Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Cytochrome oxidase histopathology in the central nervous system of developing rats displaying methylmercury-induced… Dyck, Richard Henry 1988

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1988_A6_7 D92.pdf [ 10.47MB ]
Metadata
JSON: 831-1.0097668.json
JSON-LD: 831-1.0097668-ld.json
RDF/XML (Pretty): 831-1.0097668-rdf.xml
RDF/JSON: 831-1.0097668-rdf.json
Turtle: 831-1.0097668-turtle.txt
N-Triples: 831-1.0097668-rdf-ntriples.txt
Original Record: 831-1.0097668-source.json
Full Text
831-1.0097668-fulltext.txt
Citation
831-1.0097668.ris

Full Text

Cytochrome Oxidase Histopathology in the Central Nervous System of Developing Rats Displaying Methylmercury-lnduced Movement and Postural Disorders By RICHARD H E N R Y DYCK B . S c , The University of Lethbridge, 1981 A T H E S I S SUBMITTED IN PARTIAL FULF ILLMENT O F THE REQUIREMENTS FOR THE D E G R E E O F MASTER O F SCIENCE in THE FACULTY O F G R A D U A T E STUDIES P R O G R A M IN NEUROSCIENCE W e accept this thesis as conforming to the required standard T H E UNIVERSITY O F BRITISH COLUMBIA August 1988 © Richard Henry Dyck, 1988 In present ing this thesis in partial fulfillment of the requirements for an advanced degree at the Univers i ty of Bri t ish C o l u m b i a , I agree that the Library sha l l make it freely avai lable for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representat ives. It is understood that copying or publ icat ion of this thesis for f inancial gain shal l not be al lowed without my written permiss ion. Faculty of Graduate Studies, Program in Neuroscience The University of Brit ish Co lumb ia 1956 Ma in Mal l Vancouver, Canada V6T 1Y3 Date : August 30 , 1988 i i Abstract Sprague-Dawley rats were administered daily, subcutaneous injections of methylmercuric chloride at a dose of 5 mg/Hg/kg beginning on postnatal day 5. By their fourth postnatal week, animals exhibited a constellation of neurological s igns of motor impairment which resembled the cerebral palsy syndrome of humans perinatally exposed to methylmercury. Routine histological examination of the brain revealed no gross di f ferences between methylmercury-treated (MeHg) , normal control (NC) or weight-matched littermates. The histochemical localization of the mitochondrial enzyme cytochrome ox idase (CO) was util ized in Experiment I to examine possible alterations in the metabol ic activity of motor nuclei which might contribute to the observed movement and postural disorders. A population of intensely-staining cytochrome ox idase neurons ( ICONs) in the magnocel lular portion of the red nucleus (RMC) and interrubral mesencepha lon (IRM) were conspicuously present in all M e H g animals at the onset of motor impairment. These morphological ly, histochemical ly, and anatomical ly distinct neurons did not exhibit intense C O staining in control animals. Conversely, a significant decrease was demonstrated in the oxidative metabolic activity of many neurons in the substantia nigra, z o n a reticulata of M e H g animals. In Experiment II, the postnatal appearance of ICONs was morphometrically quantified in M e H g animals sacr i f iced at P N D 14, 16, 18, 20, 22 , or 25 . The histochemical ly-def ined onset of increased metabolic activity in ICONs was first observed on P N D 16, at least one week before the onset of cl inical s igns of neurological impairment. This was the earliest manifestation of methylmercury neurotoxicity yet descr ibed in this animal model . A subsequent four-fold increase in the total number of ICONs at P N D 18 was followed by a gradual decrease in number to P N D 25. Significantly more of the ICONs were found in the IRM than in the R M C at P N D 18 & 20. The possibil ity that the increased activity of ICONs may result from disinhibition of specif ic afferents to the red nucleus was addressed by introducing either hemidecortication or hemicerebel lectomy on P N D 10 and then morphometrically determining the deviation from symmetry in the bilateral distribution of the total number of I C O N s in the R M C and IRM at P N D 22. The distribution of ICONs was symmetrical and not different in either hemidecort icate or unoperated controls. A significant (36%) decrease in the total number of ICONs was observed in both the R M C and IRM contralateral to hemicerebellectomy. The identical ipsilateral regions did not differ from control or hemidecort icate M e H g animals . In Exper iment III, the anatomical distribution of major histocompatabi l i ty complex antigens (MHC) in the brain of MeHg animals was examined using immunohistochemical methods. M H C immunoreactivity was widely distributed throughout the brain of M e H g animals. A reas with low immunoreactivity, or lack of it, stand out and include all of the h ippocampus, tha lamus, pyriform and entorhinal cortex, and lateral cerebel lar hemispheres. Moderate staining intensity was observed in neocort ical areas, basa l forebrain, caudate-putamen and cerebel lar vermis. Strong immunoreactivity was found in red nuc leus, substant ia nigra, cingulate cortex, retrosplenial cortex, presubicu lum, parasubicu lum and vest ibular nuclei . It was suggested that the increased activity of ICONs likely contributes to the movement and postural d isorders resulting from methylmercury intoxication. The increased activity in I C O N s was determined to be, at least partially, dependent upon cerebel lar input. The results are d i scussed with reference to the toxic effects of methylmercury and specif ical ly to the susceptibil i ty of G A B A e r g i c interneurons in perinatal t rauma. Poss ib le analogies are drawn between the mechanisms of methylmercury- induced cerebral palsy syndrome and those of other developmental movement and postural disorders. 1 V T a b l e of C o n t e n t s Abst rac t List of Figures v i Acknowledgement v i i Neurotoxic i ty of Methy lmercury Introduct ion 1 H is to ry 2 Neurotoxicity in the adult population 3 Neurotoxicity in the pediatr ic population 4 Exper imental Studies 6 Thes is Rationale 1 1 Experiment I. Metabolic changes in mesencephal ic neurons demonstrated with cytochrome ox idase histochemistry Introduct ion 1 3 Materials and Methods 1 4 Resul ts 1 5 D iscuss ion 1 8 Experiment II. Postnatal changes in the total number of small, intensely-staining, cytochrome oxidase-positive neurons (ICONs) and the effect of cerebral cortex or cerebellar lesions Introduction 21 Materials and Methods 2 2 Results 2 6 Discussion 3 5 Experiment III. Immunohistochemical localization of major histocompatability complex class I and class II antigens Introduction 4 2 Materials and Methods 4 3 Results 4 5 Discussion 5 3 Conclusion 5 6 Direction of future studies 5 9 Bibliography 61 vi List of Figures Figure 1. (Page 9) Postnatal changes in body weight of methylmercury-treated (MeHg) and control animals. Figure 2. (Page 10) Compar ison of posture in a M e H g and a control rat. Figure 3. (Page 16) Cytochrome oxidase (CO) histochemistry in the mesencephalon at the level of the red nucleus. Figure 4 . (Page 17) C O histochemistry in the substantia nigra of M e H g and control rats. Figure 5. (Page 29) C O histochemistry in a horizontal section of the rubral mesencephalon. Figure 6. (Page 30) Postnatal changes in the total number of ICONs in the rubral mesencephalon. Figure 7. (Page 31) C O histochemistry in horizontal sect ions of the rubral mesencephalon in M e H g rats on postnatal days 16, 18 and 25. Figure 8. (Page 32) Postnatal changes in the number of ICONs differentiated by region. Figure 9. (Page 33) Number and distribution of ICONs in the mesencephalon of unoperated, hemidecorticate and hemicerebel lectomized M e H g rats. Figure 10. (Page 34) C O histochemistry in horizontal sect ions of the rubral mesencephalon in hemidecorticate and hemicerebel lectomized M e H g rats. Figure 11. (Page 48) M H C and G F A P immunohistochemistry in frontal sect ions through the caudal mesencephalon. Figure 12. (Page 49) High power micrographs of M H C and G F A P antigen expression. Figure 13. (Page 50) M H C immunoreactivity in frontal sect ions through the forebrain. Figure 14. (Page 58) M H C I immunohistochemistry in posterior neocortex. Figure 15. (Page 60) M H C I immunohistochemistry in the cerebel lum and medul la. Acknowledgement The completion of this manuscript owes much to the warm support and generous assistance of many friends and col leagues at' the University of Lethbridge and the Kinsmen Laboratory. I am especial ly grateful to Drs. Edith M c G e e r and Pat M c G e e r for the use of laboratory space and their generous gift of M H C and G F A P antibodies and to Drs. Steve Vincent and Haru Ak iyama for their offering of antibodies and expertise in the field of immunoh is tochemis t r y . I am thankful to Dr. John O'Kusky for providing the opportunity to work in his laboratory on a fascinating topic. His critical comments and generous ass is tance through all phases of this research have been invaluable. I am most deeply indepted to my family. I could not have accompl ished my goal without the unconditional support, encouragement and love of my parents; Henry and Katharina, my son Matthew and my partner in life, Jan ine . 1 Neurotoxicity of Methylmercury Introduction The use of mercury compounds in medications and amalgams had been recorded by the Romans prior to 33 B C . Both inorganic and organic mercurial compounds were subsequently developed as antiseptics, disinfectants and germicides to aid in the control of bacter ia and fungi. The industrial application of mercurials in the 1 9 t h and 2 0 t h centur ies has resulted in increases in the concentration of mercury into aquatic and marine environments and in the atmosphere with the potential detrimental effects extended to humans (Cass idy and Furr, 1978). The effects of inorganic or organic mercury poisoning vary as a function of their lipid solubility which endows a compound with the ability to cross biological membranes. Inorganic mercury compounds are considered only moderately toxic because of their low lipid solubility (Shamoo, 1987). The toxic effects of inorganic mercurial poisoning are general ly seen following ingestion of mercury or inhalation of its vapor s ince mercury is readily, but not efficiently, absorbed from the gastrointestinal tract and lungs. Pathology is most extensive within the gastrointestinal tract but may include edema and hemorrhages of internal organs, swollen liver and kidneys plus congested lungs (Cassidy and Furr, 1978; S h a m o o , 1987). The danger of toxicity from exposure to inorganic mercury compounds is relatively low and they are not considered significant problems in contamination of the general environment (Cass idy and Furr, 1978). The organic mercur ia ls, especial ly methylmercury, are cons idered of primary concern with regard to the environment because they pose the greatest threat to animal life. Their hydrophobic character ist ics and lipid solubility enable them to c ross membrane barriers 2 easi ly (Shamoo, 1987), including the p lacenta and blood-brain barrier (Broman, 1967). Although the pathological consequences of exposure to methylmercury are not limited to the nervous system, the severity and permanence of its effect in the central nervous system are comparatively select ive and will be the focus of this thesis. History of methylmercury poisoning Owing to their effective germicidal propert ies, organic mercury compounds were widely introduced in the early 1900's for fungus control in both industry and agriculture (Cass idy and Furr, 1978). Although the toxicity of organomercur ials had been reported in 1869 (Edwards) , the first detai led cl inical and pathological descript ion of human toxicity did not surface until 70 years later (Hunter, Bomford and Russe l l , 1940; Hunter and Russe l l , 1954). In these reports, four seed-grain workers were descr ibed as having deve loped constriction of v isual f ields, ataxia, dysarthria and impaired somesthesthet ic sense 3-4 months after the onset of exposure to methylmercury. This constellation of neurological s igns, termed the "Hunter -Russel l syndrome", is character ist ic of methylmercury poisoning in adults. Autopsy findings in one of their patients conf irmed that the pathological effects of methylmercury were largely conf ined to the central nervous system and that, consistent with the clinical s igns, lesions of the cerebral cortex and cerebel lum were predominant. Reported c a s e s of methylmercury poisoning resulting from isolated inc idences of occupat ional exposure numbered only 40 prior to 1956 (Cass idy , 1 9 7 8 ) . The increasing signif icance of research concerning the neurotoxic effects of methylmercury was revealed in 1956 as a consequence of industrial d ischarge of mercury and methylmercury into Minimata Bay , Japan (Harada, 1978). This major outbreak of methylmercury poisoning resulting from environmental contaminat ion affl icted the Methy lmercury neurotoxici ty 3 general population of the area who ate contaminated f ish. A similar incident in Ni igata, with the first c a s e s reported in 1965, resulted from the industrial contaminat ion of the Agano River (Tsubaki and Irukayama, 1977). The total number of recognized cases of what came to be known as "Minimata d isease" reached 1,775 in Minimata (Tokuomi, 1982) and 520 in Ni igata (Tsubaki and Irukayama, 1977). The most extensive outbreak of methylmercury poisoning occurred in Iraq as a result of the inadvertent consumpt ion of methylmercury-treated s e e d wheat. During the winter of 1971-72, 6500 men , women and children were admitted to Iraqi hospitals after consuming bread prepared from methylmercury- t reated grain (Bakir et a l . , 1973) . Neurotoxicity in the Adult Population The symptoms of methylmercury toxicity in the adult population of the Japan and Iraq epidemics were consistent with those reported following occupat ional exposure (Hunter, Bomford and Russe l l , 1940; Hunter and Russe l l , 1954). The initial manifestations of methylmercury toxicity, which fol lowed a latent period ranging from severa l months to one year after exposure, were paresthesia of the extremities and perioral numbness (Clarkson, 1983). Appearance of the major cl inical s igns; impairment of hand movements (lack of coordination, weakness and tremor) accompanied by dysarthria, an ataxic gait and disturbances of v is ion and hearing, occurred within 1-3 weeks of onset of the initial s igns. In some instances of continued exposure, these symptoms became aggravated and resulted in general paralys is , deformity of the l imbs, difficulty in swal lowing, convuls ions and death (Ha rada , 1978) . Neuropathological observat ions revealed the cerebral cortex to be severely affected with focal les ions in the calcar ine, precentral and postcentral cort ices (Harada, 1978). Methy lmercury neurotoxici ty 4 Neuronal degeneration was predominant in cortical layers II and IV and was accompanied by degenerat ion of the myelin sheath and glial proliferation (Takeuchi , 1985). In all c a s e s studied, there was gross atrophy of the cerebel lar folia in the lateral lobes and vermis with thinning of the cerebel lar gray matter. Neuronal degenerat ion was most severe in the granule and basket cel ls while, except for chronic c a s e s , the Purkinje cel ls were resistant to toxicity. Astrocyt ic and glial proliferation was observed in both granule and Purkinje layers but not in the molecular layer (Chang, 1977, Takeuch i and Eto, 1977). Neurotoxicity in The Pediatric Population The tragic epidemic poisonings in Japan and Iraq inevitably resulted in the exposure of infants and chi ldren to methylmercury. The first indication of a differential sensitivity to the effects of M e H g was oul ined in a report by Harada (1968) based on the Minimata outbreak. He discovered that pregnant mothers who had either slight or no symptoms of poisoning gave birth to infants who later developed severe psychomotor impairment resembling a cerebral palsy syndrome. These symptoms were confirmed in a 5 year follow-up study of chi ldren prenatally exposed to methylmercury in the Iraq ep idemic (Amin-Zak i et a l . , 1979) . Although the severity of symptoms of the cerebral palsy syndrome has been demonstrated to vary in proportion to the degree of exposure (Amin-Zaki et a l . , 1979), 100% of chi ldren known to have been prenatally exposed to methylmercury during these epidemic poisonings eventually developed psychomotor retardation later in life (Harada, 1976 in C h a n g and Annau , 1984; Amin-Zaki et a l . , 1979). The cl inical s igns of prenatal methylmercury toxicity included spastici ty, ataxia, athetosis, general tonic convu ls ions and myoclonic jerking. The persistence of some primitive reflexes (Moro's reflex and asymmetr ical neck reflex) was accompanied by a general ized hyperreflexia and Methy lmercury neurotoxici ty 5 hypertonicity of limb musc les . Indicators of normal motor development, such as grasping, crawling , s tanding, and walking, were either de layed or never ach ieved (Marsh, 1987). The children had a tendency to smile, laugh, and cry without obvious provocation and displayed an exaggerated response to sensory stimuli such as sudden noise. Their cognitive development was found to be severely impaired and language development was poor or nonexistent (Amin-Zak i et a l . , 1979). The clinical signs and neuropathology observed in children who had been exposed to methylmercury as infants were similar to those seen after in utero exposure (Takeuchi et a l . , 1979). A complete loss of motor and mental functions in severe cases of infant exposure is descr ibed as resembling the decortication syndrome. The degenerative changes compromise areas of the neocortex and cerebel lum with a regional specificity which resembles, but is much more widespread than, the effects of adult neurotoxicity (Matsumoto et a l . , 1965). The neuropathology of chi ldren who suffered prenatal exposure to methylmercury was markedly different from that observed in either adult or infantile exposure. Whi le neuronal damage following infant exposure showed considerable anatomic and regional specif icity, examinat ion of prenatally exposed brains demonstrated widespread histopathology throughout the cerebrum and cerebel lum (Chang , 1977; Cho i et a l . , 1978). A reduction of brain weight was associated with general ized thinning of the corpus cal losum and cerebral cortex (Chang and Annau , 1984). In the cerebel lum, the lateral hemispheres and vermis were grossly atrophic with thinning of both the molecular and granular layers. Upon microscopic examinat ion, the prenatally exposed brains revealed widespread neuronal heterotopias in the cerebral and cerebel lar white matter. Extensive neuronal loss in cerebral and cerebel lar cort ices was accompanied by abnormal patterns of organization and distorted alignment of many of the remaining neurons (Choi et a l . , 1978; Chang and A n n a u , 1984). Methy lmercury neurotoxici ty 6 These findings suggest that the aetiology of the methylmercury- induced cerebral palsy syndrome following prenatal and infant exposure may develop as a consequence of interference with the normal migration, posit ioning and connectivity of neurons in the developing brain, rather than the focal degenerat ion of neuronal populations descr ibed following adult exposure. The observation that infant exposure results in neuropathology intermediate to that in adult and prenatal exposure confirms that the neurotoxic consequences of methylmercury poisoning vary as a function of brain maturity (Chang and Annau , 1984). When considered together, it is apparent that two mechan isms are responsib le for the neurotoxic effects of methylmercury; one is an interference with specif ic developmental p rocesses (mitosis, migration, synaptogenesis) ; and the other is a toxic degenerat ive effect resulting in structural damage and/or cell death (Annau and E c c l e s , 1987) . A relatively smal l proportion of all infants and children exposed to methylmercury in Japan and Iraq came to the attention of medical services. It is bel ieved that only those exposed to moderate or high doses of methylmercury were ever reported (Amin-Zaki , 1979). The consequences of low exposure are, therefore, relatively unknown. It is of interest to note, however, that the incidence of idiopathic cerebral palsy d iagnosed in chi ldren in the vicinity of Ni igata and Minimata was reported to be 14-17 times higher than the Japanese national average (Harada, 1978). Experimental Studies Despite extensive investigations of the effects of methylmercury in the adult nervous system (for reviews see Takeuchi et a l . , 1957; C h a n g , 1977, 1979, 1980; Chang and Reuh l , 1982; T h o m a s and Syve rsen , 1987) relatively little is known of the pathological Methy lmercury neurotoxici ty 7 events or mechan isms for the neurotoxicity of methylmercury in the developing brain. In light of the effects of methylmercury toxicity on human children exposed prenatally or neonatally, much of the animal research currently underway in this area has been directed toward the developing nervous system. Histological examination of the brains of developing animals exposed to methylmercury has demonstrated neuropathology which resembles that seen following human developmental exposure. Prenatal exposure of cats characterist ical ly resulted in incomplete granular cel l formation in the cerebel lum, as well as abnormal cytoarchitecture in cerebral neurons (Khera, 1973), while exposure during the postnatal period produced degenerative changes in the internal granule cell layer with some loss of Purkinje cel ls and abnormal myelination (Khera et a l . , 1974). Fewer cel ls in the external granular layer of the cerebel lum following a de layed migration of these cel ls has been observed in mice following prenatal exposure (Khera and Tabakova , 1973). Although severe teratogenic pathology such as fetal resorption, gross organ defects and hydrocephalus have also been demonstrated following fetal exposure (Harris et a l . , 1972; Murakami , 1972 ; Khe ra , 1973 ; Fuyuta, et a l . , 1979; Cho i et a l . , 1988), methylmercury intoxication has also resulted in b iochemical , neurophysiological , and behavioural changes which occur at low dose levels in the absence of histopathology or marked physical debilitation (Spyker et a l . , 1972; Hughes and Sparber , 1978; Ecc les and A n n a u , 1982a, 1982b, 1987 review; Slotk in, 1985, Thomas and Syve rsen , 1987 review; Komula inen and Tuomis to , 1987 review). A recent ser ies of experiments has consistently demonstrated that the postnatal administrat ion of methylmercur ic chloride to neonatal rats (5 mg/kg/day) results in the development of movement and postural disorders by the fourth postnatal week (O'Kusky, 1985; O 'Kusky and M c G e e r , 1985; O 'Kusky et a l . , 1988a, 1988b). Of signif icant importance is the fact that the constel lat ion of cl inical s igns of neurological impairment in Methy lmercury neurotoxici ty 8 this animal model resembles that of the cerebral palsy syndrome descr ibed in human infants perinatally exposed to methylmercury (Chang , 1979; Amin-Zak i et a l . , 1979). The first s ign of impairment in methylmercury-treated (MeHg) rats was an abnormal rate of weight gain when compared to normal control animals (NC), which began one week after the onset of exposure to methylmercury (O'Kusky, 1985; O 'Kusky et a l . 1988a). An imals cont inued to gain weight, but less rapidly than N C animals, until their maximum body weight was attained between postnatal days 18-21 (Fig.1). The cl inical s igns of neurological impairment appeared 3-5 days later and are the following. Upon pass ive manipulat ion of the l imbs, M e H g animals exhibited apparent spastici ty, with hypertonicity in the flexor musc les of all 4 l imbs. Al l M e H g rats exhibited flexion deformities which were most easily v isual ized when the animal was suspended by the tail (Fig. 2). M e H g animals demonstrated hyperreflexia in the c rossed extensor reflex and in response to auricular startle. Many , but not all animals d isp layed myoclonic jerking of the hindl imbs which somet imes progressed to grand mal se izures. Visual impairment was conf irmed in all M e H g animals upon examining performance on a visual cliff and visual placing response. No gross apparent pathology was noted upon examination of the brain. Although brain weights of M e H g animals were significantly lower than those of N C animals, littermates which were weight-matched by nutritional deprivation had similar brain weights. However, they did not manifest any behavioural s igns nor did they differ significantly from N C animals in the neurochemical and ultrastructural indices which are descr ibed below. Neurochemical assays at the onset of neurological impairment have shown a significant reduction in the specif ic activity of glutamate decarboxy lase, the enzyme which synthesizes gamma-aminobutyr ic acid ( G A B A ) , in the neocortex and striatum (O'Kusky and M c G e e r , 1985; O 'Kusky et a l . , 1988b). Ultrastructural examinat ion of the brain has revealed a relatively select ive degenerat ion of aspinous and sparsely-spinous stellate neurons in the cerebral cortex (O'Kusky, 1985). Most neocort ical neurons with these morphological Methy lmercury neurotoxici ty 9 Figure 1. Postnatal changes in body weight for methylmercury-treated (MeHg) , weight-matched control (WMC) and normal control (NC) animals. Arrows I and II indicate two subcl inical changes in body weight. Arrow III indicates the onset of neurological impa i rmen t . Methylmercury neurotoxicity Figure 2. In MeHg animals the hypertonicity of limb flexors was most easily detected when the animal was suspended by the tail (A). The hindlimbs were abnormally adducted and held in full flexion with the digits tightly clenched. This hypertonicity, determined by manipulation of the limb, was symmetrical and more pronounced in the hindlimbs than in the forelimbs. When a normal control (NC) animal was suspended by the tail (B), all four limbs were held in full extension with the digits splayed. Methy lmercury neurotoxici ty characterist ic have been demonstrated to contain G A B A (Houser et a l . , 1983; Ribak, 1978; Ribak et a l . , 1981) and function as inhibitory interneurons in the cerebral cortex. The degeneration of symmetrical (e.g. type II) synapses , which has also been descr ibed in MeHg rats is consistent (O 'Kusky, 1985), s ince such morphology is ascr ibed to inhibitory boutons. The results of these studies suggest that inhibitory G A B A e r g i c interneurons are extremely suscept ib le to the neurotoxic effects of methylmercury. Thesis Rationale Abnormal i t ies of G A B A e r g i c neurotransmission in the cerebral cortex and neostriatum have been implicated in several models of movement disorder in adult animals. It has long been known that compounds that elevate brain G A B A tend to be anticonvulsants, whereas those that interfere with G A B A neurotransmission are often convulsants (Loscher, et a l . , 1983; Loscher , 1984). Recent studies have suggested that inhibition of G A B A e r g i c neurons may be involved in a genetic form of epi lepsy in rats (Roberts et a l . , 1984) and gerbils (Loscher, 1984). Morphological studies have demonstrated a preferential loss of G A B A e r g i c synapses in motor cortex at sites of epileptic foci (Ribak et a l . , 1979, 1981, 1984). Chronic, intracortical infusion of G A B A produced a complete blockade of the paroxysmal d ischarges and associated clinical s igns (Brai lowsky et a l . , 1987). In the neostriatum, injections of G A B A antagonists into the caudate nucleus increase motor activity, while corresponding injections of G A B A or G A B A agonists decrease activity (Wachtel and A n d e n , 1978). G iven the role that G A B A e r g i c neurotransmission serves in both pyramidal and extrapyramidal motor control in the adult, an impaired development of G A B A e r g i c neurons in neonates is consistent with the motor impairment observed in M e H g -induced cerebral palsy. Methy lmercury neurotoxici ty In addition to M e H g , other etiologic factors such as hypoxia and viral encephal i t is in neonates may precipitate cerebral palsy-l ike symptoms. Hypoxia in neonatal monkeys produces a selective degeneration of symmetric synapses in the motor cortex which probably originate from G A B A e r g i c interneurons (Sloper et a l . , 1980) and reduced brain levels of G A B A in rats have been shown to result from chronic hypoxia (Arregui and Barer, 1980; Franc is and Puls inel l i , 1982). Us ing neocort ical explants from developing rat brain, Romjin et a l . (1988) conf i rmed that G A B A e r g i c neurons were indeed the first neurons to die during hypoxia. Experimental models of viral encephal i t is in adult animals have shown decreased G A B A concentrations and G A D activities in the cortex and neostriatum (Bonil la et a l . , 1977; Ka taoka et a l . , 1979). Thus , it appears that an impairment of G A B A e r g i c neurons may represent a common denominator in these developmental disorders. Disinhibition of cortical efferents by means of a specif ic reduction in cortical G A B A synapt ic activity may be sufficient to provoke motor impairment by increasing activity of efferent neurons. The studies indicating that inhibitory G A B A e r g i c interneurons in the neocortex of the rat are highly suscept ible to developmental methylmercury toxicity may provide a model to investigate this hypothesis. To address this issue, the primary object ives of the exper iments which follow were to: 1. identify increases in activity of motor nuclei which receive projections from the neocortex. Metabol ic activity of neurons was histochemical ly demonstrated using a technique which stains the mitochondrial enzyme cytochrome ox idase (CO) . 2. correlate changes in CO-act iv i ty with the development of methylmercury- induced movement and postural disorders. It was hypothesized that the neocortical G A B A e r g i c dysfunction preceding the onset of neurological impairment would result in a parallel increase in metabol ic activity. 3 . confirm the source of abnormal activity. The activating influence of disinhibited projections should be eliminated by lesions which include the origin of these projections. 1 3 Experiment I: Metabolic Changes in Mesencephalic Neurons Demonstrated with Cytochrome Oxidase Histochemistry Introduction The neurotoxicity of methylmercury in humans during prenatal and early postnatal development can result in neurological d isorders resembl ing cerebral palsy (Amin-Zaki et a l . , 1979; C h a n g and A n n a u , 1984). Chron ic postnatal administration of methylmercury to developing rats has been shown to produce a similar syndrome, including spasticity, ataxia and myoclonic jerking, which appears during the animal 's fourth postnatal week (O'Kusky, 1985; O 'Kusky and M c G e e r , 1985; O 'Kusky et a l . , 1988a; O 'Kusky et a l . , 1988b). Examination of the brain at the onset of neurological sign has revealed the relatively select ive degenerat ion of inhibitory G A B A e r g i c interneurons in the neocortex and striatum (O'Kusky, 1985; O 'Kusky and M c G e e r , 1985; O 'Kusky et a l . , 1988). The possibil i ty that the resulting disinhibit ion of cort icofugal sys tems may increase activity in brainstem nuclei and thereby contribute to the motor impairment seen in these animals was addressed in this experiment. Severa l studies have demonstrated that changes in activity within discrete neuronal populations can be detected using an histochemical technique which marks the mitochondrial enzyme cytochrome oxidase (CO) . Monocular eyel id suture or enucleation results in a marked reduction in staining for C O in the lateral geniculate nucleus and corresponding areas of the v isual cortex of cats and primates (Wong-Ri ley, 1979; Wong-Riley and Carrol l , 1984). A similar decrease in C O staining is observed in the dorsal lateral geniculate nucleus of rats following neonatal or adult enucleat ion (Land, 1987). That intensity of C O staining accurately discr iminates the metabol ic activity in the brain has been conf irmed by other methods in vivo and in vitro (Darriet et a l . , 1986). The C O histochemical method was applied in the present study to investigate abnormal neuronal I. Cytochrome oxidase histopathology activity in the brain of methylmercury-treated rats at the onset of neurological impa i rmen t . Materials and Methods Each of three litters of Sprague-Dawley rats was cul led to three triplets of weight- and gender-matched pups on postnatal day (PND) 3. Individual pups within each triplet were randomly ass igned as methylmercury-treated (MeHg) , weight-matched control ( W M C ) , or normal control (NC). Beginning on P N D 5, M e H g animals received daily subcutaneous injections of 0.01 M methylmercuric chlor ide in physio logical sal ine (5 mg Hg/kg). Control animals received injections of an equivalent volume of physiological sal ine. W M C animals were periodically isolated in an incubator at 37 °C to maintain body weight within ± 5 % of M e H g rats. At the onset of neurological impairment (PND 21-26), an individual M e H g rat and its matched controls were injected i.p. with a lethal dose of sodium pentobarbital and perfused through the ascending aorta with ice-cold fixative containing 4 % paraformaldehyde and 4 % sucrose in 0.1 M phosphate buffer (pH 7.4) at a pressure of 120 mm Hg for 15 min. The brains were immediately removed and postfixed in the same solution for 45 min; they were then transferred to 0.1 M phosphate buffer containing 4 % sucrose (PBS) for 30 min. The three brains from each triplet were blocked and mounted on a brass freezing platform and quickly frozen with powdered dry ice. Corona l sect ions (40 u.m) from all three brains were cut simultaneously through the brainstem and col lected in cold P B S . Sect ions were processed for the presence of cytochrome oxidase using the histochemical method of Wong-Riley (1979). Sect ions were incubated at 37 °C for two hours in a P B S solution containing 0 .03% cytochrome C (Sigma# C-2506) and 0 .05% 3,3 ' -d iaminobenzid ine (Sigma# D-5637), washed in three changes of P B S for 15 min each and then mounted on chrome-alum coated s l ides. The mounted sections were dehydrated briefly in absolute ethanol, followed by 5 min in xylene and then coversl ipped with Permount. Representat ive sect ions were stained for N iss l substance using 0 .1% thionin in acetate buffer (pH 3.7). I. Cytochrome oxidase histopathology Results M e H g rats exhibited a mixed spastic/dyskinetic syndrome which appeared between P N D 20-26. Cl in ica l s igns of neurological impairment were the same as those reported in previous studies (O'Kusky, 1985; O 'Kusky and M c G e e r , 1985; O 'Kusky et a l . , 1988a; O 'Kusky et a l . , 1988b), including hypertonicity of l imb musc les , f lexion deformit ies, myoclonic jerking of hindlimbs and general ized motor convuls ions. Histological examination of sect ions stained for C O revealed a striking difference between M e H g and control animals in the mesencephalon at the level of the magnocellular red nucleus ( R M C ) . In N C and W M C animals, cel l bodies of the R M C neurons were darkly stained and surrounded by neuropil of moderate staining intensity (Fig. 3A) . There was a moderate decrease in the staining intensity of the neuropil and fewer of the larger C O -positive cell bodies were observed in the R M C of M e H g rats. In all M e H g animals (N=9) a population of smal l , very intensely CO-posi t ive neurons ( ICONs) was observed within the R M C , extending into the interrubral mesencepha lon (IRM) to the midline (Fig. 3B) . At higher magnif ication, ICONs were confirmed to be neurons by virtue of stained dendritic processes on many profiles (Fig. 3C) . These neurons were not observed in any N C or W M C an ima ls (N=18). Cytochrome ox idase staining in the substantia nigra of normal rats (Fig. 4B,D) was character ized by many medium-s ized neuronal profi les with elaborate p rocesses in the z o n a reticulata (SNr) and pars lateralis (SNI). In M e H g rats, examination of the S N r revealed an apparent decrease in the number of CO-posi t ive neurons while staining in the SNI appeared normal (Fig. 4A) . A few neuronal profiles are seen in the S N r at higher magnif icat ion, but smal l , CO-pos i t ive profi les appear to predominate (Fig. 4 C ) . There were no apparent differences between M e H g and N C animals in cytochrome-oxidase staining in any other brainstem nuclei and examination of Niss l stained material failed to reveal gross degenerative changes in those areas showing cytochrome oxidase histopathology. I. Cytochrome oxidase histopathology Figure 3. Cytochrome oxidase histochemistry demonstrating intensely-staining, cytochrome oxidase-positive neurons (ICONs) in frontal sections of the mesencephalon at the level of the magnocellular red nucleus (RMC) in MeHg (A) but not WMC (B) animals. ICONs in the interrubral mesencephalon (IRM) indicated by the arrow in A are shown in C at higher magnification. The midline is oriented along the right-hand side of each figure. Calibration bars in u.m. Calibration for A and B are the same. I. Cytochrome oxidase histopathology SNr SNr • ® *T ® —22fi K - v Figure 4. Cytochrome ox idase histochemistry in the substant ia nigra, z o n a reticulata (SNr) and pars lateralis (SNI) of M e H g (A) and W M C (B) animals. Neurons throughout the S N r but not SNI of M e H g animals exhibited a marked decrease in staining intensity compared to controls. At higher magnification, the elaborate p rocesses of the many cytochrome ox idase (CO)-posit ive neurons in the control S N r (D) are contrasted and replaced by C O - d e n s e , punctate profiles in the M e H g S N r (C). Cal ibrat ion bars in urn. Magnification in A is the same as B; C the same as D. I. Cytochrome oxidase histopathology Discussion Using cytochrome oxidase histochemistry we have demonstrated changes in the oxidative metabol ic activity of neurons in the substant ia nigra, red nucleus and the interrubral mesencephalon in an animal model of developmental movement and postural disorders. A decrease in C O levels in the large neurons of the R M C with a concomitant increase in a population of smal ler neurons was found to occur at the onset of neurological impairment. The presence of a normal complement of large neurons in the R M C of M e H g animals, as seen on Niss l sect ions, suggests that the decrease in C O staining intensity was not simply the result of atrophy or degenerat ion. The increase in staining of the ICONs may result as a consequence of two processes : 1). a direct neurotoxic effect of methylmercury which accelerates the metabol ic activity ' of these neurons, or 2). an increase in excitatory activity along axons which comprise the afferent input to these neurons. In the developing nervous system, the neurotoxicity of methylmercury at extremely low doses has been shown to result in a significant increase in the number of neuronal mitochondria in the neuropil of the cerebral cortex (O'Kusky, 1983). The observed increase was suggested to be a mechanism employed by degenerating neurons to compensate for a reduced efficiency of aerobic metabol ism resulting from methylmercury toxicity. Simi lar descr ipt ions of a transient accumulat ion of mitochondria during early stages of Waller ian degenerat ion have been reported in sterile lesions (Webster, 1962) and allergic encephalomyel i t is (Condie and G o o d , 1959; Luse and McDouga l , 1960). Thus , a direct toxic effect of methylmercury on I C O N s may account for an increased cytochrome ox idase staining intensity in these neurons. A n increased activation of afferents to the R M C might lead to the observed increase in cytochrome oxidase staining of ICONs in the red nucleus. The R M C receives its primary inputs from neurons in the contralateral cerebel lar interpositus nucleus (Dekker, 1981) and ipsilateral motor cortex (Giuffr ida et a l . , 1988). Excitatory neocort ical projections converge, v ia speci f ic cort icorubral f ibers and/or col laterals of pyramidal tract axons , I. Cytochrome ox idase histopathology onto rubrospinal neurons and inhibitory interneurons (Tsukahara and K o s a k a , 1968). The cerebel lar interpositus nucleus exerts a monosynapt ic excitatory inf luence on rubrospinal neurons (Toyama et a l . , 1970). Cerebel lar neuropathology has not yet been confirmed in this animal model , but the specif ic degenerat ion of G A B A e r g i c interneurons in the neocortex (O'Kusky, 1985) might result in a consequent disinhibition of cort icorubral axons and an increase in activity of rubral neurons. The pathology of G A B A e r g i c and somatostatin-immunoreactive populations in the striatum (O'Kusky et a l . , 1988b) might contribute to a decrease in activity of neurons in the S N r by compromis ing striatonigral projections. Alternatively, a direct effect of methylmercury toxicity on the intrinsic G A B A e r g i c neurons of the S N r might account for a decrease in their cytochrome oxidase staining. G A B A e r g i c neurons have been demonstrated to maintain a metabolic activity higher than most other neurons in the C N S (Arregui and Barer , 1980; Dunwiddie et a l . , 1983). Their part icular susceptibi l i ty to t rauma resulting from disruption of oxidative metabol ism, such as occurs in hypoxia, has been conf i rmed (Francis and Puls inel l i , 1982; Romjin et a l . , 1988). The seemingly independant alterations in the metabol ic activity of rubral and nigral neurons which likely contribute to the movement and postural disorders observed in developing rats following methylmercury exposure may, thus, have a common underlying mechan ism, namely the degenerat ion of inhibitory G A B A e r g i c neurons. The appearance of ICONs within the interrubral mesencepha lon is particularly interesting in that they do not correspond to any previously def ined anatomical or functional group. Their morphological and histochemical similarity, and anatomical proximity to ICONs in the R M C suggest they may be parts of a functionally integrated population. This functional integration might be based on a shared afferent input whose abnormal activation affects the entire populat ion. In addit ion, if the increased activity of these neurons results from disinhibit ion of efferent cort ical projections, then reduction of the specif ic activity of G A D in the neocortex, which has been documented to appear at a subcl in ical stage of methylmercury toxicity (O'Kusky et a l . 1988b) may be reflected in the subclinical appearance of ICONs in both the R M C and IRM. I. Cytochrome oxidase histopathology 2 0 The time course with which this histochemical ly-def ined population of neurons appears and its relationship to the onset of neurological impairment during postnatal development was quantif ied in the following experiment. In addit ion, the effects of eliminating rubral afferent pathways on their appearrance was assessed . 21 Experiment II: Postnatal Changes in the Total Number of ICONs and the Effects of Lesions of the Cerebral Cortex and Cerebellum Introduction The appearance of movement and postural disorders in developing rats neonatally exposed to methylmercury has been demonstrated in previous studies (O'Kusky, 1985; O 'Kusky and M c G e e r , 1985; O'Kusky et a l . , 1988a; O'Kusky et a l . , 1988b; Dyck and O 'Kusky , 1988). Using cytochrome-oxidase histochemistry we have observed an increase in the oxidative metabol ic activity of a population of smal l neurons within the magnocel lu lar red nucleus (RMC) and interrubral mesencepha lon (IRM) in this animal model at the onset of neurological impairment (Dyck and O'Kusky, 1988). The age at which these intensely staining cytochrome oxidase-posit ive small neurons ( ICONs) appear and the mechan isms responsible for their increased activity are not known. The first objective of the present study was to demonstrate the time course with which ICONs appear and the extent to which postnatal changes in their number correlate with the development of neurological s igns. The intense cytochrome oxidase staining of these neurons could result from increased oxidative metabol ism due to abnormal activation by afferent inputs. A common origin for this afferent input may be the factor responsible for their h istochemical similarit ies. The major afferents to the adult red nucleus are from the ipsilateral motor cortex and the contralateral deep cerebel lar nuclei (Guiffrida et a l . , 1988). It seemed conceivable that unilateral cort ical les ions, which would deprive neurons in the ipsilateral R M C of a major afferent, would differentially reduce the staining intensity of I C O N s in the ipsilateral but not contralateral red nucleus. Converse ly , unilateral cerebel lectomy, which would deprive neurons in the contralateral R M C of a major afferent, might differentially reduce the II. Postnatal changes and effect of lesions 2 2 staining intensity of ICONs in the contralateral but not ipsilateral red nuc leus. Furthermore, if ICONs in the R M C and IRM share cerebel lar or cortical afferents, then unilateral lesions might produce similar effects in the IRM neurons. The second objective of this study was , therefore, to examine the effects of unilateral les ions of the cerebral cortex or deep cerebel lar nuclei on the pathogenesis of cytochrome oxidase histopathology. Postnatal day 10 was chosen as the time for surgery in order to reduce the possibility of aberrant projections forming from the intact hemisphere. The formation of such aberrant contralateral corticorubral projections (Leong and Lund, 1973; Nah and Leong , 1976a; Nah and Leong , 1976b; Fujito, 1984; Murakami et a l . , 1982; Naus et a l . , 1987; V i l l ab lanca et a l . , 1982, 1984) or ipsi lateral cerebel lorubral project ions (Lim and Leong , 1975; Cas t ro , 1978; Leong , 1978) has been demonstrated following unilateral les ions in young animals, however, the potential for reorganization of rubral innervation during development is inversely related to age and appears to be minimal after postnatal day 10 (Nah and Leong , 1976b; Leong, 1978; Gramsbergen and I jkema-Paasen, 1984). Materials and Methods Exper iment 1 Each of four litters of Sprague-Dawley rats was cul led to 10 pups per litter on postnatal day (PND) 3 and organized into 5 pairs of weight- and gender-matched pups. Individual pups within a pair were randomly ass igned as methylmercury-treated (MeHg) or normal control (NC) . Beginning on P N D 5, M e H g animals received daily, subcutaneous injections of 0.01 M methylmercuric chloride in physiological sal ine at a dose of 5 mg Hg/kg. N C animals received daily injections of an equivalent volume of physiological sa l ine . Four M e H g - N C pairs were anaesthet ized on each of P N D 14, 16, 18, 20, and 25 with a lethal dose of sodium pentobarbital and perfused through the ascending aorta for 15 min II. Postnatal changes and effect of lesions 2 3 with a 0.1 M phosphate buffer solution containing 4 % sucrose ( P B S , pH7.4) and 4 % paraformaldehyde . The brains were immediately removed and immersed in the fixative for 45 min; then in P B S for 30 min. Ser ia l horizontal sect ions through the red nucleus were cut frozen at 40 u.m, col lected in P B S and immediately p rocessed with the cytochrome-oxidase histochemical method of Wong-Ri ley (1979). The histochemical methods have been descr ibed previously in detail (Dyck and O'Kusky, 1988; see Sect ion I). Morphometr ic analysis of ICONs in the magnocel lular red nucleus (RMC) and interrubral mesencepha lon (IRM) was performed using Bioquant Sys tem IV digitizing morphometry software (R & M Biometr ics, Inc., Nashvi l le , TN) instal led in an I B M - P C . The image analys is system was interfaced with a compound microscope v ia a digitizing tablet posit ioned below the drawing tube attachment. This al lowed simultaneous visual izat ion of the histological section and the light emitting diode mounted in the cursor. A rea measurements of the R M C and IRM were made on each serial section at a final magnification of X 4 0 . The boundary of the R M C was identified by the position of magnocel lular neurons and the contrasting staining intensity between the R M C and surrounding mesencephalon. Lines drawn from the rostral and caudal poles of the R M C perpendicular to the midline defined the boundaries of the IRM. Stained sect ions were v iewed at a final magnification of X200 for neuron counting. ICONs within the red nucleus and intervening mesencephalon were identified by their s ize , staining intensity and distinct morphological appearance. Individual counts were performed bilaterally for both R M C and IRM on each section in the ser ies. The average diameter of I C O N s was estimated by measuring 30 randomly selected neurons from each brain at a final magnification of X400 . The outline of the cell body was traced with the cursor and the computer was programmed to measure the profile area and circumference, and to calculate from these the diameter and shape factor (F). II. Postnatal changes and effect of lesions 2 4 The diameter was calculated from: area = n * rad ius 2 . The shape factor, calculated from the formula: 4 TJ * area F = per imeter 2 results in a fraction indicating the profile's deviat ion from a circle (1.000=circle, 0 . 0 0 0 = l i n e ) . The numerical density of ICONs ( N v , the number of neurons per m m 3 ) was estimated bilaterally for the R M C and IRM using the Abercrombie method (Abercrombie, 1949). This value was determined according to the following formula: N A N V = T + D , where Ny\ is the number of neuronal profiles per unit area of sect ion, T is the section thickness, and D is the average profile diameter measured on the histological sect ions. The total volume (V) of the R M C and IRM region was calculated from the following formula : V = A j * T , where A j is the sum of area measurements from all serial sect ions, and T is the section thickness. The total number of ICONs in R M C and IRM was calculated from the estimates of N y and V . It should be noted that estimates of total neuron number are independent of t issue shrinkage during histological processing. Consequent ly , no attempt was made to determine systematically the degree of t issue shrinkage at the var ious postnatal ages employed in this study. The total number of ICONs within the R M C and IRM on postnatal days 14, 16, 18, 20, 22, and 25 was compared using a two-factor (age X region) analysis of var iance (ANOVA) for repeated measures. The Tukey (hsd) test was used for post hoc pairwise compar ison of means. II. Postnatal changes and effect of lesions 2 5 Exper iment 2 Four litters of Sprague-Dawley rats were cul led to 10 pups per litter on PND 3 and organized into matched pairs and injected with either methylmercuric chloride or physiological sal ine beginning on PND 5 as descr ibed in Experiment 1. MeHg - N C pairs were ass igned to the following three surgical groups on PND 10: 1) left uni lateral decort icat ion (n=8), 2) right unilateral hemicerebe l lec tomy (n=16), and 3) sham-operated controls (n=16). The pups were p laced in a cotton-lined box and hypothermic anaesthes ia was induced by cooling for 10-15 min at -10 °C. An imals were monitored until a toe-pinch no longer elicited limb withdrawal. Using this technique animals were completely immobi l ized for 6 to 8 minutes, while the surgical procedure required less than 4 minutes. Aspiration lesions were performed using a g lass pipette under v isual guidance with the aid of a surgical microscope. For hemidecortication, the scalp was incised and the frontal and parietal bones above the left hemisphere were removed with iris sc issors . The overlying dura was exc ised and the exposed cortex was aspirated from frontal to caudal poles of the hemisphere and extended from the midline laterally to the rhinal f issure. For hemicerebel lectomy, the right cerebel lar hemisphere was exposed by removing the overlying occipital bone and beginning at the midline of the vermis, the cerebel lar cortex, interpositus and lateral nuclei were aspirated. Following surgery, the wound was c losed and the rats were al lowed to warm under a heat lamp before returning to the home cage. Unoperated control, hemidecorticate and hemicerebel lectomized M e H g - N C pairs were sacri f iced on PND 22 and the brains were processed for cytochrome-oxidase histochemistry as descr ibed in previous sect ions. Morphometr ic analys is was performed as detailed in Experiment 1 except that the data for each side was kept separate. The experimental protocol included measurements of: 1) the volume of R M C and IRM, 2) the Nv of ICONs , and 3) the total number of ICONs in these regions. II. Postnatal changes and effect of lesions 2 6 Compar ison of the number of ICONs in the right and left R M C and IRM of unoperated controls and animals sustaining unilateral cerebel lar or cortical lesions was accompl ished using a three-factor (lesion X side X region) A N O V A for repeated measures. The Tukey (hsd) test was used for post hoc pairwise compar ison of means. Results Exper iment 1 The smal l , intensely-staining cytochrome ox idase neurons ( ICONs) were observed in the R M C and IRM of MeHg but not N C animals (Fig. 5). ICONs were not detected on P N D 14 but were observed in all M e H g animals from P N D 16-25. Their histological appearance was remarkably consistent across these ages. The ICONs were character ized by a very intense staining, with an ovoid to circular accumulat ion of dense reaction product, often surrounded by a halo of decreased staining intensity. Cytochrome oxidase positive p rocesses were often seen emanating from the neuronal profiles (Fig. 5, inset). The average diameter of ICONs did not vary significantly between R M C and IRM or among the var ious age groups, ranging from 13.5-14.4 u.m. The average shape factor did not vary consistent ly from P N D 16-25, ranging from 0.856-0.879. This indicates that cel l bodies of ICONs deviate somewhat from a perfect sphere (where F=1.000), but their shape was consistent throughout the course of the experiment. Although the application of the Abercrombie method is designed for use with spherical objects, the deviation of ICONs from a true sphere is relatively minor and bears no consistent relationship with the age of an animal . Any error in determining the total number of neurons would, therefore, be distributed uniformly ac ross all age groups. There were significant changes (F (4 j3 ) = 13.689, p < .0001) in the total number of ICONs in the R M C and IRM of MeHg animals as a function of age (Fig. 6). No ICONs were II. Postnatal changes and effect of lesions 2 7 observed on P N D 14. Morphometr ic analysis revealed approximately 300 ICONs in the mesencepha lon on P N D 16, fol lowed by a four-fold increase to approximately 1300 by P N D 18 (p <.05). There was a significant decrease in their number between P N D 18-25 (83%, p <.01). Despite the marked changes in the number of ICONs across these age groups, their individual appearance was remarkably consistent (Fig. 7). The staining intensity of these neurons did not increase gradually with age (and therefore, with increasing methylmercury accumulat ion). On the contrary, there was an abrupt appearance of increasing numbers of intensely stained neurons between P N D 16 and 18. Similar ly, the decrease from P N D 18-25 did not result from a gradual decrease in staining intensity, but rather from a gradual decrease in the number of intensely-stained neurons. The initial increase and subsequent decrease in the total number of ICONs in the mesencephalon was observed in both R M C and IRM (Fig. 8). A significant regional difference in number was noted (F(4 13) = 3.318, p<.05). The number of ICONs was significantly greater in the IRM on P N D 18, (62%,p<.001) and P N D 20 (55%, p<.05), but not at P N D 16, 22 , or 25 . No significant bilateral dif ferences were observed in either the R M C or IRM. Exper iment 2 The effects of unilateral cortical or cerebel lar surgery were determined by compar ing the total number of ICONs in the contralateral vs ipsilateral R M C and IRM. It was expected that any effect of the lesions would be manifested in an asymmetry in their bilateral distribution. In both lesion groups, those sustaining left hemidecort icat ion or right hemicerebel lectomy, the left red nucleus would be deprived of one of its normal afferent p ro jec t ions . Examinat ion of cytochrome oxidase-sta ined sect ions revealed that the distribution of ICONs in the R M C and IRM of unoperated control animals was bilaterally uniform. The number of I C O N s did not differ significantly on either side in the R M C or the IRM (Fig. 9). II. Postnatal changes and effect of lesions 2 8 The bilateral distribution of ICONs in the mesencephalon of hemidecort icate animals (Fig. 10A) was indist inguishable from that observed in unoperated controls (see F ig . 7B) . The total number of ICONs in the deafferented, ipsilateral R M C was not significantly different from the number in the contralateral R M C . Similarly, the total number of I C O N S in the ipsilateral and contralateral IRM did not differ (Fig. 9). An asymmetr ical distribution of I C O N s in M e H g animals sustaining cerebel lar lesions was , however, evident in histological sect ions. More ICONs were evident in the ipsilateral than the contralateral IRM (Fig. 10B, right panel) . Statist ical analys is revealed a significant bilateral difference in the number of neurons in both the R M C and IRM fol lowing hemicerebel lectomy (F(2,16) = 7.963, p<.01). The total number of ICONs in the ipsilateral R M C or IRM did not differ from the equivalent region of unoperated control or hemidecort icate animals. However, there was a significant decrease (36%, p < .05) in the total number of I C O N s in the deafferented (left) R M C compared to the intact ipsilateral R M C (Fig. 9). There was a similar effect of cerebel lar lesions on the number of ICONs in the IRM; significantly fewer ICONs were located in the IRM contralateral to hemicerebel lectomy (37%, p < .05). Changes in the total number of ICONs in each region were not found to result from deviat ions in neuron diameter or staining intensity. The histological and morphological appearance of ICONs was not found to differ between any of the groups in this study. Their average diameter (14.1 u.m) and shape factor (.866) were not found to vary significantly within these groups. The results of this experiment are further val idated when compared to those of Experiment 1. The total numbers of ICONs in the mesencephalon of unoperated controls and hemidecort icate animals were not significantly different in their distribution or total numbers when compared to the same regions in animals of equivalent age (PND 22) in Exper iment 1. II. Postnatal changes and effect of lesions 2 9 Figure 5. Cytochrome oxidase (CO) histochemistry of horizontal sect ions in the mesencephalon at the level of the magnocellular red nucleus ( R M C ) . The intensely-staining cytochrome ox idase neurons (ICONs) in the R M C and interrubral mesencepha lon (IRM) of methylmercury-treated rats (A) were never apparent in the control an imals (B). Many ICONs were observed to have CO-posi t ive processes emanating from the cel l body (A, inset). Cal ibrat ion bar in microns. Magnif ication is the s a m e for A and B. II. Postnatal changes and effect of lesions 3 0 Figure 6. Postnatal changes in the combined number of ICONs in the R M C and IRM of rats chronical ly exposed to methylmercury beginning on postnatal day 5. The initial appearance on postnatal day (PND) 16 was followed by a four-fold increase on P N D 18 and then gradually decl ined to the last age group studied (PND 25). II. Postnatal changes and effect of lesions 3 1 Figure 7. Cytochrome oxidase histochemistry of horizontal sect ions through the R M C and IRM of M e H g animals. The initial appearance of ICONs on P N D 16 is apparent in A . A significant increase in the number of ICONs on P N D 18 (B) is fol lowed by a gradual decrease to P N D 25 (C). Calibration bar in C is in u.m. Magnif ication for A , B, and C are the same. II. Postnatal changes and effect of lesions 3 2 Figure 8. Postnatal changes in the total number of ICONs in M e H g rats are differentiated by region. The number of ICONs in R M C and IRM were not significantly different on P N D 16, 22, or 25 . At P N D 18 and 20 significantly more ICONs were observed in the IRM. II. Postnatal changes and effect of lesions 3 3 Figure 9. The bilateral distribution of ICONs in the R M C and IRM of unoperated MeHg rats and M e H g animals sustaining left hemidecortication or right hemicerebel lectomy on postnatal day 10. No left-right differences in the number of ICONs were observed in either the R M C or IRM of unoperated control or hemidecorticate animals. The total number of ICONs in the ipsilateral R M C and IRM of animals receiving right hemicerebel lectomy was not found to differ from control or hemidecorticate animals. A significant decrease was demonstrated in the contralateral R M C (36%) and IRM (37%). II. Postnatal changes and effect of lesions 34 Figure 10. Cytochrome ox idase histochemistry in horizontal sect ions of the rubral mesencephalon in hemidecorticate (A) and hemicerebel lectomized (B) M e H g animals. The distribution of ICONs in animals sustaining left hemidecortication on postnatal day 10 was symmetr ica l . In an imals undergoing right hemicerebel lectomy (B), signif icantly fewer ICONs are apparent in the contralateral (left) IRM. Cal ibrat ion bar in u.m. Magnif icat ion is the same in A and B. II. Postnatal changes and effect of lesions 3 5 Discussion In this study, we have quantified the number of intensely-staining cytochrome ox idase neurons ( ICONs) in the magnocel lular red nucleus (RMC) and interrubral mesencepha lon in rats of different ages , neonatally exposed to methylmercury (MeHg). The time course for their appearance was biphasic over the time course of this study, beginning with a tremendous increase in the total number of ICONs followed by a gradual decrease. The increased activity of these neurons was determined to be influenced in part by cerebel lar p ro jec t ions . ICONs were first demonstrated on P N D 16. An increased activity in this specif ic population of neurons in 16 day-old MeHg-treated rats, is the earliest demonstrat ion so far d iscovered , of a subcl in ical manifestation of methylmercury- induced toxicity. Their onset of appearance in the mesencephalon preceeds the onset of movement and postural d isorders by almost 10 days (O'Kusky, 1985; Dyck and O 'Kusky , 1988). The increasing numbers of ICONs may be correlated to the development and arrival of excitatory afferents to the R M C . The age at which the red nucleus in the rat becomes synaptical ly mature has not been unequivocally demonstrated. A single ultrastructural report by Pov l i shock (1978) indicates that cerebel lorubral projections in the rat, identified by their synaptogenesis onto the soma and proximal dendrites of the magnocel lular neurons in the R M C , are still immature on P N D 10 and do not reach full maturity until P N D 15. Investigation of rubral development in the oppossum, demonstrates the arrival of major afferents to the red nucleus along a distinct temporal gradient (Martin et a l . , 1988). Cerebel lar afferents arrive in the red nuc leus postnatally, yet well before those originating in the cerebral cortex. The increased activity of ICONs which has been demonstrated to originate at least in part from the cerebel lum, may therefore, occur in paral lel to the development of excitatory cerebel lar afferents. II. Postnatal changes and effect of lesions 3 6 The gradual decrease in the total number of ICONs with age may result as a consequence of a direct toxic action of methylmercury. The effects of methylmercury poisoning in developing rats have been demonstrated to result in a significant increase in the number of mitochondria which preceeds degeneration (O'Kusky, 1983). The observed increase in number was suggested to be a mechanism by which degenerating neurons compensate for a reduced efficiency of aerobic metabol ism. Similar descript ions of a transient accumulat ion of mitochondria during early stages of Waller ian degenerat ion have been reported in spinal lesions (Webster, 1962) and allergic encephalomyel i t is (Condie and G o o d , 1959; Luse and McDouga l , 1960). Thus , a direct toxic effect of methylmercury on mitochondria may account for the perceived transient increase in cytochrome ox idase staining intensity in ICONs. The toxic effects of methylmercury have been demonstrated to interfere with many aspects of normal neuronal function as in inhibition of enzymes (Clarkson, 1983) and of R N A and protein synthesis (Syversen 1977, 1982; Omata et a l . , 1980). These indices of methylmercury toxicity have been demonstrated in neuronal populat ions widely distributed throughout the brain (Syversen et a l . , 1981; Syve rsen , 1982). Methylmercury has also been demonstrated to affect cell division and axonal transport by interfering with microtubules in a manner similar to colchic ine (Abe et a l . , 1975; Miura et a l . 1978 (in Cho i et a l . , 1981); Sage r et a l . , 1981). The survival of developing neurons following colchic ine- induced disruption of microtubules has been shown to be directly related to their maturity (Dyck et a l . , 1984), with neuronal degenerat ion being widespread throughout the forebrain of rats at birth but minimal after P N D 20. The increased activity of ICONs may reflect a disinhibition of afferents to the red nucleus. The red nucleus in mammals typically consists of two parts dist inguished by topography, intrinsic cel l morphology, and afferent input. The rostral, parvocel lular portion ( R P C ) , so cal led because it is made up of mostly smal l - to medium-s ized neurons, receives afferents which originate in the ipsilateral motor cortex. The second major input II. Postnatal changes and effect of lesions 3 7 to the mammal ian red nucleus ar ises v ia excitatory projections from the deep nuclei of the cerebel lum (Tsukahara and K o s a k a , 1968). Cerebel lar activation of the red nucleus is topographical ly differentiated into two distinct pathways; the nucleus lateralis projects to the anterior, parvocel lu lar nuc leus, onto rubro-olivary neurons; and the nucleus interpositus (IN) projects mostly to the cauda l , magnocel lu lar red nuc leus, onto rubro-spinal neurons (Ito, 1984). A modification in the activity of neurons in the R M C could, therefore, be expected to result from abnormal activation of its cerebel lar afferents. Using cytochrome-oxidase histochemistry we have demonstrated a decrease in the number of I C O N s by approximately one-third following cerebel lar but not neocort ical les ions. In light of the differential compartmental izat ion of rubral afferents from the cerebral cortex and cerebel lum, it is not surprising that the effects of decort ication (Dyck and O 'Kusky , 1988) or hemidecortication were not apparent in the R M C . The marked changes in activity following hemicerebel lectomy could reflect abnormal activity in the cerebel lum caused by methylmercury. Neuropathology in the cerebel lum of immature humans (Chang, 1977; Cho i et a l . , 1978; Chang and Annau , 1984) and animals (Khera, 1973; Khe ra et a l . , 1974; Khera and Tabakova , 1973) is character ist ic of methylmercury poisoning. Neurons in the granular and molecular layer are primarily suscept ible while a heterotopic organizat ion of Purkinje cel ls has also been descr ibed (Choi et a l . , 1978). Abnormal increases in activity of neurons in the red nucleus may result from disinhibition of the excitatory cerebel lorubral projection following direct damage to Purkinje cel ls or following degenerat ion of the granule cel ls which exert an excitatory influence on Purkinje neurons. Although evidence of neuropathology in the cerebel lum in this animal model has yet to be examined, significant decreases in the specif ic activity of glutamate decarboxylase (GAD) in the cerebral cortex and striatum (O'Kusky and M c G e e r , 1985; O 'Kusky et a l . 1988b) correlated with degenerat ion of neocort ical and striatal interneurons (O 'Kusky, 1985; O 'Kusky et a l . 1988b) and indicate a susceptibil ity of G A B A e r g i c neurons to II. Postnatal changes and effect of lesions 3 8 methylmercury toxicity. Whether the increased susceptibi l i ty of G A B A e r g i c neurons is restricted to the neocortex and striatum or includes cerebel lar Purkinje cel ls and G A B A e r g i c neurons in other areas has not yet been demonstrated. The apparent increase in activity of ICONs may be a consequence of the direct toxic effect of methylmercury on G A B A e r g i c neurons intrinsic to the red nuc leus, which is further potentiated by the excitatory cerebel lar input. The presence of G A B A in intrinsic neurons of the red nucleus of humans, monkeys, cats, and rats has been inferred from measurement of enzyme activities, uptake studies, release from R N s l ices, e lectrophysiological ev idence and immunohistochemistry (Muller and Langemann , 1962; Perry et a l . , 1971; Al tmann et a l . , 1976; V a n der Heyden et a l . , 1979; Nieoul lon and Dust ic ier, 1981 ; Kubota et a l . , 1983; Murakami et a l . , 1983; Vui l lon-Cacciut to lo et a l . , 1984). Direct project ions from the ipsi lateral sensor imotor cortex to intrinsic G A B A e r g i c neurons (Katsumara et a l . , 1984) mediate cort ical inhibition of rubrospinal neurons (Nieoul lon et a l , 1988). Disinhibition of R M C neurons with bilateral (Dyck and O 'Kusky , 1988) or unilateral lesions of the neocortex was not sufficient to demonstrate an increased activity of rubral neurons. However, an increase in activity of ICONs in the R M C following intrinsic disinhibition may have been greatly potentiated when combined with a disinhibit ion of the excitatory input from the deep cerebel lar nuclei (Nieoullon and Dus t i c ie r , 1981) . A modifiable activity of rubral neurons in response to a change in cerebel lar output has recently been demonstrated. A significant reduction in the discharge activity of rubrospinal neurons has been sa id to result from lesions of the inferior olivary nuclei (Bil lard et a l . , 1988). Th is les ion- induced loss of c l imbing fibres re leases the Purkinje ce l ls from the inhibitory influence of the olivary nuclei (Col in et a l . , 1980). Disinhibit ion of Purkinje cel ls facil itated the inhibition of neurons in the deep cerebel lar nuclei and thus diminished the excitatory cerebel lorubral projection (Toyama et a l . , 1970). It is possib le that MeHg- induced pathology of Purkinje neurons would result in II. Postnatal changes and effect of lesions 3 9 the disinhibition of the deep cerebel lar nuclei which would increase activity in the cerebe l lo rubra l pro ject ion. The observation that the number of ICONs in MeHg-treated rats seen at P N D 22 were reduced in number by less than 4 0 % following cerebel lar hemispherectomy on P N D 10, indicates that their activation was partially, but not exclusively, dependent on innervation from the contralateral cerebel lar hemisphere. The ICONs which remain may have been maintained by the formation of aberrant excitatory projections from intact s i tes. Although several studies have demonstrated that synaptic plasticity in the developing red nucleus is minimized when lesions of afferent projections occurs on P N D 10, at least one report indicates that a smal l aberrant projection from the contralateral interpositus nucleus is still poss ib le up to P N D 15 (Leong, 1978). In addit ion, electrophysiological studies have demonstrated that aberrant inputs to the contralateral red nucleus following unilateral hemicerebel lectomy, may arrive from three independent sources ; the intact cerebel lar hemisphere; the ipsilateral motor cortex; or the contralateral motor cortex (Tsukahara et a l . , 1974, 1975, 1981, 1983). The magnitude and origin of this synapt ic plasticity has been demonstrated to vary with age. The formation of aberrant projections from all three sources was only seen in the contralateral red nucleus following neonatal hemicerebel lectomy (Tsukahara et a l . 1981, 1983; Fujito et a l . , 1984). Plast ici ty of rubral afferents in the adult was limited only to the synapt ic reorganization of normal, ipsilateral cort ical projections onto synaptic sites vacated by the degenerat ing cerebel lorubral axon terminals (Tsukahara et a l . , 1974, 1975). Al though we attempted to minimize these inf luences by performing lesions on P N D 10, the effect of a compensatory innervation which continues through adulthood cannot be discounted. The plasticity of rubral circuitry has also been demonstrated to involve neurotransmitter-specif ic pathways. A significant increase in G A D activity has been descr ibed in the red nucleus contralateral to cerebel lar hemispherectomy in adult cats (Nieoul lon and Dusticier, 1981 ; Vui l lon-Cacciut to lo, 1986). Th is increase, measured by II. Postnatal changes and effect of lesions 4 0 G A D immunohistochemistry and enzyme activities, was first detected 4 days postlesion and reached a maximum 7-10 days after surgery. It was also associated with an increased density of GAD-pos i t ive terminals and the appearance of an anomalous population of smal l , GAD-pos i t ive neurons dorsolateral to the red nucleus which were not detected in unoperated control animals. A suppression of ICONs in the contralateral R M C and IRM following hemicerebel lectomy in M e H g animals may, therefore, result from a compensatory increase in local G A B A e r g i c neurotransmission. The conspicuous appearance of an anomalous population of neurons within the IRM is of significant interest. The parallel reduction in number of I C O N s descr ibed within both the R M C and the IRM following elimination of the contralateral cerebel lar projection and their concurrent postnatal appearance suggests that these neurons may form a functionally homogenous population. This homgeneity appears to be a reflection of common cerebellar input. The primary afferents from the nucleus interpositus to the R M C originate from the anterior division (NIA). However , a smal l projection from the posterior division of the interpositus nucleus (NIP) to a smal l region just medial to the caudal red nucleus was initially descr ibed in the cat by Voogt (1967). Further analys is by Angaut (1970) revealed that axons of the PIN supply abundant thin terminals within a narrow strip, just medial to the caudal two-thirds of the red nucleus. A homologous projection may exist in the rat but has yet to be descr ibed. The analysis of NIP and AIP projections in the rat may reveal a similar distribution of projections to the IRM and R M C which would account for the appearance of ICONs in the IRM. Converging evidence appears to suggest that increases in the activity of neurons descr ibed as ICONs may result as a consequence of cerebellar pathology. Although pathology in the cerebel lum has not yet been demonstrated in this animal model , its contribution to the histopathology in the mesencephalon necessi tates further anatomical , neurochemical and physiological studies. The primary objective of the experiment which follows was to II. Postnatal changes and effect of lesions 41 examine the pathology of developmental methylmercury toxicity in the brain at the light mic roscop ic leve l . 4 2 Experiment III: Immunohistochemical Localization of Major Histocompatability Complex Class I and Class II Antigens Introduction The consequences of methylmercury (MeHg) intoxication in the human nervous system have been demonstrated to vary with the age of exposure. M e H g neurotoxicity in the mature nervous system results in c i rcumscr ibed lesions of the cerebral and cerebel lar cort ices. In the cerebrum, focal lesions are found in the calcar ine, postcentral and precentral cor t ices; while cerebel lar degenerat ion is limited to neurons in the granular and molecular layers ( C h a n g , 1977, 1979, 1980). The effects of prenatal exposure to M e H g are much more severe , resulting in widespread neuronal degenerat ion throughout the cerebral cortex and cerebel lum (Chang and Annau , 1984). Many of the surviving neurons exhibit an heterotopic or atopic organization (Choi et a l . , 1978). Early postnatal exposure results in widespread degenerat ion similar to fetal pathogenesis yet resembling adult toxicity in local izat ion (Matsumoto et a l . , 1965). The developmental neuropathology of methylmercury has recently been examined in neonatal rats chronically exposed to M e H g . Investigations into the neuropathology of these animals have revealed ultrastructural (O 'Kusky, 1985) and b iochemica l ind ices (O'Kusky and M c G e e r , 1985; O 'Kusky et a l . , 1988b) of a relatively select ive degenerat ion of G A B A e r g i c interneurons in the neocortex and striatum. Using cytochrome oxidase histochemistry, we demonstrated an early physiopathology in the mesencephalon at the level of the magnocel lular red nucleus (Experiment I; Dyck and O 'Kusky , 1988). The neuronal selectivity of degenerat ion in the neocortex and striatum, combined with indications of a widespread pathology which include specif ic neurons in the mesencephalon, have not been descr ibed in previous studies of early postnatal toxicity. The primary III. M H C Immunohistochemistry 4 3 objective of the present study was to examine the extent of methylmercury- induced neuropathology at the light microscopic level in preparation for subsequent ul trastructural a n a l y s e s . An immunohistochemical method using antibodies to major histocompatabil i ty complex (MHC) molecules has recently been suggested to demonstrate an immune response to neuropathological p rocesses in controlled lesions of the cerebral cortex (Akiyama et a l . , 1988). The expression of two different M H C antigens, c lass I and c lass II, were demonstrated to mark a distinct subset of affected areas over the time course of pathogenesis. M H C class II antigens have been detected in some animal models of demyel inat ing d i sease (Matsumoto and Fuj iwara, 1986; Matsumoto et a l . , 1986) Addit ionally, the proliferation of an M H C c lass II antigen (HLA-DR) has been demonstrated to occur in human neurodegenerat ive disorders (McGeer et a l . , 1987a, 1987b, 1987c; Traugot t , 1987) . The present experiment descr ibes the immunohistochemical distribution of M H C c lass I and M H C c lass II antigens in the brain of developing rats neonatally exposed to m e t h y l m e r c u r y . Materials and Methods One litter of Sprague-Dawley rats was cul led to 10 pups on postnatal day (PND) 3. Six pups (MeHg) were randomly selected to receive daily, subcutaneous injections of 0.01 M methylmercuric chloride in physiological sal ine (5 mg Hg/kg) beginning on P N D 5 and continuing to P N D 21 . Normal control (NC) rats received daily injections of an equivalent vo lume of physiological sal ine. On P N D 22 , methylmercury-injected (MeHg) and normal control (NC) rats were anaesthet ized with a lethal dose of sodium pentobarbital and perfused transcardially with 40-50 ml of 0.1 M phosphate buffer ( P B , pH 7.4) containing 0 .9% sodium chlor ide. The III. MHC Immunohistochemistry 44 brains were immediately removed, cut in 0.5 cm coronal pieces, and immersed for 24 hours in fixative containing 2% paraformaldehyde and 1% picric acid in PB. The tissue was cryoprotected in PB containing 15% sucrose for 24 hours before 30 urn frozen sections were cut on a sliding microtome and collected in PB containing 0.3% Triton-X-100 (PB-TX). Sets of five consecutive sections were taken at 1 mm intervals throughout the cerebrum and cerebellum. One section in each set was immediately processed for cytochrome oxidase histochemistry using the method of Wong-Riley (1979) and one was stained for Nissl substance with cresyl violet. The remaining three sections in each set were treated for 20 min in PB containing 0.2% hydrogen peroxide in preparation for immunohistochemistry. Sections for immunohistochemistry were incubated at 4°C for 4 days in monoclonal antibody OX18 or OX6 (Sera -lab, England) against monomorphic determinants of rat MHC class I and class II antigens respectively; or polyclonal rabbit anti-glial fibrillary acidic protein (GFAP) antiserum (Dako, Denmark). OX 18 was used at a dilution of 1:300 in PB-TX, OX 6 at 1:100 and GFAP at 1:10,000. Immunostaining was performed using the ABC method as described by Akiyama et al. (1988). Sections were incubated at room temperature for 2 hours with biotinylated anti-mouse or anti-rabbit IgG (Vector) and then 1 hour in ABC solution (Vector, 1:400). The peroxidase labelling was developed in a 0.05M Tris-HCl buffer solution (pH 7.6) containing 5% 1M imidazole, 0.02% 3,3'-diaminobenzidine tetrahydrochloride (DAB), 0.01% hydrogen peroxide and 25 mM nickel ammonium sulfate. Sections were mounted on chrome-alum coated glass slides and coverslipped with Permount. The nomenclature used in this study is based on the works of Paxinos and Watson (1986) and Zilles (1985). III. M H C Immunohistochemistry 4 5 Results The photographs from the caudal mesencephalon in Figure 11 reveal the typical distribution of M H C I, M H C II and G F A P observed at all levels in M e H g and N C animals. In both groups, M H C I (Fig. 11 A , MeHg & B, NC) expression was more intense than that of M H C II (Fig. 11 C & D). Despite the difference in magnitude, examination of adjacent sect ions at higher magnif ication revealed that all areas exhibiting M H C ll-positive cel ls were mirrored by a greater number of MHC-I positive cel ls (Fig. 12 A , M H C I; B M H C II). M H C II positive cel ls were not apparant at any level in N C animals (Fig. 11D) while those expressing M H C I were confined to vascular endothelial cel ls and glial cel ls in the white matter (F ig. 11B) . GFAP-pos i t i ve astrocytes were diffusely distributed throughout both groups (Fig 11E , M e H g & 11F, NC) but at higher magnification were demonstrated to be more prominent in M e H g (Fig. 12C) than in N C animals (Fig. 12D). Although major dif ferences in topographical localization were not apparent, a greater intensity of G F A P staining seemed to overlap with areas staining intensely for M H C antigens as well (Fig. 11E , note RN) . M H C immunoreactivity in the brainstem nuclei was most intense in the magnocel lular red nucleus (RN) and substant ia nigra (SN) . The apparently similar staining intensity in the S N of M e H g and N C animals was deceptive as many more M H C l-positive cel ls were present in the S N of M e H g animals. This was more apparent when M e H g and N C were compared with M H C II staining (Fig. 11B & C ) . A moderate staining intensity was observed in the medial geniculate nucleus (MG) while a laminar distribution of M H C immunoreactivity was apparent in the superf icial, intermediate and deep grey layers of the superior col l iculus (SC) . Diffuse staining, which was inconsistent between M e H g animals, was observed in several other brainstem nuclei . In the remaining descr ipt ion, M H C + will refer to combined localization of M H C I and M H C II although only M H C I is depicted. III. M H C Immunohistochemistry 4 6 A s can be seen in Fig.13A, C , & D and Fig. 14A, M H C + staining was observed at all rostro-caudal levels of the cortex within distinct laminar boundar ies. A representative sect ion from control animals (Fig. 14B) demonstrates the typical lack of M H C + staining except in heavily myel inated areas. At rostral levels (Fig. 13A), intense M H C + staining was observed in laminae ll-IV and lamina VI of anterior cingulate cortex (Cg). Moving caudal ly , the same bi laminar pattern was observed in retrosplenial cortex (Rs) . In parietal cortex (Par) M H C + cells were conf ined to the deeper laminae V and VI with an intense band just above the corpus cal losum. At higher magnification, adjacent sect ions through Par stained with M H C II (13B) and M H C I (13C) showed a similar pattern but reflected a significant difference in magnitude of staining. The same pattern was maintained caudal ly in temporal cortex (Te) but with an addit ional band in lamina III. Occipi tal cortex (Oc) exhibited a diffuse distribution of M H C + cel ls throughout layers IV-VI. Cort ical areas which exhibited little or no M H C + express ion inc luded pyriform (Pir) and entorhinal (Ent) cor t ices. Severa l subcort ical nuclei were observed to exhibit M H C + immunoreactivi ty (F ig. 13). The caudate-putamen (CPu) contained a diffuse distribution of M H C + staining which was most intense dorsomedial ly at rostral levels and throughout the more caudal levels. A diffuse localization of M H C + cel ls was also observed throughout the basal forebrain and extending caudal ly to the anterior hypothalamus. The ventro-(VMH) and dorso-medial (DM) hypothalamic nuclei and the lateral hypothalamus (LH) all exhibited intense MHC+ immunoreactivity. A less intense, diffuse pattern was apparent in the z o n a incerta (ZI). The conspicuous presence of very intense staining was consistently observed in the amygda la but only the posteromedial (PM) and posterolateral (PL) cortical amygdalo id nuclei . The presubiculum (PrS) and parasubiculum (PaS) were consistently among the most intensely staining^ nuclei in M e H g animals. M H C + staining was conspicuously absent from all thalamic nuclei except for a few sparse cel ls in the dorsal lateral geniculate (DLG) and, as mentioned above, the medial III. M H C Immunohistochemistry geniculate nucleus. At posterior thalamic levels, the pretectal nuclei (PT) were the only other d iencephal ic nuclei which exhibited M H C + staining (Fig. 13E) . Except for the scattered appearance of M H C + staining in a few brainstem reticular formation nuclei (Fig. 15), the lack of M H C + immunoreactivity in the medul la stood out. The absence of staining in the lateral cerebel lar hemispheres and deep nuclei (IntA, Lat) was even more blatant. M H C + immunoreactivity in the cerebel lum was essential ly restricted to the cortical a reas of vermal lobules 1, 2 & 3. In contrast, the superior (SuVe) and lateral (LVe) vestibular nuclei exhibited very intense M H C + staining. Discussion Immunohistochemical localization of M H C c lass I and c lass II antigens in developing rats neonatally exposed to methylmercury was demonstrated to varying degrees at all levels of the brain examined in this study. Intense MHC-pos i t ive (MHC+) staining was consp icuous in the posterior cort ical amygdalo id nucleus, red nucleus, substant ia nigra, parasubicu lum, presubiculum and vest ibular nuclei . A moderate laminar distribution of M H C + immunoreactivity was observed in laminae V & VII of all neocortical areas which extended to lamina III only in temporal and lamina IV in occipital cortex. Moderate staining was also shown to occur in the cerebel lar vermis, caudate-putamen, nuclei of the basal forebrain and in var ious nuclei throughout the hypothalamus. Low levels or no immunoreactivity was demonstrated in the h ippocampus, entorhinal and pyriform cortex, most thalamic nuclei , and cerebel lum. The apparent lack of M H C antigen expression in normal control animals contrasted with the w idespread , yet regional ly local ized distribution in the methylmercury-treated animals indicated that antibodies to these antigens do reflect specif ic neuropathological p rocesses. It has been suggested that the M H C class I (0X18) and the M H C c lass II (0X6) III. M H C Immunohistochemistry 48 Figure 11. Comparison of MHC I (A & B), MHC II (C & D) and G F A P (E & F) immunoreactivity in adjacent coronal sections through the caudal mesencephalon of MeHg (A, C, E) and normal control rats (B, D, F). In the control, MHC antigen expression was limited to class I staining on vascular endothelium and heavily myelinated areas. Although all areas staining positive for MHC II (C) antigens were also positive for MHC I (A), the MHC I expression was always more robust. The appearance of GFAP staining cells did not differ markedly between MeHg (E) or control (F); however, areas which exhibited greater MHC expression appeared to stain more intensely with G F A P (note RN). Magnification X10. M H C Immunohistochemistry 4 9 Figure 12. High power magnification of M H C and G F A P antigen express ion. M H C I (A) and M H C II (B) positive cel ls exhibited similar morphological appearance and regional specificity in these adjacent sect ions through the cingulate cortex. The marked difference in the magnitude of M H C I vs M H C II expression is again apparent in these micrographs. Although significant dif ferences in the localization of G F A P staining were not apparent between M e H g (C) and control sect ions (D), in those areas highly M H C immunoreact ive, GFAP-pos i t i ve astrocytes appeared more numerous and hypertrophic in M e H g animals. Plates A , B . and C are adjacent serial sect ions. Calibration bar in B is in urn and is the same for all p la tes. III. M H C Immunohistochemistry 5 0 Figure 13. Photographic montages showing M H C staining in frontal sect ions through the forebrain of an M e H g rat. M H C immunoreactivity was observed at all rostrocaudal levels (A, D & E) . In adjacent cortical sect ions, both M H C I (B) and M H C II (C) positive cel ls were observed to respect distinct laminar boundaries. M H C staining was diffusely distributed throughout the basa l forebrain but respected nuclear subdiv is ions of the hypothalamus. Cort ical areas (Cg, R s , Par, Te) were observed to exhibit moderate M H C immunoreactivity while very intense staining was evident in the presubicu lum (PrS) and posteromedial cortical amygdaloid nucleus (Pm). M H C staining was consp icuous ly absent in severa l cort ical (Ent, Pir) and subcort ical areas (thalamus, h ippocampus) . Cal ibrat ions bars in u.m. Magnif ication in B & C and A , D, & E are identical. III. M H C Immunohistochemistry 5 1 Figure 14. M H C I immunohistochemistry in frontal sect ions through posterior neocortex in M e H g (A) and N C (B). Photomicrographs X 1 7 . III. M H C Immunohistochemistry 5 2 Figure 15. M H C I immunoreactivity in the cerebel lum and medul la of an M e H g rat. M H C posit ive cel ls were mostly limited to lobules 1-3 of the cerebel lar vermis while the lateral hemispheres and deep nuclei (Int, L) exhibited little or no immunoreact ivi ty. Intense staining was found in the vestibular nuclei (SuVe , LVe) . M H C immunoreactivity was inconsistent and apparent in only a few areas of the medul la. Magnif ication X 1 6 . 53 III. M H C Immunohistochemistry 5 4 ant ibodies posit ively identify cel ls which are morphological ly similar to reactive microgl ia (Ak iyama, et a l . , 1988), in which case these antibodies may reflect an acute immune response in pathologically compromised areas of the brain. Anatomical studies compar ing the pathogenesis of methylmercury toxicity in the fetal, postnatal, and adult brain have demonstrated that the immature nervous system exhibited a widespread pathology which became reduced and focussed with increasing maturity (Matsumoto et a l . , 1965; Chang and Annau , 1984). At all ages , though, the cerebel lar and cerebral cortex were descr ibed as being compromised. In this study we have confirmed the appearance of a pathogenetic process distributed throughout the cerebral cortex, but conspicuously missing from the cerebel lum. Although pathogenesis in the presubiculum, parasubiculum and cingulate cortex, among the most intensely M H C + immunoreact ive areas in this study, had not been previously descr ibed, recent studies have confirmed the presence of degenerat ing neurons in cingulate and retrosplenial areas following neonatal exposure to methylmercury (O 'Kusky , unpubl ished results). Perhaps the most puzzl ing discovery in the present study was the apparent lack of M H C immunoreactivity in the cerebel lum given the ubiquitous neuronal degenerat ion in cerebral and cerebel lar cort ices which has been descr ibed in cl inical and experimental reports of neuropathology in humans and animals (for review see C h a n g , 1977, 1988; Chang and Annau , 1984). If MHC+ immunoreactivity does in fact reveal an immune response in the brain which underlies a neuropathological state, one would expect its localization there to be extensive. Of particular interest to the preceding ser ies of experiments was the demonstration of strong M H C + immunoreactivity in the red nucleus and substantia nigra at an age when corresponding physiopathology had been descr ibed with cytochrome ox idase histochemistry (see Sect ions I & II). It is possible that a direct (MeHg-induced) or indirect (via afferents) toxic effect on intensely-staining cytochrome oxidase-posi t ive neurons accounts for the reduction in total number seen at an age equivalent to that in this study. III. MHC Immunohistochemistry Ultrastructural analysis of the MHC + regions demonstrated in this study are required future studies. Conclusion The histochemical localization of cytochrome oxidase (CO) revealed the presence of metabolical ly and presumably functionally distinct neurons in the brainstem of developing rats neonatal ly exposed to methylmercury. Intensely-staining CO-pos i t i ve neurons ( ICONs) were identifed by their distinct histochemical and morphological appearance, but more importantly by their anatomical local ization. A dense , ovoid deposit ion of cytochrome oxidase chromogen in the cell body was characteristic of ICONs . The CO-posi t ive area was approximately 14 u.m in diameter and was often surrounded by a clear halo. Occasional ly , very long (up to 100 u.m) CO-posi t ive processes were observed to emanate from the cell body. I C O N s were distinctly and consistently distributed within the magnocel lular red nuc leus ( R M C ) and the interrubral mesencepha lon (IRM) of methylmercury-treated animals (MeHg). The vague term IRM was intentionally coined to descr ibe the distribution of ICONs which were not confined to traditionally-defined boundaries of the red nucleus yet were always demonstrated in caudorostral , dorsoventral and mediolateral directions to lie between the cauda l , magnocel lular part of the red nuclei . In addit ion, their distribution was not known to correspond to any previously defined anatomical boundaries. A n increase in the oxidative metabolic activity of ICONs was initially demonstrated at the onset of methylmercury-induced movement and postural d isorders. Subsequent investigations revealed that an increase in activity was also reflected by an increase in the total number of ICONs , and the onset of their appearance on postnatal day (PND) 16 was demonstrated to preceed the clinical s igns of neurological impairment by more than one week. A four-fold increase in the total number of ICONs on P N D 18 was followed by a gradual decl ine to P N D 25. In MeHg-treated rats receiving unilateral lesions of the cerebral cortex or cerebel lum on P N D 10, a significant decrease in the total number of Conclus ion 5 7 ICONs in both the R M C and the IRM was only seen following cerebellar lesions, indicating that their h istochemical appearance was , at least partially, due to cerebel lar activation. Converging evidence suggested that a major contribution to the development of movement and postural disorders in the MeHg- induced cerebral palsy syndrome in neonatal rats may be a disinhibited output of the cerebel lum. This abnormal activation was manifested in the present ser ies of studies by the intense C O staining in neurons of the rubral mesencepha lon . The relatively high susceptibi l i ty of inhibitory G A B A e r g i c neurons to infectious, metabolic and toxic disturbances in the brain has been implicated in the aetiology of animal models of var ious human motor disorders (see Thes is rationale). The susceptibil i ty of G A B A e r g i c cerebel lar Purkinje neurons to viral and toxic insult has been reliably demonstrated. Lymphatic choriomeningit is viral infection in rats has been shown to result in the atrophy of Purkinje neurons (Amaral et a l . , 1982). Oral consumpt ion of diphenylhydantoin in cats induced widespread degenerat ion of Purkinje and granule cel ls (Utterback et a l . , 1958). Chronic hypoxia induced by exposure to carbon monoxide has been reported to cause severe degeneration of Purkinje, basket and stellate cel ls in the cerebel lum (Chan-Pa lay et a l . , 1976). In addit ion, Purkinje neurons are congenital ly compromised in several spec ies of mutant mice. Nervous, Purkinje cell degeneration (Roff ler-Tarlov et a l . , 1979), staggerer (Bradley and Berry, 1978), leaner and tottering , (Meier and M c P i k e , 1970; Herrup and Wi lczynsky , 1982), and stumbler (Caddy et a l . , 1981) mice all exhibit ultrastructural ev idence of degenerat ion of Purkinje cel ls. Histogenesis and morphogenesis of the cerebel lum are normal in the Purkinje cell degeneration mutants until postnatal day 15. Degeneration reached a peak at 20-28 days and resulted in almost a complete loss of Purkinje cel ls (Roffler-Tarlov et a l . , 1979). A concurrent degenerat ion of some other G A B A e r g i c neurons, specif ical ly the retinal photoreceptors and olfactory bulb mitral cel ls was also observed in these animals (Greer. Conclus ion 5 8 and Shepherd , 1982), suggesting that the degenerative effects were not specif ic to Purkinje cel ls but were extended to several c lasses of G A B A e r g i c neurons. In cl in ical studies of spasticity (Davidoff, 1978) and epi lepsy (Fariel lo and T icku , 1983), the administration of G A B A receptor agonists has demonstrated some efficacy in relief of symptoms. The most striking cl inical support for the efficacy of accentuated G A B A e r g i c neurotransmission was revealed in the use of chronic cerebel lar stimulation for relief of se izure d isorders and spasticity in cerebral palsy patients (Cooper , 1978). Signif icant suppress ion of seizures was reported in 18 of 27 patients undergoing chronic cerebel lar cortical stimulation ( C C S ) for 13 to 53 months. Whi le C C S did not appear to alleviate some symptoms of cerebral palsy - specif ical ly reflexes and muscle strength, significant improvement of spasticity and athetosis was reported. Al l improvements were functionally manifested in the ability to control larger body segments rather than in fine skil led or bimanual acts. The improvements were also reported to increase as a function of t ime. A cl inical ly useful improvement in v isual acuity was reported in 1 8 % of follow-up patients. Many of the symptoms which were alleviated were not regarded as arising from cerebel lar dysfunct ion but were nevertheless amel iorated by the "prosthetic mobil ization of the inhibitory potential of the cerebel lum". The author conc ludes that the "judicious use of cerebel lar stimulation is indicated for the functional and symptomat ic relief of cerebral palsy pat ients". B a s e d on the studies presented herein it appears that the relative susceptibil ity of G A B A e r g i c interneurons following perinatal traumatic insult or in congeni ta l abnormalit ies may provide a common aetiology for many developmental movement and postural d isorders . Conclus ion 5 9 Direction of future studies Even though the results of the series of experiments presented in this thesis seem to address several important issues in the aetiology of MeHg- induced movement and postural d isorders, they seem only preliminary in compar ison to the end less possibil i ty of future research quest ions. A description of several possible avenues follows in point form: 1) Light- and electron microscopic evaluation of neuropathology in the central nervous sys tem of MeHg-treated rats especial ly in: A) Cerebel lum B) Red nucleus C) Substant ia nigra. Had the ICONs not turned out to be such an interesting anomaly, I would have studied the effects of M e H g on the S N with more vigor. D) A reas demonstrating M H C immunoreactivity. 2) Character izat ion of the anatomy, neurochemistry and physiology of I C O N s : A) Ultrastructural examination using C O histochemistry at the electron microscop ic level . B) Insight into the connectivity and function of ICONs could arise from knowledge of their neurotransmitter specif ici ty. Prel iminary attempts to identify an intrinsic neurotransmitter using ant ibodies against dopamine-p -hydroxy lase , tyrosine hydroxylase, chol ine acetyl t ransferase, glutamate decarboxy lase, glutamate and noradrenaline have all y ie lded negative results (unpublished results). Studies should continue with other antibodies including those against neuropept ides. Conclus ion 6 0 C) Use of retrograde and anterograde tracing techniques to examine the projections from and afferents to ICONs . D) Electrophysiological examination of I C O N s to examine the correlation of neuronal activity to observed C O activity. 3) Therapeut ic intervention to ameliorate MeHg- induced movement and postural d isorders in rats us ing: A) G A B A agonists or benzodiazepines B) glutamate antagonists such as the N M D A receptor blocker M K 801 4) Application of the above techniques to other animal models of movement and postural d isorders : A) Hypoxia- induced B) V i ra l - induced C) Mutant strains 61 B ib l i og raphy Abercrombie, M. , Estimation of nuclear population from microtomic sect ions, Anat. R e c , 94 (1946) 2 3 9 . Ak i yama, H., Itagaki, S . , M c G e e r , P.L. , Major histocompatibil ity complex express ion on rat microgl ia following epidural kainic acid les ions, J. Neurosci. Res., (1988, in press). Al tmann, H., Ten Bruggencate, G . , P ickelmann, P. and Steinberg, R., Effects of G A B A , g lyc ine, picrotoxin and bicucull ine on rubrospinal neurons in ca ts , Brain Res., 111 (1976 ) 3 3 7 - 3 4 5 . Amara l , D .G. , Foss , J . , Monjan, A .A . , and Cerro, M. , A Golgi analysis of cerebel lar immunopathy in neonatal rats after infection with lymphocyte choriomeningit is v i rus, J. Neurol., 17 (1982) 71 -82 . Amin-Zak i , L , Majeed, M.A., E lhassan i , S . B . , C larkson, T.W., Greenwood, M.R. , and Doherty, R .A. , Prenata l methylmercury po ison ing, Am. J. Dis. Child., 133 (1979) 1 7 2 - 1 7 7 . Annau , Z . , and Ecc les , C . U . , Sensory deficits caused by exposure to methyl mercury, In C U . Ecc les and Z . Annau (Eds.) , The Toxicity of Methylmercury, John Hopkins Univ. P ress , Ba l t imore , 1987 , pp. 104 -113 . Angaut, P. , The ascending projections of the nucleus interpositus posterior of the cat cerebe l lum: A n experimental anatomical study using silver impregnation methods, Brain Res., 24 (1970) 377 -394 . Arregui , A . , and Barer , Chron ic hypoxia in rats: Alterations of striato-nigral angiotensin converting enzyme, G A B A , and glutamic acid decarboxylase, J. Neurochem., 34 ( 1980 ) 7 4 0 - 7 4 3 . 6 2 Bakir , F., Damluj i , S . F . , Amin -Zak i , L , Mur tadha, M. , Khal id i , A . , A l - R a w i , N.Y. , Tikrit i, S . , Dhahir, H.I., C la rkson , T .W. , Smi th , J . C , and Diherty, R.A. , Methylmercury poisoning in Iraq, Science, 181 (1973) 230 -241 . Bi l lard, J . M . , Batini , C , and Danie l , H., The red nucleus activity in rats deprived of the inferior ol ivary complex , Behav. Brain Res., 28 (1988) 127-130. Boni l la , E . , Neurochem. Res., 5 (1980) 209-215. Bradley, P. , and Berry, M. , The Purkinje cell dendrit ic tree in mutant mouse cerebel lum. A quantitative Golg i study of Weaver and Staggerer mice, Brain Res., 142 (1978) 1 3 5 - 1 4 2 . Brai lowsky, S . , Menin i , C , Si lva-Barrat , C , and Naquet, R., Epi leptogenic y-aminobutyr ic acid-wi thdrawal syndrome after chron ic , intracort ical infusion in baboons , Neurosci. Lett., 74 (1987) 75-80. Broman, T., On bas ic aspects of the blood-brain barrier. In J . Minckler (Ed.), Pathology of the Nervous System, Blakiston, New York, 1967, chap. 33 . Caddy , K.T., and S idman, R . L , Purkinje cel ls and granule cel ls in the cerebel lum of of the Stumbler mutant mouse , Dev. Brain Res., 1 (1981) 221-236. Caddy , K.T., S idman , R . L , and Eicher, E . M . , Stumbler, a new mutant mouse with cerebel lar d i s e a s e , Brain Res., 208 (1981) 251-255 . Cass idy , D.R. and Furr, A . , Toxicity of inorganic and organic mercury compounds in animals. In F.W. Oehme (Ed.), Toxicity of Heavy Metals in the Environment, Dekker, N e w York , 1978 , 303-330 . Cast ro , A . , Project ions of superior cerebel lar peduncle in rats and the development of new connect ions in response to neonatal hemicerebel lectomy, J. Comp. Neurol.,MS ( 1978 ) 6 1 1 - 6 2 8 . C h a n g , L.W., Neurotoxic effects of mercury-A review, Environ. Res., 14 (1977) 329-3 7 3 . 6 3 C h a n g , L.W., Pathological effects of mercury poisoning. In J . O . Nriagu (Ed.), Biogeochemistry of Mercury, E lsevier , New York, 1979, pp. 518-580 . C h a n g , L.W., Neurotoxic effects of mercury. In P . S . Spencer and H.H. Schaumberg (Eds.), Experimental and Clinical Neurotoxicology, Wi l l iams and Wi lk ins, Bal t imore, 1980, pp . 5 0 8 - 5 2 6 . C h a n g , L.W., and Annau , Z. , Developmental neuropathology and behavioral teratology of methylmercury. In J . Yana i (Ed.), Neurobehavioral Teratology, E lsev ier , Amste rdam, 1984 , pp . 4 0 5 - 4 3 2 . C h a n g , L.W., and Reuhl , K.R. , Mercury effects on human and animal health. In J . Rose (Ed.), Trace Elements and Health, IPC S c i . Technol . P ress , New York, 1982. Chan-Pa lay , V. , Palay, S . L , and McCroskey , L , The effects of carbon monoxide on neurons of the cerebe l lum, Neuropath. Appl. Neurobiol., 2 (1976) 293-312. Cho i , B., C h o , K .H . , and Lapham, I.W., Effects of methylmercury on human fetal neurons and astrocutes in vitro: a time lapse cinematographic, phase and electron microscopic study, J. Neuropathol. Exp. Neurol., 38 (1981) 307. Cho i , B .H . , K im, R .C . , and Peckham, N.H. , Hydrocephalus following prenatal methy lmercury po ison ing , Acta Neuropathol., 75 (1988) 325-330 . Cho i , B., Lapham, L.W., Amin-Zak i , L., and S a l e e m , T., Abnormal neuronal migration, deranged cerebral cortical organizat ion, and diffuse white matter astrocytosis of human fetal brain: A major effect of methylmercury poisoning in utero, J. Neuropath. Exp. Neurol., 37 (1978) 7 1 9 - 7 3 3 . C la rkson , T .W. , Methylmercury toxicity to the mature and developing nervous sys tem: Poss ib le mechan isms. In B. Sarkar (Ed.), Biological Aspects of Metals and Metal-Related Diseases, Raven Press , New York, 1983, pp. 183-197. 6 4 Co l in , F., Mani l , J . , and Descl in , J . C , The ol ivocerebellar system. I. Delayed and slow inhibitory effects: an over looked salient feature of cerebel lar c l imbing f ibers, Brain Res., 187 (1980) 3 -27 . Cond ie , R .M . , and G o o d , R.A., Experimental allergic encephalomyel i t is: its production, prevention, and pathology as studied by light and electron microscopy, Prog. Neurobiol., 4 (1959) 321 -392 . Cooper , I.S., Cerebellar Stimulation in Man, Raven P ress , 1978. Darriet, D., Der, T., and Col l ins , R . C , Distribution of cytochrome ox idase in rat brain: studies with diaminobenzidine histochemistry in vitro and [ ^ 4 C] cyan ide t issue labeling in v ivo, J. Cerebral Blood Flow Metab., 6 (1986) 8-14. Davidoff, R .A. , Pharmacology of spasticity, Neurology, 28 (1978) 46 -51 . Dekker , J . J . , Anatomica l ev idence for direct fiber projections from the cerebel lar nucleus interpositus to rubrospinal neurons. A quantitative E M study in the rat combining anterograde and retrograde intra-axonal tracing methods, Brain Res., 205 (1981) 2 2 9 - 2 4 4 . Dunwiddie, T., Muel ler, A . , Palmer, M. , Stewart, J . , and Hoffer, B., Electrophysiological interactions of enkephal ins with neuronal circuitry in the rat h ippocampus. I. Effects on pyramidal cel l activity, Brain Res., 184 (1980) 311 -330 . Dyck, R . H . and O'Kusky, J . R . , Increased activity of mesencephal ic neurons in developing rats displaying methylmercury- induced movement and postural d isorders, Neurosci. Lett., 89 (1988) 721 -726 . Dyck, R .H . , Whishaw, I.Q. and Suther land, R . J . , Dramatic neurotoxicity of neonatal intracerebral injection of tubulin-binding agents, Soc. Neurosci. Abstr., 10 (1984) p . 1 1 9 8 . Ecc les , C . U . , and Annau , Z. , Prenatal methylmercury exposure, I: Alterations in neonatal activity, Neurobehav. Toxical. Terat., 4 (1982a) 371-376. 6 5 Ecc les , C . U . , and Annau , Z. , Prenatal methylmercury exposure, II: Alterations in learning and psychotropic drug sensitivity in adult offspring, Neurobehav. Toxical. Terat., 4 ( 1 9 8 2 a ) 3 7 7 - 3 8 2 . Edwards , G . N . , St. Bartholomew's Hosp. Rep., (London), 141 (1) 1865. Fariel lo, R . G . , and Ticku, M.K., The perspective of G A B A replenishment therapy in the ep i leps ies : A crit ical evaluat ion of hopes and concerns , Life Sci., 33 (1983) 1629-1 6 4 0 . Francis , A . , and Pulsinel l i , W., The response of G A B A e r g i c and cholinergic neurons to transient cerebra l i s chemia , Brain Res., 243 (1982) 271-278 . Fujito, Y . , Watanabe, S . , Kobayash i , H. and Tsukahara , N. , Lesion- induced sprouting in the red nucleus at the early developmental stage, In S . Finger and C R . Almli (Eds.) , Early Brain Damage, Neurobiology and Behavior, Vol. 2 , New York, Academic P ress , 1984, pp . 3 5 - 4 7 . Fuyuta, M. , Fujimoto, T., Kiyofuji, E., Teratogenic effects of a single oral administration of methy lmercur ic ch lor ide in mice, Acta. Anat. (Basel), 104 (1979) 356-362 . Giuffrida, R., Palmer i , A . , Raffaele, R., R i cca , G . , and Sap ienza , S . , Convergence pattern of cortical and interposital inf luences on rubrospinal neurons in the cat, Behav. Brain Res., 28 (1988) 109-113 . Gramsbergen , A . and I jkema-Paassen, J . , The effects of early cerebel lar hemispherectomy in the rat: Behavioural , neuroanatomical and electrophysiological seque lae , In S . Finger and C R . Almli (Eds.), Early Brain Damage, Neurobiology and Behavior, Vol. 2, New York, A c a d e m i c P r e s s , 1984, pp. 155-177. Greer , C .A . , and Shepherd , G . M . , Mitral cell degeneration and sensory function in the neurological mutant mouse Purkinje cel l degenerat ion ( P C D ) , Brain Res., 235 (1982 ) 1 5 6 - 1 6 1 . 6 6 Gwyn , D .G. , and Flumerfelt, B.A. , A compar ison of the distribution of cortical and cerebel lar afferents in the red nucleus of the rat, Brain Res., 69 (1974) 130-135. Harada , M. , Methyl mercury poisoning due to environmental contamination ("Minimata Disease") . In F.W. Oehme (Ed.), Toxicity of Heavy Metals in the Environment, Dekker, N e w York , 1978, pp. 261-302 . Harada, Y . , Congeni ta l (or fetal) Minimata d isease. In M. Kutsuna (Ed.), Minimata Disease, Study group of Minimata Disease, Kumamoto University P ress , J a p a n , 1968, pp . 9 3 - 1 1 7 . Herrup, K., and Wi lczynsk i , S .L . , Cerebel lar cel l degenerat ion in the leaner mutant mouse, Neurosci., 7 (1982) 2185 -2196 . Harr is, S . B . , Wi l son , J . G . , and Pr inz, R .H . , Embryotoxicology of methylmercuric chloride in go lden hamsters , Teratology, 6 (1972) 139-142. Houser, C . R . , Hendry, S . H . C . , Jones , E .G . , and Vaughn, J . E . , Morphological diversity of immunocytochemical ly identified G A B A neurons in the monkey sensory-motor cortex, J. Neurocytol., 12 (1983) 617-638 . Hughes, J .A . , and Sparber, S . B . , cJ-Amphetamine unmasks potential consequences of exposure to methylmercury in utero: Methods for studying behavioral teratogenesis, Pharmacol. Biochem. Behav., (1978) 365-375 . Hunter, D., Bomford, R.R. , and Russe l l , D.S. , Poisoning by methylmercury compounds, Q. J. Med., 9 (1940) 193-214. Hunter, D. and Russe l l , D.S. , Focal cerebral and cerebel lar atrophy in a human subject due to organic mercury compounds . J. Neurol. Neurosurg. Psychiat., 17 (1954) 235-2 5 4 . Ito, M. , The Cerebellum and Neural Control, Raven Press , New York, 1984. 6 7 Kataoka , K., Bak, I.J., and Markham, C . H . , Alteration in tyrosine hydroxylase, glutamic acid decarboxy lase and choline acetyltransferase in basal gangl ia following herpes s implex virus inoculat ion in rat neostr iatum, Brain Res., 169 (1979) 401-405 . Katsumara, H., Murakami , F., W u , J . Y . , and Tsukahara , N., G A B A e r g i c intrinsic interneurons in the red nucleus of the cat demonstrated with combined immunocytochemistry and anterograde degeneration methods, Neurosci. Res., 1 (1984) 3 5 - 4 4 . Khera , K .S . , Teratogenic effects of methylmercury in the cat. Note on the use of this spec ies as a model for teratogenicity studies, Teratology, 8 (1973) 293-304. Khera , K . S . , Iverson, F., Hierlihy, L , Tanner , R., and Trivett, G . , Toxicity of methyl mercury in neonatal ca ts , Teratology, 10 (1974) 69-76. Khera , K .S . , and Tabacova , S .A . , Effects of methylmercuric chloride on the progeny of mice and rats treated before or during gestat ion, Fd. Cosmet. Toxicol., 11 (1973) 243-2 5 4 . Kolb, B .E . , and Whishaw, I.Q., Decortication in rats in infancy or adulthood produced comparable functional losses on learned and species-typical behaviors, J. Comp. Physiol. Psychol., 95 (1981) 468 -483 . Komula inen, H., and Tuomisto, J . , The neurochemical effects of methyl mercury in the brain, In C U . Ecc les and Z . Annau (Eds.), The Toxicity of Methylmercury, John Hopk ins Univ. P r e s s , Bal t imore, 1987, pp. 172-188. Kubota, M. , Sakaguch i , H. and Tsukahara, N., Release of endogenous G A B A from the cat red nuc leus s l i ces , Brain Res., 270 (1983) 190-192. Land, P.W., Dependence of cytochrome oxidase activity in the rat lateral geniculate nucleus on retinal innervat ion, J. Comp. Neurol., 262 (1987) 78-89. Leong , S . K . , Plasticity of cerebel lar efferents after neonatal lesion in albino rats, Neurosci. Lett., 7 (1978) 281-289 . 6 8 Leong , S .K . , and Lund, R.D. , Anomolous bilateral corticofugal pathways in albino rats after neonata l les ions , Brain Res., 62 (1973) 218-221 . L im, K .H . and Leong, S .K . , Aberrant bilateral projections from the dentate and interposed nuclei in albino rats after neonatal les ions, Brain Res., 96 (1975) 306-309. Loscher, W. , Frey , H.H., Reiche, R., and Schul tz, D., High anticonvulsant potency of gamma-aminobutyr ic ac id (GABA)mimet ic drugs in gerbi ls with genetical ly determined ep i lepsy , J. Pharmacol. Exp. Ther., 226 (1983) 839-844. Loscher , W. , Ev idence for abnormal sensitivity of the G A B A system in gerbils with genetically determined epi lepsy. In R . G . Fariel lo, P.L. Morsel l i , K . G . L loyd, L.F. Quesney , and J . Engel Jr . (Eds.), Neurotransmitters, Seizures, and Epilepsy II, Raven P r e s s , 1984, New York, pp 179-190. Luse , S .A . , and McDouga l , D.B., Electron-microscopic observat ions on allergic encephalomyel i t is in the rabbit, J. Exp. Med., 112 (1960) 735-742 . Marsh , D.O., Dose- response relationships in humans: methyl mercury epidemics in Japan and Iraq, In C U . Ecc les and Z . Annau (Eds.), The Toxicity of Methylmercury, John Hopk ins Univ. P r e s s , Bal t imore, 1987, pp. 45-53 . Martin, G . F . , C a b a n a , T., and Hazlett, J . C , The development of selected rubral connect ions in the North Amer ican opossum, Beh. Brain. Res., 28 (1988) 21-28. Matsumoto, H., Koya , G . , and Takeuchi , T., Fetal minimata d isease : A neuropathological study of two c a s e s of intrauterine intoxication by a methyl mercury compound, J. Neuropathol. Exp. Neurol., 24 (1965) 563-574 . Matsumoto, Y . , and Fujiwara, M. , In situ detection of c lass I and c lass II major histocompatabil i ty complex antigens in the rat central nervous system during exper imental al lergic encephalomyel i t is . A n immunohistochemical study, J. Neuroimmunol., 12 (1986) 265-277 . 6 9 Matsumoto, Y . , Hara , N. , Tanaka , R., and Fujiwara, M. , Immunohistochemical analysis of the rat central nervous system during experimental al lergic encephalomyel i t is , with spec ia l reference la-posit ive cel ls with dendrit ic morphology, J. Immunol., 135 (1986 ) 3 6 6 8 - 3 6 7 6 . Meier, H., and MacP ike , A .D . , A neurological mutation of (msd) of the mouse causing a def ic iency of myelin synthes is , Exp. Brain Res., 10 (1970) 512-525 . M c G e e r , P . L , Itagaki, S . , Tago, H., McGeer , E . G . React ive microglia in patients with senile dement ia of the Alzheimer type are positive for the histocompatibil i ty glycoprotein H L A - D R , Neurosci. Lett, 79 (1987a) 195-200. McGeer , P . L , McGeer , E . G . , Itagaki, S . , Mizukawa, K., Anatomy and pathology of the basal gang l ia , Can. J. Neurol. Sci., 14 (1987b) 363-372 . McGeer , P . L , Itagaki, S . , Boyes , B .E . , McGeer , E . G . , React ive microglia are positive for H L A - D R in the substant ia nigra of Park inson's and Alzheimer 's d isease brains, Neurology, (1987c, in press) . Muller, P .B . , and Langemann, H., Distribution of glutamic acid decarboxy lase activity in human bra in, J. Neurochem.,2 (1962) 399 -401 . Murakami , F., Katsumaru, H., Sai to, K., and Tsukahara , N. , A quantitative study of synaptic reorganization in red nucleus neurons after lesion of the nucleus interpositus of the cat: and electron microscopic study involving intracellular injection of horseradish pe rox i dase , Brain Res., 242 (1982) 41-53 . Murakami , F., Katsumaru, H., W u , J . -Y . , Matsuda, T. and Tsukahara , N., Immunocytochemical demonstration of G A B A e r g i c synapses on identified rubrospinal neurons , Brain Res., 267 (1983) 357-360 . Murakami , U., The effect of organic mercury on intrauterine life, Adv. Exp. Med. Biol., 27 ( 1972 ) 3 0 1 - 3 3 6 . 7 0 N a h , S . H . , and Leong , S .K . , Bilateral corticofugal projection to the red nucleus after neonatal les ions in the albino rat, Brain Res., 107 (1976a) 433-436 . N a h , S . H . , and Leong, S .K . , A n ultrastructural study of the anomolous corticorubral projection fol lowing neonatal les ions in the albino rat, Brain Res., 111 (1976b) 1 6 2 - 1 6 6 . Naus , C . G . , Flumerfelt, B.A. and Hrycyshyn, A .W. , An anterograde H R P - W G A study of aberrant cort icorubral projections following neonatal les ions of the rat sensor imotor cortex, Exp. Brain Res., 59 (1985) 365 -371 . Nieoul lon, A . and Dusticier, N. , Decrease in chol ine acetyltransferase and in high affinity glutamate uptake in the red nucleus of the cat after cerebel lar les ions, Neurosci. Lett, 24 (1981) 2 6 7 - 2 7 1 . Nieoul lon, A . and Dusticier, N. , Glutamate decarboxy lase distribution in discrete motor nuclei in the cat brain, J. Neurochem., 37 (1981) 202-209. Nieoul lon, A . , Kerker ian, L. and Dusticier, N. , High affinity glutamate uptake in the red nucleus and ventrolateral thalamus after lesion of the cerebel lum in the adult cat: b iochemical ev idence for functional changes in the deafferented structures, Exp. Brain Res., 55 (1984) 409 -419 . Nieoul lon, A . , Vui l lon-Cacciut to lo, G . , Dusticier, N., Kerker ian, L., Andre , D., and Bosler , O. , Putative neurotransmitters in the red nucleus and their involvement in post lesion adapt ive mechan i sms , Behav. Brain Res., 28 (1988) 163-174. Ohno , T., Kernicterus: effect on choline acetyltransferase, glutamic acid decarboxy lase and tyrosine hydroxylase activit ies in the brain of the Gunn rat, Brain Res., 196 (1980) 2 8 2 - 2 8 5 . Ohno, T., Kernicterus: enzymat ic evidence for difference between the development of chol inergic and G A B A e r g i c innervations in the brain of the Gunn rat, J. Neurochem., 3 7 (1981) 1 0 6 1 - 1 0 6 3 . 71 O 'Kusky , J . R . , Methylmercury poisoning in the developing nervous sys tem: morphological changes in neuronal mi tochondr ia, Acta Neuropathol. (Berk), 61 (1983) 116-122. O 'Kusky , J . R . , Synapt ic degenerat ion in rat v isual cortex after neonatal administration of methy lmercury , Exp. Neurol., 89 (1985) 32-47 . O 'Kusky, J .R . , Boyes , B .E . , and McGeer , E . G . , Methylmercury-induced movement and postural disorders in developing rat: regional analysis of brain catecholamines and indo leamines , Brain Res., 439 (1988a) 138-146. O 'Kusky, J .R . , and McGeer , E . G . , Methylmercury poisoning of the developing nervous system in the rat: decreased activity of glutamic acid decarboxy lase in cerebral cortex and neostr iatum, Dev. Brain Res., 21 (1985) 299-306. O 'Kusky , J . R . , Radke , J . M . , and Vincent, S .R . , Methylmercury- induced movement and postural d isorders in developing rat: loss of somatostat in- immunoreact ive interneurons in the str iatum, Dev. Brain Res., 40 (1988b) 11-25. Omata , S . , Horigome, T., Momose , Y . , Kambayash i , M. , Mochizuki , M. , and Sugano, H., Effect of methylmercury chloride on the in vivo rate of protein synthesis in the brain of the rat: Examinat ion with the injection of a large quantity of [ 1 4 C ] val ine, Toxicol. Appl. Pharmacol., 56 (1980) 207 -215 . Pax inos, G . , and Watson, C , The Rat Brain in Stereotaxic Coordinates, Academic P ress , Toronto , 1986. Perry, T.L. , Berry, K., Hansen , S . , Diamond, S . , and Mok, C , Regional distribution of amino ac ids in human brain obtained at autopsy, J. Neurochem., 18 (1971) 513-5 1 9 . Povl ishock, J .T . , Taylor, J . J . and Se ibe l , H.R., Electron microscopic study of the maturing rat red nuc leus. I. The large-neuron populat ion, Acta Anal 102 (1978)129-146. 7 2 R a m e l , C , Methylmercury as a mitosis disturbing agent, Jpn. J. Exp. Med., 61 (1968) 1 0 7 2 - 1 0 7 6 . Ribak, C . E . , Asp inous and sparsely-spinous stellate neurons in the v isual cortex of rats contain glutamic ac id decarboxy lase , J. Neurocyte-/., 7 (1978) 461-478. Ribak, C . E . , Harris, A . B . , Vaughn, J . E . , and Roberts, E. , Inhibitory G A B A e r g i c nerve terminals dec rease at si tes of focal epi lepsy, Science, 205 (1979) 211-214. Ribak, C . E . , Harris, A . B . , Vaughn, J . E . , and Roberts, E. , Immunocytochemical changes in cortical G A B A neurons in a monkey model of epi lepsy. In P.L. Morsel l i , K . G . LLoyd, W . Loscher , B. Meldrum, and E . H . Reynolds (Eds.), Neurotransmitters, Seizures, and Epilepsy, Raven Press , New York, P P . 11-22. Ribak, C . E . , Vaughn, J . E . , and Barber, R .P . , Immunocytochemical localizatin of G A B A e r g i c neurones at the electron microscopic level , Histochem. J., 13 (1981) 555-582. Roberts, R . C , Ribak, C . E . , Peterson, G . M . , and Oertel , W . H . , Increased numbers of G A B A e r g i c neurons in the inferior col l iculus of the genetically epi lepsy prone rat, Soc. Neurosci. Abstr., 10 (1984) 410 . Rob in , M . M . , Pal f reyman, M .G . , Zra ika, M .M. , and Schechter , P . J . , Mapping of dyskinetic movements induced by local application of picrotoxin or (+)-y-acetylinic G A B A on the rat motor cortex, Eur. J. Pharmacol., 65 (1980) 411-415 . Romi j in, H . J . , Ruijter, J . M . , and Wolters, P . S . , Hypox ia preferentially destroys G A B A e r g i c neurons in developing rat neocortex explants in culture, Exp. Neurol., 100 (1988) 3 3 2 - 3 4 0 . Roffler-Tarlov, S . , Beart, P . M . , O 'Gorman, S . , and S idman, R.L. , Neurochemical and morphological consequences of axon terminal degeneration in cerebel lar deep nuclei of mice with inherited Purkinje cel l degenerat ion, Brain Res., 168 (1979) 75-95 . Sager , P .R . , Doherty, R.A. , and Olmstead, J . B . , Interaction of methlmercury with m i c r o t u b u l e s , J. Cell. Biol., 91 (1981) 3 3 0 a . 7 3 Shamoo , A . E . , Biological target sites of mercurials, In C U . Ecc les and Z . Annau (Eds.), The Toxicity of Methylmercury, John Hopkins Univ. P ress , Bal t imore, 1987, pp. 13-23. S loper , J . J . , Johnson , P., and Powel l , T . P . S . , Select ive degenerat ion of interneurons in the motor cortex of infant monkeys following controlled hypoxia: a possible cause of e p i l e p s y , Brain Res., 198 (1980) 204-209 . Slotkin, T.A. , Pachman , S . , Kavlock, R . J . , and Bartolome, J . , Early b iochemical detection of adverse effects of a neurobehavioral teratogen: influence of prenatal methylmercury exposure on ornithine decarboxylase in brain and other t issues of fetal and neonatal rat, Teratology, 32 (1985) 195-202 . Spyker , J . M . , Sparber , S . B . , and Goldberg, A . M . , Subtle consequences of methylmercury exposure : Behav iora l deviat ions in offspring of treated mothers, Science, 177 (1972) 6 2 1 - 6 2 3 . Syve rsen , T . L . M . , Effects of methylmercury on in vivo protein synthesis in isolated cerebra l and cerebel lar neurons, Neuropath. Appl. Neurobiol., 3 (1977) 225-236. Syversen , T . L . M . , Changes in protein and R N A synthesis in rat brain neurons after a single d o s e of methy lmercury , Toxicol. Lett, 10 (1982) 31-34. Syversen , T . L . M . , Tot land, G . , and F lood, P.R. , Early morphological changes in rat cerebel lum c a u s e d by a single dose of methylmercury, Arch. Toxicol., 47 (1981) 1 0 1 - 1 1 1 . Takeuch i , T., Human effects of methylmercury as an environmental neurotoxicant. In K. Blum and L. Manzo (Eds.), Neurotoxicology, Dekker, New York, 1985, pp. 345-367. Takeuchi , T., and Eto, K., Pathology and pathogenesis of Mimata d isease. In T. Tsubaki and K. I rukayama (Eds.) , Minimata Disease, E lsev ier , New York , 1977, pp .103-141. Takeuchi , T., Eto, N. , and Eto, K., Neuropathology of chi ldhood cases of methyl mercury poisoning: Minimata d isease with prolonged symptoms, with particular reference to the decort icat ion syndrome, Neurotoxicol., 1 (1979) 1-20. Takeuch i , T., Mor ikawa, N., Matsumoto, H., and SHira ish i , A . , Pathologic study of Minimata d i sease in J a p a n , Acta. Neuropathol. (Berl), 2 (1957) 40-57. Thomas , D.J . , and Syversen , T .L .M . , The alteration of protein synthesis by methyl mercury, In C U . Ecc les and Z . Annau (Eds.), The Toxicity of Methylmercury, John Hopk ins Univ . P r e s s , Bal t imore, 1987, pp. 131-171 . Tokuomi, H., Uchino, M. , Imamura, S . , Yamanaga , H, Nakanishi , R., and Ideta, T., Minimata d isease (organic mercury poisoning): Neuroradiologic and electrophysiologic studies, Neurology, 32 (1982) 1369 -1375 . Toyama, K., Tsukahara , N. , Kosaka , K., and Matsunami, K., Synapt ic excitation of red nucleus neurons by f ibers from interpositus nuc leus, Exp. Brain Res., 11 (1970) 1 8 7 - 1 9 8 . Traugott, U., Multiple sc leros is : relevance of c lass I and c lass II MHC-exp ress ing cel ls to les ion deve lopment , J. Neuroimmunol., 16 (1987) 283-302. Tsubak i , T., and Irukayama, K., Minimata Disease, Elsevier , New York, 1977. Tsukahara , N. , and Fujito, Y . , Neuronal plasticity in the newborn and adult feline red nucleus, In H. Flohr and W . Precht (Eds.) Lesion-induced neuronal plasticity in sensorimotor systems, Spr inger, Ber l in, Heidelberg, New York, 1981, pp. 64-74. Tsukahara , N. , Fujito, Y . and Kubota, M. , Specif ici ty of the newly-formed corticorubral s y n a p s e s in the kitten red nuc leus, Exp. Brain. Res., 51 (1983) 45-56 . Tsukahara , N. , Hultborn, H. and Murakami , F., Sprouting of cortico-rubral synapses in red nucleus neurones after destruction of the nucleus interpositus of the cerebel lum, Experientia, 30 (1974) 57 -58 . Tsukahara , N., Hultborn, H., Murakami , F. and Fujito, Y . , Electrophysiological study of formation of new synapses and collateral sprouting in red nucleus neurons after partial denerva t ion , J. Neurophysiol., 38 (1975) 1359-1372. 7 5 Tsukahara , N. and Kosaka , K., The mode of cerebral excitation of red nucleus neurons, Exp. Brain Res., 5 (1968) 102-117. Utterback, R.A. , O jeman, R., and Malek, J . , Parenchymatous cerebel lar degenerat ion with Dilantin intoxicat ion, J. Neuropathol. Exp. Neurol., 17 (1958) 516. Van der Hyden, J . A . M . , DeKloet, E.R., Korf, J . and Versteeg, D.H.G. , G A B A content of discrete brain nuclei and spinal cord of the rat, J. Neurochem., 33 (1979) 857-861 . Vi l lab lanca, J . R . , Burgess , J .W. and Sonnier , B . J . , Neonatal cerebral hemispherectomy: a model for post lesion reorganization of the brain, In S . Finger and C R . Almli (Eds.) , Early Brain Damage, Neurobiology and Behavior, Vol. 2, New York, Academic P ress , 1984 , pp . 1 7 9 - 2 1 0 . Vi l lablanca, J .R . , O lmstead, C h . E . , Sonnier, B . J . , McAll ister, J . P . and G o m e z , F., Evidence for a c rossed corticorubral projection in cats with one cerebral hemisphere removed neonata l ly , Neurosci. Lett., 33 (1982) 241-246 . Voogd , J . , The Cerebellum of the Cat. Structure, and Fibre Connexions, Van Gorcum, A s s e n , 1 9 6 7 . Vui l lon-Cacciuttolo, G . , Bosler, O. and Nieoul lon, A. , G A B A neurons in the cat red nucleus: a b iochemica l and immunohis tochemical demonstrat ion, Neurosci. Lett, 52 (1984) 1 2 9 - 1 3 4 . Vui l lon-Cacciut tolo, G . , Bosler , O . and Nieoul lon, A . , Immunohistochemical ev idence of plasticity of y-aminobutyric acid neurons in the red nucleus and adjacent reticular formation after contralateral cerebel lectomy in the adult cat, Neurosci. Lett, 70 (1986 ) 3 0 8 - 3 1 3 . Wachte l , H., and A n d e n , N. , Motor activity of rats following intracerebral injections of drugs influencing G A B A mechanisms, Naunyn-Schmiedeberg's Arch. Pharmacol., 302 (1978 ) 1 3 3 - 1 3 9 . 7 6 Webster , H.F., Transient focal accumulat ion of axonal mitochondria during the early stages of Wal le r ian degenera t ion , J. Cell. Biol., 12 (1962) 361-383 . Wong-Ri ley , M.T.T., Changes in the visual system of monocularly sutured or enucleated cats demonstrab le with cytochrome ox idase histochemistry, Brain Res., 171 (1979) 1 1 - 2 8 . Wong-Ri ley , M. , and Carrol l , E.W., Effect of impulse blockade on cytochrome oxidase activity in monkey v isual sys tem, Nature, 307 (1984) 262-264 . Zi l les, K., The Cortex of the Rat, Spr inger -Ver lag , Ber l in , 1985. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0097668/manifest

Comment

Related Items