Abstracts of the 5th Meeting of the

International Neurotoxicology Association

THE HUMAN GENOME PROJECT: NEW TECHNOLOGIES FOR DNA EXPLORATION, OR LOOKING FOR THE GEORGE BURNS GENOTYPE
Deborah A. Nickerson
Department of Molecular Biotechnology, University of Washington, Seattle, WA 98195
One of the most challenging problems in the study of biology and medicine is related to defining the connection between phenotype, the observable characteristics of an individual, and genotype, the DNA sequence of the individual. Over the next decade, the Human Genome Project will provide a composite sequence of the 24 different human chromosomes and will uncover the information stored in these structures, i.e. the 100,000 or so genes responsible for human structure, function, growth and development. The Human Genome Project is unique, based not only on its size and scope but also based on its dependence on the development of new technologies. Current and newly emerging technologies from this project will be presented and discussed in regard to their role in the next generation of medicine and its implications in understanding the diversity of the human phenotype.
THE LATENT ROOTS OF NEURODEGENERATIVE DISEASE
Bernard Weiss
Department of Environmental Medicine, University of Rochester School of Medicine and Dentistry, Rochester, NY
Satchel Paige, along with his extraordinary athletic accomplishments, is also remembered as a philosopher. Although no one ever thought of him as an expert on neurodegenerative disease, he was perspicacious enough to enunciate a useful guiding principle: "Don't look back; something may be gaining on you." That something is always gaining on us is certain, and we all try to elude it, but looking back is what scientists, in our arrogance, do. We all accept the Paige principle that neurodegenerative diseases begin long before they erupt into perceptible disability. We cannot always state whether the process is inherently abnormal, or an acceleration of natural, universal aging, or a confluence of the two. The puzzle of how the process advances is one of the most profound challenges to neuroscience, but many clues tell us that aging must be accorded at least a secondary role: 1. Neurobehavioral function declines with age. 2. The incidence of neurodegenerative disease shows a clear age-related acceleration. 3. Key areas of the brain lose neurons with age, restricting their ability to compensate for damage. 4. Apparent recovery from an earlier neurological disability may revert with age. Any research agenda aimed at the etiology of neurodegenerative disease must deal with both onset and process. For investigators, it means no acceptable substitute for meticulous longitudinal studies in both the clinic and the laboratory, carried out with the most sensitive instruments available.
ORGANOPHOSPHATE-INDUCED DELAYED NEUROTOXICITY (OPIDN): FACTS AND FANTASIES
S. Padilla 1 P. Tandonand 2, M. Ehrich 3 and C.N. Pope 4 1Neurotoxicity Division, U.S. Environmental Protection Agency, Research Triangle Park, NC
2 Harvard University, Boston, MA
3 Virginia Tech. University, Blacksburg, VA
4 Toxicology Program, School of Pharm., N. L.U., Monroe, LA
OPIDN is one of the few neuropathic conditions where the association between exposure to a neurotoxic compound and the development of neurodegeneration is well-recognized. Exposure to a certain subset of organophosphate anticholinesterase compounds (some are environmentally relevant) produces a central-peripheral distal axonopathy in man and many experimental animals. This axonopathy does not occur immediately after exposure; the development of the clinical signs is usually delayed approximately 14 to 21 days after an acute exposure. Although over 60 years has elapsed since the initial identification of OPIDN, there has yet to be an elucidation of the mechanism of this neuropathy. It is known that the neuropathy can be prevented or potentiated by other esterase inhibiting compounds (generally depending on the relative sequence of administration of both compounds) and that young animals are particularly insensitive to the development of the neuropathy. OPIDN will be discussed in the light of these developments as well as more recent data and hypotheses regarding the balance of neuronal degeneration and regeneration normally maintained by growth/trophic factors.
EVIDENCE FOR DELAYED METHYLMERCURY NEUROTOXICITY IN MONKEYS
D. C. Rice
Toxicology Research Division, Department of Health, Ottawa, Ontario
As a consequence of the tragic outbreaks of human organic mercury poisoning in Japan and Iraq, substantial research effort has focused on characterizing the developmental effects of exposure to methylmercury in animal models, including studies in macaque monkeys (Macaca fasicularis) at the Canadian Health Protection Branch. One group of five monkeys, presently 19 years old, was dosed with 50 µg/kg/day of mercury as methylmercuric chloride from birth to 7 years of age; blood mercury concentrations during the period of dosing were approximately 0.75ppm. In another study examining the effects of in utero plus postnatal exposure in the same species of monkey, females were dosed with 10, 25, or 50 µg/kg/day of mercury as methylmercuric chloride; blood mercury levels averaged 0.37, 0.75, or 1.42 ppm during pregnancy. One, two, and five infants were born from the three dose groups, respectively. One infant in the high-dose group was born with signs of methylmercury poisoning resembling those of human infants, including motor impairment and nystagmus. When the group of monkey exposed only postnatally until 7 years of age was 13 years old, individuals began exhibiting clumsiness not present previously. Further exploration revealed that treated monkeys required more time to retrieve treats than did non exposed monkeys and displayed abnormalities on a clinical assessment of sense of touch in hands and feet, despite the fact that clinical examinations performed routinely during the period of dosing had not yielded abnormal results. This observation was pursued in both groups of monkeys by objective assessment of somatosensory function in the hands: both groups of monkeys exhibited impaired vibration sensitivity. These results are strongly suggestive of a delayed neurotoxicity manifested when these monkeys reached middle age. These data are consistent with data in mice exposed developmentally to methylmercury (J.M. Spyker, Fed. Proc. 34, 1835, 1975), in which abnormalities including kyphosis, obesity, and severe neurological deficits were observed only as the animals aged.

NEUROTOXINS AND NEURODEGENERATIVE DISEASE
J. William Langston
During the last decade there has been increasing interest in the relationship between neurotoxins and neurodegenerative disease. There are several reasons for this. First, a number of toxins have been discovered which are remarkably selective for the same neural populations that are affected in human neurodegenerative diseases, including Parkinson's disease and Huntington's chorea. Such compounds can be used to model these diseases, an approach that has wide-ranging experimental and clinical implications. Secondly, there is the important possibility that investigating the mechanism of action of these compounds may provide insights into the mechanisms underlying the actual process of neuronal degeneration. Finally, in at least some instances, there is the provocative possibility that similar compounds in the environment might even be a cause of the disease. In this presentation, 1-methyl-4-phenyl- 1,2,3,6- tetrahydropyridine or MPTP will be discussed as the prototype of such a toxin. This compound meets all of the criteria for what might be called a "neurodegenerative neurotoxin": it induces degeneration of specific neuronal populations after systemic administration which are nearly identical to those which are affected in a neurodegenerative disease (in this case Parkinson's disease). The discovery of the biological effects of this compound, which induces degeneration of dopaminergic neurons of the substantia nigra, and a corresponding profound dopamine depletion in the striatum of human and non-human primates, has already lead to the first good animal model for the disease. But there are many other lessons that have been learned as a results of studies on the mechanism of action of MPTP as well. For example, it is now weel known that MPTP is biotransformed to its toxic metabolite, 1-methyl-4-phenypyridinium ion, via MAO B. This process appears to take place in astrocytes. It could be argued that this a precedent-setting phenomenon, in that it may be the first example of the brain's own enzymatic machinery generating a selective neurotoxin from a non-toxic compound. Since MAO B can be easily blocked pharmacologically, this observation has already lead to the development of a neuroprotective strategy for Parkinson's disease. Other aspects of MPTP neurotoxicity, such as its effects on Complex I of the mitochondrial respiratory chain, have greatly enhanced interest in the role of energy metabolism and neurodegenerative disease. These and other features of MPTP neurotoxicity, and its broader implications for neurotoxins and neurodegenerative disease, will be reviewed, along with implications for future research in this interesting and rapidly developing field.
POLYCHLORINATED BIPHENYLS DECREASE DOPAMINE AND ELEVATE GABA CONTENT IN STRIATAL SLICES FROM ADULT RATS
R.F. Seegal, M.A. Chishti and G. Battaglioli
Wadsworth Center, New York State Department of Health, Albany, NY
Polychlorinated biphenyls (PCBs) are widespread environmental neurotoxicants associated with behavioral changes in human infants, behavioral and neurochemical changes in experimental animals and reductions in dopamine (DA) content of pheochromocytoma (PC12) cells in vitro. To study the neurochemical effects of PCBs in a more complex ex-vivo system, we exposed 350-µm thick striatal slices from adult rats to varying concentrations of a 1:1 mixture of Aroclors 1254 and 1260 and measured slice content of DA and gamma-aminobutyric acid (GABA) by HPLC. PCBs reduced slice DA content in a time and dose dependent manner with a 65-80% reduction observed after 6 h at 200 ppm. Accompanying these reductions in slice DA content were significant dose-related increases in slice GABA content. Maximal elevations in slice GABA content (a doubling compared to vehicle-exposed slices) were observed at PCB concentrations in media equal to or greater than 100 ppm. GABA and DA in the basal ganglia are mutually inhibitory, complicating interpretation of the PCB-induced alterations in these neurotransmitters. Reductions in in-vivo striatal DA content by neuroleptics or 6-OH DA elevate GABA content. However, in preliminary studies, a-methyl-_-tyrosine (an inhibitor of DA synthesis) reduced slice DA content by 50% but did not alter slice GABA content. This result suggests that the elevations in GABA are a direct consequence of PCB exposure and not secondary to PCB-induced reductions in slice DA content. A cause and effect relationship, if any, between changes in GABA content and changes in DA content remains to be investigated. Supported by NIEHS Grant #ES04913.

DEVELOPMENTAL EXPOSURE TO AROCLOR 1254 CAUSES HYPOTHYROIDISM AND LOW-FREQUENCY HEARING LOSS IN RATS: ATTENUATION OF EFFECTS BY THYROXIN REPLACEMENT
E. S. Goldey, L. S. Kehn & K. M. Crofton
Heightened interest has focused on the possible role of endocrine disruption in mediating the effects of developmental neurotoxicants. The auditory system is dependent upon thyroid hormones for normal development, and we reported that developmental PCB exposure caused low-frequency hearing loss and hypothyroidism in rats. These findings lead us to suggest that PCB-induced disruption of this endocrine axis may have contributed to the observed auditory deficits. To further investigate this possibility, primiparous Long-Evans rats received daily oral doses of corn oil (control) or 8 mg/kg of Aroclor 1254 from gestation day (GD) 6 through postnatal day (PND) 21. In addition, from PND 4-21, all pups received daily, subcutaneous injections of saline or 100 µg/kg thyroxin to yield four groups of litters: corn oil plus thyroxin (CO-T4), corn oil plus saline (CO-S), Aroclor 1254 plus thyroxin (PCB-T4), and Aroclor 1254 plus saline (PCB-S). We measured thyroid hormone (T4 and T3) concentrations in serum collected from pups on PND 7, 14 and 21. On PND 7 and 21, we also monitored the kinetics of the hormones following T4 injection in the CO-T4 and PCB-T4 groups at 1, 3, 5, 8 and 24 hours post injection. On PND 7, 14 and 21, T4 levels were dramatically depleted in the PCB-S and PCB-T4 groups compared to the CO-S group. The thyroid hormone levels in the CO-T4 pups remained significantly elevated over all other groups on each sample day, although no other physical or functional alterations were seen in this group compared to CO-S. The kinetics studies revealed that thyroxin injection raised circulating hormone levels in the PCB-T pups to near CO-S levels for approximately 6 hours post-injection, and that levels fell precipitously thereafter. As in our previous studies, reflex modification audiometry revealed low-frequency (1 kHz), auditory threshold deficits in adult offspring of Aroclor 1254-exposed dams. Importantly, the auditory effects of PCBs were significantly attenuated, although not eliminated, by thyroxin replacement therapy. Further, the T3 and T4 kinetics indicate that the incomplete attenuation of the auditory deficit in the PCB-T4 group may be due to the relatively short half-life of the injected T4. These findings strongly support our earlier suggestion of a causative link between PCB-induced hypothyroidism and disruption of cochlear development and function.

PRENATAL EXPOSURE TO POLYCHLOROBIPHENYLS: PCB METABOLISM, THYROID HORMONE HOMEOSTASIS AND BRAIN DEVELOPMENT IN THE RAT
Dennis C. Morse1 , Abraham Brouwer1, Kor J. van den Berg2 , Richard F. Seegal 3
1 Department of Toxicology, Agricultural University, Wageningen, The Netherlands
2 ITV-TNO, Rijswijk, The Netherlands
3 New York State Department of Health, Albany New York
The underlying mechanisms of PCB-induced developmental neurotoxicity are unclear, although
it is a plausible hypothesis that pre- or postnatal PCB exposure indirectly affects brain development by transiently reducing the amount of thyroid hormone in the brain. We therefore examined the effects of pre- and postnatal PCB exposure on thyroid hormone levels in the plasma and brain of developing rats, the mechanisms involved in altered thyroid hormone homeostasis and which brain regions and cell types were affected. The results indicate that when pregnant rats are exposed to Aroclor 1254, there is a substantial accumulation of the hydroxylated PCB metabolite, 2,4,5,3',4'-pentachloro-4-biphenylol (4-OH-pentaCB) in the fetal plasma and brain. The accumulation of 4-OH-pentaCB in the plasma is probably responsible for the dramatic reductions in plasma and brain T4 concentrations by blocking the transport of thyroid hormone to the fetus. However, the fetal brain may be able to compensate for the decreases in T4, by increasing the conversion of T4 to T3. Despite the lack of an observed effect of maternal PCB exposure on brain T3 levels in the offspring, the levels of a neurotypic protein (synaptophysin) and a glial protein (glial fibrillary acidic protein, GFAP) as well as serotonin metabolism were altered in the brain of adult offspring in a complex fashion. The most prominent neurochemical alterations were found in the lateral olfactory tract, prefrontal cortex and the brainstem. The neurochemical data could be interpeted as the result of a primary lesion in the brainstem early in development.

EARLY DEVELOPMENTAL PCB-EXPOSURE AND NEUROLOGICAL/ NEUROBEHAVIORAL OUTCOME: AN INTERIM REPORT FROM THE GERMAN PERINATAL STUDY*
G. Winneke1, A. Bucholski1, U. Krämer1, B. Heinzow2, B. Seidel3, J. Walkowiak1, S. Weipert3, A. Wiener4, E. Schmidt3, and H.J. Steingrüber4
1 Medical Institute of Environmental Hygiene, University of Düsseldorf
2 Laboratory of Environmental Toxicology of Schleswig-Holstein, Kiel
3 Pediatric Clinic, University of Düsseldorf
4 Institute of Medical Psychology, University of Düsseldorf
In order to clarify uncertainties regarding spectrum and persistence of developmental PCB-effects, an European coordinated prospective study was initiated with study cohorts from the Netherlands, Denmark and Germany. The ongoing German study is based on PCBs 138, 153 and 180 in cordblood serum and maternal milk. Measures taken at ages 15-20 d, 7 and 18 mo include neurodevelopmental status (TOUWEN/PRECHTL) BAYLEY II, the FAGAN-test, as well as vocalization-analyses. In addition TSH-levels are measured in cordblood and at age 5 d. With PCBs ranging from 0.05-0.64 µg/g fat in serum preliminary analyses based on 103 infant- and 45 7 mo data sets do not yet suggest significant exposure/outcome associations, but are indicative of a borderline positive association between cordblood PCB and TSH-elevation, suggestive of subtle hypothyroid dysfunction.
*Partly supported by the CEC (Brussels), contract No. EV5V-CT92-0207

METALLOTHIONEIN IN ASTROCYTES: WHAT POSSIBLE SIGNIFICANCE?
M. Aschner
Department of Physiology and Pharmacology, Bowman Gray School of Medicine, Winston-Salem, NC
The biological functions of metals and their accumulation within cells are invariably linked to the existence of specific metal-binding proteins. Novel kinds of -SH based metal clusters that have recently come o attention are the metallothioneins (MTs). MTs are 6-7 kD in MW, and they contain some 60 amino acid residues, among them 20 cysteines. Recent in vivo and in vitro studies suggest that methylmercury (MeHg) neurotoxicity may be associated with astrocyte dysfunction, leading to their failure to adequately control the extracellular microenvironment. To explore the possibility that astrocytic MTs afford a protective role in heavy metal-induced neurotoxicity, we investigated the effects of MeHg on the induction of MTs, and the ability of cultures which express elevated MT levels to protect against the cytotoxic effects of MeHg. MT mRNA was detected in untreated cells, suggesting constitutive MT expression in astrocytes. Expression of MT-I mRNA in astrocyte monolayers exposed to 2 x 10-6 M MeHg for 6 hours was increased approximately 2-fold over MT-I mRNA levels in controls. Western-blot analysis revealed a time-dependent increase in MTs. Consistent with the constitutive expression of MTs both at the mRNA level and protein level, a time-dependent increase in MT-immunoreactivity was noted in MeHg-treated cells. The cytotoxic effects of MeHg were measured by the rate of astrocytic [3H]-D-aspartate uptake. Pre-exposure to CdCl2, a potent inducer of MTs, completely reversed the inhibitory effect of MeHg on [3H]-D-aspartate uptake which occurs in MeHg-treated astrocytes with constitutive MT levels. In summary, astrocytes constitutively express MTs; treatment with MeHg increases MT expression, and increased MT levels attenuate MeHg-induced toxicity. Increased MT expression may represent a generalized response to heavy metal exposure, thus protecting astrocytes, and perhaps also indirectly, juxtaposed neurons.

SCHWANN CELLS AS TARGETS OF NEUROTOXICANTS*
P. Morell and A.D. Toews
University of North Carolina, Chapel Hill, NC 27599
Schwann cells envelop axons of the peripheral nervous system. Many are further specialized to produce and maintain myelin. Schwann cells communicate with neurons both during development (e.g., through a nerve growth factor and receptor system) and even after maturation (ATP receptors on Schwann cells presumably respond to neuronal activity). Schwann cells, and the process of myelination, are preferentially vulnerable to neurotoxic insults during the developmental period of maximal accumulation of myelin. Part of the pathophysiology of various disease states, and of certain neurotoxic insults, is a direct effect on Schwann cells and/or the myelin they maintain, leading to primary segmental demyelination. Dissection of the biochemical events involved is often complicated by the fact that the metabolic insult involved also affects neurons. An exception is the neuropathy caused by feeding tellurium to young rats; a highly synchronous primary demyelination results, but cessation of tellurium exposure is followed closely by a period of rapid remyelination. The lack of detectable axonal degeneration allows examination of changes in gene expression specific to the processes of demyelination and remyelination. We have elucidated the primary metabolic lesion involved (tellurium blocks cholesterol biosynthesis) and many of the steps coupling this metabolic lesion to demyelination. A useful byproduct of the study is characterization of a sensitive marker for any neuropathy resulting in even subtle alterations in the normal relationship between the Schwann cell-myelin unit and the axon it ensheathes. Upregulation of mRNA for the low affinity nerve growth factor receptor is a sensitive marker of nerve damage which may be useful as a screen for potentially neurotoxic compounds. *Supported by USHS grants ES-01104 and NS-11615.

POTENTIAL ROLE OF GLIAL CELLS IN PARKINSON'S DISEASE
D. A. Di Monte
The Parkinson's Institute, Sunnyvale, CA
Although degeneration of dopaminergic cells of the nigrostriatal pathway represents the main neuropathologic feature of Parkinson's disease, other cells, including glial cells, may play a role in this process. A clear example supporting this concept arises from studies on the mechanism of toxicity of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP). MPTP causes a parkinsonian syndrome in humans which is clinically indistinguishable from idopathic Parkinson's disease. In order to exert its toxic effects, however, MPTP needs to be metabolically activated to its fully oxidized pyridinium metabolite MPP+. This conversion is catalyzed by monoamine oxidase (MAO) B and is likely to occur within glial cells. MPP+ is then released from glial cells and actively accumulated into dopaminergic neurons (the ultimate target of MPTP/MPP+ neurotoxicity) via the catecholamine uptake system. It is noteworthy that glial MAO B activity increases with age, paralleling an age-dependent enhancement of MPTP-induced neurotoxicity. The contribution of glial cells to nigrostriatal degeneration may also result from their participation in the metabolism of dopamine. Both MAO A and MAO B are present in glial cells ad catalyze the oxidative deamination of dopamine. This reaction generates H2O2 which in glial cells could be scavenged by the glutathione/gluatathione peroxidase system. H2O2 may also cross cell membranes, however, and cause greater damage to neuronal cells because of their lack of defense mechanisms against oxygen radical accumulation. A final potential route by which glial cells may play a role in the pathophysiology of Parkinson's disease is related to the age-dependent intraglial accumulation of iron. Iron could act as a catalyst for the bioactivation of MPTP-like neurotoxicants as well as for the conversion of dopamine into reactive toxic metabolites. Thus, (I) the possible involvement of glial cells in neurodegenerative processes requires greater consideration that it has been given in the past, and (II) changes in glial cell number and/or functions may contribute to the only evident risk factor in neurodegenerative disorders such as Parkinson's disease, namely aging.

THE ROLE OF MICROGLIA IN BRAIN INJURY
Wolfgang J. Striet
Department of Neuroscience, University of Florida, Gainesville, FL 32610
Recent research has resulted in a concept which views microglial cells as immunocompetent cells indigenous to the central nervous system. As such, microglia are known to be critical for phagocytosis, antigen presentation, and cytotoxicity. At the same time, evidence has been accumulating which supports a neurotrophic role of microglial cells, particularly in paradigms of acute brain injury . This presentation will review work which has examined microglial activation in various brain injury paradigms, including peripheral nerve lesions, cerebral ischemia, and neurotoxicant-induced brain injury. The immuno- and lectin histochemical observations will be discussed with reference to proposed neurotoxic or neurotrophic functions of microglial cells.

FDA PROPOSED GUIDELINES FOR NEUROTOXICOLOGICAL TESTING OF FOOD CHEMICALS
T. J. Sobotka
Center for Food Safety and Applied Nutrition, Food and Drug Administration, Washington, D.C.
That some chemicals adversely affect the nervous system is certainly not news in regulatory toxicology. In 1982, the Food and Drug Administration (FDA) issued testing guidelines for the safety evaluation of proposed direct food and color additives. Assessment of nervous system toxicity was included as part of the general toxicological profile. However, these guidelines provide only broad and nonspecific recommendations for carrying out such an assessment. Little more than the description of very apparent adult nervous system toxicity associated with general neuropathology and overt neurological dysfunction can be derived from toxicity screening studies conducted according to these guidelines. Little consistent or systematically documented information is typically developed about other equally important types of neurotoxic effects such as behavioral dysfunction and developmental neurotoxicity. Concern about these more subtle types of neurotoxicity has become a prominent public health issue. This has resulted in demands for increased assurance that efforts are being made to further minimize the risks of neurotoxicity from human exposure to chemical substances. In an effort to address this concern, FDA is revising its testing guidelines for food chemicals to include a more careful evaluation of the structural and functional measures of neurotoxicity. These are to be routine components of safety assessment in adult and developing organisms. Such a focus will generate the type of information needed for a more effective assessment of the full spectrum of neurotoxic hazards. The revised guidelines for neurotoxicity testing will be discussed in terms of FDA's overall approach to safety assessment.

Neuropathology in Risk Assessment of Xenobiotics
H. B. Jones
Safety of Medicines Department, Zeneca Pharmaceuticals, Alderly Park, Macclesfield, Cheshire SK10 4TG, United Kingdom
The fundamental requirements for registration of a new chemical entity are that it should be efficacious and that exposure or administration to man should prevent an acceptable risk. Evaluation of the safety of potential new drugs and chemicals (xenobiotics) is of critical importance in determining their progression from industrial development to registration and subsequent marketing. The responsibility for making this evaluation rests with the manufacturer and depends upon information derived from a diverse array of in vitro and in vivo experiments in bacterial and mammalian species that are demanded by Regulatory prescription. Assessment of the neurotoxicological phenomena associated with exposure to certain xenobiotics differs substantially according to their type. An insecticide will be evaluated in stringent tests defined by national agencies whereas a dopamine receptor agonist intended for treatment of Parkinson's Disease may not. These significant difference in scientific practices are incidental if the determination of product safety in animals is a true predictor of its safety when given to man. Neuropathological assessment represents a vital component of this neurotoxicological evaluation as it provides qualitative and quantitative evidence of morphological perturbation which may be correlated with functional deficits. This evidence is pivotal in weighing the risks against the benefits associated with treatment by or exposure to a xenobiotic. The focus of this presentation is to discuss the positive and negative aspects of neuropathological assessments in animal studies and address their benefits in prediction of potential human risk posed by xenobiotics.

INVESTIGATIONS OF NEUROTOXICITY BEYOND SCREENING
L.A. Ivens
A. G. Bayer, Department of Toxicology, 42096 Wuppertal, Germany
The nervous system is an extremely complex "organ" with many subsystems which all have to function to together. Other organs and tissues like the heart, muscle, kidneys, lung, and even the immune system are interconnected with the nervous system. Xenobiotics may affect the whole nervous system or only sub-systems like certain cell populations or even act indirectly through effects on other organs. Neurotoxicity screening batteries were developed to answer the question: are effects of a novel compound seen which indicate impairments of nervous system functions. Generally they are not able to answer specific questions about the type of effect or the part of the nervous system which is impaired. As soon as specific information about the area and type of change in the nervous system is needed, this results in the need for specific methods investigating specific parts and functions of the nervous system. The most suitable method or methods have to be selected, since this type of specific investigation is generally time consuming. Methods possible may vary as wide as behavioral techniques, electrophysiology, histology and biochemistry. The selection of methods depends on the area in which the neurotoxic is suspected and/or most prominent. Further, information of the possible neurotoxicity of a compound may also come from human case studies, from animal experiments or from structural similarities with known neurotoxic compounds. Examples are given of investigations of learning and memory by Morris Maze and Active Avoidance. Biochemistry, and electropysiology in vivo and in vitro are discussed as possible methods beyond screening tests.

AN APPROACH TO RISK ASSESSMENT
L. Simonsen, S.P. Lund and U. Hass
National Institute of Occupational Health, Denmark
A strategy for delineating risk factors from use of neurotoxic chemicals was applied to the Danish working environment1 . This analysis disclosed the need for internationally adopted criteria for neurotoxicity, and consequently a working group was established by the Nordic Council of Ministers with the task to suggest criteria for neurotoxicity2 . Effects on the nervous system, e.g. reduction in memory and learning ability, decrease in attention, and alteration of behavior due to toxic chemicals in the environment is now being acknowledged as an important public health problem. This change in concern from obvious effects of high dose exposure to the more subtle effects of environmental exposure to neurotoxicants was incorporated in the criteria. The approach for evaluating neurotoxicity data suggested by the working group has subsequently been used on 80 common industrial chemicals. The results indicate that numerous persons are exposed occupationally and in the environment in general to several chemicals, for which almost no data on the effect on more subtle neurophysiological functions are available. Development of a risk assessment approach dealing with this problem is a major challenge of neurotoxicology in the nineties.
1 Simonsen L, Lund SP: Amer J Ind Med 21:773-792, 1992
2 Simonsen L, Johnsen H, Lund SP, Matikainen E, MidtgÜrd U, Wennberg A: Scand J Work Environ Health 20:1-12, 1994

PROSPECTS FOR USING NEUROTROPHIC FACTORS IN THE TREATMENT AND PREVENTION OF NEUROTOXICITY: AN OVERVIEW
S.C. Apfel
Departments of Neurology and Neuroscience, Albert Einstein College of Medicine, Bronx, NY
Toxic neuropathy represents one of the most likely clinical settings for the early therapeutic application of neurotrophic factors. These are proteins which promote the survival, differentiation, and gene expression of specific populations of peripheral and central neurons. Numerous pre-clinical studies, in-vitro and in-vivo, have demonstrated the potential usefulness of growth factors, and clinical trials for some of them are either underway or are being planned. We plan to briefly review some of the basic biological features of a variety of neurotrophic factors which may have particular importance in both the pathophysiology and the treatment of toxic neuropathy. We will also review the important pre-clinical studies, and assess their current status as it pertains to their potential clinical use. We will include discussions of neurotrophin gene family members, ciliary neurotrophic factor, the insulin like growth factors, and others.

THE DEVELOPMENT OF RECOMBIANT HUMAN NERVE GROWTH FACTOR
AS A TREATMENT FOR PERIPHERAL NEUROPATHIC DISEASE
B.C. Rogers
Genetech, Inc., South San Francisco, CA
Since its discovery over thirty years ago, the role played by nerve Growth Factor (NGF) in the development and maintenance of sympathetic and selected sensory neurons of the peripheral nervous system has been well established and extensively studied. In addition, the existing data suggest that NGF may offer neuroprotection to dorsal root ganglion neurons against drug- and diabetes-induced neurotoxicity. Based on its neurotrophic and neuroprotective activities, the systemic administration of NGF has been proposed as a treatment for several types of neurodegenerative diseases including diabetic peripheral neuropathy where sensory and sympathetic neurons are affected early in the course of the disease. Until recently, painstaking purification and separation techniques were required to obtained minimal quantities of NGF. The cloning of the human gene and the subsequent large-scale production of recombinant human NGF (rhNGF) has allowed for the initiation of comprehensive preclinical animal studies and the initiation of human clinical trials. Growing evidence from these and other studies suggest that the physiological actions of NGF may not be strictly limited to the nervous system but may also include other organs and systems. Results from recent preclinical studies also indicate no safety findings which would preclude the chronic system administration of rhNGF in clinical trials in humans. Results from on-going human trials with rhNGF should greatly advance our understanding of the biology and therapeutic potential of this molecule for treating peripheral neuropathic disease.

INSULIN-LIKE GROWTH FACTOR-I (IGF-I) PREVENTS THE PERIPHERAL NEUROPATHY INDUCED BY VINCRISTINE, PACLITAXEL AND CISPLATIN
P. Contreras, C. Steffler, J. A. Gruner, S. C. Apfel, C. Brosnan, S. Dennis, J. C.Arezzo, A. Yee, J. Kessler and J. L. Vaught
Cephalon Inc., WestChester, PA and Albert Einstein College Medicine, Bronx, NY
IGF-I is a neurotrophic peptide that enhances the survival of motoneurons, induces neuronal sprouting and enhances the rate of regeneration of damaged nerves. To assess whether IGF-I can also prevent or ameliorate the neuropathy induced by anti-tumor agents, IGF-I was administered daily to mice injected with vincristine (1.7 or 2 mg/kg administered twice per week for 8-10 weeks), cisplatin (10 mg/kg administered once per week for 16 weeks) or paclitaxel (21.6 mg/kg administered daily for 6 days). Vincristine treatment produced a sensorimotor neuropathy characterized by decreases in conduction velocity in caudal nerves, decreases in gripping ability, alterations in gait, reduced responses to noxious stimuli, an increase in the number of degenerating nerve fibers and abnormalities in myelination in the sciatic nerve. IGF-I-treatment prevented or ameliorated the effects of vincristine in a dose-dependent manner with the dose of 1 mg/kg providing significant neuroprotection. A sensory neuropathy was induced by 16 weeks of treatment with cisplatin or 6 days of treatment with paclitaxel. Cisplatin treatment decreased the conduction velocity of caudal nerves, altered gait and reduced the ability to respond to noxious stimuli. IGF-I (1mg/kg) administered daily prevented development of the cisplatin neuropathy. A milder sensory neuropathy induced by paclitaxel was characterized by reduced responses to noxious stimuli and concentrations of calcitonin gene-related peptide in cervical DRG. Doses of IGF-I greater or equal to 0.3 mg/kg prevented or ameliorated development of the paclitaxel neuropathy. These data strongly suggest that IGF-I may be useful therapeutically in preventing the peripheral neuropathy induced by a variety of anti-tumor agents.

PRECLINICAL PROFILE OF RECOMBINANT-METHIONYL HUMAN BRAIN-DERIVED NEUROTROPHIC FACTOR (r-metHuBDNF): ALTERNATIVE ROUTES OF DELIVERY WITH THIS POTENTIAL THERAPEUTIC FOR NEURODEGENERATIVE DISEASES
Carl P. LeBel, Ph.D.
Department of Toxicology, Amgen, Inc.
Amyotrophoic Lateral Sclerosis (ALS) is a progressive degenerative disorder of motor neurons in the spinal cord, brain stem and motor cortex, and is the most common motor nervous system disorder. Unfortunately, despite decades of research on this disease, the etiology and pathogenesis of ALS remains unclear. The well known survival-promoting actions of trophic factors in various neuronal cell culture systems essentially forms the foundation for the rationale to use trophic factors in the treatment of neurodegenerative disorders. Brain-derived neurotrophic factor (BDNF) is a member of the nerve growth factor (NGF) family, of which NGF is the prototype. Specificity and proof of concept of the survival-promoting actions of BDNF in vitro have been demonstrated in embryonic and neonatal motor neurons. Additionally, BDNF treatment has been shown to prevent the decrease of cholinergic function in an adult rat facial motor neuron lesion model. Recent evidence has also shown that systemically administered BDNF slows the progressions of motor neuron degeneration in transgenic Wobbler mice. Despite the limited utility of animal models to predict treatment for ALS, it is know that patients with this disease eventually succumb to limb and speech paralysis, and finally to death due to respiratory insufficiency and /or complications therefrom. Given the inevitably fatal outcome of ALS, combined with the database of the effects of BDNF on motor neurons, a Phase I/II clinical study in patients with ALS was proposed. Preclinical safety testing was undertaken in order to support the clinical research. The preclinical development of BDNF, starting with in vitro data, and moving towards the preclinical safety strategy used to support the clinical studies of systemically administered BDNF to treat ALS will be presented. Additionally, data generated in other preclinical studies with intrathecal delivery of BDNF will be presented.

COMPUTERIZED BEHAVIORAL TESTING OF HUMANS
W. Kent Anger
Center for Research in Occupational and Environmental Toxicology
no abstract

NEUROTOXICITY TESTING: ARE WE ON THE RIGHT TRACK?
Abby Li
Monsanto Comapny - Environmental Health Laboratory
Speakers representing industry, public interest groups, government, and academia will engage in a lively debate on strategies for neurotoxicity testing. This session will build upon and bridge two early sessions (Monday and Thursday morning) and question some assumptions that may have been taken for granted. Questions to be debated included: How big a problem is neurotoxicity? Two what extent to industrial chemicals contribute to this problems? Are regulators proposing too much or too little testing? Are academic research approaches used in neurotoxicology inappropriate for regulatory testing or not used enough? Come join us and voice your opinions. (Discussants included: Joel Mattsson, Werner Classen, Barbara McElgunn, Steven G. Gilbert).

RELEVANCE OF NEUROTROPHIC FACTORS TO NEUROTOXICOLOGY
S. Barone Jr.
Neurotoxicology Division, HERL, U.S. Environmental Protection Agency, RTP, NC
This panel discussion is to inform and instigate discussion among participants about the impact of advances in neurobiology of neurotrophic factors and the relevance to neurotoxicology. The topics to be discussed will be loosely arranged around three general areas which are the role of neurotrophic factors in, (1) development of the nervous system, (2) compensatory responses toinjury, and (3) as a therapeutic strategy following neural injury or in treatment of neurodegenerative disorders. After a brief review of the biological role of neurotrophic factors, a discussion will follow regarding what types of neurotrophic responses may be indicative of injury and the possible utility of these alterations as biomarkers of neural injury. This session is meant to be informal and anyone interested in presenting pertinent data should contact the moderator (B.Y.O.D.).

EFFECTS OF 1,3-DINITROBENZENE ON MITOCHONDRIAL FUNCTION IN CULTURED ASTROCYTES
I.A. Romero, R.J. Rist and N.J. Abbott
Physiology Group, King's College, London, UK
1,3-Dinitrobenzene (DNB) is an industrial chemical causing methaemoglobinaemia, testicular damage and neurotoxicity with primarily glio-vascular lesions (Romero et al., 1991, Neuropath. Appl. Neurobiol., 17, 498). Increased glucose consumption and lactate production associated with intracellular depletion of glutathione, have been observed in primary cultures of astrocytes after exposure to DNB over 24 h (Romero et al., 1995, Free Rad. Biol. Med., in press). These changes have been related to oxidative stress damage due to futile redox cycling of the parent compound. DNB and/or its oxygen radical metabolites may have a direct action on the normal mitochondrial respiratory activities leading to changes in cell metabolism. The present study examined mitochondrial function in cultured rat astrocytes following exposure to DNB. During the first 4 h of incubation with DNB, there was a dose-dependent increase in the reduction of the tetrazolium dye MTT in cultured astrocytes, suggesting either an enhanced mitochondrial enzyme activity or an increased mitochondrial membrane permeability to the dye. After incubation with DNB for 24 h, MTT reduction dose-response curves reflected cytotoxic effects at 2 mM DNB. There was a dose-dependent decrease in the uptake of Rhodamine 123, an index of mitochondrial membrane potential, by astrocytes following exposure to DNB for 2 h, with a threshold DNB concentration of 0.1 mM. Changes in cell autofluorescence (exc. wavelength 340 nm, emission wavelength 400- 510 nm) were monitored at 37°C, as an index of mitochondrial NAD(P)H/NAD(P) ratio. Blockade of electron transport by potassium cyanide (0.5 mM) induced an increase in cell autofluorescence, while the uncoupler FCCP (20 µM) decreased the signal, as a result of increased oxygen consumption. Concentrations of 1 mM DNB induced a reduction in the autofluorescence signal, indicating a decrease in mitochondrial NAD(P)H/NAD(P) ratio. This effect could be related to cell metabolism since the decrease in the autofluorescence signal after DNB exposure could not be observed when astrocytes were incubated at 4°C. This results suggest that one of the earliest effects in DNB intoxication in astrocytes is mitochondrial impairment, and supports the hypothesis that metabolism of the parent compound by futile redox cycling leading to the production of free radicals is involved in DNB-induced neurotoxicity._

SENSITIVITY TO 2,5-HEXANEDIONE OF NEUROFILAMENTS IN NEUROBLASTOMA CELL LINE SK-N-SH INCREASES DURING DIFFERENTIATION
E. Heijink1, P. A. Bolhuis2, G. B. van der Voet3, F. A. de Wolff 1
1Coronel Laboratory and 2Department of Experimental Neurology,
University of Amsterdam, Academic Medical Centre, 1105 AZ Amsterdam
3Toxicology Laboratory, University Hospital Leiden, 2300 BC Leiden, The Netherlands
Chronic occupational exposure to n-hexane may cause peripheral neuropathy via the metabolite 2,5-hexanedione (2,5-HD). The most characterizing features are axonal swelllings containing aggregates of neurofilamental (NF) proteins in the preterminal axon and nerve degeneration distally. Mechanistic studies have demonstrated that 2,5-HD can react directly with the NF-proteins resulting in covalent-crosslinking. However, this reaction is observed within other proteins as well and can not exclusively explain the high vulnerability of NF-proteins for 2,5-HD. Using the human neuroblastoma cell line SK-N-SH in culture as a model, we studied the direct effects of 2,5-HD on the distribution of the NF-proteins with immunocytochemical methods. Retinoic acid induced differentiation into neuronal cells expressed by the outgrowth of processes and the detection of NF-proteins in the majority of the cells. Cells were exposed to 0- 10 mM 2,5-HD for 3 days. A concentration-dependent accumulation of NF-proteins was detected as a spherical structure in the perikaryon. NF in the differentiated neuronal cells were more susceptible to 2,5-HD than in undifferentiated cells, as effects occurred at much lower 2,5-HD concentrations. At present the dynamics of NF-proteins during 2,5-HD exposure are examined in both undifferentiated and differentiated cells using metabolic labeling studies. These experiments may be helpful in determining which NF-features are important in 2,5-HD induced neurotoxicity.

IMMUNOHISTOCHEMICAL LOCALIZATION OF NEURONAL AND GLIAL CALCIUM-BINDING PROTEINS IN HIPPOCAMPUS OF CHRONICALLY LOW LEVEL LEAD EXPOSED RHESUS MONKEYS
Sigrid Noack1, Gisela Stoltenburg1, Hellmuth Lilienthal2 and Gerhard Winneke 2
1Institute of Neuropathology, Freie Universität Berlin
2 Institute of Environmental Hygiene at the University of Düsseldorf, Germany
The purpose of this study was to investigate the distibution of the neuronal proteins parvalbumin, calbindin D28k, calretinin and the glial protein S100 in the hippocampus of lead exposed rhesus monkeys. It has been suggested that lead may exert its toxic effects by perturbing the intracellular calcium homeostasis. Lead is able to increase the intracellular Ca2+concentration and can serve as a calcium substitute. Some calcium-binding proteins such as calmodulin or renal and intestinal calbindin are capable of binding lead. We tried to find out a putative dose-dependent relation between long-term low level lead exposure and the expression of the proteins investigated. Rhesus monkeys were pre- and postnatally exposed to 600mg-300mg-0mg lead-acetate in diet for 9 years, as described by Lilienthal et al. 1986. Hippocampal paraffin sections were stained for parvalbumin (PV), calbindin D28k (CB), calretinin (CR) and S100 with immunohistochemical methods. The distribution of the neuronal calcium-binding proteins was almost identical for the different exposure groups. The most striking observation was a marked decrease of S100 immunoreactivity of astrocytes in the high lead group. Considering a protective role against high Ca2+concentration and Pb2+accumulation respectively the unchanged expression of PV, CB, CR remains to be clarified. The apparent difference of S100 expression supports the hypothesis, that glial cells are the main target of lead toxicity. The reduced expression may indicate a developmental retardation of astroglia.
EFFECTS OF CHRONIC LOW LEVEL LEAD EXPOSURE ON THE EXPRESSION OF FIVE DIFFERENT NMDA-SUBUNITS IN HIPPOCAMPUS OF YOUNG RATS: A mRNA IN SITU HYBRIDIZATION STUDY
S. Partl1, G. Stoltenburg1, M. Hummel2, H. Herbst2, L. Altmann3 and H.Wiegand3
1 Institute of Neuropathology, Freie UniversitÑt Berlin, Germany
2 Institute of Pathology, Freie UniversitÑt Berlin, Germany
3 Institute of Environmental Hygiene, DÅsseldorf, Germany
The purpose of this investigation was to study the effect of chronic low level lead exposure on the mRNA expression of different n-Methyl-D-Aspartate (NMDA) receptor subunits in the hippocampus of young rats. The effects of lead were investigated in a total of seventy 17 to 20 days old male Wistar rats exposed to lead during gestation and postnatally until weaning (postnatal day 20). The control animals were the offspring of and suckled by unexposed mothers. Radioactive mRNA hybridization was performed with P33 labelled oligonucleotides for the different subunits of the NMDA receptor channel complex: NR1, NR2A, NR2B, NR2C and NR2D. Frozen horizontal sections were used. The expression of NR1 and NR2A in the CA1 region of the hippocampus was significantly higher in lead exposed animals compared to controls. The differences in expression of NR2B were less pronounced. According to normal developmental distribution, the expression of NR2C was restricted to cerebellum and thalamus. The NR2D expression was no longer to be detected. The increased level of NR1 and NR2A expression may indicate an upregulation to compensate the functional alteration by lead on the postsynaptic or presynaptic site. The anatomical and developmental distribution of NR2D was in physiological range. There was no indication for developmental delay in the postnatal time interval investigated. The overexpression of the NR1 subunit correlates with the increased MK 801 binding and the reduction of long term potentiation (LTP).

2.5-HEXANEDIONE ACCUMULATES GFAP-POSITIVE INTERMEDIATE FILAMENTS IN NONNEURONAL CELLS IN IN-VITRO CELL CULTURES OF THE NERVOUS SYSTEM
W. GrÅning1, G. Stoltenburg2, F. Boegner 3 and H. Altenkirch4
1Department of General Medicine and Nephrology, 2Institute of Neuropathology, 3Department of Neurology, Klinikum Benjamin Franklin, Freie UniversitÑt Berlin 4Department of Neurology, Spandau Hospital, Berlin, FRG
In organic solvent neurotoxicity 2.5-hexanedione (2.5-HD) plays an important role as the most potent neurotoxic metabolite of n-hexane and 2-hexanone. It has been the toxic agent in a wealth of experimental studies. The well identified histopathological correlate of the axonotonic effect consists in a massive accumulation of 10nm neurofilaments proximal to Ranvier's nodes. We have already shown, that nerve cell cultures of chick embryos exposed to 2.5-HD develop specific neurotoxic alterations. Different nervous system compartments displayed different vulnerability reflecting clinical conditions. Apart from topical divergencies neurotoxicity depended on concentration of 2.5-HD in a linear manner. However, neurons failed to develop classic neurofilament accumulation. In contrast, we found gliatoxicity exceeding neurotoxicity significantly. Due to high proliferation rates gliatoxicity occurred with threshold dose resulting in abrupt pancytotoxicity. Now we demonstrated by immunohistochemistry for glial fibrillary acidic protein (GFAP) that subpancytotoxic concentrations produced juxtanuclear accumulations of intermediate filaments in nonneuronal cells. The alteration observed indicates breakdown of cytoarchitecture. These results corroborate the general character of intermediate filament pathology and cytoskeleton dysfunction under 2.5-HD exposure.

FACTORS DETERMINING THE SENSITIVITY OF THE BRAINSTEM TO GLIOVASCULAR DAMAGE
J. L. Holton, C. C. Nolan, A. Sale & D. E. Ray
Medical Research Council Toxicology Unit, Leicester, England
The gliovascular lesions produced in the brainstem by 1,3-dinitrobenzene (DNB) can be unilaterally decreased in severity by reducing auditory input to one ear (Neurotoxicol. 13:379-388; 1992). This suggests that ongoing sensory input may modulate the brainstem damage caused by this agent. A total dose of 8-12g/kg of the antibiotic metronidazole given to 10 rats over 10-15 days produced lesions with a similar morphology and topography to those produced by DNB. They also developed asymmetrically in animals with a unilateral left tympanic membrane perforation. Mean lesion severity scores (out of 3) were 0.5 : 1.1 in the left : right cochlear nuclei (p=0.016, Wilcoxon test), and 2.1 : 1.2 in the left : right inferior colliculi (p=0.008), suggesting that the sensory modulation effect is not unique to DNB. In rats given DNB (3x 9 mg/kg at 0, 4 & 24 hours), we examined the effect of increasing motor activity by inducing tremor by co-intoxication with the pyrethroid bifenthrin (20 + 10 mg/kg at 6 & 24 hours).This treatment produced a fine generalised tremor which began 12 hours after the first dose of DNB and lasted until 42 hours. The bifenthrin significantly increased the severity of DNB damage in motor areas such as the cerebellar roof nuclei and red nuclei. As expected, bifenthrin alone caused no morphological damage in the central nervous system at this (maximal tolerated) dose. The red nuclei and cerebellar roof nuclei are major motor areas in the rat, and show increased glucose utilisation during pyrethroid-induced tremor. We therefore conclude that changes in motor, as well as sensory activity, can modulate the severity of these lesions.

MYELIN BASIC PROTEIN IN CEREBROSPINAL FLUID AS A MONITOR OF ACTIVE DEMYELINATING LESIONS IN THE RAT CNS
X. Liu, P. Glynn & D. E. Ray
Medical Research Council Toxicology Unit, Leicester, England
A number of chemical agents produce direct myelin damage leading to demyelination, and others produce secondary demyelination consequent on axonopathy. We hypothesised that myelin breakdown would lead to liberation of myelin basic protein (MBP) into the extracellular space, and thence into the cerebrospinal fluid (CSF). As CSF turnover is relatively rapid, then the concentration of MBP might provide a measure of the current status of a lesion (i.e. active or inactive). In order to test this, rats were implanted with a chronic catheter in the cisterna magna and concentrations measured in 100 µ1 samples of CSF by radioimmune assay using antibodies developed against purified rat MBP. Following a test injection of MBP in 3 rats, CSF concentrations fell with an initial half life of about two hours. The detection threshold was approximately 2 ng/ml MBP. Standard demyelinating lesions were created by microinjection of 25 µg of lysolecithin in 2.5 µ1 saline bilaterally into the cerebellar white matter. Lesions had an approximate diameter of 1.5 mm at 72 hours, and were characterised by extensive demyelination. They produced no obvious motor signs, but cerebellar auditory evoked response amplitude decreased reversibly, reaching a minimum at 3 days. The demyelination had largely resolved when examined at 15 days. Prior to injection no MBP could be detected in CSF samples, but within 6 hours of lesioning MBP concentration reached 84±19 ng/ml, remained elevated at 24 hours, and fell to 33±13 ng/ml at 2 days (n=5, mean±S.E.). MBP could be detected at 3 and 4 days, but not at 6 days or later. Rats given control injections of saline developed only minor needle track artefacts and no elevation in MBP (n=5). We conclude that measurement of MBP in CSF can be used to monitor the progress of chemical demyelination, provided sufficiently large CSF samples are available.

NEURONAL DYSFUNCTION ASSOCIATED WITH GLIAL LESIONS IN THE RAT BRAINSTEM
D. E. Ray, J. L. Holton, T. Lister, M. Mulheran and C. C. Nolan
Medical Research Council Toxicology Unit, Leicester, England
The astrocytic and oligodendrocytic death seen in brainstem nuclei following systemic administration of 1,3-dinitrobenzene (DNB) is associated with vacuolation and perivascular údema. The údema resolves over 8 days, giving place to astrocytic and microglial proliferation. These lesions can produce severe ataxia and a degree of secondary neuronal death, but are primarily glial. We investigated the relationship between lesion severity and neuronal function. Eleven rats were implanted with platinum recording electrodes in the inferior colliculi. Auditory evoked responses were measured in response to 40 kHz tone bursts, and local blood flow was measured by hydrogen polarography using the same electrodes, both before and over 8 days following administration of 30 mg/kg DNB. Lesion severity was assessed in rats killed by arterial perfusion with fixative at 8 days. Lesions in the auditory pathway (cochlear nuclei, superior olives and inferior colliculi) were scored on a qualitative 0-3 scale based on the proportion of the nucleus involved. Overall auditory pathway lesion score ranged from 0/9 to 6/9. Inferior collicular blood flow showed a large increase of from 0.95±0.03 ml/g/min (mean ±S.E.) to a peak of 3.10±0.42 at 12 hours. This preceded development of lesions, which develop from 12 hours onwards. Flow returned to predose values after 72 hours. The magnitude of the flow increase correlated with the severity of the subsequent lesion as assessed at 8 days, linear regression giving a correlation coefficient of 0.676 (p=0.022). The amplitude of the collicular auditory evoked response to 100 dB stimuli was significantly depressed over 24 to 120 hours, but returned to normal at 8 days. At 24 hours it was 49±7% of the predose value. The peak auditory deficit was well correlated with the overall auditory pathway lesion, linear regression giving a correlation coefficient of 0.756 (p=0.007), although less well correlated with the local lesion in the inferior colliculus (r=0.627, p=0.039). We conclude that morphological measures of gliotoxicity correlate well with neuronal dysfunction, as measured in the auditory pathway.

LOW LEVEL EXPOSURE TO ORGANIC SOLVENTS IN RATS MAY INDUCE A LONG-LASTING INCREASE IN THE AUDITORY EXCITABILITY
S.P. Lund, L. Simonsen and U. Hass
National Institute of Occupational Health, Denmark
Several studies demonstrate reduced auditory sensitivity and/or permanent threshold shift in rats exposed to various organic solvents, e.g. toluene, trichloroethylene, xylene, styrene. The hearing impairment is most pronounced in the mid-frequency range and is probably related to outer hair cell loss in the cochlea. For toluene and trichloroethylene, the reduction in auditory sensitivity is not detectable until the dose of exposure has exceeded a certain threshold. However, less attention has been focused on changes in the auditory function at low exposure doses, i.e. before the outer hair cells are lost and the auditory thresholds are increased. To test the effect of organic solvent exposure on the auditory function, rats were exposed to trichloroethylene (100 ppm, 1000 ppm), n-heptane (800 ppm, 4000 ppm), p-cymene (50 ppm, 250 ppm), and 4-tert.butyltoluene (40 ppm, 20 ppm) 6 hours/day for 28 days. Two months after the end of exposure the rats were anesthetized and the auditory brain stem responses to 25-95 dBlin SPL 16 kHz tonepips were measured in steps of 10 dB. The group exposed to 4000 ppm n-heptane showed a reduction in the auditory sensitivity together with an increase in the auditory threshold, while the auditory sensitivity in the groups exposed to 1007 ppm trichloroethylene and 800 ppm n-heptane showed increase without shifts in the auditory thresholds. These results demonstrate the possibility of long-lasting increase in the auditory excitability and suggest a bellshaped dose-effect relationship of organic solvent exposure on the auditory function in rats.

ACUTE EFFECTS OF INHALED TRICHLOROETHYLENE (TCE) ON RAT VISUAL FUNCTION DEPEND ON EXPOSURE CONCENTRATION AND NOT DURATION
W.K. Boyes and V.T. Griffin
Neurotoxicology Division, U.S.E.P.A., and ManTech Environmental Technology, Research Triangle Park, North Carolina
Risk assessments traditionally have focused on averaged lifetime exposures, and have not considered hazards of brief exposures. These experiments explore the relationships between exposure concentration (C) and time (T) to provide guidance for assessing the risk of short-term solvent exposures. The null hypothesis was that the product of CxT would produce a constant health effect (Haber's Law). Awake, adult, male Long-Evans rats were exposed to TCE vapors in a head-only exposure chamber while recording pattern onset/offset visual evoked potentials (VEPs). Experiment 1 involved exposure to 0, 500, 1,000, 2,000 or 4,000 ppm TCE (n=5/concentration) and testing 0.5, 1, 2, and 4 hr after exposure began, for CxT products ranging from 0-16,000 ppm-hr. Results showed a selective reduction in amplitude of the VEP frequency double component (F2) related to C, but not to T or CxT product. Experiment 2 focused on CxT products of 0 ppm-hr (0 ppm-4 hr) or 4,000 ppm-hr created through 4 exposure scenarios: 1,000 ppm-4 hr; 2,000 ppm-2 hr; 3,000 ppm- 1.3 hr; or 4,000 ppm-1hr (n=9-10/concentration). Again, the results showed a significant decrease in F2 amplitude as a function of C, but not T or CxT product. These limited results imply that for short term exposure health risks, exposure concentration is more important than exposure duration. The outcome may differ, however, for other endpoints, compounds, or exposure scenarios.

QUANTITATIVE EEG AND CSF CYTOLOGY FINDINGS IN INDIVIDUALS WITH CHRONIC OCCUPATIONAL EXPOSURE TO MIXTURES OF ORGANIC SOLVENTS
Kiti M I Muller Mari Antti-Poika and Tero Kovala
Finnish Institute of Occupational Health, Helsinki, Finland
Quantitative electroencephalography (QEEG) recordings and cerebrospinal fluid (CSF) cytological analysis were performed in 70 subjects (63 men and 7 women) referred to our Institute for examinations because of significant occupational exposure to mixtures of organic solvents and symptoms suggestive of organic brain disease. The mean age of the subjects was 49.4 years (median 51, range 31-61yrs) and the mean/median exposure time 26 yrs (range 8-45yrs). The EEG recordings were performed with Cadwell Spectrum 32 equipment. The quantitative analyses were made within the frequency band of 1.5-20 Hz with the Neurometrics program and reference values of our own laboratory. CSF cytological analysis was performed from Millipore filtration specimens and cytocentrifuged cell smears as earlier described.1 An increase in the absolute or relative power values of delta and theta EEG activity or a slowing in the mean frequency of EEG was observed in 39 subjects (56%). CSF cytological abnormalities were found in 55 subjects (79%). The most frequently detected abnormality in CSF (in 36%) was an abnormal differential cell count with a predominance (>60% of CSF cells) of mononuclear phagocytes, the presence of lipophages (37%) and an increased proportion of enlarged stimulated lymphoid cells (>31% of lymphoid cell population). Additionally the CD4+/CD8+ T cell ratio was elevated in the CSF of 14 subjects (22%). In only 8 subjects (11%) both the QEEG analysis as well as their CSF finding were normal while 32 subjects (46%) were abnormal for both their QEEG and their CSF cytological findings. Thus findings indicative of diffuse brain dysfunction were detected in a large proportion of subjects with chronic exposure to mixtures of solvents. 1 Muller et al. J Neuroimmunol. 1990;30:219-227.

THE APPLICATION OF NEUROBEHAVIORAL TESTING METHODS TO THE OCCUPATIONAL HEALTH CARE SETTING: THE EFFECTS OF TOLUENE IN WORKERS IN THE PRINTING INDUSTRY1
H.H. Emmen2, B.M. Kulig2, W.F.M. Rooijendijk3, J. Hofstee3, H. Muijser 2 and J. Hooisma2
2 Department of Neurotoxicology & Reproduction Toxicology,
TNO Nutrition and Food Research, Rijswijk
3 The Occupational Health Service Midden Ijssel, Deventer, The Netherlands
Occupational exposure to organic solvents has been associated with a number of neurobehavioral effects ranging from acute subjective feelings of inebriation to severe and persistent changes in cognitive functioning and personality changes. Despite the growing evidence for solvent effects in occupationally exposed workers, no generally accepted approach to worker monitoring has been adopted. In the present study, neurobehavioral functioning was evaluated in toluene-exposed printers tested in the context of routine physical examinations conducted in the occupational health care clinic. The purpose of the study was to examine the effects of solvent exposure in workers exposed to a single solvent and the exposure levels at which effects occur and to evaluate the feasibility of implementing neurotoxicity monitoring procedures in the occupational health care setting. Onehundred and thirty one toluene-exposed workers and 69 non-exposed workers in the printing industry were tested using a battery of automated neurobehavioral tests developed as part of the European EURONEST project. The EURONEST test battery consisted of tests designed to assess changes in memory (digit spans forward and backwards), psychomotor slowing (simple reaction time), attentional processes (color work vigilance test), perceptual coding (digit symbol substitution), and visuomotor performance (ballistic tracking). In addition, subjective symptoms were evaluated using a standardized neurotoxic symptom questionnaire (NSC-60). Analysis of the symptom questionnaire results indicated a higher prevalence of subjective complaints related to memory difficulties, sleep disturbances, and sensorimotor changes (paresthesias and reduced neuromuscular strength) in the group of tolueneexposed workers. In addition, covariance analyses corrected for age and educational level also indicated a significant increase in simple reaction time in the exposed group. These results demonstrate the feasibility of the application of neurobehavioral testing techniques in the health care setting and suggest a mild degree of psychomotor slowing workers with a mono-exposure to toluene. Further evaluation of toluene-related effects using individual exposure data is currently in progress in order to assess the significance of these results. 1 This work was supported, in part, by the Netherlands Ministry of Social Affairs and Labour (SoZaWe) and forms part of the EC Concerted Action EURONEST

THE ROLE OF CONFOUNDING VARIABLES IN THE ASSESSMENT OF NEUROBEHAVIOURAL EFFECTS OF CHRONIC SOLVENT-EXPOSURE
Ann M. Williamson
National Institute of Occupational Health and Safety,Worksafe Australia, Sydney, Australia
The presence and relative contribution of confounding variables is an issue in the assessment of toxic effects on neurobehavioural function. Factors like education level and alcohol use are well-known to exert effects on the performance of neurobehavioural tests. In a prospective study of a cohort of vehicle spray painters, measures of neurobehavioural function, demographic characteristics, daily habits and general health status were taken from the onset of their exposure on four occasions for three full years. This study provided the opportunity to assess the relative contribution of confounding variables and the study factor, solvent exposure, on neurobehavioural function in a study design in which the effects of confounding would be expected to be reduced. Multiple linear regression analysis was used with three confounding variables, education level, alcohol use and occupational experience as well as solvent exposure as predictor variables of neurobehavioural test performance. Although the level of solvent exposure was considerably less than the composite threshold limit value for the solvent mixtures encountered, the results indicated that psychomotor performance deteriorated with increasing solvent exposure, but only on the hand steadiness test. The confounding variables showed even greater influences on performance. Education level affected performance on both psychomotor and cognitive tests in the expected direction. Occupational experience also influenced test performance but only for psychomotor tests. With increasing time in the trade, spray painters showed significantly superior performance on reaction time, hand steadiness and visual pursuit tests. Since training in spray painting focuses on psychomotor coordination, this result would be expected. Alcohol use was associated with mixed effects on neurobehavioural function. Painters who reported using greater amounts of alcohol showed significantly poorer performance on the visual test, critical flicker fusion, but better performance on the short term memory and learning measures of the paired associates test. These findings reinforce the importance of separating effects due to confounders from effects due to toxic exposures in studies of neurobehavioural function. They emphasise the dilemma of test selection, choosing tests that are sensitive enough to detect effects due to toxic exposure but which are not affected by confounding variables.

PERIPHERAL MARKERS OF NEUROCHEMICAL EFFECTS AMONG STYRENE-EXPOSED WORKERS
E. Bergamaschi, A. Mutti, S. Cavazzini, M.V. Vettori, F.S. Renzulli and I. Franchini
Laboratory of Industrial Toxicology, University of Parma Medical School, Italy
Several parameters involved in neurotransmission represent selectively vulnerable targets for environmental and occupational neurotoxicants. In order to evaluate possible changes in biochemical events occurring in the central nervous system, monoamine oxidase B (MAO) and dopamine-b-hydroxylase (DBH) activities were measured in accessible biological media during a cross-sectional investigation in workers occupationally exposed to styrene. The study group consisted of 53 workers (33 men and 20 women) exposed to styrene, employed for 9.3 years on average (range 1-22) in reinforced plastics plants. On the basis of indices of internal dose (i.e., the sum of styrene metabolites in the "next morning" urinary samples) two subgroups of workers were identified, whose median metabolite levels were 88.6 and 340.2 mg/g creatinine, respectively. Sixty industrial workers with no known exposure to chemicals and comparable as to age, sex and confounding variables were recruited as controls. The activities of MAO in platelet-rich plasma and of DBH in serum were measured within the same run for both groups using methods based on HPLC-ECD. A lower DBH activity was found in exposed as compared to control workers (GM: 10.11 U/ml serum vs. 7.25 U/ml serum; p < 0.01), whereas MAO activity was affected in the heavily exposed subgroup (13.8 vs. 10.1 U/107 platelets; p= 0.05). After adjustement for confounding variables, DBH correlated with styrene metabolites (R2= -0.182, p=0.005); a dose-response relationship between styrene metabolites and DBH activity was also apparent (c2 =13.3, p< 0.01). This study confirms that long-term exposure even to relatively low levels of styrene can affect DBH activity. Since DBH is expression of catecholamine firing, its decreased activity could represent an indirect index of an altered turnover rate of the physiological substrate (i.e. dopamine) at neuronal level. However, a direct consequence of styrene metabolites on enzyme activity cannot be ruled out. Platelet MAO B activity does not parallel that of DBH and seems to be less sensitive to styrene exposure.

VARIATIONS OF URINARY CATECHOLAMINES AND EXPOSURE TO STYRENE
M. P. Sassine, G. Truchon, R. Savard, S. Bélanger & D. Mergler
Université du Québec a CINBIOSE; Institut de recherche en santé et en sécurité du travail du Québec, IRSST
Styrene, a well-known neurotoxin, is frequently used in industry. Animal studies suggest that styrene targets the catecholaminergic system. The aim of this study was to examine pre-shift (Monday) and end-shift (Friday) variations in urinary catecholamines in relation to exposure among actively employed workers. The study population included 47 workers, chosen randomly, from three Québec fiber glass reinforced plastics factories; 9 were excluded for the following reasons: history of disease that could influence catecholminergic functioning; consumption of medication or drugs; organic solvent exposure outside of working hours. The time weighted average (TWA) for environmental styrene, measured for each worker in their respiratory area, varied between 666.0 mg/m3 and 8.0 mg/m3 (geometric mean: 100.1 mg/m3) and exposure was constant throughout the week. The TWA of workers not wearing a mask was strongly related to end shift urinary metabolites (mandelic and phenylglyoxylic acids); the summation of these metabolites was used as bioindicators for the level of styrene exposure. The level of catecholamines were adjusted for confounding factors such as age, scolarity, consumption of coffee, cigarette and alcohol, and weight. End-Shift dopamine levels decreased with increasing level o urinary metabolites (r2=0.17; p=0.05), factoring in seniority. Monday to Friday increase of urine norepinephrine levels was significantly correlated to the level of exposure (r2=0.20; p=0.01). On the other hand, changes of epinephrine were not associated to the level of exposure. Further studies should examine the relation between performance on neurobehavioral test batteries an catecholamines excretion in order to better understand the mechanisms that underlay neurotoxic manifestations in humans.(This study was financed by grants from the IRSST #PE-92-06 and FRSQ-CQRS #92-0756).

SYMPTOM BASE TATES AFTER CHEMICAL EXPOSURE FOR WHITE, HISPANIC AND AFRICAN AMERICANS
R.M. Bowler1, Guy Huel2 , D. Mergler3 , J.E. Cone4, S . Rauch1
1 San Francisco State University, CA
2 I.N.S.E.R.M., Paris, France
3University of Quebec, Montreal
4 University of California San Francisco, CA
The results of a symptom checklist of three matched-pair studies of exposed groups of: 1) a primarily white community (N=220) environmentally exposed to metam sodium, a toxic pesticide, 2) a Hispanic group who worked in a microelectronics plant and had extensive past (M 6.7yrs) exposure to multiple organic hydrocarbon solvents (N=180); and 3) an African/American environmentally exposed group (N=168) who had sulfuric acid exposure are presented. Each exposed group was compared to a matched (race, age + 3yrs, gender and education + 2yrs and the nearest number of children) unexposed control group, resulting in 90 matched pairs for the metam sodium group, 61 matched pairs for the organic solvent group and 77 pairs for the sulfuric acid group. MANOVA results for all three exposed and unexposed groups indicate no differences in symptoms for: weight loss, heart palpitations, anxiety, depression, chest tightness, increase or decrease in sleep, dark vision. Differences were obtained at p=<.001 for: Neurological symptoms: tingling, muscle twitching, tingling, numbness, tremors, loss in muscle strength in hands and feet, fainting, perspiring, blurred vision and change in personality; at p=<.01 for: incoordination and diarrhea; at p=<.05 for irritability, lower alcohol tolerance and loss of smell. MANOVA results indicate no significant differences between all three exposed and control groups for age, race, education and gender. Significant differences by education alone were obtained for weightloss, and for age on depression and irritability. Prevalence rates and odds ratios of symptoms will be reported by race, exposure group and individual symptoms. These results suggest a robust symptom complex following chemical exposure and appear to generalize as a specific symptom pattern regardless of specific chemical.

CLINICAL DATA ON THREE CASES OF OCCUPATIONALLY INDUCED PCB- INTOXICATION
H. Altenkirch1, G. Stoltenburg2, D. Haller1, G. Walter1 , O. Hopmann
1 Department of Neurology, Spandau Hospital
2 Department of Neuropathology, Benjamin Franklin Hospital Free University Berlin, Germany
Since the ban of PCBs at the end of the 1970s, extensive measures have been undertaken in Germany to dispose of PCB contaminated transformers. We report on three patients with considerable skin exposure to PCBs, in particular to Clophen A-30, while repairing or dismounting transformers. The periods of exposition range from 4 and 5 to 20 years. All patients presented with distal- symmetrical sensorimotor polyneuropathy as well as encephalopathy. One case of hepatopathy was observed, chloracne occurred in all three cases. In two cases the neurophysiological examination indicated an axonal, in one case a mixed axonaldemyelinating neuropathy. A nerve- muscle biopsy revealed axonopathy as well as chronic neurogenic muscular impairment. In one of the cases, the neuropathy and encephalopathy progressed for a period of over 8 years after termination of exposure. In the two other cases the neurological deficits persisted over an observed period of 2-3 years. The reported clinical results may suggest that the long half- life of Clophen and its accumulation in fatty tissue can lead to persistence of PNS and CNS impairment long after the period of exposure.

NEUROLOGICAL INVESTIGATIONS IN 23 CASES OF PYRETHROID INTOXICATIONS REPORTED TO THE GERMAN FEDERAL INSTITUTE FOR HEALTH PROTECTION
H. Altenkirch, D. Hopmann, B. Brockmeier, G. Walter
Department of Neurology, Spandau Hospital, Free University Berlin
In 1993, 64 cases of chronic pyrethroid intoxications were reported to the Federal Health Office in Germany. Shortly afterwards the media spoke of thousands of cases of pyrethroid intoxications in homes. Twenty-three of the persons reported were examined in a neurological department on an in-patient basis using clinical neurological, neurophysiological (NCV, EMG, VEP, AEP, SEP, EEG), neuroradiological (CCT, SPECT, MRI) and laboratory investigations, including the examination of pyrethroid values in blood and urine. The pyrethroid exposition involved carpets, moth killers, pesticide sprays and wood preservatives. Nine of the cases presented with severe somatic disorders with completely different clinical diagnoses, such as pituitary tumor, radiogenic lumbosacral plexus paralysis, Guillain-Barre syndrome, spinal muscular atrophy, with no plausible relationship to exposition. Eight cases presented with multiple chemical sensitivity syndrome (MCS) and normal somatic findings. In six of the cases, the complaints could be attributed to pyrethroid exposition. There was, however, not a single case in which evidence for irreversible PNS or CNS lesions could be found.

PROMOTION OF ORGANOPHOSPHATE INDUCED DELAYED POLYNEUROPATHY: THE TARGET IS NOT A PHENYL VALERATE ESTERASE
D. Milatovic, K.M. Osman, A. Moretto and M. Lotti
Istituto di Medicina del Lavoro, Universita degli Studi di Padova, Padua, Italy
Certain esterase inhibitors exacerbate organophosphate induced delayed polyneuropathy (OPIDP) and the effect is called promoion. Target for OPIDP but not for promotion is neuropathy target esterase (NTE). Although the target for promotion shares some characteristics with NTE, it remains unknown (Moretto et al., Toxicol Appl Pharmacol 1994, 129:133-137). Since NTE is a neural phenyl valerate (PV) esterase (defined as PV activity resistant to 40 uM of the non-neuropathic paraoxon and sensitive to 50 uM of the neuropathic mipafox; Johnson, Arch Toxicol 1977, 37:113-115), the aim is to ascertain whether the target for promotion is a PV esterase too. No correlation was found between the ability of a given compound to promote OPIDP in hens and to inhibit paraoxon+mipafox resistant PV esterases in brain and sciatic nerve. Therefore, the promotion site is not among paraoxon+mipafox resistant PV esterases. About 70% of total PV esterases from hen brain and sciatic nerve is not inhibited by 5xNTE I50 of non-promoter neuropathic inhibitors such as diiso-propylfluoro phosphate (DFP, 5 uM) and mipafox (50 uM) (20 min, pH 8.0, 37 oC). Titration of these PV esterases with the promoter phenylmethanesulfonyl fluoride (PMSF) showed that a fraction (about 10%) was resistant to high concentrations of PMSF (1 mM). Therefore, the promotion site does not belong to this fraction. Inhibition by PMSF followed a first order kinetic. The calculated I50s for PMSF were 50 uM in both brain and sciatic nerve, and the activity was 6,065+681 and 1,035+172 nmol of PV hydrolysed/min/g of tissue (mean+SD) in brain (n=19) and peripheral nerve (n=11), respectively. However, other promoters such as L-(-) and D-(+) methamidophos, and di-n-pentyl phenyl phosphinate, did not significantly inhibit this enzyme (>10 mM for methamidophos isomers and 1 mM for di-n-pentyl phenyl phosphinate, same conditions as above). Moreover, no inhibition of this enzyme in brain and sciatic nerve was found when the same promoters were given to hens at effective doses (50 mg/kg p.o. for methamidophos isomers, 5 mg/kg s.c. for di-n-pentyl phenyl phosphinate). We conclude that the target for OPIDP promotion is not a PV esterase.
NEUROBEHAVIORAL FOLLOW- UP STUDY IN WORKERS EXPOSED TO MERCURY
W. G|nther1, B. Sietmann2, A. Seeber2
1 Health Office, Bitterfeld, Germany
2 Institute for Occupational Physiology at the University of Dortmund, Germany
It is known that slight subjective symptoms could be observed after chronic exposure to nearly 0.01 mg/m3 of in-organic mercury in the air, and psychomotor dysfunctions after chronic exposure to concentration above 0.1 mg/m3. The present follow-up study investigated effects due to chronic exposure to mercury vapours between 0.06 mg/m3 and 0.1 mg/m3 in the room air. Exposed workers, included in the study after an average exposure of 12 years, were observed 3 times during a period of about 4 years. High-exposure (n = 19, 125 100 5g/l mercury in urine), low- exposure (n = 39, 22 12 5g/l) and control (n=37, non-exposed) groups were defined. Eighteen different scales concerning neurotoxic complaints and personality were considered for statistical analyses, as well as 4 variables concerning memory performance and 19 for perceptive, sensomotor, interference and balance performances. Partial correlations (corrected by age and verbal intelligence) among mercury in urine and dependent variables showed no stable pattern within the correlation coefficients in all three observations. Analyses of covariance (corrected by age and verbal intelligence) detected group differences in one of the memory tasks, in which a decrease of correct word reproductions corresponded to an increase of the exposure. Finger tapping and finger dexterity are related to the exposure in the same manner. For the other performance data, the neurotoxic complaints and personality traits no stable relations to the exposure could be established.

VISUAL CONTRAST SENSITIVITY DEFICITS IN BOHEMIAN CHILDREN
H.K. Hudnell1, I. Skalik2, D. Otto1 , D. House1 and R. Sram2
1 Health Effects Research Laboratory, U.S. EPA, Research Triangle Park, NC 27711
2 Regional Institute of Hygiene, Czech Republic
Visual contrast sensitivity (VCS) tests have been used successfully in medical diagnosis and subclinical neurotoxicity detection. Recently, VCS was measured in two studies of 2nd grade children in the Czech Republic. Study 1 compared children in standard schools and schools for the learning disabled (LD). Study 2 compared children in Teplice, an area in which soft-brown coal combustion produced high levels of pollutants (e.g. Hg, As, SO2, NOx, and aromatic hydrocarbons), with children in an area of low air pollution, Prachatice. It was hypothesized that perinatal exposure to the combustion products disrupted neurological development (Sram, 1991). The VCS test (Stereo Optical Co.) consisted of circular fields containing sinusoidal gratings at 8 contrast levels for each of 5 spatial frequencies (1.5-18 cycles/degree). Subjects indicated orientation of the patterns by pointing left, up, or right. Visual acuity and VCS were measured in each eye of 74 children in Study 1, and 323 children in Study 2. Hair and urine samples were collected in Study 2 and analyzed for Hg and As. Children attending schools for the LD scored significantly lower than controls on VCS, whereas visual acuity was normal. The deficit was greatest at mid- to high spatial frequency. In Study 2, significant VCS deficits were seen in exposed children at low to mid-spatial frequency, even though visual acuity was slightly above control level. Regression analyses showed that VCS had no relationship to As, but an overall significant negative correlation with hair Hg. However, analyses by district showed that mean Hg concentration was higher in control than exposed children, and that the relationship of VCS to Hg was significant only in the exposed population. The results of Study 1 indicated that behavioral VCS testing is practical in young children, and suggested that vision may be compromised in LD children. In Study 2, VCS was lower in children living in Teplice, but the causal agent(s) remains unknown. Separate mechanisms are suggested by differences in Study 1 and 2 VCS effects.

MECHANISM OF DITHIOCARBAMATES POTENTIATION OF 1-METHYL-4-PHENYL
-1,2,3,6-TETRAHYDROPYRIDINE (MPTP) NEUROTOXICITY
S. Bachurin1, N. Lermontova1, E. Shevtzova1, L. Petrova1, L. Solyakov1, T. Serkova1, T. Singer2, R. Ramsay2
1Institute of Physiologically Active Substances, Chernogolovka, 142432, Russia 2University of California, San Francisco, USA
A lot of dithiocarbamate (DTC) derivatives are widely used in industry, medicine and agriculture. Recently it was shown that diethyldithiocarbamate potentiates neurotoxic effect of MPTP, which induces parkinsonism-like neurological disorder in human and in some laboratory animals. In the present work we have studied structure-activity relationships of DTC derivatives action on neurotoxic effect of MPTP in vivo and on the the key steps in the MPTP mechanism of action; particularly, on monoamine oxidase (MAO) catalysed oxidation of MPTP, yielding 1-methyl-4-phenylpyridinium (MPP) and the following interaction of MPP with dopamine uptake system and with mitochondrial respiration chain. Among DTC studied the highest potentiation effect in experiments in vivo vivo was revealed for diisopropyl-, dicyclohexyl- and diethylderivatives, whereas number of their close analogs show no influence on MPTP neurotoxicity. It has been shown, that the potentiation of MPP-metabolite intracellular toxicity by DTC was a result of mitochondria respiration alteration (inhibition and disruption) and change in dopamine uptake system functioning due to DTC derivatives membranetropic action.
SUMMARIZING THE RESULTS OF NEUROTOXICITY TESTING: PHASE ONE - DATABASE DEVELOPMENT AND CHARACTERISTICS
K.M. Crofton1, W.F. Sette2 and D.O. Norris2
1 Neurotoxicology, U.S. EPA, Research Triangle Park, North Carolina
2 Office of Polution Prevention, Pesticides and Toxic Substances, U.S. EPA, Washington, DC
In 1985 the Office of Toxic Substances of EPA promulgated a set of Neurotoxicity Testing Guidelines, which were subsequently revised and published in 1991 by the Office of Pesticide Programs. Under the auspices of test rules and consent agreements for industrial chemicals, and data call-ins for pesticides, companies have submitted over 100 studies that contain data collected accord- ing to these guidelines. This presentation outlines an effort to collate and analyze these data. The first phase will be the construction of a database that will allow tracking of datasets. The second phase will involve detailed data entry (i.e., study results) and methodological analyses (e.g., meta-analysis). Progress to date includes an alpha version of a program to track datasets, written in Microsoft Access®. The program tracks chemicals based on the CAS number, chemical name and EPA Chem Code number (for pesticides). Each dataset is recorded in a datafile and contains information on the submitting company, report title, authors, year and basic information on the testing methodology (e.g., species, duration of exposure, test methods) and a simple description of the results. Initial efforts have been limited to those datasets containing Neurotoxicity Screening Battery data (e.g., functional observational battery, motor activity and neuropathology). Future plans involve expansion to other neurotoxicity related datasets (e.g., OPIDN studies) and publications in the open literature.
EEG CHANGES CAUSED BY SUBCHRONIC LEAD TREATMENT IN RATS
L. Nagymajtenyi, H. Schulz, and I. Desi
Department of Public Health, Albert Szent-Gyoergyi Medical University,Szeged, Hungary
Lead is a well-known neurotoxic compound causing characteristic symptoms in central and peripheral nervous system in both acute and chronic intoxication. During the last decades, chronic low-level environmental lead exposure has become more important than the occupational one. Although there are some functional neurological symptoms among the signs of poisoning, the effect of lead on the bioelectric processes in the central and peripheral nervous system has not been properly investigated. Male Wistar rats were treated by gavage with 80.0, 160.0 and 320.0 mg/kg lead (in form of lead acetate) for 4, 8 and 12 weeks. EEG of the anaesthetized animals (1000.0 mg/kg urethane) was registered with silver electrodes put directly on the primary somatosensory, visual and auditory areas. EEG was on line analyzed by a numeric analyzer program (yielding mean amplitude and frequency, EEG index and frequency band power spectrum of the recorded waves) and by a Waterfall program. There were dose-dependent changes of the investigated parameters which became more expressed by the end of the experiment. Mean amplitude of the EEG had a moderate change, while changes in the frequency (mean value, frequency band power spectrum) were stronger. On the basis of these data we suppose that the chronic, low level exposure can influence electrophysiological activity within the human brain.

EFFECT OF SUBCHRONIC MERCURY EXPOSURE ON EEG OF RATS
I. Desi, L. Nagymajtenyi and H. Schulz
Department of Public Health, Albert Szent-Gyoergyi Medical University,Szeged, Hungary
Due to its neurotoxic effect, acute and chronic mercury exposure causes clinical symptoms in the central and peripheral nerve system. In the present study, we investigated the changes in EEG of rats subchronically treated with mercury. Male Wistar rats were given by gavage 0.4, 0.8 and 1.6 mg/kg Hg (in form of HgCl2) for 4, 8 and 12 weeks. EEG of the anaesthetized animals (1000.0 mg/kg) was recorded by silver electrodes placed directly on the primary somatosensory, visual and auditory areas. Parameters of the recorded EEG, calculated by a numerical analyzer computer program, were: mean amplitude, mean frequency, EEG index and power spectrum of the frequency wave bands. Changes of these parameters, depending on the dose and length of the treatment, were manifest for EEG index and power spectrum but amplitude changes were moderate. No clinical sign of mercury poisoning was seen. Based on the above observations we suppose that the chronic low-level exposure can affect brain electrical activity in animals and also in humans.

SENSITIVITY OF NEUROTOXICITY SCREENING METHODS: EFFECTS OF LEAD ON BEHAVIORAL ENDPOINTS AND BRAIN LEVELS OF GLIAL FIBRILLARY ACIDIC PROTEIN1
B.M. Kulig, J.H.C.M. Lammers, E.M.G. Hoogendijk and K.J.v.d. Berg
Department of Neurotoxicology & Reproduction Toxicology, TNO Nutrition and Food Research, Rijswijk, The Netherlands
Although the sensitivity of the developing nervous system to the effects of lead exposure is well known, effects in the adult organism have not been extensively studied. Very high exposure levels in humans have been associated with encephalopathy and signs of peripheral neuropathy and in more recent studies, reversible changes in peripheral nerve conduction velocity at present day occupational exposure levels have been described (Muijser et al., 1987). The purpose of the present study was to evaluate the effects of lead in adult animals and to obtain further information on the sensitivity of behavioral and neurochemical assessment methods for detecting neurotoxicity. Four groups of rats were dosed with lead acetate at 0, 4, 8 or 12.5 mg/kg i.p., 5 days/week for 4 weeks. Neurological and behavioral changes were evaluated using a standardized functional observational battery and automated motor activity assessment prior to exposure, at the end of the dosing period, and at 2 weeks post-exposure. In addition, concentrations of glial fibrillary acidic protein (GFAP) were measured in discrete brain areas using ELISA-based techniques 4 weeks following the termination of exposure. Results indicated that repeated lead exposure resulted in dose-related decreases both in the pattern and amount of motor activity, decreased rearing, signs of motor dysfunction, and changes in arousal. Dose-related increases in GFAP of up to 100% of control values were also found in frontal and occipital cortex, hippocampus and striatum. Taken together, these results demonstrate the sensitivity of these end-points for detecting relatively low levels of lead exposure in adult rats. 1This work was supported by the Netherlands Ministry of Social Affairs and Labour (SoZaWe) and the Netherlands Ministry of the Environment (VROM).
SOCIAL PLAY BEHAVIOR IN JUVENILE RATS AFTER IN UTERO OPIOID EXPOSURE TO MORPHINE
Raymond J.M. Niesink Louk Y.M.Y. Vanderschuren and Jan M. van Ree2
1Faculty of Natural Sciences, Open University, Heerlen and
2Department of Pharmacology, Rudolf Magnus Institute for Neurosciences, Faculty of Medicine, Utrecht University, Utrecht, The Netherlands
Changes in analgesia, play behavior, sexual behavior and responsiveness to stress and stimualants have been reported in rodents treated in utero with opiates. During development the endogenous opioid receptors are present in a tonic balance in the mammalian nervous system. The development of this balance is particularly sensitive to prenatal administration of opioid agonists and antagonists. The motivational and rewarding aspects of play behavior are controlled by endogenous opioid systems; low doses of opioid agonists stimulate play behavior, whereas administration of opioid antagonists inhibit play behavior. We have analyzed the effect of morphine during the prenatal development of endogenous opioid systems by examination of play behavior in juvenile rats. The doses of morphine used neither affected gestation of pregnant mother rats nor motor- or sensoric development of the juvenile rats. Levels of pinning and social grooming behavior in juvenile rats were elevated after prenatal opioid exposure to morphine. To study these changes in more detail, social play was investigated using a sequential analysis in prenatally morphine- and saline-exposed pairs.

GLUCOCORTICOIDS ENHANCE OXYGEN RADICAL ASSOCIATED NEUROTOXICITY
L.J. McIntosh and R.M. Sapolsky
Department of Biological Sciences, Stanford University, Stanford, CA
Modern populations are constantly exposed to a variety of compounds in the workplace and the environment that promote formation of reactive oxygen species (ROS) within susceptible tissues. Due to its high oxygen consumption, the brain may be particularly vulnerable to oxidative damage and degeneration. Agents that impact cellular oxidative homeostasis would therefore be expected to alter the toxicity of ROS generating compounds. We are testing this hypothesis using endogenous stress hormones, glucocorticoids, to perturb neuronal homeostasis, and adriamycin to generate oxygen radicals. Glucocorticoids (GCs) are hormones secreted by the adrenals in response to stress, and are also prescribed clinically to control inflammatory and autoimmune disorders in millions of people annually. Therefore, high GC levels may not be uncommon in individuals exposed to low levels of toxic compounds. Also, GCs appear to act on cellular pathways relevant to ROS, as seen by their potentiation of neurodegeneration following insults such as stroke, hypoglycemia, or seizure. Using rat primary neuronal culture, we determined neuronal susceptibility to adriamycin toxicity by cell counting (using MAP-2 staining) and biochemical correlates. Dichlorofluorescein fluorescence confirmed ROS generation after adriamycin administration. Physiological levels of GCs (up to uM concentrations) in the culture media exacerbated adriamycin toxicity. Further, we assayed antioxidant enzyme activities from brain regions of rats either GC supplemented, or adrenalectomized to eliminate circulating GCs. These tests showed that GC level affected enzyme activity, and the pattern of difference was unique for each enzyme. Our results indicate that stress hormones may directly affect pathways involved in oxygen radical toxicity, and studies of these pathways could indicate points at which antioxidant intervention would be useful.

MICROGLIAL RESPONSIVENESS AS A SENSITIVE MARKER FOR HEAVY METAL NEUROTOXICITY
F. Monnet-Tschudi, M.G. Zurich, E. Pithon, B. Pardo and P. Honegger
Institute of Physiology, University of Lausanne, Lausanne, Switzerland
In the search for early markers of neurotoxicity, the responses of microglial cells, macroglial cells and neurons were analyzed in vitro, using aggregating brain cell cultures of fetal rat telencephalon as a model. Aggregates were treated during an early developmental period with concentrations of trimethyltin (TMT) at or below the limit of cytotoxicity. Microglia were found to be the most sensitive cell type, since already at 10-9 M of TMT an increased number and clustering of Griffonia simplicifolia-positive cells could be observed. At 10-8 M of TMT, an increased staining for glial fibrillary acidic protein, characteristic of a gliosis, was found; as well as a decrease in synaptic proteins (synapsin I and synaptophysin) content in synaptosomal fractions. A decrease in neuron-specific enzymes was observed at the highest concentration tested, 10-6 M of TMT. Microglia may be the first target of TMT and their reaction may trigger the responses of other cell types. On the other hand, it cannot be excluded that discrete neuronal changes could act as an activating signal for microglial cells. A similar sensitivity of microglial cells was found after treatment with other heavy metal compounds, e.g., lead acetate, mercury chloride, and monomethylmercury chloride. The present findings show that microglial responsiveness can be detected prior to any sign of neuronal degeneration, and therefore may serve as a sensitive indicator for heavy metal neurotoxicity in the brain.

ATTENUATION OF DOPAMINERGIC ACTIVITY IN NUCLEUS ACCUMBENS OF RATS EXPOSED TO LEAD
A.L. Jadhav and S.V. Kala
Department of Pharmaceutical Sciences, College of Pharmacy & Health Sciences, Texas Southern University, Houston, Texas USA
Long-Evans hooded (21-day-old, male) rats were exposed to either sodium acetate (control) or 50 ppm lead acetate in deionized distilled water for 90 days and tissue contents of dopamine (DA) were studied in nucleus accumbens (NA). Pb measurements were carried out using atomic absorption spectrophotometry equipped with graphite furnace while DA was measured with HPLC coupled with electrochemical detection. The exposure protocol produced blood Pb levels of 18+2 mg/dl, brain Pb levels of 350+20 ng/g wet tissue; and resulted in significant reduction in DA contents (50%) in NA. These studies were extended to examine whether the reductions in DA contents were associated with alterations in basal and stimulus (100 mM K+) induced release of DA in this brain region employing the microdialysis technique. After the 90-day exposure period, rats were anesthetized with urethane and 2 mm microdialysis probes were implanted in NA (AP 2.7 mm, L 1.5 mm, V -7.5 mm) using a stereotaxic apparatus and perfused with artificial CSF (aCSF) at a rate of 2 ml/min. The perfusates were analyzed for DA. Consistent with the reductions of DA contents observed in Pb treated rats the microdialysis studies indicated significant reductions in basal and potassium induced release of DA in Pb treated rats. These results indicate that subchronic exposure to Pb results in reduced dopaminergic activity in NA. Supported by ATSDR/MHPF Grant No. U50/ATU398948-03.

NEUROBEHAVIORAL PERFORMANCE OF 4TH-GRADE CZECH CHILDREN LIVING IN DISTRICTS WITH VARYING LEVELS OF AIR POLLUTION
D. Otto1, I. Skalik2, H.K. Hudnell1, D. House1, R. Sram2
1Health Effects Research Laboratory, U.S. EPA, Research Triangle Park, NC
2Regional Institute of Hygiene, Prague, Czech Republic
Northern Bohemia is one of the most polluted areas in Central Europe. Elevated ambient levels of SO2, NOx, PAHs and heavy metals occur as a result of intensive mining and combustion of brown coal for power generation. Sram (1991) hypothesized that in utero exposure to these chemicals causes functional changes in the nervous system expressed as developmental disorders or behavioral dysfunctions. Finger tapping performance was better in 8th-grade children from the reference district of Prachatice than the mining district of Teplice. Weak associations of hair mercury and neuro-behavioral performance were observed in 2nd-grade children from these districts. 7 tests from the Neurobehavioral Evaluation System (Letz 1991)--finger tapping, hand-eye coordination, continuous performance, visual digit span, symbol-digit substitution, pattern comparison and switching attention--were administered to 504 4th-grade children (165 from Teplice, 161 from Prachatice; and 178 from Znojmo, a district where brown coal is not used). Children from Prachatice and Znojmo performed better than children from Teplice on coding (p=.001) and digit span (p=.018) tests. In conclusion, evidence of poorer neurobehavioral performance has been found in several cohorts of children living in a heavily polluted mining district of Northern Bohemia.DISCLAIMER: This is an abstract of a proposed presentation and does not necessarily reflect EPA policy.

INTERACTION BETWEEN TOLUENE AND ACETYL SALICYLIC ACID: EFFECTS ON THE AUDITORY SENSITIVITY AND TOXIKOKINETICS IN RATS
A-C. Johnson1, G. Linder1, A. Löf2 and G. Johansson2
Departments of 1Neuromedicine and 2Toxicology, National Institute Occupational Health, S-171 84 Solna, Sweden
Rats were exposed either to toluene by inhalation at two dose levels (800 ppm 14 h/day or 1000 ppm 16 h/day, 10 days), and/or to acetyl salicylic acid (ASA) by gavage (100 mg/kg, twice daily, 10 days). The loss of auditory sensitivity was measured by auditory brainstem response (ABR) or distortion product otoacoustic emissions (DPOEs). The results show that ASA permanently potentiates the loss of auditory sensitivity produced by toluene. This was seen as raised ABR thresholds and lowered DPOEs amplitudes in the rats exposed to both toluene and ASA. This interaction is most prominent at the higher level of toluene exposure but present also at the lower level. In a second experiment the toxicokinetics after exposure to toluene (1000 ppm 16 h/d) with or without ASA exposure (100 mg/kg, twice daily) was investigated in rats exposed from 1 up to 10 days to evaluate the possible metabolic interaction between the substances.

DISTAL DEMYELINATING NEUROPATHY IN RATS TREATED WITH ALORACETAM, A NOOTROPIC AGENT:MORPHOLOGICAL STUDIES IN VIVO AND IN VITRO
N. Deutschlander, U. Schindler, D. Bury, H.W. Muller
Pharma Research, HOECHST AG, Frankfurt, CASSELLA AG, Frankfurt, and Neurological Clinics, University of Dusseldorf, Germany
A nootropic agent, Aloracetam (a pyrrol-3-aldehyde-derivative), effective at low doses in learning and memory experiments in mice and rats, caused signs of clinical neuropathy when administered chronically to rats at a high dose of 400 mg/kg/day. Light- and electron-microscopical evaluation of peripheral nerves revealed a severe distal neuropathy, characterized by myelin destruction. Morphometric analysis of the axonal filament and tubule densities did not show initial focal increase of cytoskeletal components compared to controls, as observed in axonopathies. Additional in vitro studies using spinal ganglia explants of rat showed inhibition of neurite outgrowth and of adhesion to different substrates by Aloracetam, but to a lesser extent by a metabolite lacking the aldehyde function. As was shown in this system, the compound induces a dose-dependent demyelination process by myelin and Schwann cell stripping followed by axonal retraction. Thus, this neurotoxic model differs from classical examples of distal peripheral neuropathy (e.g. haxacarbons, organophosphates) since it primarily involves the myelin-axon interaction.

SUSCEPTIBILITY OF VARIOUS AREAS OF THE NERVOUS SYSTEM OF HENS TO CP-INDUCED DELAYED NEUROPATHY
W. Classen, M. Rauch, E. Weber, F. J. Krinke
CIBA-GEIGY Ltd., Toxicology, 4332 Stein, Switzerland
Sensitivity of in-life parameters and susceptibility of various areas of the chicken nervous system to delayed neuropathy induced by triorthocresylphosphate (TOCP) was assessed. Groups of hens were exposed to a singe oral dose of TOCP of 50, 200, or 500 mg/kg and the animals observed for 21 days., Perfusion fixed , paraffin embedded tissue sections stained with Bodian's silver and luxol blues or semi-thin epoxy sections stained with toluidine blue were examined. Sciatic and tibial nerves, lumbosacral, midthoracic, and upper cervical spinal cord, medulla oblongata and cerebellum were examined using a semiquantitative scoring system. Ataxia and body weight loss occurred in high-does animals only, while dose-related changes were observed in the distal tibial nerve, medulla oblongata and cerebellum. Ataxia was correlated best with neuropathy in tibial nerves, while nerve fibers in the cerebellum were the most sensitive towards TOCP-induced delayed neuropathy. The particular susceptibility of spinocerebellar neurons was recognized long ago 1, but has been neglected in routine neurotoxicity tests and respective guidelines. Optimal sensitivity of toxicity tests is a prerequisite for risk assessment, can be cost efficient, and nowadays should be a main interest of animal welfare in order to reduce the animals' suffering. Based on these data, histopathological examination of longitudinal paraffin plastic sections of upper cervical spinal cord represents an optimally sensitive and cost efficient test requirement that should be considered for future adaptation of guidelines.

NEUROTOXIC EFFECTS OF ORGANOPHOSPHATE ON NEUROTRANSMITTERS: SIGNS OF POISONING
Gudrun E. Cassel, Ann Gornasson-Nyberg, Britt Karlsson, Lena Karlsson and Stig Jacobsson
Division of Biomedicine, Department of NBC Defense, National Defense Research Establishment, S-90182 Umea, Sweden
This study was undertaken to assess any correlation between signs of poisoning, determined by behavioural observation, and striata neurotransmitter levels in two different rat strains after soman poisoning,. In the present study, the technique of Microdialysis was used to determine the extracelluar levels of some neurotransmitters following subcutaneous administration of soman (90 ug/kg). A comparison between Sprauge Dawley and Wistar rats were done. Overt signs of poisoning were noted during the experiment. Noradrenaline, dopamine, DOPAC,, HVA and 5-HIAA in the dialysate samples were analyzed using HPLC. In both group's of rats, the inhibition of AChE activity in different brain tissues were determined. Dopamine levels increased as the severity of signs of poisoning increased in Wistar rats not by Sprauge Dawely rats. The increased dopamine might be the result of seizure spread into the basal ganglia. It is possible that individual differences in energy flow following electroconvulsive shock or in the spreading of seizure activity to the basal ganglia influence DA output. This lead to the conclusion that soman-induced non-cholinergic effects are extremely individualized between different strains, probably due to different sensitivity to organophosphate-induced seizure.

TIME-COURSE OF FUNCTIONAL CHANGES AND CHOLINESTERASE INHIBITION FOLLOWING A SINGLE ORAL DOSE OF CHLORPYRIFOS IN RATS
V.C. Moser1, A.C. Nostrandt2, and S. Padilla1
1Neurotoxicology Division, U.S. EPA, Research Triangle Park, NC
2Curriculum in Toxicology, University of North Carolina, Chapel Hill, NC
For neurotoxicity screening, the time of peak effect (TOPE) should be determined before acute studies are undertaken. To accurately define the time course of behavioral effects following a single dose of chlorpyrifos (CPF), we dosed rats (adult, male Long-Evans) with either vehicle (corn oil) or CPF (80 mg/kg) via oral gavage. We then tested them using a functional observational battery, motor activity, and measured blood cholinesterase (ChE) levels at these time points: 1, 2, 3.5, 6.5, 24, 72, or 168 hrs after dosing (n=5/treatment/time point). Functional changes produced by CPF were observed as early as 1 hr, and considerable recovery was evident at 24 hr; no signs were seen at 3 and 7 days. Autonomic effects peaked at 2 hr, whereas motoric changes, depressed pain perception, tremors, and hypothermia were maximal at 3.5 hr. Pilot studies using abbreviated behavioral observations had also indicated a TOPE of 3.5 hr. Blood ChE levels were maximally suppressed from 1 to 24 hr, with only slight recovery by 3 and 7 days. Thus, the time-course characteristics of CPF depended partially on the test measure. Moreover, the clinical effects of CPF did not parallel the time-course of the blood ChE inhibition; although both were maximally affected at approximately 3.5 hr after dosing, blood ChE recovered markedly slower than did the clinical effects.

NEUROPSYCHIATRIC SYMPTOMS AMONG FORMER MERCURY EXPOSED WORKERS
T. Mathiesen*, D.G. Ellingsen, and H. Kjuus
Department of Occupational and Environmental Medicine, Telemark Central Hospital, N-3700 Skien, Norway* Present address: Section of Neuropsychology, Blakstad Psychiatric Hospital, Pb. 143, N-1371 Asker, Norway
The aim of this cross-sectional investigation was to study prevalence of diffferent neuropsychiatric symptoms among chloralkali workers after ceased mercury exposure. The index group comprised 75 previously exposed males below 65 years of age. Exposure assessment was based on more than 2300 U-Hg measurements and included cumulative dose estimates and peak values, and the mean time since cessation of exposure was 11.6 years. These workers were compared with 52 unexposed referents frequency-matched for age. All subjects completed four self-reporting questionnaires: SCL-90-R (Symptom Checklist List 90-Revised), Q16 (The Scandinavian Questionnaire), GSCL (Giessen Subjective Complaints List) and POMS (Profiles of Mood States), and a semistructured interview. The overall prevalence of the majority of self-reported symptoms was significantly higher in the mercury exposed group compared with the referents. At the interview higher prevalences were reported for dizziness, hand tremor and reduced finger accuracy. On the Q16 eight questions (on concentration, memory, mood, and fatigue) differentiated between groups. On the GSCL all four subscales and the overall pressure score discriminated between groups, as did six of nine symptom dimensions and the general severity index on the SCL-90-R. No clearcut dose-response relationship was seen, neither for cumulative dose nor for peak values. These results are consistent with prior studies on ongoing exposure to inorganic mercury. Possible factors that may explain the lack of a clearcut dose-response relationship and the high level of distress, despite the time elapsed since exposure, will be discussed.

INORGANIC AND ORGANIC MERCURY DIFFERENTLY AFFECT CENTRAL MUSCARINIC RECEPTOR BINDING AND ENTERIC CHOLINERGIC TRANSMISSION
A.F. Castoldi1, S.M. Candura1, E. Messori1, M. Tonini1, Lucio G. Costa1 2 , L. Manzo1
1University of Pavia and Toxicology Unit, Clinica del Lavoro Foundation, Pavia, Italy
2Department of Environmental Health, University of Washington, Seattle, WA
Mercurials are known to adversely affect both the central nervous system and the gastrointestinal functions by mechanisms involving alterations of the cholinergic transmission and/or a dysregulation of intracellular Ca++ homeostasis. The central effects of methylmercury (MeHg) and mercuric chloride (HgCl2) were investigated on the binding to the cholinergic muscarinic receptor subtypes M1 and M2, as labeled by their selective antagonists [3H]telenzepine and [3H]AF-DX 384, in the rat cerebral cortex. MeHg and HgCl2 inhibited the antagonist binding to M1 and M2 sites in a competitive and noncompetitive manner, respectively. HgCl2 (IC50: 5.0 ± 1.1 µM) was a more potent inhibitor of [3H]AF-DX 384 binding than MeHg (IC50: 149 ± 48 µM). The affinity of the agonist carbachol for the M1 and the M2 sites was decreased in the presence of MeHg, but not of that of HgCl2. Enteric mercury effects were studied on the cholinergic contractile activity of longitudinal muscle-myenteric plexus strips from guinea-pig distal ileum. Micromolar (0.3-300 µM) concentrations of either HgCl2 or MeHg inhibited electrically induced twitch contractions, as well as 100 nM acetylcholine-elicited submaximal contractions, suggesting that mercurials possess antimuscarinic properties also in the intestine. By contrast, only HgCl2 enhanced the basal tone of electrically stimulated preparations, and this effect was, at least in part, mediated by an increase in intracellular Ca++ concentrations, since it was noncompetitively antagonized by both nifedipine and dantrolene. In conclusion: a) Hg central and peripheral actions differ depending on its chemical form b) several Hg functional effects on the GI tract may result from its antimuscarinic properties, as described centrally.

440KD ISOFORM OF BRAIN ANKYRIN AS A SENSITIVE MARKER FOR THE NEUROTOXICITY OF METHYLMERCURY
M. Kunimoto and T. Suzuki
Environmental Health Sciences Division, National Institute for Environmental Studies,
16-2 Onogawa, Tsukuba, Ibaraki 305, Japan
440kD isoform of brain ankyrin, 440kD ankyrinB, is a neuron-specific protein and is localized in the molecular layer in rat cerebellum1. When rat cerebellar neurons matured for 7 days in vitro were exposed to methylmercury at 0.1uM for 48 hours, viability of the cells was unaffected. However, the immunocytochemical staining of 440kD ankyrinB, which is confined to axons of the neurons, almost completely disappeared, while that of MAP(microtubule-associated protein)2, which is localized in dendrites and cell bodies, and of GFAP(glial fibrillary acidic protein), a marker for astroglial cells coexisting in the culture, remained unchanged. These observations were confirmed by the determination of each protein expressed in the culture by immunoblot analysis. In cerebella from 16-day old rats treated with methylmercury (10 times at 10 mg Hg/kg body weight, p.o.) the amount of 440kD ankyrinB was significantly reduced (89% of control), whereas changes in the amount of MAP2 and GFAP were insignificant. These results clearly indicate that selective loss of axonal protein, 440kD ankyrinB, is associated with the degeneration of cerebellar neurons induced by methylmercury both in vitro and in vivo. 440kD ankyrinB, therefore, should be useful as a specific and sensitive marker for the neurotoxicity of methylmercury, which could partly be ascribed to its selective effects on neuron-specific proteins including 440kD ankyrinB. 1 Kunimoto, M. et al. (1991) J. Cell Biol. 115, 1319-1330.

CELL TYPE-SPECIFIC TOXICITY OF THE FOOD MYCOTOXINS OCHRATOXIN A (OTA) AND FUMONISIN B1 (FB1) IN AGGREGATING BRAIN CELL CULTURES
B. Schilter1, M.G. Zurich2, F. Monnet-Tschudi2, A. Huggett1 and P. Honegger2
1Department of Quality and Safety Assurance, Nestec Ltd Research Centre, Vers-chez-les Blanc,
2Institute of Physiology, University of Lausanne, Lausanne, Switzerland
OTA and FB1 are two natural mycotoxin contaminants of food crops, however their neurotoxic potential is poorly characterized. We have examined their effects in three-dimensional (aggregating) brain cell cultures prepared from 16-day fetal rat telencephalon. The cultures were treated during 10 days, either at an immature stage (day 5 to day 15) or at a differentiated stage (day 25 to 35). Cell type-specific effects were studied by measuring the enzymatic activities of choline acetyltransferase, glutamic acid decarboxylase, glutamine synthetase, and 2',3'-cyclic nucleotide 3'-phosphohydrolase. Astroglial reactions were determined by immunofluorescence studies of glial fibrillary acidic protein (GFAP), and microglial reactivity was estimated by cytochemical staining, with Griffonia simplicifolia lectin. OTA was found to be highly cytotoxic, particularly in immature cultures. A drastic decrease in cell type-specific enzyme activities was observed at concentrations = 50 nM in immature cultures and at concentrations = 2.5 µM in differentiated cultures. Concomitant with these modifications was a significant decrease in GFAP immunoreactivity. In the immature cultures, microglial cells appeared to be the most sensitive cell type, forming large clusters within the aggregates at concentrations = 2 nM. In contrast to OTA, FB1 exhibited low toxicity. In immature cultures, at the concentrations used, (up to 40 µM) none of the parameters tested was affected, whereas in the differentiated cultures, FB1 (= 1 µM) decreased GFAP immunostaining.

ELECTROPHYSIOLOGICAL CHANGES INDUCED BY ALLYLCHLORIDE (AC)
K. Xie, S. Gao & K. Sun
Institute of Toxicology, Shandong Medica University, Jinan 250012, P.R. China
AC could elicit the axonal degeneration of peripheral nerve for workers exposed long periods and at low concentration in chemical factories. The clinic examinations in chronic poisoning patients could find the symptoms of pain at legs and arms, falling easily when walking, and a sense of distance reduced. Electrophysiological tests found the motor conduct velocity of nervus peroneus communis was slowed and the distant latent period longer. EMG showed muscle contract failure and contract voltage decreased. To further understand the pathogenesis of clinic findings, the present paper explored the changes of resting membrane potentials (RMPs) of NG108-15 cell line, Na+ and K+ currents of mouse rib perineural sheath in vitro and compound action potentials (CAPs) of rat sciatic nervetrunk in vivo induced by AC. The results showed: RMPs decreased with AC added. RMPs in control were 40.50 + 4.39 (-mV), however, RMPs were equal to 34.30 + 3.69 (-mV), 30.29 + 4.33 (-mV), 25.13 + 4.29 (-mV), 9.36 + 8.75 (-mV) and 1.80 + (-mV) as AC was added to 19.2, 38.3, 76.5, 153 and 306 mmol/L and exposed 24 h respectively. Na+ currents were slightly less, but K+ currents markedly decreased. The K+ current waveform became smaller than that in control after added AC 61.2 mmol/L 10 min. When AC was added to 122.4 mmol/L for 10 min, the K+ current waveform disappeared. CAPs of rat sciatic nervetrunk were also decreased. CAPs were lessened 65% compared with control after injected AC 150 mg/kg/day and kept 9 days by subcutaneous injection. The results prove that animal experimental effects correspond with the clinic findings in patients and explain that the electrophysiological show might be correlated with RMPs and K+ channel. The recent study showed the second negative waveform is not only associated with K+ current, but also with Ca2+ inward current. So the electrophysiological findings induced by AC might be related with Na+, K+ and Ca2+ unhomeostasis inside and outside nerve on.

ASSESSMENT OF NEUROTOXIC DISORDERS AT HAZARDOUS CHEMICAL SITES IN THE UNITED STATES OF AMERICA
R. W. Amler1 , W. K. Anger2 , O. J. Sizemore2 , J. A. Lybarger1
1 Agency for Toxic Substances and Disease Registry, Atlanta, Georgia, USA 2Oregon Health Sciences University, Portland, Oregon, USA
Neurotoxicants are extremely prevalent at hazardous chemical sites in the United States. To assess the effects of site-related neurotoxicants in populations living or working near such sites, the Agency for Toxic Substances and Disease Registry (ATSDR), a Public Health Service agency of the U.S. Department of Health and Human Services, has developed consensus test batteries for adults and children. ATSDR convened experts from North America to discuss the complex issues associated with neurobehavioral testing in populations and to propose basic and focused test batteries. The Adult Environmental Neurobehavioral Test Battery (AENTB) was adapted from the World Health Organization's Neurobehavioral Core Test Battery, individual tests of human performance, and the computerized Neurobehavioral System. Pilot testing of the AENTB demonstrated a mean total testing time of 58.0 minutes (standard deviation = 9.6) and a sample size requirement of fewer than 140 participants (a = 0.05, 95% power) to detect a 20% difference between study groups. ATSDR has administered the AENTB (and other neurobehavioral test batteries) to hundreds of adults at health study sites in six states (California, Colorado, Idaho, Massachusetts, Nebraska, and North Carolina), and has prepared a field manual to foster consistent use and interpretation of the AENTB. ATSDR also has developed a similar battery (Pediatric Environmental Neurobehavioral Test Battery, PENTB) for basic screening of children 1 through 16 years of age. The PENTB is composed of informant-based and performance-based assessment procedures, emphasizing the importance of procedures appropriate to a child's developmental stage. In all age groups, it is also crucial to assess family, cultural, economic and other potential confounding variables. Direct experience with the AENTB and PENTB, in concert with other organ-specific test batteries, will enhance ATSDR's ability to identify important end-organ effects that may result from exposure to neurotoxicants in the environment.

COMPARISON OF EXCRETION KINETICS OF PYRROLE-LIKE SUBSTANCES AND FREE 2,5-HEXANEDIONE IN RAT URINE ON 5-WEEK M-HEXANE EXPOSURE
A.P. M. dos Santos and M.C.C. Batoréu
Laboratory of Toxicology, Faculty and Pharmacy, 1600 Lisboa, Portugal
Six male Wistar rats were administered 15 doses of n-hexane (46,2mmol/Kg, per dose) by gavage, during 5 weeks. The 24 h urine was collected before (control value), during and after the exposure. In the present study pyrrole-like-substances (PLS) were analyzed in every sample with a spectrophotometric method (Erhlich's reagent) and 2,5-hexanedione (2,5-HD) was analyzed by a gas-chromatographic method without acid hydrolysis (free 2,5-HD). The results show an increase of PLS and 2,5-HD, 24th after the first dose. The PLS present maximum values after the 2nd or 3rd dose, in each week, confirming the metabolic saturation of n-hexane. The period between each week is not sufficient to eliminate the PLS, so the 1st 24th urine collected after those intervals, reflect the exposure of the previous week. The observed excretion kinetics of PLS is much slower than that of 2,5-HD. Three weeks after the last exposure, PLS can still be analyzed in urine. The results obtained suggest that PLS could be a good index of the accumulation of 2,5-HD when there is a repeated exposure to n-hexane, the typical human exposure in Occupational Toxicology.

PRELIMINARY CHARACTERIZATION OF CNS LESIONS INDUCED BY BIPIPERIDYL MUSTARD IN MICE
G. Sturman1 , P. Freeman1 and M.G. Simpson2
1 Dept. of Life Sciences, University of East London, London E15 4LZ, UK
2 Zeneca Central Toxicology Laboratory, Alderley Park, Macclesfield, Cheshire SK10 4TJ, UK
Bipiperidyl mustard (BPM) has been reported to produce obesity in rodents via an inflammatory action on the hypothalamus (Caffyn, Z.E.Y., 1972, J. Path., 106, 49-56) with the animals becoming obese within 10-12 weeks. We have previously shown (Sturman, G., 1991, Human & Exp. Toxicol., 10, 505-506) that, in mice, BPM causes weight loss and hypophagia for several days before producing a dose-related increase in weight with no accompanying hyperphagia. In the current study, groups of BKTO male mice (20-30g) received BPM 20 mg/kg in 0.1 M borate buffer pH 9.0 (s.c.), and were given the dye Evan's Blue by i.v. injection (7.5 mg in 0.5ml) just prior to killing to study the leakage of the blood brain barrier (BBB) induced by BPM. The results of dye injection showed a clear "blueing" of the hypothalamus in mice injected with BPM 5 hours previously, but not with vehicle. This blueing was minimal 24 hours after injection. Other groups of mice were killed at 1, 3, or 7 days after dosing with 20mg/kg BPM or vehicle (s.c.) only. At necropsy, brains were removed and fixed by immersion in 3.7% phosphate buffered formalin. Each brain was trimmed into 7 transverse blocks, the tissues processed routinely into paraffin wax and 5um cut sections stained with alumhaernatoxylin and eosin. The neuropathological changes seen consisted of focal areas of colliquative necrosis in the dorso-lateral septal nuclei and to a lesser extent, in the ventro-lateral hypothalamic nuclei of the forebrain. Mitotic activity in capillary endothelial cells and a sparse cuffing of hypertrophic astrocytes were associated with the peripheral areas of the lesions. There was minimal inflammatory cell infiltration of the anterior hippocampal formation, mainly the dentate gyrus. These changes were not seen at 24 hours after dosing but were present subsequently at 3 and 7 days. All brains from control mice were within normal histological limits. Our data confirms the presence of a mild hypothalamic lesion and an initial leakage of the BBB. However, the marked degree of damage seen later in the septum confounds a simple interpretation of BPM causing obesity in mice by lesioning appetite control centers in the hypothalamus. Moreover, neurotoxicant induced neuronal cell necrosis in the CNS does not commonly initiate an inflammatory response. Further work is under way to characterize the lesion and study the pathogenesis of BPM-induced CNS damage in mice.

THE INFLUENCE OF EXPOSED TO MERCURY ON INTELLIGENTAL DEVELOPMENT OF CHILDREN IN PARENTS
L. Hong, Chen Chen et al.
Laboratory of Industrial Hygiene & Toxicology, Department of Preventive Medicine, Henan University, Zheng Zhouk, Henan 450052, P.R. CHINA
Test of abstract: The 156 offsprings (below 6 years old) whose parents exposed to mercury vapor (Mercury Group MG), male 79, female 77, and 48 control group's children (CG) were investigated on intelligental development by DDST (Dever Developmental Screening Test). the mercury vapor concentration of working environment: Mean 0.0185 mg/m, range 0.001-0.0060 mg/m. The results showed that intelligential development level of MG was lower than CG., The adverse effects of exposed to mercury on intelligental development of offsprings in mother were more serious than the effects in fathers. Among the four items of motor behavior, adaptive behavior, language behavior and personal development level of children whose urine mercury (UM) more than 0.01 mg/l were lower that the children whose UM less than 0.01 mg/l. Our study showed that the mercury was harmful to intelligental development of their children when the parents exposed to mercury vapor.

NEUROBEHAVIORAL PROFILE OF SUBCUTANEOUSLY ADMINISTERED MK-801 IN THE RAT
G.C. Haggerty and G.D. Brown
G.D. Searle & Co., Skokie, IL 60077
MK-801 (dizocilpine) is a NMDA antagonist known to cause vacuolization and a hypermetabolic state in the neurons of the posterior cingulate and retrosplenial cortices of the rat within 2 - 4 hours of acute dosing. The objective of the present work was to characterize the acute behavioral effects produced by doses of MK-801 known to cause vacuolization. Female rats, approximately 9 weeks of age, (8/group) received one of three doses of (+)MK-801 (0.1, 0.25, or 0.5 mg/kg). Behavioral evaluation was conducted from 15 - 120 minutes 24 and 48 hours and 7 days post dosing using a functional observational battery. Animals from the 0.1 and 0.25 mg/kg groups initially exhibited increased open field activity and stereotyped circling and head weaving. Ataxia, impaired righting response and depressed forelimb and hindlimb extension responses became evident in these groups between 15 - 30 minutes post dosing and continued to be seen out to 2 hours. At the 0.5 mg/kg dose, ataxia and impaired righting and extension responses were seen as early as 15-30 minutes post dosing with the animals becoming completely immobile by 2 hours. In conclusion, acute dosing with MK-801 produced a biphasic response with severe impairment of neuromuscular coordination evident within two hours of dosing. This neurobehavioral profile is similar to what has been reported in the rat with phencylidine (Haggerty et al., Tox Appl. Pharm. 75:444-453, 1984), a dissociative anesthetic that is also a NMDA antagonist.

MICRODIALYSIS STUDY OF EFFECTS OF TOLUENE ON THE BRAIN NEUROTRANSMITTERS AND METABOLITES OF RATS
T. Honma
Department of Occupational Diseases, National Institute of Industrial Health, Ministry of Labour, Tama-ku, Kawasaki 214, Japan
Microdialysis technique was applied to detect changes in brain neurotransmitters induced by organic solvent in freely moving rats. Brain microdialysis is used to evaluate pharmacological effects of drugs and chemicals1 on the central nervous system. Rat brain was implanted with microdialysis probe, and the probe tip was located in the striatum. The probe was perfused at the rate of 1 ul/min. Toluene in olive oil was injected to rat intraperitoneally at doses from 500 to 2000 mg/kg. Neurotransmitters and metabolites in the perfusate were detected in HPLC. We could obtain peaks of acetylcholine(ACh), DOPAC, 5-HIAA, etc. Following the injection of toluene, ACh decreased and 5HIAA increased at 1000 mg/kg or more. No changes were observed at 500 mg/kg. These results mean that toluene decreases ACh release from nerve terminals in the striatum.1 T. Honma. Industrial Health 30, 47-60, 1992.

DOPAMINE (DA) METABOLISM IN PC12 CELLS EXPOSED TO MANGANESE (Mn) AT DIFFERENT OXIDATIVE STATES
R. Alinovi, M.V. Vettori, A. Mutti, S. Cavazzini, A. Bacchini and E. Bergamaschi
Laboratory of Industrial Toxicology, University of Parma Medical School, Italy
The present study was aimed at assessing the role of Mn valency state in Mn-induced changes in DA metabolism by PC12 cells. Mn(II)Cl2, Mn(III)acetate, and Mn(IV)O2 were used for these experiments. PC12 cells were incubated for 3, 24 and 72 hours to Mn nominal concentrations ranging from 10-8 to 10-4 M in 24-well plates containing 2 x 105 cells/well. Supernatants and cellular materials were then separated and immediately processed for the analysis of dopamine (DA), its metabolite 3,4-dihydroxyphenylacetic acid (DOPAC), and tyrosine hydroxylase (TH) activity. Lactate dehydrogenase (LDH) activity was determined as an index of cell death. In parallel experiments, Mn-containing medium (10-6 M) was removed and cells incubated for further periods with Mn-free medium to evaluate the reversibility of observed changes. At the end of the experimental periods, none of Mn-exposed cultures showed appreciable reduction in cell viability as compared to their respective controls. After exposure to Mn(II) and Mn(III), irreversible and dose-dependent decreases in extra-cellular but not intra-cellular DA were apparent. 10-4 M Mn(II) caused the disappearance of DA and DOPAC from the medium.The same effect was caused by 10-5 M Mn(III), the dose-effect relationship being shifted towards lower dose levels. Mn(IV) induced a parallel and dose-dependent decrease of DA and DOPAC concentrations in both intra- and extra-cellular compartments. Such an effect was reversible after removal of Mn from the medium. Intracellular DA depletion by Mn(IV) was associated with a dose-dependent inhibition of TH activity ("nominal" ED50: 10-5 M). Multiple interferences on DA metabolism are caused by Mn. Mn(II) and Mn(III) seem to block DA secretion without affecting DA turnover rate. Mn(IV) seems to cause DA depletion and aspecific (secondary) changes in secretion rates. Further studies are necessary to understand the mechanisms underlying the differential effects of various Mn compounds on DA metabolism.

ACUTE EFFECT OF INHALATION EXPOSURE TO CARBON MONOXIDE ON SCHEDULE-CONTROLLED OPERANT BEHAVIOR IN RATS
M. Miyagawa and T. Honma
National Institute of Industrial Health, Ministry of Labour, Kawasaki, Japan
The acute effect of carbon monoxide (CO) exposure on a steady-state operant behavior (bar-pressing maintained on a VI 60" schedule of food reinforcement) was repeatedly measured in (a) rats exposed to various concentrations of CO (500, 1000, 1500 and 2000 ppm) for 1 h and (b) rats exposed to 1500 ppm for different periods (1, 2 and 4 h). Measurements were made continuously before, during and after the exposure period. Abrupt cessation of the response was produced by exposure to 1000 ppm or higher concentrations of CO. Recovery from its effects was observed as sudden resumption of responding during the post-exposure period. Duration of exposure required to produce complete response inhibition was closely correlated with the exposure concentration. The post-exposure interval required for recovery of the response (from the termination of exposure to abrupt response resumption) was also correlated with the exposure concentration. This post-exposure response recovery interval, however, was constant and independent of duration of exposure when the concentration was fixed at 1500 ppm. In order to correlate these behavioral changes with the internal index of CO exposure, blood carboxyhemoglobin (HbCO) levels were determined under several exposure conditions corresponding to those of the behavioral observations. It was found that HbCO levels were within the 33-43% range when response recovery occurred, suggesting the existence of a critical HbCO level (threshold) associated with the drastic behavioral change. Hence, these results support the view that blood HbCO is an important determinant of the acute behavioral effects of CO.

PERIPHERAL MARKERS OF CATECHOLAMINERGIC SYSTEMS AMONG WORKERS OCCUPATIONALLY EXPOSED TO TOLUENE
A. Smargiassi1, A. Mutti2, E. Bergamaschi2, S. Bélanger1, G. Truchon3 and D. Mergler1
1Centre pour l'étude des intéractions biologiques entre la santé et l'environnement (Cinbiose), Université du Québec à Montréal, Canada
2Laboratory of Industrial Toxicology, University of Parma, Italy
3Institut de Recherche en Santé et Sécurité au Travail (IRSST), Québec, Canada
In a preliminary study of 8 men occupationally exposed to toluene, whose median urinary hippuric acid level was 0.31 mmole/mmole creatinine (range 0.15-1.97) and of 10 control subjects, platelet monoamine oxidase activity (MAO), serum dopamine b-hydroxylase activity (DBH) and serum prolactin (PRL) were measured. Although MAO, DBH and PRL were similar in the 2 groups, negative dose-effect relationships were noted between urinary o-cresol and DBH (R2=0.59, p=0.03) and between urinary o-cresol or hippuric acid and PRL (R2=0.59, p=0.03; R2=0.52, p=0.04), within the exposed group. DBH catalyses the hydroxylation of dopamine (DA) into norepinephrine and PRL secretion is tonically inhibited by tuberoinfundibular DA. The concomitant decrease in these peripheral markers are consistent with the reported increase in DA content of various brain regions following inhalation exposure to toluene. Interestingly, covalent adducts of cresol with an active-site tyrosine residue, leading to a unstable oxy-radical intermediate, have been recently implicated in DBH inactivation. Owing to the limited sample size, these findings should be confirmed by further experimental and epidemiological studies.

INHIBITION OF MUSCARINIC RECEPTOR: STIMULATED PROLIFERATION OF RAT CORTICAL ASTROCYTES AND HUMAN ASTROCYTOMA CELLS BY ETHANOL
M. Guizzetti, P. Costa and L.G. Costa
Department of Environmental Health, University of Washington, Seattle, WA
Activation of acetylcholine muscarinic receptors coupled to phosphoinositide metabolism has been shown to cause proliferation of glial cells (Nature 340, 146, 1989). As inhibition of muscarinic receptor - stimulated phosphoinositide metabolism has been suggested as a mechanism involved in the developmental neurotoxicity of ethanol, we have further characterized this mitogenic response in rat cortical astrocytes. These cells express mRNA for m3 and m2 muscarinic receptors, as measured by RT-PCR. Carbachol causes a time - and dose - dependent increase in proliferation, measured by incorporation of 3H-thymidine, with a maximal stimulation of 280% of basal. The full agonist methacholine and acetylcholine had similar effects. The mitogenic effect of carbachol was antagonized by atropine and other muscarinic antagonists. The phorbol ester TPA also elicited a strong proliferative response (500% of basal). The protein kinase C (PKC) antagonist H7 and down - regulation of PKC by preincubation for 24h with 300 nM TPA, significantly inhibited carbachol - induced proliferation. Ethanol inhibited basal and serum - stimulated astrocyte proliferation by a maximum of 30-40% at concentrations of 100-200 mM. On the other hand, carbachol - stimulated astrocyte proliferation was inhibited by ethanol at much lower concentrations, the IC50 being less than 10 mM. Similar effects of muscarinic agonists and of ethanol were also observed in the human astrocytoma 132 1N1 cell line, which expresses m2, m3 and m5 muscarinic receptor mRNA. These results suggest that acetylcholine - stimulated astrocyte proliferation may be a sensitive and relevant target for ethanol and may play an important role in ethanol developmental neurotoxicity, particularly in ethanol - induced microencephaly (Supp. in part by AA-08154).

THE EARLY TIME COURSE OF L-2 CHLOROPROPIONIC ACID INDUCED GRANULE CELL NECROSIS IN THE RAT CEREBELLUM
M.G. Simpson, I. Wyatt, A.J. Gyte, P.S. Widdowson, H.B. Jones and E. A. Locke
Zeneca Central Toxicology Laboratory and Safety of Medicines Department, Zeneca Pharmaceuticals, Alderley Park, Macclesfield, Cheshire, SK10 4TJ, UK
We have recently shown that the chemical intermediate L-2 chloropropionic acid (L-2-CPA) induces selective necrosis of the granule cell layer of the rat cerebellum 36 hours after a single oral administration of 750 mg/kg L-2-CPA. The histological and ulstrastructural characteristics of the cerebellum appeared normal at 24 hours after the above dosing regime. We have defined a critical morphological window between apparent normality and obvious neuropathological change of between 24 and 36 hours. In the current study, groups of male Alderley Park rats (Wistar derived, approximately 200g body wt.) were administered a single oral dose of 750 mg/kg L-2-CPA. Groups of rats were sampled for neuropathology at 24, 27, 30, 33 and 36 hours after dosing. A control group received water alone. Rats were killed by a lethal dose of barbiturate, followed by perfusion with Karnovsky's fixative (pH7.4). Multiple longitudinal blocks of cerebellum were embedded in araldite resin following post fixation with osmic acid semithin 1-2um sections cut and strained with toluidine blue. The cerebellar samples from control and the 24 hour L-2-CPA treated rats appeared histologically normal. Cerebellar granule cell necrosis characterized by nuclear condensation and karyorrhectic change was obvious and multifocal at 30 hours after dosing, the necrotic foci appearing mostly in the vermis. No necrosis was seen at 27 hours. The earliest ultrastructural change which was regarded as being truly representative of treatment-related damage to the granule cells and not of artefactual origin occurred at 36 hours. At earlier times, the swelling of astroglial cells was observed in both control and L-2-CPA treated animals and in the latter was not qualitatively different to that seen in the former.

CYCLOATE INDUCED NEUROPATHOLOGICAL AND NEUROCHEMICAL CHANGES IN RAT BRAIN
M.G. Simpson, P.S. Widdowson, K. Tseng, Y. Scanion, and E.A. Lock
Zeneca Central Toxicology Laboratory, Alderley Park, Macclesfield, Cheshire SK10 4TJ, UK
Our laboratory has established that the herbicide cycloate (an S-alkylmonothiocarbarnate), given as a high single oral dose (2000mg.kg) to rats, causes selective neuronal cell death in two regions of the rat forebrain, the pyrifom cortex and in the posterior region of the ventrolateral dentate gyrus. No other parts of the rat nervous system are affected. We report the time course of these changes together with some preliminary neurochemical observations. Groups of 10 male Alderly Park rats (Wistar derived, approximately 200g. body wt., 6-8 weeks old) were given a single oral dose of either 0 or 2000mg/kg cycloate and killed for neuropathological study at 2,3,4,8, 15, 29 days after dosing using a regime of perfusion fixation with modified Karnovsky's fixative, followed by routine paraffin methodology. Seven transverse levels of brain were examined from each rat. Cycloate-induced neuronal cell death was seen in the pyriform cortex from day 2 and persisted through to day 29, the lesion appearing to arise in the most anterior region of the pyriform cortex and subsequently spread in a posterior direction. Neuronal cell death was also seen in the dentate gyrus from day 2, persisting until day 15. In the early stages, the neuronal cell death resembled apoptosis, characterized by condensation of nuclear material, cell shrinkage and strong cytoplasmic eosinophilia, at time points of 2,3,4 and to a lesser extent at 8 days. At day 15 and 29 and to a lesser extent at day 8, the cell death resembled necrosis, ie karyorrhectic nuclei with pale irregular cytoplasm. Modest microglial accumulation was associated with the neuronal cell lesions. In control brains, an occasional apoptotic body was seen in both the pyriform cortex and dentate gyrus. Our results display a novel neurotoxicant induced brain lesion, which analyzed temporarily, indicates that a re-triggering of apoptosis may contribute to the pathogenesis of cycloate induced CNS neurotoxicity. Neurochemical measurements were made at day 14 following cycloate administration. We could detect no changes in the density of kairate, NMDA or muscarinic receptors in the hippocampus using qualitative receptor autoradiography. Measurement of amino acid concentration using HPLC were made in a number of brain regions at 0.5 hrs, 3 days, and at 14 days following cycloate administration. We demonstrated a transient increase in glutamine concentration at day 3 which fell back to control values by day 14. No other amino acids were observed to be altered as compared to controls following cycloate administration. These data suggest that cycloate does not target excitatory amino acids receptors in the hippocampus nor is there evidence for excitatory amino acid mediated cell death as a result of the increased glutamate concentrations.

SURVEILLANCE OF EARLY NEUROTOXIC DYSFUNCTION
D. Mergler1, G. Huel 2, S. Bérlanger1, R.M. Bowler3, G. Truchon4, D. Drolet4 and C. Ostiguy4
1CINBOISE , Université du Québec à Montréal, Montréal, Québec
2U-169, Institut national de la santé et de la recherche médicale (INSERM), Villejuif, France 3San Francisco State University, San Francisco, CA
4Institute de recherche en santé et en sécurité du travail, Montréal, Québec
Surveillance of early neurotoxic alterations was undertaken in 3 reinforced plastics plants, with a view to preventive intervention. Using a longitudinal study design, exposure parameters (environmental styrene in the respiratory zone at jobsite and end-shift mandelic acid (MA)) and neurobehavioral performance (Neurobehavioral Core Test Battery and Field assessment : Sensory Tests), were assessed at time zero (T0); recommendations were made to reduce exposure at jobsites with the highest risk. Reassessment was made two years later (T2). Of the 128 workers (85% of total population at T0); 81 were still employed at T2; of these, 63 (78%) returned for testing. Those who returned were older (p<0.05), more senior (p<0.001) with higher MA at T0 (p<0.05) than the others. Analyses, performed on the T0 - T2 differences, showed improvement in exposure parameters in one factory (F1), where lower levels were observed at T2 for styrene (p<0.05) and MA (p<0.001). Workers in F1 (n=30) performed better (p<0.05) at T2 for short term memory, perceptuo-motor speed, motor precision and manual dexterity; they reported more vigor (p<0.05) and less aggressiveness (p=0.07). This was not the case for the workers from the other plants. The T0 - T2 difference in color vision was correlated to the difference in MA (F=4.13; p<0.05). These findings suggest that group surveillance of early nervous system changes for jobs with exposure to neurotoxins, using a sensitive neurofunctional test battery, may be useful for preventive intervention.

NEUROBEHAVIORAL EFFECTS OF SOLVENTS AND CIRCADIAN RHYTHMS
E. Kiesswetter, A. Seeber, M. Blaszkewicz, R.R. Vangala
Institute of Occupational Physiology at the University of Dortmund, Germany
From chronotoxicological studies it is known that the susceptibility of the human organism to several chemical agents reveals a circadian variation, but neurobehavioral aspects are less investigated. Shiftwork is a field which offers the possibility to study the interaction between time and exposure. Investigations in this area are of practical importance because time-related synergistic effects of shiftwork and exposure may lead to critical conditions for the safety and health of the workers. Two shiftwork studies were performed with (1) 8 shiftworkers exposed to a mixture of organic solvents and 8 controls working on morning, and afternoon shifts, and (2) 8 shiftworkers exposed to acetone and 8 controls working on morning, afternoon, and night shifts. Repeated measurements of exposure, body temperature, well-being (tension, tiredness, annoyance,complaints), and performance (simple reaction, color word vigilance) were taken during each shift and during several shift cycles. The air concentrations (1) of the single components of the solvent mixture were clearly below and (2) of acetone were near the German occupational exposure limit values. In both studies body temperature revealed circadian patterns. Ratings of well-being and performance showed circadian and exposure effects. However the solvents contributed mainly in an additive way to the negative effects of shiftwork. There was no convincing evidence that the neurobehavioral exposure effects vary with time of day.

THE EFFECT OF CALCIUM, CALMODULIN (CaM) AND Ca2+/CaM-DEPENDENT PROTEIN KINASE (Ca2+/CaM-PK) IN NERVE CELL INDUCED BY ALLYL CHLORIDE (AC)
K. Xie, S. Gao, L. Zhang & K. Sun
Institute of Toxicology, Shandong Medical University, Jinan 250012, P.R. China
The workers for long periods and in low concentration exposed to AC could elicit chronic toxic peripheral neuropathy. Pathological examinations found that microtubules (MT) and neurofilaments (NF) in axon of patient's nervus peroneus communis were changed in morphology, and the numbers of MT and NF decreased markedly in per 10 um2. To deeply understand the pathogenesis of toxic peripheral neuropathy induced by AC, the present paper explored intracellular free calcium ([Ca2+]i), CaM and Ca2+/CaM-PK in vitro, utilizing ebryon brain cell primary culture. The results showed: [Ca2+]i was increased sharply with AC added after 3 min. [Ca2+]i in control was 135.10 + 6.28 nmol/L. As AC was added to 40.72, 81.17, 161.27 and 318.35 mmol/L, [Ca2+]i rose up to 255%, 423%, 600% and 1200% compared with control respectively. CaM, using ELISA method to determine chicken embryon brain cells after added AC 24 h, was decreased as AC concentrations added. CaM in without AC group 45.83 + 19.14 µg/g, 1.225 mmol/L AC group 26.65 + 11.08 µg/g, 2.45 mmol/L group 23.98 + 10.02 µg/g. Ca2+/CaM-PK activities were increased sharply as AC was added. Control group was 148.5 + 39.9 pmol/mg protein min, 1.02 mmol/L AC group was equal to 573.5 + 185.2, 2.04 mmol/L AC rose to 890.7 + 345.7 pmol/mg protein min respectively. CaM dependant on intracellular free calcium, Ca2+/CaM-PK rely on Ca2+ and CaM in cell. Increased Ca2+ in cell could activate CaM and Ca2+/CaM-PK. Activated Ca2+/CaM-PK could make microtubule associated protein 2 (MAP2) and t factor phosphorylation and result in microtubule disassembly. The study reveals that chronic toxic peripheral neuropathy induced by AC might be associated with the changes of Ca2+ and the activity of Ca2+/CaM-PK increased in nerve cells.

EFFECTS OF TOLUENE AND TRICHLOROETHYLENE ON VIGILANCE IN RATS: CxT RELATIONSHIPS FOR SENSITIVITY,AND RESPONSE LATENCY
P. J. Bushnell W. M. Oshiro and B. Padnos
Neurotoxicology Division, U.S. EPA, Research Triangle Park, North Carolina 27711
The risk of inhaling organic solvents is presently assessed on the basis of lifetime exposure to average vapor concentrations, in which short-term high-level exposure episodes are ignored. This strategy leads to rational predictions of risk if the product of the concentration (C) and time (T) of exposure yields constant effects on health. The validity of this assumption for neurotoxicological endpoints has not been evaluated. Thus, the behavioral effects of acute toluene and trichloroethylene (TCE) inhalation were assessed in rats in terms of C and T. Rats were trained in a vigilance task to press either of two retractable levers for food reward. A press on one lever produced food on trials containing a target (a brief, unpredictable auditory or visual stimulus); a press on the other lever produced food on trials lacking a target. Response time (RT) and the signal detection index of sensitivity (SI) were calculated across 25-min blocks during repeated daily 40 to 75-min tests conducted in air with 0, 1000, 1500, or 2000 ppm toluene or 0, 400, 600, 800, 1200, 1600, 2000 or 2400 ppm TCE. In air, both measures remained stable across blocks. In vapors of both solvents, SI declined and RT increased as functions of both C and T. Plotting these endpoints as a function of the CxT product showed that SI depended more upon C than T. In contrast, RT appeared to increase monotonically as a function of the CxT product. These results indicate that the nature of the endpoint used can affect the relative contribution of C and T to the outcome, and that the effect of short-term, high-level exposure to solvents may be adequately described as the CxT product for some endpoints, whereas C may be more important than T for others.

THE ION CHANGES IN NERVE CELLS AND AXON INDUCED BY ALLYL CHLORIDE (AC)
K. Xie, L. Zhang, S. Gao & K. Sun
Institute of Toxiocology, Shandong Medical University, Jinan 250012, P.R. China
Chronic poisoning induced by AC could damage peripheral nerve system and result in "axonal degeneration" of peripheral nerve among exposed workers. Electrophysiological examination could find that the motor conduct velocity (MCV) of nervus peroneus communis in chronic poisoning patients was slowed, and the distant latent period became longer. In toxic rat the wave of sense evoked potentials (SEPs) fell down, and the latent period longer. In order to deeply study the abnormal electrophysiological phenomena induced by AC, the present paper, using fluorescence molecular probe Fura-2/AM, SBFI/AM and electron probe X-ray microanalysis, determined the changes of cytosolic ionized calcium ([Ca2+]i) and sodium ([Na+]i) in chicken embryon brain nerve cells after added AC 3 min, analyzed the variations of Ca, Na, K, Mg, Cl and P elements in rat sciatic nerve axon and brain stem after injection AC 8 days. The results showed: [Ca2+] i was increased sharply with AC added. [Ca2+]i in control was 135.10 + 6.28 nmol/L. As AC was 40.72, 81.17, 161.27 and 318.35 mmol/L, [Ca2+]i rose up to 345.62 + 15.24**, 572.50 + 30.50**, 810.21 + 75.86** and 1671.33 + 161.33 nmol/L** respectively, P <0.01 compared with control. [Na+]i was also added with AC increased [Na+]i fluorescence intensity values (FIV) without AC were 1.045 + 0.009, however, the FIV of 40.72, 81.17, 161.27 and 318.35 mmol/L AC were 1.048 + 0.014**, 1.175 + 0.011** and 1.280 + 0.034** respectively. In rat sciatic nerve axon the element variations of Ca, Na, K, Mg, Cl, P with AC concentration added. Ca increased markedly. In control group was 2.758 + 0.143%, 100 mg/kg group 3.229 + 0.132%*, 200 mg/kg group 3.669 + 0.158%**. Na slightly increased. K decreased sharply, control was 1.177 + 0.045%, 100 mg/kg group 0.910 + 0.033%**, 200 mg/kg group 0.908 + 0.021%**, Mg increased and P element decreased. In rat brain stem the variation tendencies of Na, K, Ca, Mg, Cl, P elements were the same as the changes in rat sciatic nerve axon. Na increased sharply with AC added. Control group was 5.404 + 0.134%, 100 mg/kg group 6.093 + 0.190%**, 200 mg/kg group 6.175 + 0.174%**. Ca was also increased. Ca were 2.698 + 0.132% in control; 100 mg/kg group 3.034 + 0.159%*, 200 mg/kg group 2.988 + 0.083%*. K and P were lower. The results reveal that electrophysiological changes caused by AC might be associated with the variations of Na, K, Ca elements in nerve axon.
* P<0.05, ** P<0.01 compared with control.

SELECTIVE NEUROTOXICITY IN RAT DISSOCIATED CEREBRAL CULTURES CAUSED BY THE IONOPHORE LASALOCID
N. Safran 1, R. Haring2, D. Gurwitz2, A. Shainbergand2 and A. Shahar2
1 Koret School of Vet. Med. Hebrew University of Jerusalem 76100
2 Israel Institute for Biological Research, Ness-Ziona 70450 and Bar-Ilan University, Ramat-Gan 52100, Israel
An in vitro model of dissociated cerebral cultures, prepared from g15 rat fetuses, was used to further characterize the neurotoxic effects caused by the antibiotic ionophore Lasalocid - X-537A (Safran et al, 1993). Following a short exposure (2-4 h) to Lasalocid, cells were examined morphologically (by phase contrast microscope and scanning electron microscopy (SEM)) and biochemically. The morphological damage caused by lasalocid (1-2 µM) consisted of swelling of perikarya, followed by cytolysis of most neurons present in the cultures. The neuronal damage was dose dependent, noticeable above the concentration of 0.5 µM, and was much more pronounced in established cultures (14 days in vitro - div) than in younger ones (7 div). Unlike neurons, glia and other non-neuronal cells present in the cultures were not damaged by exposure to 2 µM Lasalocid. Moreover, the drug was not toxic, for cultures of rat astrocytes and C6 glioma cells. In contrast to Lasalocid, another calcium ionophore A-23187 (calcimycin), applied at a concentration of 1 µM, virtually destroyed both neuronal and non-neuronal cells, within 1 h. Biochemical analysis showed an increase of Ca2+ influx in cultures exposed to 1.5 µM Lasalocid. The Lasalocid neurotoxic effects mediated by Ca2+ influx were not inhibited by 10 µM nimodipine (a calcium channel antagonist), but exclusively blocked by 10 µM MK-801 (a non-competitive NMDA receptor/channel antagonist). The neurotoxicity induced by Lasalocid was confirmed by measurements of lactate dehydrogenase (LDH) release. Moreover, release of arachidonic acid (AA), which has been implicated in NMDA receptor functioning and toxicity, was also studied. Lasalocid (1.5 µM) induced the release of both LDH and AA (8 and 4 fold of control values, respectively), and this was blocked by MK-801. In contrast, MK-801 did not block the release of either LDH or AA mediated by the calcium ionophore A-23187 (1µM) in these cultures. Our observations suggest that lasalocid mediates selective neurotoxicity of cultured cerebral neurons by increasing Ca2+ influx, and implies an involvement of the NMDA receptor/channel in its mechanism of action. Safran et al, (1993). Toxic. in vitro Vol. 7, No. 4, pp. 345-352.

SUPEROXIDE ANION PRODUCTION AND TRANSENDOTHELIAL PERMEABILITY CHANGES IN AN IN VITRO MODEL OF THE BLOOD-BRAIN BARRIER
P. Lagrange1, I.A. Romero2 C. Denizotand1, and P.A. Revest3
1 Centre du Medicament, Nancy, France
2 Physiology Group, King's College, London, UK
3 Physiology Department, Queen Mary & Westfield College, London, UK
In phase I drug metabolism, NADPH-cytochrome P450 reductase transfers, simultaneously, two electrons from the cofactor NADPH+ to cytochrome P450 which can then reduce xenobiotics and produce more polar metabolites susceptible to be conjugated and eliminated by phase II enzymes. For some compounds, the transfer of the two electrons by NADPH-cytochrome P450 occurs in two steps, giving rise to a free radical intermediate by acceptance of one electron. In the presence of oxygen, the free radical intermediate can produce superoxide anion radicals (O2.-) through a univalent reduction mechanism and regenerate the parent compound, giving rise to "futile redox cycling". Rat brain capillary endothelial cells are able to produce O2.- via a redox cycling of quinone, nitroheterocycle and iminium compounds (Lagrange et al., 1994, Free Rad. Biol. Med., 17, 335). The O2.- produced target protein, polyunsaturated fatty acids and DNA and may lead to cell damage and, ultimately, death. We used a cell culture model of the blood-brain barrier consisting of the immortalised endothelial cell line RBE4 (Roux et al., 1994, J. Cell Physiol., 159, 101) grown on semipermeable filters (Falcon, U.K.) in co-culture with C6 glioma cells grown in the base of the culture wells. We measured the clearance of a radiolabelled tracer, [3H] sucrose, of these monolayers, according to the method of Dehouck et al. (1992, J. Neurochem., 58, 1790), after 30 min pretreatment with xenobiotics which undergo "futile redox cycling" in brain endothelial cells (menadione, benzylviologen, methylviologen and nitrofurazone). Cultured brain endothelial cells had an increased production of O2.- compared to control levels when incubated with all compounds. Maximum production of O2.- was observed after incubation with 4 µM menadione. Benzylviologen, methylviologen and nitrofurazone did not induce any changes in the permeability of the cell monolayers to sucrose, but a 30 min incubation with menadione gave rise to a 50% increase in sucrose clearance. The increase in permeability induced by menadione was related to superoxide production since pre-incubation with the enzymes superoxide dismutase and catalase reduced sucrose clearance to control levels, but not to cell death, as assessed by the trypan blue exclusion method. We conclude that, amongst the chemicals tested, menadione has the highest affinity for NADPH-cytochrome P450 reductase. The changes in permeability observed with menadione could be related to the production of O2.- by redox cycling of the parent compound._

ACTIONS OF FLUOROCITRATE AND 1,3-DINITROBENZENE ON BRAIN ENDOTHELIAL CYTOSKELETON AND PERMEABILITY
R.J. Rist, I.A. Romero and N.J. Abbott
Physiology Group, Biomedical Sciences Division, King's College London, The Strand, London WC2R 2LS, UK
Experimental energy-deprivation syndromes are known to have toxic effects on brain astrocytes in vivo (Cavanagh, 1988, Toxicology 49, 131-136). Some of these toxins are proposed to act on astrocytes alone (fluorocitrate) whilst others are associated with changes in the permeability of the BBB (1,3-dinitrobenzene). Fluorocitrate has been shown to alter metabolism in glial cells and this is proposed to be its main site of action (Hassel et al., 1994, J. Neurochem. 62, 2187-2194). In contrast, 1,3-dinitrobenzene has been shown to alter BBB permeability in vivo (Romero et al., 1991, Neuropath. Appl. Neurobiol. 17, 495-508). The RBE4 cell line (Durieu-Trautmann et al., 1993, J. Cell. Physiol. 155, 104-111) has proved to be a useful in vitro model of the BBB. It has been reported that a marginal F-actin distribution is required to maintain the tightness of the tight junctions in brain endothelial cells (Rubin et al., 1991, J. Cell Biol. 115, 1725-1735). In this study, the effects of fluorocitrate and 1,3-dinitrobenzene on permeability of RBE4 cell monolayers grown on filters was measured and correlated with changes in the metabolism, and F-actin cytoskeleton in the cells. Treatment of RBE4 cells with fluorocitrate and 1,3-dinitrobenzene leads to a decrease in F-actin distribution at the cell margin and a reduction in the F-actin content of the cells at concentrations >0.5mM. Fluorocitrate significantly reduced and 1,3-dinitrobenzene significantly increased the glucose consumption and lactate production of the RBE4 cells. Fluorocitrate (0.1 to 1mM) had no significant effect on RBE4 cell monolayer permeability measured by FITC-dextran or 14C-sucrose. However, 1,3-dinitrobenzene (0.5mM) significantly increased the permeability of RBE4 cell monolayers. These results show that whilst both fluorocitrate and 1,3-dinitrobenzene have significant effects on the RBE4 cell F-actin cytoskeleton and cellular metabolism, only 1,3-dinitrobenzene treatment produces increases in endothelial cell monolayer permeability. These data demonstrate that profound toxic effects on endothelial cell structure and metabolism do not necessarily indicate changes in monolayer permeability.

OXIDATIVE STRESS INDUCES NEUROTOXICITY IN RAT BRAIN REAGGREGATE CULTURES THAT CAN BE PREVENTED BY a-TOCOPHEROL
R.M. Fox, J.E. Davenport Jones and C.K. Atterwill
CellTox Centre, Division of Biosciences, University of Hertfordshire, Hatfield, Herts. AL10 9AB, UK
Oxygen-derived free radical species play a role in the pathogenesis of various neurodegenerative disorders producing 'oxidative stress' within a particular tissue. In the brain oxidative free radicals may injure neurons directly prompting a glial response or attack glial cells directly causing 'reactive gliosis', a marker of which is glial fibrillary acidic protein (GFAP). The biochemical mechanisms underlying excitotoxicity and oxidative stress may co-operate in the genesis of neuronal damage. Glutamate synthetase (GS) is an enzyme instrumental to glutamate cycling and synaptic 'buffering' since it converts glutamate to glutamine within glia cells. It has been shown that GS can be inactivated by oxidative stress, supporting the theory that it may be important in the development of excitotoxicity following oxidative stress and/or accumulation of glutamate. Whole rat brain reaggregate spheroidal cultures are organotypic (neurons, astrocytes and oligodendroglia) which display correlates with aspects of in vivo CNS development and maturation. The aim of this study was to assess the ability of oxidative stress to cause reactive toxic insults to glial cells within reaggregates and address reversibility of the responses. Oxidative-mediated damage to brain reaggregate cultures resulted in glial responses indicative of both inactivated 'buffering' enzymes such as GS and an elevation of GFAP in intact astroglia. Excitotoxicity following exposure to such toxicants may be mediated primarily by inactivation of GS. Fe2+-induced neurotoxicity in terms of these glial responses could be blocked by exposure to a-tocopherol. The role of exogenous and endogenous antioxidant protection of GS in stroke and free-radical neurodegenerative damage could be further investigated in vitro using the brain reaggregate culture model.

OXIDATIVE STRESS INDUCES NEUROTOXICITY IN RAT BRAIN REAGGREGATE CULTURES THAT CAN BE PREVENTED BY a-TOCOPHEROL
R.M. Fox, J.E. Davenport Jones and C.K. Atterwill
CellTox Centre, Division of Biosciences, University of Hertfordshire, Hatfield, Herts. AL10 9AB, UK
Oxygen-derived free radical species play a role in the pathogenesis of various neurodegenerative disorders producing 'oxidative stress' within a particular tissue. In the brain oxidative free radicals may injure neurons directly prompting a glial response or attack glial cells directly causing 'reactive gliosis', a marker of which is glial fibrillary acidic protein (GFAP). The biochemical mechanisms underlying excitotoxicity and oxidative stress may co-operate in the genesis of neuronal damage. Glutamate synthetase (GS) is an enzyme instrumental to glutamate cycling and synaptic 'buffering' since it converts glutamate to glutamine within glia cells. It has been shown that GS can be inactivated by oxidative stress, supporting the theory that it may be important in the development of excitotoxicity following oxidative stress and/or accumulation of glutamate. Whole rat brain reaggregate spheroidal cultures are organotypic (neurons, astrocytes and oligodendroglia) which display correlates with aspects of in vivo CNS development and maturation. The aim of this study was to assess the ability of oxidative stress to cause reactive toxic insults to glial cells within reaggregates and address reversibility of the responses. Oxidative-mediated damage to brain reaggregate cultures resulted in glial responses indicative of both inactivated 'buffering' enzymes such as GS and an elevation of GFAP in intact astroglia. Excitotoxicity following exposure to such toxicants may be mediated primarily by inactivation of GS. Fe2+-induced neurotoxicity in terms of these glial responses could be blocked by exposure to a-tocopherol. The role of exogenous and endogenous antioxidant protection of GS in stroke and free-radical neurodegenerative damage could be further investigated in vitro using the brain reaggregate culture model.

EVIDENCE OF A POLYAMINE SITE ON AN NMDA RECEPTOR MACROCOMPLEX ON MAST CELLS: A ROLE IN EXCITOXICITY?
W.M. Purcell, K. Doyle and C.K. Atterwill
CellTox Centre, Division of Biosciences, University of Hertfordshire, Hatfield, Herts. AL10 9AB, UK
Natural polyamines induce histamine release from rat peritoneal mast cells (RPMC) with a rank order of potency of spermine > spermidine > putrescine; compound 48/80 (C48/80), a synthetic polyamine, is regarded as a classic mast cell secretagogue. We have shown that spermine and
C48/80 also trigger histamine release from rat brain mast cells (RBMC) dispersed from thalamic tissue. However, the mechanism of action remains to be elucidated. We have explored the possibility that spermine triggers histamine release from RPMC via interaction with a polyamine site associated with an NMDA receptor macrocomplex using arcaine (a competitive inhibitor at the polyamine site on the NMDA receptor macrocomplex), ifenprodil (an NMDA receptor antagonist) and MK801 (an open-channel blocker). Arcaine, ifenprodil and MK801 co-added with the agonist inhibited histamine secretion from RPMC challenged with spermine or C48/80 in a concentration dependent manner. Ifenprodil and MK801 were more potent inhibitors of spermine-induced histamine release than C48/80-triggered histamine secretion. This data would support the hypothesis that spermine induces histamine release from RPMC by interaction with a polyamine site associated with an NMDA receptor macrocomplex located on mast cells. Whether this site is distinct from that described in brain tissue remains to be elucidated, since neither glutamate nor glycine were added to the cells. The fact that RBMC are likewise responsive to spermine and C48/80 would support the presence of an NMDA receptor macrocomplex on these cells. Mast cells offer the potential to test drug interactions at the various ligand binding sites on the NMDA receptor macrocomples. In addition, mast cells may be involved in mediating excitotoxic events.

RAT BRAIN MAST CELLS: AN IN VITRO PARADIGM FOR ASSESSING THE TOXIC EFFECTS OF NEUROTROPHIC THERAPEUTICS
W.M. Purcell, C. Westgate and C.K. Atterwill
CellTox Centre, Division of Biosciences, University of Hertfordshire, Hatfield, Herts. AL10 9AB, UK
Neurotrophic factors (NTFs) such as nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF) and ciliary neurotrophic factor (CNTF) are currently being explored as novel therapeutics in a range of neurodegenerative disorders such as amyotrophic lateral sclerosis (ALS) and Alzheimer's disease. To this end, animal studies and clinical trials have been conducted to assess the toxic effects of recombinant NTFs. It is apparent that both NGF and BDNF induce a range of adverse effects, for example inflammation, hyperalgesia, and disturbances in CNS biogenic amine levels which variously manifest as weight loss/gain, changes in feeding behaviour and general malaise. It has been demonstrated that NGF induces release of biologically active mediators, such as histamine, from rat peritoneal mast cells (RPMC). However, whether other NTFs do likewise or indeed are able to induce secretion from
other mast cells types had not been explored. We have developed a novel protocol for dispersing mast cells from rat brain tissue, in particular the thalamus which contains the highest number of mast cells in the adult rat. Rat brain mast cells (RBMC) released histamine in a concentration dependent manner in response to NTFs, with a rank order of potency of BDNF > CNTF > NGF; in contrast RPMC were refractory to the effects of BDNF and CNTF. The ability of NTFs to induce release of histamine (a neurotransmitter and neuromodulator in the CNS) from RBMC may go some way to explain some of the adverse effects apparent in vivo upon dosing with NTFs. Mast cells in vitro, and brain mast cells in particular, offer the potential to screen novel NTFs for their neuroimmunotoxic potential relevant to detecting likely clinical adverse effects in man.
IN VITRO DETECTION OF THE PERIPHERAL NEUROTOXICITY OF URAEMIC PLASMA USING PRIMARY CULTURES OF ADULT RAT DORSAL ROOT GANGLION
S.P. Williams, C. Egan, W. Butt, C.K. Atterwill and J. Tattersall1
CellTox Centre, Division of Biosciences, University of Hertfordshire, Hatfield, Herts. AL10 9AB, UK 1 Renal Unit, Lister Hospital, Stevenage, Herts., SG1 4AB, UK
Peripheral neuropathy is a common long-term manifestation of chronic renal failure and inadequate haemodialysis, with sub-clinical neuropathy being present in most dialysis patients. Although the nature of the uraemic toxin(s) is unknown, the concentration of middle molecules
(MM; 500 - 2000 daltons) in plasma is inversely correlated with nerve function. Although MM can be purified from uraemic plasma, the toxic components have not been identified. In this study we have taken plasma samples from normal volunteers and from patients with renal failure, pre-dialysis and post-dialysis and separated MM fractions by reverse phase and size exclusion separation HPLC. In order to assess the neurotoxicity of the MM fractions we have explored the use of cultures of adult rat dorsal root ganglion sensory neurons (DRGs). The model has been validated by immunohistochemistry of neurofilament protein and estimating neurite extension and arborisation in the presence of nerve growth factor (NGF). These parameters were then assessed after lesioning with, for example, acrylamide. The purpose of the project is to eventually determine the uraemic neurotoxin(s) by characterising and quantifying its neurotoxic effect on DRGs; whether such effects can be reversed by treatment with neurotrophic factors, such as NGF, will also be assessed. Further, we aim to use this model system to develop suitable molecular neurotoxicological endpoints relevant to the detection of peripheral insults.

EFFECT ON THE CONTENT OF N-ACETYLASPARTATE, TOTAL CREATINE, CHOLINE CONTAINING COMPOUNDS, AND LACTATE IN THE HIPPOCAMPUS OF RATS EXPOSED TO AROMATIC WHITE SPIRIT FOR THREE WEEKS AS MEASURED BY NMR SPECTROSCOPY
A. Stensgaard1, G. Østergaard2, C.V. Jensen1, H.R. Lam2, S. Topp1,O. Ladefoged2, P. Arlien-Soborg3 and O. Henriksen1
1 Danish Research Centre of Magnetic Resonance, Hvidovre Hospital, DK-2650 Hvidovre, Denmark 2 Institute of Toxicology, National Food Agency of Denmark, DK-2860 S_borg, Denmark 3 Department of Neurology, Hvidovre Hospital DK - 2650 Hvidovre, Denmark
Several epidemiological studies of workers occupationally exposed to white spirit show that neuropsychiatric disorders are a frequent cause of early disability pension in this population compared with non-exposed controls. In the rat, we have demonstrated that exposure to different kinds of white spirit induces changes in neurotransmitter concentrations, indices of oxidative stress, and electrophysiological parameters. Others have confirmed that acute behavioral effects can be induced by short-term high-level exposure. With NMR spectroscopy technique it is possible to study neurochemical parameters in vivo, and to examine the same subjects repeatedly over time. NMR spectroscopy was used to study the effects of organic solvents in rats. Rats were exposed to 0, 400 ppm, or 800 ppm of aromatic white spirit 6 hr/day, 7 days/week for 3 weeks. During the first week, the rats showed signs of irritation of mucous membranes, and appeared to be sedated. Both types of effect gradually diminished during the second week. The rats were examined by single volume of interest (VOI) NMR spectroscopy. N-acetylaspartate, creatinine and phosphocreatinine, and choline containing compounds were measured in the hippocampus and surrounding regions. The concentration of N-acetylaspartate for the three groups was found to be in the range of 8.2-8.5 mM with a standard deviation of 0.6-0.9. There was no difference between the three groups. In a previous study no change in the number of neurons in hippocampus was found following exposure to white spirit for six months. Since N-acetylaspartate is thought to be a marker for neurons, the results indicate that white spirit does not produce a marked neuronal loss. The NMR technique can be applied to the rat, and it is possible to obtain reasonable signal-to-noise ratios.

THE EFFECTS OF IN UTERO METHYLMERCURY EXPOSURE IN NONHUMAN PRIMATES: AN OVERVIEW OF 10 YEARS OF RESEARCH
S. G. Gilbert, K. S. Grant-Webster, and T. M. Burbacher
Department of Environmental Health, School of Public Health and Community Medicine,
University of Washington, Seattle, WA
Methylmercury (MeHg) is both a well known neurotoxicant and a global environmental pollutant. Over a decade a study began to assess the maternal reproductive and offspring developmental effects of in utero MeHg exposure in monkeys. Twenty-four adult female Macaca fascicularis were exposed to either 0, 50, 70, or 90 ug/kg/day MeHg hydroxide prior to and throughout pregnancy and produced 11, 9, 2, and 2 infants, respectively. Maternal MeHg exposure did not effect conception rate but did significantly reduce the number of viable deliveries at blood Hg concentrations above 1.5 ppm. At birth, blood mercury levels of treated infants ranged from 1.04 to 2.46 ppm were significantly higher than maternal blood Hg concentrations at delivery. There was no effect of MeHg on offspring size at birth. Neurobehavioral evaluations of the offspring during the first year of life indicated that MeHg exposed infants showed deficits in the development of object permanence, visual recognition memory, and social behavior. Physical measurements of growth continued to be collected after infancy. Beginning at about 2.5 years of age, MeHg exposed males began to show significantly decreased body weights when compared to control males. This effect persisted until the male monkeys reached approximately 5 years of age. Studies of the learning capabilities of these animals between 5 and 10 years of age did not support treatment related deficits in adult cognitive abilities. Performance of the MeHg exposed adult animals on tests of spatial and non spatial memory, and schedule controlled behavior was within the normal range. Currently, this cohort of animals is being assessed on tests of visual contrast sensitivity and somatosensory functioning. After completion of the sensory experiments, the long range plan is to evaluate decrements in performance associated with the aging process.
Supported in part by grant # ES03745 from NIEHS.