employing tissue culture have generally failed to identify replicating viral agents in samples obtained from individuals with this disease.

There have been recent advances in the cultivation of neural cells and in the identification of viral replication within infected cells. We thus have attempted to infect A172 cells, a glioblastoma cell line, with 0.2m m filtered brain extracts obtained post-mortem from individuals with schizophrenia. Aliquots of the cells and supernatants of inoculated cells and controls were obtained at defined intervals following inoculation and analyzed for evidence of viral replication by a number of methods.

Analytical methods employed for these studies include-

< Electron microscopy to detect the presence of viral particles

< Polymerase chain reaction (PCR) using primers directed at defined viral target sequences

< PCR employing arbitrary primers and differential display to detect novel transcripts

< Product enhanced reverse transcription (PERT) to detect retroviral reverse transcriptase

The interim results of this ongoing project will be presented and discussed.

Daniel Gonzalez-Dunia, Beatrice Cubitt, Eliezer Masliah, and Juan Carlos de la Torre*.
The Scripps Research Institute
Borna disease virus (BDV) causes a central nervous system disease manifested by profound behavioral abnormalities in several vertebrate animal species. Cross-sectional seroepidemiological studies together with prospective follow-up studies on psychiatric patients have suggested that BDV might infect humans, possibly being associated with certain mental disorders. This possibility is further supported by the detection of both BDV-specific antigens and RNA in peripheral blood mononuclear cells (PBMCs) of psychiatric patients but not in PBMCs of healthy controls. Co-cultivation experiments have allowed us to recover BDV from PBMCs of three hospitalized psychiatric patients. BDV was unequivocally identified based on sequence identification of BDV open reading frames (ORFs) p24, p16, and p56, as well as of the predicted catalytic domain of the BDVL polymerase and partial sequencing of BDV ORF p40. BDV human isolates and BDV from naturally infected animals of different species showed a high degree of sequence conservation. In addition, we have also detected both BDV antigen and RNA in autopsy brain samples of patients who, based on well defined clinical records, were afflicted with psychiatric disorders that have been associated with BDV.

our results provide support for the hypothesis that BDV could be one of the environmental determinants contributing to the pathophysiology of major neuropsychiatric disorders whose etiology remains elusive.

AUTHORS: Lee E. Eiden, Dianne Rausch, Anna da Cunha, Eberhard Weihe, Todd Reinhard and Ashley Haase
AFFILIATION: Section of Molecular Neuroscience, Lab. of Cell Biology, NIMH, NIH, Bethesda MD USA (LEE, DR, AdC); Dept. of Anatomy and Cell Biology, Phillips University, Marburg, Germany (EW); Dept. of Microbiology, Univ. of Minnesota, Twin Cities Campus, MN USA(TR,AH)
TEXT:Rhesus monkeys infected with simian immunodeficiency virus (SIV) develop acquired immuno-deficiency syndrome (AIDS) and motor and cognitive impairments reminiscent of HIV-associated cognitive /motor complex, or AIDS dementia complex (E. Murray et al., Science 255:1246, 1992). Examination of the central nervous system of SIV-infected rhesus monkeys at various stages of immune disease can reveal, as human AIDS brain at autopsy cannot, whether or not various indices of brain inflammation, productive virus infection, or other neurochemical or immunochemical markers associated with ADC in human brain are coincident with the onset of encephalopathy, and thus candidates for its cause, or not.

We have observed an increase in the number of cortical interneurons expressing mRNA encoding the neuropeptide somatostatin (SOM mRNA) in SIV-infected monkeys with either motor or cognitive impairment, and apparently independently of the stage of immune disease or the level of productive viral infection in the CNS (A. da Cunha et al., PNAS 92:1371, 1995). Increased numbers of SOM+ interneurons are found in lamina IV-VI and subjacent white matter in animals with predominantly motor impairments, and in lamina IV without significant increases in deeper cortical layers in animals with predominantly cog-nitive impairments. Furthermore, reactive astrogliosis as measured by increased numbers of glial fibrillary acidic protein (GFAP) mRNA+ cells also exhibited a dual pattern in animals with motor or cognitive im-pairments, being elevated in layers IV-VI and subjacent white matter in monkeys with motor impairments, and not significantly elevated in any cortical lamina in monkeys with predominantly cognitive impairments.

Up-regulation of SOM mRNA+ cell number was not highly correlated with level of productive virus infection of CNS as assessed by in situ hybridization histochemistry, which was highly correlated with systemic virus burden (p26 viral antigenemia). Cells of the CNS infected with SIV included endothelial cells of the blood-brain barrier, microglial cells both singly and in microglial nodules, and macrophages and multinucleated giant cells.

We postulate that lentiviral invasion of the central nervous system in SIV disease initiates a process of inflammation that ultimately leads both to encephalopathy and to widespread productive viral infection of the brain. Increased expression of SOM mRNA in cortical interneurons appears to be a marker for the encephalopathic process, and attendant astrogliosis (increased numbers of GFAP mRNA+ cells) or its absence may indicate inflammatory processes initiated locally (concommitant SOM mRNA and GFAP mRNA increases) in cortex or in subcortical regions and conveyed to the cortex trans-synaptically (SOM mRNA+ cell increases without intense cerebrocortical astrogliosis)

AUTHORS: Rohan Ganguli, MD, B.S. Rabin, MD, PhD, J.S. Brar, MD, A.V. Gubbi, PhD
AFFILIATION: Univ. of Pittsburgh Medical Ctr., Department of Psychiatry
TEXT:Autoimmunity has been suggested as a pathophysiologic mechanism in some patients who suffer from schizophrenia. Among the most reproducible findings supporting this possibility, is that, during episodes of illness, lymphocytes from schizophrenic patients have a lowered capacity to produce the cytokine, Interleukin-2 (IL-2). This abnormality has been confirmed in first-episode never-treated patients, ruling out the possibility that it is an artifact of treatment. However, in common with all cross-sectional biological abnormalities that have been found to be associated with schizophrenia, it has been impossible to determine whether decreased IL-2 is a “cause” of the illness (or its symptoms), or an “effect.”

Two patients with schizophrenia were followed for one year with weekly blood draws and clinical assessments. During this time one patient had two relapses and the other had one. It was found that the quality of IL-2 produced by the lymphocytes on each occasion was highly predictive of the clinical state of the patients in the subsequent week. In a regression model, IL-2 production and serum level of IgG, against hippocampal antigen, together, were very efficient in predicting whether the patient was going to be in relapse or remission in the week following the sample. The apparent temporal precedence of immunologic alterations over changes in clinical state, favor the hypothesis that immune factors may play a pathophysiologic role in relapse in schizophrenic patients.

The possibility that immune changes may yet be the result of sub-clinical manifestations of the illness or an epiphenomenon cannot be completely ruled out at this time. It should be noted that one additional patient studied, did not show the above relationship between cytokine production and clinical state. This confirms several previous investigations which indicate that immune abnormalities are associated with the illness in only a subgroup of patients with schizophrenia.

Further studies of a similar longitudinal design are needed along with other biologic measures reflective of brain function.

AUTHORS: Ann B. Goodman
AFFILIATION: Nathan S. Kline Institute for Psychiatric Research
TEXT:In mammals vitamin A (retinoid) is an essential nutrient for development. It is involved in gene expression, immunocompetance, and in cell differentiation, proliferation, migration, and death. Retinoic acid (RA), the morphogenic derivative of vitamin A, alters patterns of neurulation, and brain development. Delivery of retinoids is precisely controlled throughout development. Rentinoid toxicity or deficit can result in psychosis, hydrocephalus and craniofacial anomalies. Alterations in the retinoid cascade may have profound implications for the etiology of schizophrenia. Retinoid transcription factors in the form of steroid hormone-like retinoic acid receptors (RARs) are incorporated into target genes by means of specific DNA motifs, termed retinoic acid response elements (RAREs). These particular sequences recognize RARs which then serve as mechanisms of gene transcription and regulation.

Viruses do not require retinoid. However, during ontogeny many viruses appear to have scavenged DNA sequences from their hosts and so contain the DNA motifs of putative RAREs. Among the many viruses containing these RARE motifs are influenza virus A England RNA, feline leukemia virus RNA, herpes simplex virus, type 1, Coxsackie virus B1, measles virus, Theiler murine encephalomyelitis, adenovirus, cytomegalovirus, Epstein-Barr virus, and HIV. It is proposed here that these viral RAREs may be incorporated into host genes that originally may not have been retinoid responsive, but become retinoid responsive upon infection by the virus. These newly incorporated RAREs may then respond inappropriately to RA, thus dysregulating the temporal and spatial activation of genes by retinoid during crucial developmental sequences. Such dysregulation, particularly that occurring during the second trimester, a time of high retinoid neurodevelopmental activity, may result in the subsequent stigmata of schizophrenia.

Beth Levine, J. Marie Hardwick, Philippe Despress, and Diane E. Griffin
Johns Hopkins School of Public Health, 615 N. Wolfe Street, Baltimore, MD 21205
Alphaviruses are causes of mosquito-borne encephalitis in the Americas. Sindbis virus (SV) is an alphavirus that causes acute encephalitis in mice. The primary target cell in the central nervous system (CNS) for viral infection is the neuron. Although infection of immature neurons leads to cell death by induction of apoptosis, infection of mature neurons is noncytolytic. Immunodeficient 3-4 week old mice inoculated with SV develop persistent CNS infection while immunocompetent mice clear infectious virus from the CNS within 7-8 days after infection. The mechanism of clearance involves antibody probably acting synergistically with interferon a /b _. Clearance of SV occurs without apparent damage to the infected neurons. RT-PCR analysis of the brains and spinal cords of infected mice reveals that SV RNA is persistent suggesting that virus replication is controlled by local production of antiviral antibody and interferon, but infected cells have not been eliminated. The potential exists for the development of chronic neurologic disease as a consequence of this persistence of viral RNA in neurons. 

John Gruzelier
Charing Cross and Westminster Medical School, London
There is extensive evidence that mood influences immune functions (Adey et al, 1991). An important substrate of mood expression is fronto-limbic asymmetry as disclosed electrophysiologically (Davidson, 1994). Animal experimentation has shown that unilateral lesions of frontoparietal neocortex have opposite effects on T-cell production and responsiveness (reviews by Renoux and Bizierre, 1986: Neveu, 1993). The direction of the lesion effects is consistent with frontal asymmetry mood relations, and with the influence of approach/withdrawal emotions on immunity (Gruzelier, 1989). These relations were tested in a longitudinal investigation of asymptomatic HIV+ men (Baldeweg et al, 1995). Neuropsychological and EEG measures at study onset correlated with CD4 and CD8 counts up to three years later, such that the better the left hemisphere functioning or the greater the left/right asymmetry, the higher the immune competence. Changes in neuropsychophysiological parameters over time also mirrored declines in immune function and progression to symptomatic stages of the disorder (Gruzelier et al, 1995). On the basis of genetic and epidemiological evidence similarities have been drawn between schizophrenia and immune disorders (Knight et al, 1987). Heterogeneity is also reported; prior to neuroleptics surveys of causes of death in US hospitals indicating striking subdiagnostic differences within schizophrenia in cancer proneness. Since 1981 we have reported evidence for two patterns of lateral asymmetry in schizophrenia (Gruzelier, 1991). These delineated positive versus negative syndromes, the character of which included affinities with the approach/withdrawal dimension. When combined with the evidence above, asymmetry patterns may provide markers with which to elucidate heterogeneity of immune response in schizophrenia.

AUTHORS: D.J. Hart (1), R.G. Heath (1), F.J.Sautter, Jr. (1), R.F. Garry (2), M.A. Beilke (3)
AFFILIATION: Tulane University School of Medicine, New Orleans, LA 70112 (1) Department of Psychiatry and Neurology, (2) Department of Microbiology and Immunology, (3) Department of Medicine.
TEXT:Results of the HIV-1 Western blot assay may have three possible interpretations: positive, negative, or indeterminate. The indeterminate interpretation may result from testing at an early stage of infection before full seroconversion. On the other hand, some individuals maintain a stable indeterminate result over time, a condition that is not indicative of HIV-1 infection. One possibility for such a case is that the patient has been exposed to another retrovirus which is antigenically related to HIV-1.

We have used the HIV-1 Western blot assay as a source of retroviral antigens to screen an archived bank of sera (n=81) Taken from schizophrenics, patients having a mental illness with schizoid features, and from patients having other conditions (including non-mental illness). A significant number of samples from the schizophrenics or from those having a mental disorder with schizoid features yielded indeterminate Western blots. Most of the reactivity was found to be directed against the group-specific antigens of HIV-1, suggesting exposure to a related retrovirus.

Additionally, 3/81 of the sera were scored as HIV-1 positive. Two of these patients (chronic schizophrenics) were traced. Their archived samples had been drawn in 1987 during an active schizophrenic episode. Both patients were now in a residual phase of schizophrenia and reported compliance with their medications. current serum samples were assayed for HIV-1 antibodies. One of the patients had seroreverted to a negative blot. The other retained reactivity to only a single HIV-1 antigen, indicating a change from positive to indeterminate. These results were confirmed by indirect immunofluorescent microscopy.

We have also noticed a trend among other serum samples taken from schizophrenics that suggests that stabilization with psychotropic drugs may affect the level of antibody reactivity to HIV-1 antigens. These data support a possible retroviral involvement in schizophrenia and suggest possible immunomodulation by psychotropic medications.

AUTHORS: G.R. Heninger, P. Rao, L. Karper, M.P. Heys and J. Krystal
AFFILIATION: Dept of Psychiatry, Yale Univ. and West Haven Veterans Administration Hospital, New Haven, CT 06510 and Lab Clin. Science, NIMH, Bethesda, MD 20892
TEXT:Many lines of evidence indicate that infectious and/or abnormal neural-immune mechanisms may be involved in schizophrenia. In order to assess whether immune abnormalities are evident in CSF of patients with schizophrenia, 6 measures sensitive to neural-immune activation were assessed in the CSF of 21 male patients (mean age: 41) with schizophrenia (SCH) in comparison to 16 healthy controls (HC) (mean age: 35). CSF was obtained from patients who were off medication and were hospitalized at the West Haven Veterans Administration Hospital and most had been ill several years. Quinolinic acid (QUIN), was measured by GCMS, and interlukins (IL1B,IL2, and IL6), tumor necrosis factor alpha (TNF alpha) and beta 2 microgloublin (B2MG) by double antibody ELISA.

There was a small increase in mean QUIN levels in SCH of 24% over those in HC (18.7 vs 15.0 nM/L respectively, P<.04 Chi square). Mean IL6 and B2MG levels did not differ between SCH and HC (2.4 vs 2.2 pg/ml and 1.3 vs 1.4 ug/ml respectively). Levels of IL1B, IL2, and TNF alpha did not differ between SCH and HC since 82vs88; 100vs100, and 71vs81% respectively, had levels below the lower detection limits of the assay.

The 24% increase in QUIN levels in SCH is much smaller than the 200-500% increase seen in early stage HIV infected patients. QUIN levels in SCH are more like patients with Alzheimer’s Disease where CSF QUIN levels are not statistically different than normals. The lack of elevations of IL2B, IL2, TNF alpha and B2MG indicate that an active infectious neural-immune process is no occurring in the large majority of these chronic SCH patients.

AUTHORS: Hornig-Rohan M*, Van Bell CT#, Sluzewska AH, Amsterdam JD*.
AFFILIATION: *Depression Research Unit, Univ. of Penn., Phila., PA#Dept. of Research, Hamot Medical Ctr., Erie PA

HDept. of Psychiatry, Univ. School of Med. Sciences, Poznan, Poland

TEXT:Background: Depression has been associated with altered immune function. In addition to evidence of in vivo cellular immune activation and increased autoimmune responses, previous investigations have noted elevated plasma positive acute phase proteins in individuals with depressive disorders. However, the association of such findings with depressive subtypes has not been well-researched. this preliminary investigation sought to determine whether differences in acute phase proteins and cytokine patterns could be detected in individuals with bipolar and unipolar subtypes of depression.

Methods: Acute phase proteins (alpha-1-acid-glycoprotein [AGP], AGP microheterogeneity, and cytokines (interleukin-1-beta [IL-1b ], IL-2, IL-6 and soluble IL-2 receptor [sIL-2R]) concentrations were measured in sera from 86 bipolar I (BPI), 22 bipolar II (BPII), 51 unipolar major depressive disorder (MDD) patients, and 13 healthy controls.

Results: AGP was increased in BPI compared to MDD patients (2.185 ” 0.503 vs. 1.995 ” 0.598, p =0.047), although neither patient group differed from controls. sIL-2R was increased in BPI vs. MDD patients (706.930 ” 298.621 vs. 512.667 ” 206.118, p , 0.0001) while IL-1b was lower in BPI vs. MDD patients (3.579 ” 12.909 vs. 10.314 ” 28.490, p =0.043). sIL-2R values were slightly higher in BPI vs. control subjects (706.930 ” 298.621 vs. 556.846 ” 134.237, p =0.060). No other differences were observed for IL-6 or IL-2 levels. There were no significant effects of age or gender.

Conclusions: Elevations in one or more acute phase immune proteins may help to differentiate BPI from unipolar MDD patients. These data further suggest a possible immune-priming etiology for manic-depression. Prospective studies assessing the influence of variables such as current mood state, severity of depression, age of onset, medication history, viral antigen exposure, and presence or absence of autoimmune disorders on circulating levels of acute phase proteins and cytokines are needed to amplify our understanding of this interesting finding.

R.F. Itzhaki, D. Shang, W-R. Lin
University of Manchester Institute of Science and Technology
Most cases of Alzheimer’s disease (AD) are sporadic. The gene for apolipoprotein E (apoE) type 4 is a risk factor but is neither essential nor sufficient to cause the disease. Environmental agents must presumably be involved too [1]. One possible factor is herpes simplex virus type 1 (HSV1). Nearly all adults harbour latent HSV1 in the PNS. The virus reactivates periodically, and in some people causes cold sores. As to the CNS, there was previously complete uncertainty about the presence of HSV1. We have used polymerase chain reaction (PCR) to search for HSV1 DNA in post mortem brain, taking great precautions against artefacts [2-4]. We have detected latent HSV1 in brain of 78% of 36 AD patients and 64% of 36 age-matched normals. We suggested that reactivation of the virus occurs in the CNS as a result of stress or immunosuppression [5], causing greater damage in AD because of a difference in host or virus factors. We have compared a host factor – the apoE genotypes of the patients and the normals. We found the E4 allele frequency of HSV-positive ADs to be much greater than that of HSV-negative ADs, HSV-positive and negative normals (p<0.0001), and than the maximum reported frequency for ADs (p<0.01). The results suggest that the combination of HSV1-positivity in brain and possession of an apoE4 allele confers a greater risk of developing AD than does either factor alone [6]. Parallel studies on the PNS independently and strongly support the hypothesis of the damaging effects of the two factors together [6].

1. Itzhaki, R.F. (1994) Molec Neurobiol 9, 1-13.

2. Jamieson, G.A., Itzhaki, R.F. et al., (1991) J. Med. Virol. 33, 224-227.

3. Jamieson, G.A., Itzhaki, R.F. et al., (1992) J. Pathol. 167, 365-368.

4. Itzhaki, R.F. et al., (1993) in Alzheimer’s Disease: Adv. in Clin. & Basic Res. (Corain, B. et al., eds.), 97-102, Wiley.

5. Saldanha, J., Itzhaki, R.F. et al., (1986) J. Neurol. Neurosurg. Psych. 49, 613-619.

6. Lin, W-R, Itzhaki, R.F. et al., (1995) Biochem. Soc. Trans. 23, 594S, & Molec. Chem. Neuropath. In Press.

AUTHORS: Naderah Jafari, E. Fuller Torrey, Robert H. Yolken
AFFILIATION: Stanley Neurovirology Laboratory, Johns Hopkins Univ. School of Medicine, Dept. of Pediatrics, 600 N. Wolfe St., Blalock 1111, Baltimore, MD 21287-4933
TEXT:Background: Schizophrenia is thought to be a disease associated with specific alterations in cerebral structure and function. However, differences in genomic DNA between the brain and other organs of an individual have not been previously characterized.

Method: Genomic DNA from the liver of obtained post-mortem from a schizophrenic patient and five different parts of the brain from the same individual were extracted and digested with Hind III, Bgl II, and Bam HI. Representational difference analysis (RDA) was used as a subtractive hybridization technique to purify restriction endonuclease fragments present in brain and not in hybridization technique to purify restriction endonuclease fragments present in brain and not in liver. Amplified difference bands were cloned and sequenced. Genomic DNAs and cDNAs prepared from DNA-free total RNA were screened by sequence-specific primer PCR.

Results: One, two and six amplified, difference bands were produced from Bam HI, Bgl II, and Hind III respectively. Sequencing of these bands, as well as others obtained from the cloning of the products of the RDA reactions, identified a number of clones homologous to known mammalian and viral sequences. Analysis of genomic DNA indicated that these sequences were present in both the brain and the liver of the donor individual. However, analysis of DNA-free RNA obtained from the same regions indicated that several of these sequences were differentially expressed in the brain but not in the liver.

Conclusions: RDA techniques employing DNA can identify organ-specific RNA expression in humans, perhaps as a result of the detection of small differences in DNA quantity. This identification of these RNAs may result in the discovery of new genes and viruses associated with schizophrenia.

AUTHORS: N Johnston, EF Torrey, RH Yolken
AFFILIATION: The Stanley Neurovirology Lab., The Johns Hopkins Univ. School of Med., 600 N. Wolfe St., Blalock 1111, Balto., MD 21287-4933, USA
TEXT:Schizophrenia and Bipolar Disease are associated with altered cerebral structure and function. In order to identify novel disease-associated brain transcripts, we isolate RNA from the visual cortex region of brains obtained post-mortem from affected individuals and controls. We reverse transcribed the RNAs and prepared libraries in phagemid vectors. Individual clones were purified, sequenced, and compared to known RNA species by means of database analyses.

Analysis of the library generated from the brains of individuals with schizophrenia identified RNAs similar to the gag, env, and rev genes of human retroviruses. We also identified a unique transcript in which the pseudogenomic RNA encoding the interleukin-9 receptor is interrupted by a retrotransposon. Analysis of the library from the bipolar brains revealed multiple clones highly homologous to the AF-17 protein. This protein is unique in that it contains both leucine-zipper and zinc-finger motifs and its coding region is involved in genomic translocation. Further analysis indicated that a portion of the AF-17 RNA in bipolar brains may arise by translocation or differential splicing. Our studies provide evidence for specific expression of RNAs encoding viral, immune response, and DNA-binding proteins in the brains of individuals with schizophrenia and bipolar disease.

AUTHORS: G. I. Kolyaskina, T. P. Sekirina, L. V. Androsova, S. G. Kushner, E. F. Vasilieva, O. A. Burbaeva, M. Ya. Tsutsulkovskaya
AFFILIATION: National Mental Health Research Center, Russian Academy of Medical Sciences, Moscow, Russia
TEXT:The research done over a number of years on the problem of immunity in psychiatric disorders, such as schizophrenia and bipolar disorders, made it possible to determine several peculiarities of the immune system:

# the change in the number of populations and subpopulations of blood lymphocytes

# the disturbance of the functional activity of T- and B-lymphocytes (proliferative activity on mitogen stimulation, immunoglobulin secretion by B-lymphocytes, interleukin-1 and interleukin-2 production)

# the formation of immune complexes

# the existence of autoantibodies.

It should be stressed that similar peculiarities in immunity are revealed in the autoimmune diseases or conditions accompanied by the development of an autoimmune component, and also in cases of a persisting virus infection.

The reasons for the appearance of immunological disturbances in psychiatric disorders are still far from being understood. They could be interpreted by many factors: lymphothropic activity of viruses, breakage of blood-brain barrier, hereditary deficiency in the functioning of immunologic competent cells and others. An attempt will be made to discuss our results confirming or refuting the above mechanisms of the immunological alteration development in psychiatric disorders.

Krister Kristensson
Dept. of Neuroscience, Karolinska Institute, Stockholm, Sweden
Certain epidemiological data indicate that infections may be involved in the pathogenesis of schizophrenia and bipolar disorders, but attempts to find an infectious agent have failed. However, i/ profound behavioral disturbances have been repeatedly described as a major or sole symptom following certain virus infections in both humans and experimental animals; ii/ through spread within neuronal connectivities a number of viruses have access to areas in the brain, which have repeatedly been implicated in the pathogenesis of these diseases, such as the medial temporal lobe (hippocampus, amygdala and rhinal cortex) and limbic circuits (hypothamamic and monoaminergic ones); and iii/ transient or latent viral infections may cause persistent or intermittent perturbations in the neuronal machinery, which due to intervening secondary events could progress to account for profound dysregulations of brain cell subsets. In addition, new pathogenic mechanisms by which infectious agents may alter the function of the brain and cause behavior disturbances are beginning to become unravelled. This presentation will focus on such mechanisms that involve both direct effects of defined virus components on neurotransmission and indirect effects related to a newly discovered cytokine-like regulatory molecule in the nervous system. These include: i/ how different viruses may selectively spread along a natural pathway of infection to the limbic and monoaminergic circuits and cause behavior disturbances; ii/ how components of viruses may reach pre- or postsynaptic sites of hippocampal neurons; iii/ how infections may interfere with release and bioactivities of an interferon-gamma-like molecule expressed in the tuberomammillary nuclei of the hypothalamus, which project to the hippocampus and suprachiasmatic nuclei. the latter serves as a pacemaker for biological rhythms including circadian sleep/wakefulness cycles and data that show how this function can be indirectly dysregulated by an extracellular infectious agent (the trypanosome of African sleeping sickness) will be given.

AUTHORS: H. Kulaga, J. Apud, A. Adams, M.A.Coggiano, D. Venable, and R.J. Wyatt
AFFILIATION: Neuropsychiatry branch, National Institute of Mental Health, St. Elizabeth’s Hospital, Washington, D.C. 20032
TEXT:The CD34+ cell fraction of bone marrow and blood contains hematopoietic stem cells which are essential for a functional immune system. Since a variety of drugs cause depletion of this compartment and clozapine induces agranulocytosis in some individuals, we conducted the following studies. Flow cytometric analysis using reagents which identify both noncommitted and lineage-committed stem cells were performed on peripheral blood. Subject groups included schizophrenic inpatients currently on clozapine, patients receiving other antipsychotic medications, and normal volunteers. The data indicate that peripheral blood CD34+ cells are significantly reduced in patients receiving clozapine. The details of this effect will be discussed.

AUTHORS: Jeng-Yang Ling1, William G. Stroop2
AFFILIATION: 1Div. of Molecular Virology, Baylor College of Med., Houston, TX; 2Dept. of Microbiology and Immunology, Univ. Arkansas for Med. Sci., Little Rock, AR
TEXT:Herpes simplex viruses, type 1 (HSV-1) and 2 (HSV-2), are double-stranded, enveloped DNA viruses that can cause very serious or fatal central nervous system (CNS) infections. HSV-1 is the most common cause of sporadic fatal encephalitis and the leading infectious cause of corneal blindness in human adults in the United States. Experimental studies have shown that during the initial infection, HSV-1 infects nerve ending at the site of infection and travels centripetally to the appropriate ganglionic or CNS center and different strains of HSV-1 can infect different anatomical sites of the CNS in humans as well as experimental animals.

During acute infection, an HSV-1 strain, termed +GC, localizes to several unique areas of the rabbit CNS after intranasal inoculation. Some of these areas are believed to be involved in seizure production. In order to determine the viral genetic elements in +GC which allow it to home to these CNS centers, we used intertypic recombination to create recombinant viruses between +GC and a relatively benign strain of HSV-2. A recombinant virus, R1, was recovered from brain homogenate of a mouse infected with the virus produced from cultures of vero cells initially co-transfected with EcoRI digested whole genomic strain +GC HSV-1 DNA plus intact genomic HSV-2 DNA. Because parental +GC causes syncytia in vitro, recombinant viruses were first selected for their syncytiagenic potential. A syncytiagenic recombinant virus, termed R1, was isolated and tested for its neuroinvasive potential in mice. R1 was neurovirulent following intraperitoneal inoculation like the +GC parent, and unlike the HG52p18 parent. When New Zealand white rabbits were intranasally infected with R1, it caused seizures and produced lesions in the same CNS regions as the +GC parent. Thus, R1 possesses nearly all of the neurovirulent and neurotropic characteristics as the parental +GC virus.

To determine what genetic regions of the +GC virus were present in the R1 recombinant, restriction mapping studies were undertaken. R1 contains an 8.4 kb DNA fragment from +GC. This fragment contains the intact glycoprotein B(gB) gene of HSV-1. Since R1 is a syncytial virus like the +GC parent, the gB gene is very likely responsible for the syncytiagenic phenotype of +Gc. Also, since R1 is neurovirulent in mice like the +GC parent, and because R1 also infects the same CNS centers in the rabbit as the parental +GC virus, the gB protein may also be the genetic determinant of neurovirulence and neurotropism. This conclusion is strengthened by the report by Goodman and Engel (j. Virol. 65:1770-8, 1991), which suggested that mutations within the gB gene can change viral CNS tropism.

Marylou v. Solbrig, George F. Koob, and W. Ian Lipkin
University of California-Irvine (M.V.S. and W.I.L.), The Scripps Institute (G.F.K.)
Viruses have been proposed to play a role in the pathogenesis of schizophrenia, however, the mechanisms by which infection could cause the affective, cognitive and movement disorders of schizophrenia are not understood. The neurotropic RNA virus, Borna disease (BD) virus, linked to schizophrenia by serologic studies, causes movement and behavior disorders in a wide variety of mammalian and bird hosts. BD rats have hyperactivity and stereotyped behaviors similar to those that follow neurotoxic or electrolytic lesions in frontal cortex or its catecholamine afferents. BD rats have hyperactivity and stereotyped behaviors similar to those that follow neurotoxic or electrolytic lesions in frontal cortex or its catecholamine afferents. BD rats have high levels of viral nucleic acid in the prefrontal cortex (PFC), abnormal mesocortical dopamine activity (elevated levels of DOPAC in PFC), yet no alterations in specific binding of D1 or D2 receptor radioligands in PFC. Since frontal lobe dysfunction is frequently reported in schizophrenia, the BD rat model may provide insights into pathogenesis and management of this debilitating psychiatric disease.

Ram Rao Malesu, K. McKenzie, R. Murray
Institute of Psychiatry, Genetics Section, London, (U.K.)
Previous research has indicated that schizophrenic patients are particularly likely to have been born during the winter months. It has been suggested that viral infections, more prevalent in the winter months, might increase pregnancy and birth complications (Machon et al, 1987); or that a chronic infectious process (of viral etiology) of the central nervous system occurring either in utero or at any time up to the age of onset is one of the causes that is considered to be responsible for this excess of winter births of schizophrenic patients (Torrey, 1988). This is also true in south africa where the seasons are reversed. We studied the seasonality effect of schizophrenia in a group of 466 schizophrenic patients born between the years 1915-1976 and treated in the only psychiatric hospital in Barbados. Barbados is a tropical island situated at 13o08’N and 59o36’W, has no pronounced winters (average monthly temperature = 78.8oF std. dev. of 1.7) with a population which is predominantly of African descent. We searched for seasonal/quarterly and monthly trends using total population as controls. We used Chi Square Test, Edward’s Test and GLIM program to analyze the data. We did not find any significant difference in the monthly or quarterly birth patterns of schizophrenics when compared to the general population. Our findings may suggest that lack of seasonality in barbados has resulted in a more uniform monthly distribution of schizophrenic births.

AUTHORS: McClure MO, Schulz TF, Aguzzi A, Chrystie I, Weiss RA.
AFFILIATION: St. Mary’s Hospital Medical School/Imperial College
TEXT:Spumaviruses, or foamy, viruses constitute one of three Retrovirus subfamilies, the other two being the lentiviruses (eg., HIV) and the oncoviruses (eg., HTLV). We have isolated a new foamy virus from blood samples taken from two apparently healthy orang-utans (Pongo pygmaeus). The older orang-utan has died with encephalopathy after a brief acute illness while the younger one, his grandson, remains well. Pathological changes and virus expression were detected in the brain, particularly in the cerebellum. The symptoms were strikingly similar to those described in transgenic mice carrying the human foamy virus (HFV) genome. (Bothe K, Aguzzi A, Lassman H, Rethwilm A & Horak I. Science 1991, 253; 555-557). Foamy virus infection of some primates might therefore on occasion lead to neurological disease, since one of the two known human foamy viruses originated from a patient who died of encephalopathy. (Cameron KR, Birchall SM, & Moses MA. Lancet ii 1978, ii; 796.) The characterization of foamy viruses associated with neurological disease merits attention.

AUTHORS: Michael B.A. Oldstone, M.D.
AFFILIATION: Head, Viral-Immunobiology Laboratory, Division of Virology, Dept. of Neuropharmacology, The Scripps Research Institute, La Jolla, California
TEXT:Realizing that viruses persist and thereby cause disease has been one of the major accomplishments in virology. Viruses that persist must, first, remain within a cell for a prolonged time without disturbing the transcription or translation of genes necessary for the infected cell’s survival or altering lysozomal or plasma membranes or cytoskeletal structures. Second, such viruses must avoid recognition by the host’s immune surveillance system. Viruses can accomplish this task by interfering with antigen presentation, MHC restriction, CTL activation and/or CTL activity. Highly differentiated cells within the host that do not have the potential for regeneration, i.e., neurons, have evolved special strategies to prevent antigen presentation and MHC restriction of viral peptides on their plasma membrane surfaces, thus rendering them invisible to killer T cells. Neurons have a defect in transcription of the % or heavy chain of MHC and of molecules that translocate peptides (TAP proteins) from the cytosol to the endoplasmic reticulum compartment where MHC molecules are synthesized. However, the host cell pays a price to ensure its survival during viral persistence. Persisting viruses can interfere subtly with cells’ ability to produce differentiated products (hormones, neural transmitters, cytokines, and immunoglobulins) without disrupting the cells’ viral functions (respiratory enzymes, cytoskeletal proteins, lysozomal and plasma membrane integrity, etc.). By this means, the virus can replicate in cells that appear histologically normal by light or high-resolution electron microscopy, although the function of the cell is altered. The end result, however, is disordered homeostasis and production of disease. Because viruses disorder the function of cells without killing them, there is an opportunity to reverse the diseases they cause by the use of antiviral therapy.

AUTHORS: B.D. Pearce*, S.C. Steffensen, A.D. Paoletti, A.H. Miller*, S.J. Henriksen, M.J. Buchmeier
AFFILIATION: The Scripps Research Institute, La Jolla, CA 92037, *Emory University, School of Medicine, Atlanta, GA 30322
TEXT:While the etiology of schizophrenia remains elusive, recent advances have provided clues to the origin of this devastating disorder. Studies in humans have discovered structural damage in the hippocampal formation, and current theories hold that schizophrenia may be caused by a viral infection during neurodevelopment. Nevertheless, a further understanding of schizophrenia pathogenesis through human studies may be hampered because the disorder is not usually manifested until adolescence, and at this stage in the disease, any causative virus may be cleared, and the succession of neuropathological events which preceded the symptoms could be complete. Thus animal studies are needed to investigate mechanisms by which a viral infection occurring during brain development could lead to neuropathological and functional changes in the adult CNS without leaving specific traces of viral proteins or nucleic acids.

To investigate possible mechanisms, we infected neonatal rats i.c. with lymphocytic choriomeningitis virus (LCMV) and measured electrophysiological and neuropathological changes in the hippocampus. In rats studied at 84-102 days post infection, virus was cleared form the dentate granule cells (DGCs) as determined by plaque assays and immunostaining for viral proteins, yet cell counts of DGCs were decreased by 67.5% (internal limb) and by 74.5% (external limb) (p<0.002). In vivo electrophysiological measures indicated that the remaining DGCs were receiving enhanced excitatory input, perhaps due to an early loss of GABAergic inhibitory interneurons. Immunohistochemical studies revealed a 72% loss of dentate inhibitory interneurons staining for parvalbumin (p<0.05). An early loss of inhibition from this subset of GABAergic interneurons could explain both the increase in excitatory responses, and the death of dentate granule cells. These data suggest the possibility of a novel neuropathogenic cascade involving an early virus-induced loss of inhibitory interneurons and a consequential unleashing of excitotoxic synapses on DGCs.

V. RAVI, Abdur Rakib, D.N. Gangadhar and N. Janakiramiah
Depts. of Neurovirology and Psychiatry, NIMHNS, Bangalore, India
A number of studies have been conducted on the possible viral etiology of Schizophrenia. These studies have yielded variable results largely due to methodological differences. In this study we have investigated the occurrence of common neurotropic viral infections in 30 schizophrenic patients (22 males and 8 females; mean age 28 yrs; mean duration of illness 53 months) and an equal number of control subjects. Half of the schizophrenic patients were in the life-time first episode of illness and were drug naive, the remaining were in a relapse or had an exacerbation of psychotic symptoms. these too were off neuroleptic drugs for at least 3 months. except for the duration of the illness the drug naive patients did not differ from the previously medicated group in terms of demographic variables or psychopathology. The control subjects were age and sex matched individuals attending a general hospital and undergoing spinal anesthesia for minor surgical ailments, with no present or past history of psychotic illness. Serum and CSF samples from the patients and controls were tested for the presence of IgM antibodies by using an ELISA to eight common neurotropic viruses viz, Measles, Mumps, Rubella, Herpes simplex type 1 (MSV-1), Cytomegalovirus (CMV), Varicella zoster virus (VZV), Japanese encephalitiv virus (JEV) and IgG antibodies to Human Immunodeficiency Viruses 1 and 2 (HIV). Additionally, IgM antibodies to Toxoplasma as well as autoantibodies to two neural antigens myalin basic protein (MBP) and neurofilament proteins (NFP) were tested using an ELISA.

Close to half of the schizophrenics (47%) had serum anti-viral IgM antibodies indicating a recent infection and a majority of them (58%) being positive for Mumps virus followed by CMV (16%) and JEV (10%). In contrast, only 23% of the controls had IgM antibodies in the serum. Although this difference did not reach levels of statistical significance, recent viral infections were twice as common in schizophrenics as compared to controls. It was also observed that 10% of patients were positive for anti-MBP antibodies in the CSF while none of the controls were positive and there was good correlation between the presence of antibodies to MBP and IgM antibodies to CMV.

Janusz K. Rybakowski1 and Jay D. Amsterdam2
1Department of Adult Psychiatry, University of Medical Sciences 2Depression Research Unit, Dept. of Psychiatry University of Pennsylvania Med. Ctr., 3600 Market St. 8th floor, Philadelphia, PA 19104-2649
Previous retrospective epidemiological study on American and Polish populations of affective patients showed a significant reduction of the recurrences of labial herpes during lithium prophylaxis (Rybakowski & Amsterdam, 1991). All 28 patients of Polish group who had recurrent labial herpes prior to starting lithium, were followed-up for the period 1987-94. There were 9 male and 19 female, aged 31-67 (mean 9 years), who had received lithium for 2-16 years (mean 9 years). Two female patients dropped out from the study in 1992. One died of circulatory insufficiency at the age of 70, and in another, aged 67 years, lithium was discontinued on account of severe polyuria. The quality of the prophylactic effect of lithium on affective and labial herpes recurrences in two time periods: I (before 1987), and II (1987-1994), is illustrated in the table below. Good response to lithium was defined as an absence of recurrences throughout the study period and partial response as a decrease of their frequency and insensity. The table shows numbers and percentages of patients with given effect of lithium.Lithium Affective Recurrences Labial Herpes Recurrences

Effect I II I II

good 10 (35.7%) 10 (38.5%) 13 (46.4%) 21 (80.8%)

partial 10 (35.7%) 11 (42.3%) 7 (25.0%) 3 (11.5%)

poor 8 (28.6%) 5 (19.2%) 8 (28.6%) 2 (7.7%)

Analysis by Bartlett’s test showed a significant relationship in individual patients between the magnitude of lithium effect on the reduction of affective and that of herpes recurrences.

The results show that lithium effect on affective recurrences in bipolar patients is pretty stable over time. On the other hand, the effect of lithium on labial herpes is more pronounced with longer duration of lithium administration. In individual patients, the quality of prophylactic lithium effect on affective recurrences may parallel such effect on the recurrences of labial herpes.

AUTHORS: Yoshitatsu Sei1*, Luke Whitesell2, Yelena Kustova1, Ian A. Paul1, Herbert C. Morse III3, Phil Skolnick1, and Anthony S. Basile1
AFFILIATION: 1Laborabory of Neuroscience, NIDDK, 2Tumor Cell Biology Section, Clinical Pharmacology Branch, NCI, 3Laboratory of Immunopathology, NIAID.
TEXT:Mice infected with the replication defective virus (BM5def) in the LP-BM5 murine leukemia virus (MuLV) mixture develop an immune deficiency syndrome and an encephalopathy characterized by impaired spatial learning and memory as demonstrate in the Morris Water maze. However, the molecular mechanism(s) of this cognitive deficit remains unknown. Here we report that brain fyn protein tyrosine kinase (PTK), which has been proposed to be involved in spatial learning and memory, was unresponsive to glutamatergic stimulation in mice with MAIDS. Thus, glutamate (100m m) significantly increased fyn kinase activity within 90 seconds in hippocampal slices from control mice, attaining maximal levels by 5 minutes. Fn autophosphorylation paralleled the increase in kinase activity and was maximal within 5 minutes after glutamate application. These changes in fyn kinase appeared to result from increases in specific activity, since the amount of fyn in hippocampal lysates remained unchanged and tyrosine phosphorylation of a 59 kDa protein was elevated in parallel with fyn kinase activity. In contrast to normal mice, glutamate-induced increases in fyn PTK activity were significantly impaired in mice infected with LP-BM5 MuLV for 12 weeks, a time when spatial learning and memory deficits are clearly significantly elevated in hippocampal slices from LP-BM5 MuLV-infected mice with MAIDS showed a heterogenous expression of fyn in the soma, with less intense immunoreactivity in the apical dendrites of a subset of CA1 to CA4 pyramidal neurons while fyn in control mice was homogenously expressed in pyramidal neurons throughout the hippocampal CA1 to CA4 regions, with an even distribution throughout the cell bodies and apical dendrites. These findings suggest that virus-associated disruption of fyn kinase-mediated signalling contributes to spatial memory deficits observed in mice with MAIDS, and possibly the cognitive abnormalities and personality changes in patients infected with HIV-1.
A. Sluzewska 1, J.K. Rybakowski 1, M. Sobieska 2, J.D. Amsterdam 3
1 Department of Adult Psychiatry and 2 Department of Immunology, University of Medical Sciences in Poznan, Poland, 3 Depression Research Unit, University of Pennsylvania, Philadelphia, USA

There is now string evidence that mood disorders and schizophrenia may be accompanied by abnormal immune reactivity. 

We have studied changes in plasma concentration of three positive acute phase proteins (apps): C-reactive protein (CRP), alpha-1-acid glycoprotein (AGP) and alpha-1-antichymotrypsin (ACT) and major micro-heterogeneity of AGP and ACT in 81 major depressed (MD) inpatients and 16 inpatient schizophrenics from Department of Adult Psychiatry in Poznan in 60 MD outpatients from Depression Research Unit from Philadelphia. Diagnosis was assessed according to DSM IV and ICD-10 criteria. All patients from Philadelphia fulfilled criteria for major depression recurrent. Between affective patients from Poznan 66 were as well diagnosed, as major depressive recurrent and 15 as bipolar disorder I. Mean age of inpatients with

MD was 43.9 ” 9.3 and of schizophrenics 30 ” 8 and of MD outpatients 39.5 ” 10.6. Gender of MD patients 88 women and 53 men and of schizophrenics 8 women and 8 men.

Subjects were drug free for at least 7 days before blood sampling. Plasma concentration of apps as well as clinical status: HDRS and PANSS were performed during acute episode, before treatment was started.

Concentration of CRP, AGP and ACT was measured by rocket immunoelectrophoresis and reactivity coefficient (RC) of their microheterogeneity by crossed-affinity immunoelectrophoresis (CAIE) with free concanavaline A as a ligand. CAIE reveals four microheterogeneity variants of AGP and ACT.

Concentrations of AGP and ACT were elevated in MD from both centres. There were no differences in the concentration of apps between MD and bipolar disorder patients. All schizophrenics had AGP values in the range of health controls and decreased values of ACT and CRP.

In 75% of MD, both inpatients and outpatients, we have observed two types changes in glycosylation of AGP and ACT. One group (44 patients) with 1 type of gycosylation: consisted of high AGP. ACT and high values of AGP-RC and ACT-RC and second group (53 patients) with II type of gylcosylation consisted of high AGP, ACT and low values of AGP-RC and ACT-RC.

All schizophrenics patients had decreased values of AGP-RC and values of ACT-RC in the range of healthy controls.

Chances in apps observed in MD may suggest the existence of immune activation (acute inflammatory process, expressed in elevated levels of apps and increased values of RC) in these patients.

The results obtained in schizophrenic patients may suggest that there is no evidence of acute inflammatory process but the chronic immunological process.

AUTHORS: Natasha Broude and Cassandra L. Smith
AFFILIATION: Center for Advanced Biotechnology and Departments of Biomedical Engineering, Boston University, 36 Cummington Street, Boston, MA 02215
TEXT:Conventional analysis of candidate DNA sequences in human diseases focuses on positional cloning approaches where candidate regions of the genome are identified by genetic experiments. Then, molecular experiments are used to isolate and test individual candidates. This approach is very powerful, but quite laborious because large numbers of DNA sequences must be individually analyzed. Another approach to identifying genetic contributors to disease uses the principles of parallel processing to analyze a large number of candidate DNA sequences simultaneously. We have developed a robust method that isolates and analyzes genome subsets likely to contain contributors to human disease phenotypes. The method has been applied first to the isolation of DNA restriction fragments containing trinucleotide repeating sequences. These restriction fragments, which originate from many locations in the genome, are captured by hybridization to an immobilized probe containing a DNA sequence complementary to the trinucleotide repeating sequence. The captured restriction fragments are released from the probe and displayed by size on an high resolution automatic fluorescent DNA sequencing instrument. The degree of similarity of the displayed restriction fragments correlates with the relatedness of the individual DNAs being compared. This method was validated by demonstrating its ability to display Huntington’s Disease allele size differences. The method is now being used to compare genomic samples from monozygotic twins discordant for Schizophrenia. This comparison minimizes genomic differentiation unrelated to the schizophrenia. This method also detects insertions (e.g. viruses) or deletions in the displayed restriction fragments as well as changes in DNA methylation that might lead to changes in gene expression. The method can be used with any interspersed genome repeat to display different genome subsets and can be used to display suspect restriction fragments from candidate genomic regions identified by other methods. This method also allows pooling of genome data from schizophrenic and non-schizophrenic populations, an alternative method for identifying important molecular differences between these groups.

AUTHORS: Martin Schwemmle, Axel Pagenstecher*, Wiebke Hallensleben, Hanns Ludwig#, Lothar Stitzx , Benedikt Volk*, and Peter Staeheli
AFFILIATION: Dept. of Virology and *Inst. of Neuropathology, University of Freiburg; #Inst. of Virology, Free University of Berlin; x Inst. of Virology, Univ. of Giessen, Germany
TEXT:Borna disease virus (BDV) causes lethal encephalitis in horses and some other natural animal hosts, and it is believed to be associated with psychiatric disorders in humans. BDV establishes persistent infections in the brains of experimentally infected rats and mice. While the persistent infections in the brains of experimentally infected rats and mice. While the immunopathology in rats is severe, mice were reported to be naturally resistant to BDV-induced neurological disease. We now found that BDV frequently caused disease and death when injected into the brains of 2-day-old BALB/c or 129 mice. Disease symptoms developed 2 to 5 weeks post infection, and included hyperactivity and timidity followed by apathic behaviour, impaired coordination of hind leg movement, sloping position of head and body, and bowed position of the back. Histological analysis of the brains of sick mice revealed severe meningitis and encephalitis. Animals with mild or severe neurological disease symptoms both showed extensive destruction of neurons in the CA3 region of the hippocampus. By contrast, only animals with severe disease showed high levels of viral antigen and neuron destruction in the brain stem. Neuron destruction in hippocampus and brain stem occurred in the absence of noticeable infiltration of inflammatory cells into these brain regions. These results demonstrate that the mouse model system can be used to study BDV-induced brain pathology. As mouse strains with defects in various branches of the immune system have recently become available, a more rapid progress in the understanding of the neuropathological processes involved in Borna disease can now be expected.

Janice Stevens* and Ingrid Phillips**
*Oregon Health Sciences University, Portland, OR 97201 **National Institute of Mental Health, St. Eliz. Hosp., Washington, DC 20032
Experiments were undertaken to determine whether cerebrospinal fluid (CSF) with or without CSF lymphocytes from schizophrenic patients contains a transmissable factor that could be expressed in lethality for or brain pathology in experimental animals. Under cold anesthesia, 105 newborns C57B1/6J mice were injected intracerebrally with 1-10ul fresh or frozen CSF. Half the pups in each litter were inoculated with CSF from a schizophrenic patient (Sch CSF), half from a control subject or with artificial CSF. Seven pups were inoculated with cell free media from a neuroblastoma cell line that was transformed following incubation with Sch CSF and six pups were inoculated with cell free media from the same cell line that was not exposed to CSF. Twelve animals served as anesthesia controls. All CSFs and mice were coded and investigators were blind to type of CSF inoculated throughout subsequent behavior observations and final pathologic examinations. Many pups disappeared from their cages in the first month of the experiment, apparently cannibalized by the mother. No abnormal behaviors were noted in surviving pups. Only 4 of 12 pups inoculated with fresh Sch CSF, 25 of 39 (64.1%) inoculated with frozen control CSF and 72% of mice inoculated with artificial CSF. Differences in survival between mice inoculated with fresh and frozen CSF are significant (chi square: p<.05). Dilated ventricles were found in 75% of animals that survived 1 year after fresh Sch CSF, and 16% of frozen control CSF inoculation. However, 50% of animals subjected only to cold anesthesia also had dilated ventricles at autopsy 1 year later. No correlations could be made between mortality, behavior or neuropathology and the type of CSF inoculated.

Lothar Stitz and Thomas Bilzer
Institut für Virologie, Justus-Liebig-Universität Giessen, and Institut für Neuropathologie, Heinrich-Heine-Universität, Düsseldorf, Germany
Borna Disease (BD) is a virus-induced immunopathological disease of the central nervous system in a variety of species from birds to primates. Recent work from other groups indicate that the single-stranded RNA BD virus may also infect humans and may be involved in neuropsychiatric disease. Antiviral antibodies and virus-specific genomic transcripts have been detected in human patients with various neurological disorders, including schizophrenia. In the experimental rat model severe inflammation and degenerative brain cell lesions can be found, resulting in organ atrophy and chronic debility. Clinically, somnolence and signs of chronic debility can be observed. Immunohistological analysis of the brains of rats with BD uniformly revealed the presence of CD8+ T cells in encephalitic lesions as well as CD4+ T cells and macrophages. CD8+ T cells can be found in the brain parenchyma and MHC class I antigens are expressed on BDV infected neurons whereas CD4+ T cells predominate perivascularly. Treatment with anti-CD8 antibodies inhibits the immunopathological reaction, neuronal lesions are minimal and no loss of brain substance is found. In vitro testing of T cells isolated from the brain reveals the presence of MHC class I-restricted cytotoxic activity. Adoptive transfer of a virus-specific CD4+ T cell line results in typical BD, associated with neurological symptoms. However, recipient rats treated with antibodies against CD8+ T cells neither develop encephalitis nor disease after transfer. CD4+ T cells appear to accumulate in the brain and cause perivascular inflammatory lesions, which alone obviously do not cause disease. In contrast, the presence of CD8+ cells in the brain parenchyma appears to directly correlate with the upregulation of perforin mRNA and the development of neurological symptoms. Experimental BD provides a promising in vivo model for the study of immune-mediated tissue destruction resulting in brain atrophy and functional abnormalities such as dementia and chronic debility.

Martha M. Coetzee and Felix Strumwasser
Uniformed Services University of the Health Sciences
Platelets and lymphocytes are used as peripheral markers in the study of depression. There are b 2-adrenergic and corticosteroid receptors on lymphocytes and %2-adrenergic and 5HT2 receptors on platelets. Over the last six years there have been a number of papers examining intracellular Ca2+-release in the platelets and lymphocytes of patients with affective disorders. In general, there is a consensus that ligands such as serotonin or thrombin induce a larger Ca2+-release in platelets and PHA a smaller increase in lymphocytes of patients with major depression than in age- and sex-matched controls. This talk reviews what we know about Ca2+-regulation in order to examine the possible reasons for this dysregulation of Ca2+-responses in depressed patients. Platelets respond to the natural ligands serotonin/thrombin by an increase of intracellular Ca2+ which mediates the secretion of serotonin (5HT) and other compounds. Released 5HT acts on the 5-HT2 receptor to release more intracellular Ca2+, thus amplifying the actions of thrombin/serotonin. the release of [Ca2+]1 in platelets is mediated by inositol 1, 4, 5-trisphosphate (IP3) generated by activation of phospholipase C.

10 FM 5HT induced Ca2+ responses in platelets is a stable and sex/age independent parameter in control subjects. Different workers have confirmed intracellular calcium mobilization to be significantly increased in patients with major depression. When subgroups of depression were studied, significant increases in calcium responses in melancholic and bipolar depressed groups were found when compared to normal controls and unipolar depressed patients (Kusumi, 1994). The same significant changes were demonstrated in manic as well as depressed bipolar patients and were shown to be state dependent.

AUTHORS: Yeping Sun, Shuojia Li, Flora Leister, Martin Mracna, E. Fuller Torrey, and Robert H. Yolken
AFFILIATION: Stanley Neurovirology Laboratory, Johns Hopkins Univ. School of Medicine, Dept. of Pediatrics, 600 N. Wolfe St., Blalock 1111, Baltimore, MD 21287-4933
TEXT:Schizophrenia and bipolar diseases are postulated to be associated with changes in genomic DNA, occurring as a result of either genetic inheritance or viral integration. Identical twins discordant for these diseases provide the opportunity for the direct assessment of this hypothesis. We applied the technique of representational difference analysis (RDA) to analyze differences in genomic DNA obtained from the lymphocytes of 2 sets of monozygotic twins discordant for schizophrenia and 2 sets discordant for bipolar disease as well as from 1 set of control twins. Representations of the DNA were generated using three different restriction endonucleases: HindIII, BamHI, and BgIII. Candidate RDA products obtained following different rounds of subtraction and amplification were cloned and analyzed by Southern blot hybridization and nucleotide sequencing.

Initial screening resulted in the identification of sequences highly homologous to a range of known human structural and functional genes. However, further analysis indicated that all of these target genes were equally present in the genomic DNA of both the affected and unaffected twin. The results of these studies thus do not support the hypothesis that the discordance of schizophrenia or analyses using additional restriction enzymes and subtraction strategies are ongoing.

AUTHORS: Daniel M. Tucker, M.D., James F. Leckman, M.D., and Paul Lombroso, M.D.
AFFILIATION: Yale Child Study Center, Yale University School of Medicine
TEXT:In Sydenham’s chorea, a post-streptococcal disorder, autoimmunity is thought to effect movement, speech, cognitive, and affective neuropsychiatric symptoms. These can overlap and resemble some features of Tourette’s syndrome (TS), obsessive compulsive disorder (OCD), and psychotic disorder. Observations have been reported of an association between prior streptococcal infection and TS, and OCD. Findings from our preliminary study reported here are supportive. Prophylaxis with penicillin and immune modulative treatments such as plasmapheresis suggest that post-streptococcal factors may be etiologically related to the clinical course. These formulations call for prospective, controlled study, and require suitable assay procedures.

Our preliminary study applied immunocytochemical methods to patients groups assembled openly and retrospectively. Western blot analysis investigated whether there was an interaction between patients’ sera against brain as well as M6 beta-hemolytic streptococcal protein homogenates. We have sought to: Clarify the specificity of serum-binding to central nervous system by Western blot analyses; investigate whether proteins from specific brain regions demonstrate higher immunorcactivity with patients’ sera; determine the odds ratio for elevated serum indicators for prior streptococcal infection (ASO and antistrep DNAse B) relevant to those cross-reactions; and to identify distinguishing group characteristics on Western blots between TS or OCD patients with evidence for streptococcal infection (throat culture, ASO, anti DNase B) and from controls without suspected infectious provocation.

In work supported by the Stanley Foundation, prospectively we will extend those comparative studies beyond the TS and OCD populations to include patients with schizophrenia, as well as comparison groups with pharyngitis, rheumatic fever and Sydenham’s chorea.

Grateful acknowledgement to Abel Bult, Ph.D., Debra Bessen, Ph.D., and Laurence Scahill, M.S.N., M.P.H.

AUTHORS: van Kammen DP, McAllister CG, Kelley ME, Gurklis JA, Yao JK
AFFILIATION: Department of Veteran Affairs Medical Center, 7180 Highland Drive, Pittsburgh, PA 15206
TEXT:Schizophrenia, which is a chronic but episodic disorder of unknown etiology, shares many features with organ specific autoimmune diseases such as multiple sclerosis (MS), rheumatoid arthritis or insulin-dependent diabetes mellitus. However, these similarities may be the result of a stress induced dysregulation of CNS cytokine production, rather than a permanent autoimmune disturbance. Our recent findings have shown that the cytokine interleukin-2 (IL-2) in the CSF was elevated in haloperidol treated patients who relapsed within 6 weeks following drug withdrawal (relapse prediction). In a logistic regression model of CSF monoamine data, CSF but not plasma IL-2 replaced CSF metabolites of dopamine and serotonin in the model. That CSF IL-2 rather than plasma IL-2 was predictive of relapse status and was the only remaining biological measure in the relapse prediction model suggests that the brain is the target organ. Cytokines are synthesized not only by cells involved in immune response but also by brain cells. They also play a role in neuromodulation (e.g., dopamine release). Our further studies of cytokines have included both those produced by Th1 (IL-1%, IL-6, IL-2) as well as Th2 (TGFB, Il-10) cells.

AUTHORS: 1,2Waltrip RW, 3Torrey EF, 1,2King LR, 3Weinberger DR, 3Rickler KC, 2Rubin SA, 2Carbone KM
AFFILIATION: 1Maryland Psychiatric Research Ctr., P.O. Box 21247, Baltimore, MD 21228; 2Viral Immunopathogenesis Laboratory, Johns Hopkins University, Dept. of Medicine; 3Clinical Brain Disorders Branch, Intramural Research Program, National Institute of Mental Health.
TEXT:Borna disease virus (BDV) is a neurotropic and nonlytic agent that causes an immunopathological meningoencephalitis in a rapidly expanding range of recognized natural hosts (cattle, rabbits, goats, deer, llamas, alpacas, horses, sheep, ostriches, and domestic cats). BDV infects experimental hosts ranging from avians to primates. BDV, or BDV-like agent is hypothesized to infect humans and to cause neuropsychiatric disease.

In a previous study we found anti-BDV seropositivity and both neuroanatomical abnormalities on MRI and Deficit subtype. In the present study of monozygotic twins we compared anti-BDV seropositive rates between groups of twins with and without schizophrenia and between groups who are weighted toward environmental etiological factors (discordant for disease) and groups who are weighted toward genetic etiological factors (concordant for disease).

Using a criterion of antibodies to 2 or more BDV proteins for seropositivity (S+), we assayed anti-BDV antibodies in groups of monozygotic tin pairs who were either concordant (C) or discordant (D) for psychiatric disease. S+ cases were clustered in the twin group D for schizophrenia (Sz) S+ rates were: 1/26 C for Sz (13 pairs); 0/2 C for other psychosis (1 pair); 0/8 affected and 0/8 unaffected discordant (D) for Bipolar disorder (8 pairs); 0/16 normal (8 twin pairs); 9/25 affected and 5/25 unaffected discordant for schizophrenia (25 pairs). All of the 5 S+ unaffected subjects had S+ Sz affected cotwins. Clinical, quantitative MRI and neurological data will be presented.

P Wright, N Takei, L Rifkin, RM Murray
Psychiatry, De Crespigny Park, London, SE5 8AF, UK
We studied the reported association between prenatal influenza and adult schizophrenia in individual schizophrenic patients, and investigated any relationship to obstetric complications (OCs). Infections during pregnancy, OCs, gestational age, and birth weight, were assessed by interviewing the mothers of 121 DSM-IIIR schizophrenic patients using a new structured instrument. A significant excess of infections was reported in the 2nd compared with the combined 1st and 3rd, trimesters of the patients’ gestations (p=0.005). Influenza accounted for 70% of 2nd trimester infections (p=0.004). Schizophrenic patients whose mothers suffered influenza during the 2nd trimester were almost 5 times more likely to experience at least one definite OC (p=0.01), and weighed 210 g less at birth (p=0.2). We propose that 2nd trimester maternal influenza may perturb foetal development, and may thus predispose to OCs and reduced birth weight, in a proportion of schizophrenic individuals. 

 Gestational Infections Reported by the Mothers of 121 Schizophrenic Patients


Trimester of Pregnancy


p value




2nd vs 1st+3rd


14 (11.6%)


16 (13.2%)


resp. infection


















candidiasis (oral)












dental abscess






AUTHORS: F Yee, EF Torrey, RH Yolken
AFFILIATION: The Stanley Neurovirology Lab., The Johns Hopkins Univ. School of Med., 600 N. Wolfe St., Blalock 1111, Balto., MD 21287-4933, USA
TEXT:Previous studies have identified RNAs that are expressed in the brains of individuals with schizophrenia. However, RNA species differentially expressed in the brains of individuals with schizophrenia have not been previously characterized in a systematic manner. we have adapted differential display polymerase chain reaction using arbitrary primers (AP-PCR) to identify RNAs expressed in cortical regions of brains obtained post-mortem from individuals with schizophrenia and controls.

Several RNAs have been identified in the frontal and occipital cortex of brains from individuals with schizophrenia which are differentially expressed as compared to controls. One of these RNAs is highly homologous (>98% predicted amino acid similarity) to the DB1 zinc finger protein, which interacts with the tax protein of human T-cell leukemia virus (HTLV-1) to increase interleukin-3 transcription. Another differentially expressed RNA has 99% homology to the RIZ zinc finger protein, which binds to the RB oncogene and shares an epitope with the analogous E1A protein of human adenovirus. A third RNA displays 77% homology to the reverse transcriptase of SRV-2, a primate type-D retrovirus. Our studies indicate that viral and virus associated RNAs are expressed in the brains of some individuals with schizophrenia and may play a major role in disease pathogenesis.

RH Yolken, F Yee, N Johnston, F Leister, L Bobo, N Jafari, C Ojeh, EF Torrey
Stanley Neurovirology Laboratory, Johns Hopkins Univ. School of Medicine, Dept. of Pediatrics, 600 N. Wolfe St., Blalock 1111, Baltimore, MD 21287-4933
Epidemiological studies have indicated that viral infections may play a role in many cases of schizophrenia. the recent availability of well preserved post-mortem brain samples from diseased individuals and controls has allowed us to apply molecular techniques to address the relationship between viral infections and schizophrenia. 

Nucleic Acids were extracted from defined regions of cases and controls, and analyzed by the following methods:

#RNA libraries expressed in phagemid messages.

#DNA subtractive hybridization.

#Polymerase Chain Reaction (PCR) using arbitrary primers directed at coding exons.

#PCR using primers directed at conserved viral sequences.

We have identified DNA and RNA species present in the brains of individuals with schizophrenia but absent or present in lower concentrations in analogous samples from control individuals.

These studies resulted in the identification of viral and viral-associated sequences which may be associated with disease pathogenesis. Viral sequences which are identified include those homologous to enzymatic proteins of retroviruses similar to HIV and HTLV-1 as well as other mammalian viruses. We also have identified viral-associated RNAs including ones homologous to viral-induced transcriptionally active proteins and to proteins involved in the immune response. These studies point to a central role for viral infections in the pathogenesis of schizophrenia.