Theory of mind deficits following acute alcohol intoxication.

Acute alcohol consumption is associated with socially inappropriate behaviour. Such behaviour could in part reflect the potential of alcohol to interfere with social cognition. In this experiment we tested the hypothesis that acute alcohol consumption by regular heavy social drinking young adults would compromise an aspect of social cognition, namely theory of mind (understanding intentions, emotions and beliefs). Participants who had consumed 6-8 units of alcohol showed specific impairments on two theory of mind tests: identification of faux pas and emotion recognition. This result suggests that alcohol consumption could lead to social problems secondary to difficulties in interpreting the behaviour of others due to theory of mind impairments.

The selective vulnerability of striatopallidal neurons.

The different types of striatal neuron show a range of vulnerabilities to a variety of insults. This can be clearly seen in Huntington's disease where a well mapped pattern of pathological events occurs. Medium spiny projection (MSP) neurons are the first striatal cells to be affected as the disease progresses whilst interneurons, in particular the NADPH diaphorase positive ones, are spared even in the late stages of the disease. The MSP neurons themselves are also differentially affected. The death of MSP neurons in the patch compartment of the striatum precedes that in the matrix compartment and the MSP neurons of the dorsomedial caudate nucleus degenerate before those in the ventral lateral putamen. The enkephalin positive striatopallidal MSP neurons are also more vulnerable than the substance P/dynorphin MSP neurons. We review the potential causes of this selective vulnerability of striatopallidal neurons and discuss the roles of endogenous glutamate, nitric oxide and calcium binding proteins. It is concluded that MSP neurons in general are especially susceptible to disruptions of cellular respiration due to the enormous amount of energy they expend on maintaining unusually high transmembrane potentials. We go on to consider a subpopulation of enkephalinergic striatopallidal neurons in the rat which are particularly vulnerable. This subpopulation of neurons readily undergo apoptosis in response to experimental manipulations which affect dopamine and/or corticosteroid levels. We speculate that the cellular mechanisms underlying this cell death may also operate in degenerative disorders such as Huntington's disease thereby imposing an additional level of selectivity on the pattern of degeneration. The possible contribution of the selective death of striatopallidal neurons to a number of clinically important psychiatric conditions including obsessive compulsive disorders and Tourette's syndrome is also discussed.

Reversal of parkinsonian symptoms in primates by antagonism of excitatory amino acid transmission: potential mechanisms of action.

Parkinsonism is characterised by overactive glutamatergic transmission in the cortico-striatal and subthalamo-medial pallidal pathways. Local blockade of glutamatergic transmission in these pathways can alleviate parkinsonian symptoms. The effectiveness of the treatment, however, is often limited by the simultaneous appearance of unwanted side-effects. These side-effects, including ataxia and dissociative anaesthesia, are particularly problematic when N-methyl-D-aspartate (NMDA) antagonists are used. In an attempt to overcome these problems we have attempted to manipulate excitatory amino acid (EAA)-mediated neurotransmission indirectly by targeting the NMDA receptor associated modulatory sites. We review evidence which demonstrates that antagonists for both the NMDA associated glycine and polyamine sites can reverse parkinsonian symptoms when injected intra-cerebrally in both MPTP-treated and bilateral 6-OHDA lesioned marmosets without eliciting unwanted side-effects. We further review preliminary data which suggest that ifenprodil, a polyamine site antagonist, has striking anti-parkinsonian actions in the marmoset. Potential mechanisms of action underlying these effects are discussed in terms of NMDA receptor subtypes and the neuroanatomical locus of action. The anti-parkinsonian efficacy of intra-striatally administered EAA antagonists leads us to question the view of dopamine acting in the striatum as a simple neuromodulator.

Regional brain uptake of 2-deoxyglucose in N-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-induced parkinsonism in the macaque monkey.

This preliminary report describes application of the 2-deoxyglucose (2DG) autoradiographic technique to study regional changes in brain metabolism in experimental parkinsonism, induced in the monkey by administration of the neurotoxin MPTP. In one monkey, rendered severely parkinsonian by MPTP, there was a marked increase in the uptake of 2DG in the globus pallidus (both medial and lateral segments) and in the ventral anterior and ventral lateral thalamic nuclei, in comparison to non-parkinsonian animals. Increased uptake of 2DG in the globus pallidus may reflect increased activity of striatopallidal synapses secondary to loss of nigrostriatal dopaminergic neurones. The findings are in sharp contrast to our observations on regional brain metabolism in experimental choreiform dyskinesia in the monkey.

Contralateral head movements produced by microinjection of glutamate into superior colliculus of rats: evidence for mediation by multiple output pathways.

One of the major efferent pathways of the superior colliculus crosses midline to run caudally in the contralateral predorsal bundle, innervating targets in the brain stem and eventually reaching the cervical spinal cord. A variety of evidence suggests that this tecto-reticulo-spinal pathway may mediate the orienting movements that can be evoked by tectal stimulation. However, we have recently found that orienting head movements can still be obtained in rats after section of the tecto-reticulo-spinal pathway, implying that additional pathways are also involved. The present study sought to test this implication, by taking advantage of the fact that in rats the cells of origin of the tecto-reticulo-spinal pathway are largely segregated within the lateral part of the stratum album intermediate. It is thus possible to find out whether orienting head movements can be produced by a cell-excitant from tectal regions that contain few cells of origin of the tecto-reticulo-spinal pathway. Hooded rats in an open field were filmed during microinjections of sodium L-glutamate (50 mM, 200 nl) into the superior colliculus, and the films analysed for the appearance of contralaterally directed movements of the head and body. Subsequent histological reconstruction of the injection sites indicated that such movements could be obtained from widespread areas within the superior colliculus, including not only lateral stratum album intermediale but also the deep layers, and parts of the medial superficial and intermediate layers. Moreover, sites in or close to lateral stratum album intermediate often gave circling movements with downward pointing head, whereas some sites outside lateral stratum album intermediale gave sustained immobility with the head pointing contralaterally and upwards. This evidence supports the view that tectal efferent pathways besides the tecto-reticulo-spinal pathway are involved in the control of head movement. In addition, at least some of these pathways are not collaterals of the tecto-reticulo-spinal pathway, since the movements were obtained from collicular regions with few tecto-reticulo-spinal pathway cells. Finally, the results are consistent with the view that different collicular output pathways mediate movements that have different functions.

Responses resembling defensive behaviour produced by microinjection of glutamate into superior colliculus of rats.

Electrical stimulation of the superior colliculus in rats elicits not only orienting movements, as it does in other mammals, but also behaviours resembling such natural defensive responses as prolonged freezing, cringing, shying, and fast running and jumping. To investigate the location of the cells mediating these behaviours, the superior colliculus was systematically mapped with microinjections of sodium L-glutamate (50 mM, 200 nl), and the resultant behavioural changes as assessed in an open field were analysed for defence-like responses. The main regions that gave defensive behaviour were (i) rostromedial superior colliculus (all layers), and (ii) both medial and lateral parts of the caudal deep layers. Cells in these areas project into the ipsilateral descending pathway. However, the cells of origin of this pathway are also found in collicular regions, such as rostral intermediate gray and parts of far caudal colliculus, that did not give defensive movements in response to glutamate stimulation. It is unclear whether this is because only parts of the ipsilateral pathway mediate defensive behaviours, or because glutamate is a relatively inefficient stimulating agent for these systems. An unexpected feature of the results was that at a number of collicular sites the nature of the defensive response changed with successive (up to three) injections of glutamate, often appearing to become more intense. Whether the mechanism underlying this potentiation is related to the conditioning of natural defensive behaviour is unknown.

Death of neurons in the neonatal rodent and primate globus pallidus occurs by a mechanism of apoptosis.

We have examined the developing rat, mouse and marmoset globus pallidus for evidence of cells dying by a process of "naturally occurring" or programmed cell death. We have demonstrated that cells in the developing mammalian globus pallidus die by a process of apoptosis and that by day 7 after birth many of the apoptotic cells possess a neuronal phenotype. Light microscopic and ultrastructural evidence of apoptotic cell death included cell shrinkage, blebbing of the extracellular membrane and condensation of the nuclear chromatin. Additionally we used an in situ nick translation method to assess the integrity of the DNA within the dying cells. This revealed that cells with the morphological characteristics of apoptosis also possessed fragmented DNA typical of cells undergoing Type 1 programmed or apoptotic cell death. The lack of lysosomal enzyme activity within the dying cells and the frequent observations of phagocytosis by neighbouring cells also suggest that the form of programmed cell death is apoptosis and not Type 2 autophagic degeneration. We found no evidence for cells dying by Type 3 non-lysosomal degeneration since all dying cells examined under the electron microscope possessed intact intracellular organelles and cell membranes. We developed a sensitive silver stain which detected balls of condensed chromatin within the apoptotic cells. This enabled identification of apoptotic cells in the developing globus pallidus at low magnification and so allowed us to map the numbers and distribution of dying cells with time. The incidence of apoptotic cells in the neonatal globus pallidus was greatest at birth and then declined such that few cells were detected at one week and none was seen in the adult rat. Although the loss of large numbers of cells in the developing nervous system is a well documented phenomenon, there are only a limited number of reports of the mechanism by which neuronal cells die, and few of these are in the developing mammalian brain. There are at least four different morphological categories of neuronal cell death which are discriminated on morphological and biochemical criteria. Our analysis suggests that apoptotic or Type 1 cell death is the major form of programmed cell death occurring in the mammalian globus pallidus in the first week of life. This report also describes the use of two methods for the ready identification of apoptotic cells at the light microscope level. Because these methods are suitable for use on tissue sections they provide a means to assess the incidence of apoptotic cell death, in parallel with other analyses of the expression of gene products which control cell fate.

Acute administration of haloperidol induces apoptosis of neurones in the striatum and substantia nigra in the rat.

Chronic administration of typical neuroleptics is associated with tardive dyskinesia in some patients. This dyskinetic syndrome has been associated with loss of GABAergic markers in the basal ganglia but the cause of these GABAergic depletions remains uncertain. Haloperidol, a commonly prescribed typical neuroleptic, is known to be toxic in vitro, possibly as a consequence of its conversion to pyridinium-based metabolites and potentially by raising glutamate-mediated transmission. We report here that the in vivo, acute administration of a large dose of haloperidol resulted in a microglial response indicative of neuronal damage. This was accompanied by an increase in the number of apoptotic cells in the striatum (especially in the dorsomedial caudate putamen) and in the substantia nigra pars reticulata. These apoptotic cells were characterised by the stereotaxic injection of a retrograde neuroanatomical tracer into the projection targets of the striatum and substantia nigra pars reticulata prior to the systemic injection of haloperidol. This procedure confirmed that the dying cells were neurones and demonstrated that within the striatum the majority were striatopallidal neurones though relatively high levels of apoptotic striatoentopeduncular neurones were also seen.The possibility that chronic administration of haloperidol could induce cumulative neuronal loss in the substantia nigra pars reticulata and thereby induce the pathological changes which lead to tardive dyskinesia is discussed.

A semi-quantitative atlas of 5-hydroxytryptamine-1 receptors in the primate brain.

The regional distribution of 5-hydroxytryptamine-1 receptors in the primate brain was studied by semi-quantitative autoradiographic analysis of tritiated ligand binding. Areas showing the highest density of 5-hydroxytryptamine-1 receptors (greater than 200 fmol [3H]5-hydroxytryptamine bound per mg tissue), included the cerebral cortex (laminae I-II), claustrum, posterior cell group of the basal nucleus of Meynert, the infracommissural part of the globus pallidus, cortical amygdaloid nucleus, hippocampal formation (CA1-subiculum region, the anterior CA2, CA3 and CA4 regions and the molecular layer of the dentate gyrus), thalamic nuclei (parafascicular, parataenial, paraventricular and superior central lateral nuclei), substantia nigra pars reticulata, dorsal raphe nucleus and choroid plexus. The distribution of 5-hydroxytryptamine-1 receptors is compared to the distribution of both 5-hydroxytryptamine receptors and terminal fields of serotonergic projections as previously described in subprimates.

Dexamethasone induces limited apoptosis and extensive sublethal damage to specific subregions of the striatum and hippocampus: implications for mood disorders.

It has been shown previously that the synthetic corticosteroid dexamethasone induces apoptosis of granule cells in the dentate gyrus and striatopallidal neurons in the dorsomedial caudate-putamen. We investigated whether or not dexamethasone can induce damage to other neuronal populations. This issue was addressed using OX42 immunohistochemistry to visualise activated microglia and thereby gauge the extent of dexamethasone-induced neuronal death. A single dose of dexamethasone (20mg/kg, i.p.) administered to young male Sprague-Dawley rats induced a strong microglial reaction which was restricted to the striatum, the dentate gyrus and all of the CA subfields of the hippocampus. Some OX42-immunoreactive cells were also seen in the lateral septal nucleus. Subsequent quantitative analysis of silver/methenamine-stained sections confirmed that acute administration of dexamethasone induced apoptosis in the striatum and all regions of the hippocampus at doses as low as 0.7mg/kg. In contrast, dexamethasone failed to induce apoptosis in the lateral septal nucleus at doses up to 20mg/kg. The levels of dexamethasone-induced striatal and hippocampal apoptosis were attenuated by pretreatment with the corticosteroid receptor antagonist RU38486 (Mifepristone), which implies that the cell death was mediated by a corticosteroid receptor-dependent process. We further determined whether dexamethasone induced sublethal damage to neurons by quantifying reductions in the number of microtubule-associated protein-2-immunoreactive striatal and hippocampal cells following injection of the corticosteroid. Dexamethasone induced dramatic decreases in the striatum, with the dorsomedial caudate-putamen being particularly affected. Similar damage was seen in the hippocampus, with the dentate gyrus and CA1 and CA3 subfields being particularly vulnerable.Equivalent corticosteroid-induced neuronal damage may occur in mood disorders, where the levels of endogenous corticosteroids are often raised. Corticosteroid-induced damage of striatal and hippocampal neurons may also account for some of the cognitive deficits seen following administration of the drugs to healthy volunteers.

Phencyclidine and corticosteroids induce apoptosis of a subpopulation of striatal neurons: a neural substrate for psychosis?

Phencyclidine, a non-competitive N-methyl-D-aspartate receptor antagonist and indirect dopamine agonist, has neuroprotective properties. Phencyclidine, however, can also exert toxic effects and causes degeneration of neurons in the retrosplenial cortex. In this paper we demonstrate that acute administration of a high dose of phencyclidine to rats, (80 mg/kg), also causes death of a subpopulation of striatal neurons. The dying cells exhibited many of the morphological and biochemical features of cells undergoing apoptosis as revealed by a silver methenamine stain, propidium iodide fluorescence histochemistry and a TUNEL procedure. The majority of the dying cells tended to be clustered within the dorsomedial aspect of the striatum. The type of striatal cell undergoing apoptosis was determined by stereotaxically injecting a colloidal gold retrograde anatomical tracer into the major areas of striatal termination prior to the administration of phencyclidine. This procedure demonstrated that phencyclidine induced striatal apoptosis is almost exclusively limited to striatopallidal neurons. A similar series of experiments was conducted to determine whether the synthetic corticosteroid, dexamethasone, also induces apoptosis of striatal neurons. Corticosteroids are known to be toxic to hippocampal neurons and interact with striatal dopamine transmission. Acute administration of dexamethasone, (20 mg/kg), induced apoptosis of a subpopulation of striatal cells. As was the case with phencyclidine, most of the dexamethasone-induced apoptotic striatal cells were striatopallidal neurons located within the dorsomedial striatum. The pathology during the early stages of Huntington's disease is restricted to an equivalent subpopulation of striatal neurons. Many Huntington's patients are extremely psychotic during this stage in the progression of the disease. Psychosis is also associated with the acute administration of both phencyclidine and dexamethasone to humans. We accordingly speculate that the selective loss of striatopallidal neurons in the dorsomedial striatum may represent the neural substrate of many forms of psychosis.

17 Beta-oestradiol attenuates dexamethasone-induced lethal and sublethal neuronal damage in the striatum and hippocampus.

Abnormal corticosteroid release is extensively associated with mood disorders. This association may result from the toxic actions of endogenous corticosteroids which can induce apoptosis of hippocampal neurons. Similarly, dexamethasone, a synthetic corticosteroid, can induce lethal and sublethal damage to rat hippocampal and striatal neurons and can result in steroid-induced psychoses in humans. The experiments reported here tested the hypothesis that pre-treatment with oestrogen would also attenuate dexamethasone-induced neuronal damage as oestrogens have neuroprotective actions against a variety of insults and falling levels of oestrogen are associated with increased vulnerability to mood disorders. Male Sprague-Dawley rats received three systemic injections which were a combination of vehicle, 17-beta-oestradiol (0.2 mg/kg, s.c.), the oestrogen receptor antagonist tamoxifen (10 mg/kg, s.c.) and dexamethasone (0.7 mg/kg, i.p.) and were killed 24 h after the final injection. Injections of dexamethasone (when preceded by vehicle injections) resulted in elevated levels of apoptosis and sub-lethal damage, as demonstrated by reduced levels of microtubule-associated protein-2-immunopositive neurons, in the striatum and hippocampus. This damage was regional with the dorsomedial caudate putamen and the dentate gyrus and CA1 and CA3 hippocampal sub-fields being particularly affected. Pretreatment with oestrogen substantially attenuated the dexamethasone-induced neuronal damage. This oestrogen-induced neuronal protection was in turn virtually eliminated by giving an initial injection of tamoxifen. These results suggest, therefore, that oestrogens can protect from corticosteroid-induced neuronal damage via an oestrogen receptor-mediated process.

Neural mechanisms underlying parkinsonian symptoms based upon regional uptake of 2-deoxyglucose in monkeys exposed to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine.

The 2-deoxyglucose metabolic mapping technique has been used to investigate the neural mechanisms which underlie the symptoms of Parkinsonism in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine primate model of Parkinson's disease. In six cynomolgus monkeys, 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine was either (a) administered intravenously to induce generalized Parkinsonism, or (b) infused into one carotid artery to induce unilateral Parkinsonism. Post-mortem examination revealed profound cell loss from the substantia nigra, pars compacta either bilaterally or unilaterally in the two groups, respectively. In addition, there was pathological involvement of the ventral tegmental area and locus coeruleus in animals receiving intravenous 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine. 2-Deoxyglucose autoradiography revealed widespread changes in 2-deoxyglucose uptake in the brains of parkinsonian animals when compared to controls. Most of these changes were in basal ganglia and related structures and were qualitatively similar in the two groups of experimental animals. Prominent increases in 2-deoxyglucose uptake were observed in the lateral segment of the globus pallidus (24-27%), the ventral anterior and ventral lateral nuclei of the thalamus (14-22%) and the nucleus tegmenti pedunculopontinus of the caudal midbrain (17-69%). A profound decrease (17-26%) in 2-deoxyglucose uptake was observed in the subthalamic nucleus. We propose these data to indicate that in 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-induced Parkinsonism there is the following pattern of abnormal neuronal activity in basal ganglia circuitry: (i) increased activity in the projection from the putamen to the lateral segment of the globus pallidus; (ii) decreased activity in the projection from the putamen to the medial segment of the globus pallidus; (iii) decreased activity in the projection from the lateral segment of the globus pallidus to the subthalamic nucleus; (iv) increased activity in the projection from the subthalamic nucleus to the globus pallidus; and (v) increased activity in neurons of the medial segment of the globus pallidus projecting to the ventral anterior/ventral lateral thalamus and the pedunculopontine nucleus. These results are compared to the 2-deoxyglucose uptake findings in previous studies from this laboratory in hemiballism and hemichorea in the monkey. The central importance of the subthalamic nucleus in all three conditions is proposed, and supportive evidence for the excitatory nature of subthalamic efferent fibres is adduced.

Glutamate-induced apoptosis results in a loss of striatal neurons in the parkinsonian rat.

The motor symptoms of Parkinson's disease are caused by an increase in activity of striatal neurons which project to the globus pallidus. The discharge activity of these striatal cells is normally regulated by a balance between an inhibitory nigral dopamine input and an excitatory cortical glutamate input. The loss of nigrostriatal dopamine in Parkinson's disease allows the cortical glutamatergic input to dominate (see Fig. 1). Pharmacological or surgical manipulations which redress this imbalance in activity in the striatum, or prevent its propagation throughout the basal ganglia, alleviate the motor symptoms of Parkinsonism. We present evidence to suggest the existence of an endogenous mechanism which compensates for the striatal imbalance during the early stages of Parkinsonism. In the rat rendered parkinsonian by systemic administration of reserpine, selective deletion of striatal neurons was observed. The dying striatal neurons exhibited all of the morphological and biochemical hallmarks of apoptosis. This apoptotic cell death was blocked by either administration of glutamate antagonists or decortication. Our data demonstrate that unchecked endogenous glutamate can induce apoptosis of striatal projection neurons in vivo. This observation may have relevance to the neurophysiological mechanisms which maintain the balance of neural activity within the CNS and to the pathology of neurological diseases.

Cellular localisation of enkephalin gene expression in MPTP-treated cynomolgus monkeys.

Cellular sites of enkephalin gene expression were investigated using the technique of in situ hybridization in the normal striatum and in the denervated striatum of monkeys depleted of dopamine by pretreatment with the neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP). Animals received MPTP by either (a) intravenous injection to induce generalized parkinsonism, or (b) infusion into one carotid artery to induce unilateral parkinsonism. The animals which received systemic injections of MPTP were found to have an essentially total loss of nigral dopamine cells whereas the intracarotid MPTP treatment was found to destroy approximately 95% of the dopamine neurons in the ipsilateral substantia nigra. A double-stranded cDNA probe encoding the human preproenkephalin (PPE) gene was isotopically labelled with 35S and used to detect PPE mRNA within striatal tissue sections. Application of this radiolabelled cDNA probe to lightly fixed striatal sections from both groups of animals revealed an increase in expression of PPE mRNA within denervated striatal enkephalinergic neurons relative to control tissue. An increase in the number of detectable enkephalinergic mRNA-positive neurons relative to control tissue was also noted. These results suggest that the nigral dopaminergic neurons tonically inhibit PPE gene expression in the striatum.

Neurochemical and behavioural investigations of the NMDA receptor-associated glycine site in the rat striatum: functional implications for treatment of parkinsonian symptoms.

The glutamatergic cortico-striatal and subthalamo-entopeduncular pathways are both overactive in parkinsonism. Previous behavioural investigations have shown that intra-entopeduncular injection of either NMDA-site or glycine-site antagonists results in alleviation of parkinsonian symptoms, although injection of the former is associated with the appearance of anaesthetic-like side effects. These behavioural differences may be mediated by action on different NMDA receptor subtypes. Recent neurochemical and molecular pharmacological studies have indicated the existence of NMDA receptor subtypes which display differential modulation by glycine. In the present study, three potential modes of NMDA antagonism were differentiated in vitro by effects on [3H]-glycine binding to striatal sections. Specific [3H]-glycine binding was totally displaced by the glycine partial agonist (R)-HA-966; the NMDA-site antagonist D-CPP had no effect; and the NMDA-site antagonist D-AP5 displaced [3H]-glycine binding in a subpopulation of glycine sites. The anti-parkinsonian effects of (R)-HA-966, D-CPP and D-AP5 were assessed by intra-striatal injection in reserpine-treated rats and 6-OHDA-lesioned rats. Injection of (R)-HA-966 and D-CPP resulted in alleviation of parkinsonian akinesia, although the latter elicited anaesthetic-like side effects; D-AP5 was ineffective as an anti-parkinsonian agent. (R)-HA-966 was also effective as an anti-parkinsonian agent when administered systemically in the reserpine-treated rat. These data suggest that different classes of NMDA antagonist mediate different behavioural responses within the parkinsonian striatum. The behavioural response produced may depend on the exact nature of the conformational change induced by the antagonist and the location of the subtype most sensitive to that class of compound. Selection of a specific mode of NMDA receptor antagonism or targeting of striatal NMDA receptor subtypes may form the basis of a novel therapeutic approach to Parkinson's disease.

Cell suicide in the developing nervous system: a functional neural network model.

A computational model of programmed cell death (PCD) in the nervous system is described. A neurobiologically realisable method for identifying and removing the least useful cells from a network is developed, and it is shown by simulation that an artificial neural network can solve difficult problems efficiently if it is given more neurons initially than it needs subsequently. The least useful neurons die off gradually after learning is complete, and the learned solution can then be maintained with a smaller number of units than were needed for initial learning. The research suggests a functional role for PCD, and how self-limiting PCD could be achieved in real neural systems.

In defence of optical density ratios in 2-deoxyglucose autoradiography.

The use of optical density ratios to describe changes in [14C]2-deoxyglucose uptake in neuroanatomical mapping experiments has recently been criticized. It has been argued that a fixed ratio of tissue isotope concentration does not yield a constant optical density ratio but is dependent on the exposure time and the absolute amounts of isotope used. Here it is demonstrated that such variations in optical density ratios are due to an artifact in calculating the optical density ratio, which can easily be corrected provided that the film is not approaching saturation and not due to the non-linearity of an exposure-density curve as has previously been suggested.

Plasticity of behavioural response to repeated injection of glutamate in cuneiform area of rat.

Whereas a single microinjection of L-glutamate (10 nmol) into the cuneiform area of rats gives freezing, a second or third injection (delivered at 4-min intervals to the same site)can produce fast running. To examine whether this plasticity of response was caused by a simple increase in the amount of glutamate present, 30 nmol of glutamate were given in a single injection. In 93% of sites in the cuneiform area this procedure gave only freezing, although subsequent testing with repeated injections produced fast running in 53% of these sites. Thus, response potentiation to glutamate appears to require repeated stimulation, and may therefore be related to processes underlying the natural conditioning of defensive responses.

Further evidence for segregated output channels from superior colliculus in rat: ipsilateral tecto-pontine and tecto-cuneiform projections have different cells of origin.

Two of the targets of the ipsilateral descending pathway from the superior colliculus are the cuneiform area (immediately ventral to the inferior colliculus), and the dorsolateral basilar pons. The cells of origin of the projections to these targets in rat were studied with a retrograde double-labelling technique, using the fluorescent tracers True blue and Diamidino yellow. Although many tectal cells were single-labelled by injections into basilar pons or the cuneiform area, less than 5% were double-labelled. The two projections thus appear to arise mainly from separate populations of cells within the superior colliculus.