Faculty Bibliography

A technique has been developed to record from 16 different brain sites of the freely moving rat using subminiature MOSFET preamplifiers. The high input impedance, small size, durability and light weight of the amplifiers and connecting cable allows high quality multisite recording of field potentials and unit activity. In addition, a movable headstage for positioning multiple microelectrodes is described. The compact recording system permits one to construct neocortical EEG maps, instant depth profiles of evoked and spontaneous field data, and to study neuronal synchrony of distant cell populations

All subcortical afferents to the dorsal hippocampus, running in the fimbria-fornix and supracallosal path, were removed by aspiration. Three to 5 months later the rats were implanted with chronic recording electrodes in the dentate gyrus and CA1 region, and stimulating electrodes in the angular bundle. In non-lesioned rats, high-frequency trains delivered to the angular bundle gave rise to a sustained increase of the evoked population spike in the dentate gyrus. In lesioned animals, high-frequency stimulation resulted in only short-lasting changes, and by 15 min after the conditioning trains the amplitude of both the population spike and field postsynaptic potentials returned to baseline. In lesioned rats large amplitude interictal spikes (less than 40 ms, 3-8 mV) occurred spontaneously. These findings suggest that either (1) coactivation of entorhinal and subcortical inputs is essential for the induction of long-lasting plastic changes in the dentate gyrus, or (2) the long-term potentiation mechanism is saturated by the chronically occurring interictal discharges in the subcortically denervated dentate gyrus

Rats with unilateral lesions of either the supracallosal regions (including the dorsal cingulate cortex) and the fimbria-fornix either on the same (S) or the opposite (O) sides of the brain were studied in a 16-hole open field without pharmacologic intervention and, subsequently, after 0.1 and after 1.0 mg/kg scopolamine HBr. Their performances were compared with those of unoperated control animals subjected to the same testing regime. Certain of their behaviors were compared with those of a larger number of animals with bilateral hippocampal destruction (and their control groups) from prior studies. Unilateral lesions of fimbria-fornix and supracallosal afferents to the hippocampal formation produced a decrease in hole poking activity relative to control animals. A further decrease in hole-poking behavior, coupled with increased locomotion, was observed in rats with fimbria-fornix and cingulate cortex lesions on opposite sides of the brain (group O). The smaller dose of scopolamine accentuated these effects. Indeed, the behavior of group O after scopolamine treatment was similar to animals with large bilateral hippocampal lesions. The large dose of scopolamine induced stereotyped rearing or hole poking in the brain-damaged animals but not in the control group. These findings suggest that both the fimbria-fornix and the supracallosal pathway is necessary for normal hippocampal function and that the behavioral deficit is greater when these structures are damaged on the opposite sides of the brain

Adult rats with unilateral fimbria-fornix lesion received fetal hippocampal grafts into the lesion cavity. Seven to ten months after the transplantation the graft was examined for long-term potentiation (LTP) in response to host hippocampus and direct graft stimulation. High frequency tetanizing trains delivered to either the host hippocampus or the graft resulted in augmented field potentials and prolonged neuronal discharges in the graft lasting several hours. Very low currents (10-30 microA) were required to induce LTP by direct graft stimulation. In addition to the enhancement of evoked responses, the frequency of occurrence of spontaneously occurring EEG spikes and concurrent population neuronal bursts in the graft increased significantly after the tetanizing trains. These findings suggest that the physiological-biochemical mechanisms required for plastic changes of synaptic efficacy are present in the grafted hippocampus

Consequences of transient (15-20 min) ischemia on the neuronal activity of the dentate gyrus and hippocampal CA 1 region were investigated in chronically implanted Sprague-Dawley rats. Forebrain ischemia was produced by occlusion of the carotids for 15 or 20 min, following cauterization of the vertebral arteries. Following the release of the carotids, both spontaneous and evoked activity showed a steady but partial recovery, reaching a maximum 12 to 24 h after the ischemic insult. From this plateau, both the power of rhythmic slow activity recorded during walking and the power of slow delta activity obtained during alert immobility decreased monotonically, with large changes occurring between postischemic days 2 and 4. The changes in spontaneous activity were accompanied by a decrease and eventual disappearance of the Schaffer collateral evoked responses in CA 1. Perforant path volleys were less efficient in activating the granule cells following ischemia compared to baseline levels. This decreased responsiveness was paralleled by a relative impairment of paired pulse depression. Neurophysiological signs of spontaneous or evoked neuronal hyperexcitability were not observed at any time point during the 8 postischemic days. Neuronal damage in the CA 1 region varied from moderate to complete loss of pyramidal cells. In addition, degenerating neurons were also observed in the hilus of the dentate gyrus. These findings do not support the 'overwork' version of the excitoxic hypothesis of delayed neuronal damage and indicate that the cause of ischemic cell death should be sought in factors other than neuronal hyperactivity

Recent progress has been made in defining the requirements for survival, growth and function of damaged cholinergic neurons of the central nervous system. In particular, the responsiveness of cholinergic neurons to nerve growth factor (NGF) in the regulation of development, cell survival, axon elongation, and response to injury has led to the formulation of the Neurotrophic Hypothesis, a unifying hypothesis of neuronal responsiveness to growth-promoting substances. NGF-mediated effects on cholinergic neurons in culture as well as in the septum, basal nucleus, striatum, and hippocampus, and the ability of NGF to prevent lesion-induced cell death and to ameliorate the effects of aging, provide the foundation for this work. A potential role for glia and microglia in mediating the effects of NGF is proposed

Single unit studies indicate that increased activity in the cholinergic nucleus basalis (NB) correlates with behavioral activation and neocortical desynchronization. Lesions of the NB result in neocortical slow delta waves, similar to the action of antimuscarinic drugs, and the lesion releases the oscillation of GABAergic neurons in the reticular nucleus of the thalamus, resulting in high voltage neocortical spindles. Extensive damage of the thalamus does not produce slowing of neocortical activity but it abolishes neocortical spindles. We suggest that the NB plays a key role in neocortical activation by a) blocking the afterhyperpolarizations and accommodation in neocortical pyramidal neurons and b) suppressing the rhythm generation in the reticular nucleus-thalamocortical circuitry. We further suggest that the NB system may serve as a structural basis for the concept of the generalized activation described by Moruzzi and Magoun (1949)

A miniature multiple thin-film recording sensor was used to measure simultaneously the electrical activity, oxygen content and temperature of brain tissue. The chamber-type potential sensor was an Ag/AgCl electrode covered by an Si3N4 (silicon nitride) chamber. The chamber-type oxygen sensor consisted of an Au-Ag/AgCl two-electrode electrochemical cell embedded in an electrolyte-filled Si3N4 chamber. The temperature sensor was a thin-film germanium resistor. The different sensors were spaced 300 microns apart. Anaesthetics (pentobarbital, chloral hydrate, chlornembutal, halothane) were shown to depress electrical activity and to increase local oxygen tension in the hippocampus. Halothane, but not the other anaesthetics, also increased the current output of the oxygen sensor when tested in saline bath, indicating that the apparent increase in measured oxygen levels during halothane anaesthesia was partly due to an electrochemical effect of halothane on the oxygen sensors. The decrease of tissue oxygen consumption produced by the other anaesthetics is likely to be the result of metabolic depression. Cerebral ischemia, evoked by cauterization of the vertebral arteries and occlusion of the carotid arteries for 30 min, resulted in the disappearance of both spontaneous and evoked electrical activity in the hippocampus and a decrease of both local temperature and oxygen tension. There was a marked overshoot of the oxygen tension to above preocclusion level following the release of the carotid arteries. As soon as electrical activity returned, the oxygen tension fell again, often below the lowest level seen during the ischemic period. This secondary decrease of oxygen level could be reversed by administration of supplementary small doses of anaesthetic. The anaesthetic-induced increase in oxygen tension was accompanied by a marked decrease in electroencephalogram amplitude and frequency. During electrically induced seizures a decrease in hippocampal oxygen content occurred and was accompanied by an increase of local temperature. Since the rectal temperature was kept constant, the changes in temperature are likely to reflect changes in blood perfusion of the recorded area. These findings are in agreement with previous observations made with conventional electrodes. In addition, the miniature size of the chamber-type microelectrode assembly allows a correlated monitoring of parallel physiological changes with high spatial and temporal resolution during anaesthesia, ischemia and epilepsy

Spontaneous and evoked field potentials and cellular discharges were studied in the subcortically denervated hippocampus of the freely moving rat. The fimbria fornix, the ventral hippocampal commissure, and the supracallosal afferent fibers were removed by aspiration, and recordings were made 3-5 months after the lesion. Two types of spontaneous interictal spikes were observed. Type 1 interictal spike had identical depth distribution to physiological sharp waves but they were shorter in duration (less than 40 ms), larger in amplitude (greater than 2.5 mV) and population spikes were riding on the main deflection. Type 2 interictal spikes were negative in the stratum oriens and positive in the pyramidal layer and stratum radiatum of both CA1 and CA3. The amplitude of both types of interictal spikes could exceed 6 mV. We suggest that interictal spikes were initiated randomly in different subpopulations of the CA2-3 region and the location of the initiating population burst determined the polarity and amplitude of the extracellular interictal spike. Repetitive stimulation of the perforant path (5 Hz, 6 s) evoked markedly uniform afterdischarges in both intact and fimbria fornix-deprived rats. The threshold of afterdischarges was significantly lower, the seizure spread to the contralateral hippocampus was slower, and secondary afterdischarges lasted significantly longer in the lesioned rats. We suggest that under physiological conditions the electrical stability of the hippocampus is ensured by the feed-forward inhibitory action of subcortical afferents. Removal of tonic inhibitory influences and/or sprouting of local axon collaterals allows extreme synchronization and reverberation of information in the entorhinal-hippocampal-entorhinal cortex circuitry. The presence of interictal spikes and increased susceptibility to seizures for several months after the lesion offers the fimbria-fornix-deprived hippocampus a useful chronic preparation to study the mechanisms of limbic epilepsy

Review of the normally occurring neuronal patterns of the hippocampus suggests that the two principal cell types of the hippocampus, the pyramidal neurons and granule cells, are maximally active during different behaviors. Granule cells reach their highest discharge rates during theta-concurrent exploratory activities, while population synchrony of pyramidal cells is maximum during immobility, consummatory behaviors, and slow wave sleep associated with field sharp waves. Sharp waves reflect the summed postsynaptic depolarization of large numbers of pyramidal cells in the CA1 and subiculum as a consequence of synchronous discharge of bursting CA3 pyramidal neurons. The trigger for the population burst in the CA3 region is the temporary release from subcortical tonic inhibition. An overview of the experimentally explored criteria of synaptic enhancement (intensity, frequency, and pattern of postsynaptic depolarization, calcium influx, cooperativity, threshold) suggests that these requirements may be present during sharp wave-concurrent population bursts of pyramidal cells. Experimental evidence is cited showing that (a) population bursts in CA3 may lead to long-term potentiation in their postsynaptic CA1 targets, (b) tetanizing stimuli are capable of increasing the synchrony of the sharp wave-burst, and (c) activity patterns of the neocortical input to the hippocampus determine which subgroup of CA3 neurons will trigger subsequently occurring population bursts (initiator cells). Based on the experimental evidence reviewed a formal model of memory trace formation is outlined. During exploratory (theta) behaviors the neocortical information is transmitted to the hippocampus via the fast-firing granule cells which may induce a weak and transient heterosynaptic potentiation in a subgroup of CA3 pyramidal cells. The weakly potentiated CA3 neurons will then initiate population bursts upon the termination of exploratory activity (sharp wave state). It is assumed that recurrent excitation during the population burst is strongest on those cells which initiated the population event. It is suggested that the strong excitatory drive brought about by the sharp wave-concurrent population bursts during consummatory behaviors, immobility, and slow wave sleep may be sufficient for the induction of long-term synaptic modification in the initiator neurons of the CA3 region and in their targets in CA1. In this two-stage model both exploratory (theta) and sharp wave states of the hippocampus are essential and any interference that might modify the structure of the population bursts (e.g. epileptic spikes) is detrimental to memory trace formation