Functional deficits after sustained stimulation of the perforant path

Several reports have implicated the overactivity of hippocampal glutaminergic systems in neurodegenerative conditions including Senile dementia of the Alzheimer's type (SDAT). The neurobiological effects of hippocampal glutaminergic hyperactivity were studied by perforant pathway stimulation. Forty-five minutes of sustained perforant pathway stimulation produced a 50% or greater increase in motor activity 1, 2, and 3 weeks after stimulation. Robust retention deficits in a 48-h step-through passive avoidance task were evident 2 weeks post-stimulation. Furthermore, animals receiving stimulation were impaired in the acquisition of a spatial task in the Morris water maze. Stimulated animals exhibited little reduction in their escape latencies over the testing period. The learning and memory deficits were associated with a loss of CA1 and CA3 pyramidal cells and pretreatment with the N-methyl-D-aspartate antagonist MK-801 reduced this cell loss, particularly in the CA1 region of the hippocampus. These results suggest that sustained stimulation of the perforant pathway may be useful in studying neurological deficits associated with glutaminergic hyperfunction.     

Comparative characteristics of the direct influences of the perforant pathway on neurons in hippocampal fields CA1 and CA3 in vitro]

The effect of the perforant path stimulation on the CA1 and CA3 neurons was investigated in incubated slices of the guinea pig hippocampus. Spike generation was observed in both fields during stimulation of the perforant path. The majority of the CA1 neurons followed rhythmic stimulation up to 30-80/c. The CA3 neurons responded only to low-frequency stimulation (up to 5/c). The posttetanic potentiation of responses to the perforant path stimulation was observed in both hippocampal fields.     

Activation of perforant path neurons to field CA1 by hippocampal projections

Previous evidence showed that single-shock stimulation of dorsal hippocampal commissure (PSD) fibers to the entorhinal cortex led to sequential activation of perforant path neurons to the dentate gyrus, dentate granule cells, pyramidal neurons of hippocampal fields CA3 and CA1, and, through reentrant hippocampal impulses, neurons of deep and superficial layers of the entorhinal cortex. The aim of the present study was to ascertain whether perforant path neurons to CA1 are activated by the PSD input and/or by the reentrant hippocampal impulses in this model. Field potentials evoked by single-shock (0.1-Hz) or repetitive (1-4 Hz) PSD stimulation were recorded in anesthetized guinea pigs from the entorhinal cortex, dentate gyrus, fields CA1 and CA3, and subiculum. A current source-density analysis of the evoked potentials was used to localize the input to field CA1 and dentate gyrus. After either single-shock or repetitive PSD stimulation, an early current sink was found in the molecular layer of the dentate gyrus, but no sink was present in CA1. With low-frequency PSD stimulation, a late (approximately 40-ms) surface positive wave occurred in field CA1 alone. During this wave, a current sink was found in the stratum lacunosum-moleculare of CA1, but no sink was present in the dentate gyrus. The late wave had threshold and magnitude related to the building up of the response evoked by reentrant hippocampal impulses in layer III of the entorhinal cortex and was abolished by selective interruption of the perforant path to CA1. The results show that the commissural input to the entorhinal cortex activates perforant path neurons to the dentate gyrus, but not those to field CA1 which are recruited by repetitive hippocampal impulses. These findings show different frequency-dependent patterns of loop operation that might be related to different behaviors.     

Limbic seizures, but not kindling, reversibly impair place learning in the Morris water maze

We investigated the effects of kindling and kindled seizures in different limbic structures on place and cue learning in the Morris water maze. The triggering of seizures by stimulation of the perforant path, septum, or amygdala prior to daily training impaired place learning, but had little effect on visible platform training or swim speed. Seizures triggered by stimulation of the medial perforant path after daily training also impaired place learning. Conversely, place learning proceeded normally in rats tested 24 h after kindling triggered by stimulation of the perforant path, septum, or amygdala, indicating that kindling per se does not affect place learning. Each group was able to learn the location of a reversed platform when pretraining seizures were discontinued; and perforant path and septal kindled rats, but not amygdaloid kindled rats, were impaired at learning the location of a reversed platform when seizures were triggered before training. The results confirm previous reports that limbic seizures produce amnesia, but they contradict the finding that hippocampal kindling impairs learning on tasks sensitive to hippocampal lesions.     

Sensory modulation of hippocampal transmission. I. Opposite effects on CA1 and dentate gyrus synapsis

Neuronal transmission through hippocampal subfields exhibits a high degree of modulation and appears dependent on the behavioral state and hippocampal EEG. Sensory inputs, which profoundly modify the hippocampal EEG, may be involved in modulating hippocampal excitability. Field responses of the CA1 region, evoked by ipsilateral CA3 or perforant path stimulation, as well as dentate gyrus potentials evoked by perforant path stimulation were recorded in paralyzed and locally anesthetized rats and studied before, during and after sensory stimulation, consisting of gentle stroking of the animal's fur. On some occasions the CA1 was also antidromically driven from the posterior alveus in order to study the recurrent inhibitory loop and paired pulses were applied to the perforant pathway to study recurrent inhibition in the dentate gyrus. Evoked responses were averaged and field excitatory postsynaptic potential (EPSP) slope and population spike (PS) amplitude measured. In addition the positive wave which follows the population spike, which corresponds in part to the recurrent IPSP, was also evaluated. Sensory stimulation, which evoked a high-amplitude 5-6 Hz theta (theta)-rhythm in the hippocampal EEG, drastically depressed the efficacy of Schaffer collateral volleys in discharging the CA1 cells. The EPSP-PS curves, however, were not altered revealing that cellular excitability was unaffected. The inhibitory CA1 loop appeared to be unaltered. In contrast, the dentate gyrus responses to perforant pathway stimulation were enhanced during periods of sensory stimulation and the cellular excitability increased, as judged by the shift to the left of EPSP-PS relation. In addition, the recurrent inhibition appeared to be reduced during sensory stimulation. Present results demonstrate that sensory stimulation causes modulation of information transfer through the hippocampus. This modification of hippocampal transmission may serve to properly gate the information reaching the CA1 and dentate gyrus.     

Incorporation of [H]fucose in rat hippocampal structures after conditioning by perforanth path stimulation and after LTP-producing tetanization

The contribution of glycoprotein synthesis to functional synaptic changes and to the formation of memory traces was investigated by autoradiographic determination of the incorporation of [³H]fucose into the hippocampal structures of rats. In the first experiment, the fucose incorporation was measured after induction of post-tetanic long-term potentiation (LTP) in granular cell synapses by repeated tetanization (200 cps) of the perforant path, and after stimulation of this hippocampal input by the same number of impulses with very low frequency (0.2 cps) not producing LTP. In the second experiment, the incorporation of fucose was determined after an active avoidance training using the stimulation of the perforant path by impulse trains of 15 cps as conditioning stimuli, and after a session of corresponding unpaired stimulations of the perforant path. Unstimulated animals were used in both experiments to measure the basal glycosylation. LTP-producing tetanization resulted only in a slight increase of incorporation into the ipsilateral hippocampal structures without significant differences to similar changes after the corresponding control stimulation with single impulses. After a session of unpaired stimulation of the perforant path with impulse trains of 15 cps only slight and inconsistent changes of incorporation occurred in the hippocampus too. However, after conditioning by the corresponding perforant path stimulation as conditioned stimulus, considerable increases of incorporation were observed in all structures of the ipsilateral hippocampus, when compared to the unpaired control stimulation. An enhanced labeling occurred also in some structures of the contralateral hippocampus mainly receiving commissural inputs. The results suggest again, that the activation of one single hippocampal afferent, even if producing LTP, would not be sufficient to induce an increased glycosylation of neuronal proteins. The increase of glycoprotein formation seems to require the convergence of several inputs, which can be assumed to occur during learning. Therefore, LTP of a single synaptic population seems not to represent the complete long-lasting memory trace, but only one of its components, or a preceding transient storage mechanism.     

Long-term potentiation recruits a trisynaptic excitatory associative network within the mouse dentate gyrus

Granule cells of the hippocampal dentate gyrus receive two powerful excitatory inputs: the perforant path, originating from the entorhinal cortex, and the associational pathway, originating from mossy cells, the principal neurons of the dentate gyrus hilus. We examined the electrophysiological properties of the less well-studied associational pathway and its interaction with the perforant path in the intact mouse hippocampus and then tested homosynaptic, trans-synaptic and associative long-term potentiation of these pathways. The associational pathway was either monosynaptically activated by stimulation within the inner molecular layer or trisynaptically activated after stimulation of the perforant path. Laminar profiles of extracellularly recorded associational pathway field potentials demonstrated a bell-shaped curve with a peak in the inner molecular layer. Tetanization of the perforant path induced not only homosynaptic potentiation of the perforant path (162.4 +/- 6.7% at 0.5-1.5 h after tetanus) but also heterosynaptic potentiation of the associational pathway (115.7 +/- 4.9%). Direct tetanization of the associational pathway within the inner molecular layer was ineffective in either the septo-temporal (97.2 +/- 4.5%) or temporal-septal (104.4 +/- 4.6%) direction. In contrast, conjoint tetanization of the associational pathway with the perforant path potentiated the associational pathway responses in both the septo-temporal (123.4 +/- 5.8%) and the temporal-septal (124.8 +/- 7.3%) directions. Paired-pulse facilitation was attenuated by long-term potentiation in the perforant path and the associational pathway, suggesting pre-synaptic involvement. These results demonstrate that long-term potentiation of the associational pathway and the perforant path is a product of the network properties of the dentate gyrus rather than of each monosynaptic input alone. The architecture of this neural network may be designed for flexible dynamic associations of the afferent perforant path inputs to configure encoded information within hippocampal neuronal ensembles.     

“Epileptic” brain damage in rats induced by sustained electrical stimulation of the perforant path. I. Acute electrophysiological and light microscopic studies

Sustained electrical stimulation of the perforant path in urethane-anesthetized rats evoked hippocampal granule cell population spikes and epileptiform discharges. After stimulation, recurrent inhibition in the granule cell layer was abolished. Light microscopic analysis revealed a highly reproducible pattern of hippocampal damage to dentate pyramidal basket cells, hilar cells in general and CA3 and CA1 pyramidal cells. CA2 pyramidal cells and dentate granule cells were relatively unaffected. When perforant path stimulation on one side of the brain evoked bilateral granule cell discharges, damage was bilateral. Unilateral hippocampal seizures were associated with unilateral hippocampal damage. Rapid Golgi-stained hippocampi exhibited spherical dendritic swellings at the sites of termination of excitatory entorhinal afferents to the hippocampus and in the mossy fiber region. Electrical stimulation of a single excitatory afferent to the hippocampus appears to reproduce the “epileptic” pattern of hippocampal damage without using convulsant drugs and without causing motor convulsions. It is suggested that seizure-associated brain damage in caused by excessive pre-synaptic release of excitatory transmitter that induces intracellular post-synaptic changes that lead to dendritic swelling and cell death.     

The perforant path projection from the medial entorhinal cortex layer III to the subiculum in the rat combined hippocampal–entorhinal cortex slice

Intracellular recordings were performed to examine the perforant path projection from layer III of the entorhinal cortex to the subiculum in rat combined hippocampal–entorhinal cortex slices. Electrical stimulation in the medial entorhinal cortex layer III caused short latency combined excitatory and inhibitory synaptic responses in subicular cells. In the presence of the GABA_A antagonist bicuculline and the GABA_B antagonist CGP-55845 A inhibition was blocked and isolated AMPA- or NMDA receptor-mediated EPSPs could be elicited. After application of the non-NMDA antagonist NBQX and the NMDA antagonist APV excitatory responses were completely blocked indicating a glutamatergic input from the neurons of the medial entorhinal cortex layer III. By stimulation from a close (< 0.2 mm) position in the presence of NBQX and APV and either CGP-55845 A or bicuculline we could record monosynaptic fast GABA_A or slow GABA_B receptor-mediated IPSPs, respectively. We compared synaptic responses in subicular cells induced by stimulation in the medial entorhinal cortex layer III with responses elicited by stimulation of afferent fibres in the alveus. The EPSPs of subicular cells induced by stimulation of alvear fibres could be significantly augmented by simultaneous activation of perforant path fibres originating in the medial entorhinal cortex layer III, while delayed activation of alvear fibres after stimulation of the perforant path resulted in a weak inhibition of the alveus evoked EPSPs. Thus, the perforant path projection activates monosynaptic excitation of subicular neurons. Therefore the entorhinal cortex does not only function as an important input structure of the hippocampal formation but is also able to modulate the hippocampal output via the entorhinal–subicular circuit.     

Stimulation-induced reset of hippocampal theta in the freely performing rat

Previous research has suggested that visual and auditory stimuli in a working memory task have the ability to reset hippocampal theta, perhaps allowing an organism to encode the incoming information optimally. The present study examined two possible neural pathways involved in theta resetting. Rats were trained on a visual discrimination task in an operant chamber. At the beginning of a trial, a light appeared over a centrally located lever that the rat was required to press to receive a water reward. There was a 30-s intertrial interval before the next light stimulus appeared. After learning the task, all rats received surgical implantation of stimulating electrodes in both the fornix and the perforant path and recording electrodes, bilaterally in the hippocampus. After surgery, theta was recorded before and after the light stimulus to determine whether resetting to the visual stimulus occurred. During the intertrial interval, rats received single-pulse electrical stimulation of either the fornix or perforant path. Theta was recorded both before and after the electrical stimulation to determine whether resetting occurred. In this experiment, hippocampal theta was reset after all three stimulus conditions (light, perforant path, and fornix stimulation), with the greatest degree of reset occurring after the fornix stimulation. The results suggest that activation of the perforant path and fornix may underlie theta reset and provide a mechanism by which the hippocampus may enhance cognitive processing.     

Perforant pathway-evoked long-term potentiation of CA1 neurons in the hippocampal slice preparation

Previously, we have presented electrophysiological evidence reaffirming the existence of a controversial hippocampal pathway. These fibers are part of the perforant pathway and terminate directly on the CA1 cells. We now report that, in the hippocampal slice preparation, tetanic stimulation of the perforant pathway produces long-term potentiation (LTP) of CA1 cell responses. LTP of population spikes varied from 150% to 500%. The results were of interest because these axons synapse at distal sites on the apical dendrite. This location is usually thought to be a difficult site to evoke action potentials.     

Uptake of D-[3H]aspartic acid by hippocampal slices: the influence of low and high frequency activation of nerve endings

Schaffer collaterals (Sch. coll.) projection and perforant path (PP) fibers in the hippocampal slices were stimulated in the presence of 0.8 microM D-[3H]aspartic acid--a marker of glutamergic nerve terminals. The activation of these fibers evoked a frequency and sodium-dependent increase of D-[3H]aspartate uptake. The increase of D-aspartate uptake during Schaffer collaterals stimulation was much higher than during perforant path stimulation. The stimulation of Sch. coll. nerve endings, which was preceded by a high frequency stimulation of these fibers (potentiation), evoked a decrease of D-aspartate uptake as compared to stimulated controls.     

Heterosynaptic interactions in fields CA1 and CA3 of the hippocampus in vitro]

Heterosynaptic interactions between the perforant path and Schaffer's collaterals in the field CA1 and between the perforant path and dentate mossy fibres in the field CA3 were investigated in guinea pig hippocampal slices. Using the method of paired stimuli, space-time summation (for 20--50 ms) was observed in both systems with stimuli sub-threshold for spike generation. Spike responses of the neurons to testing stimulation of afferents synapsing upon the terminal parts of apical dendrites (perforant path) were depressed after spike discharge to conditioning stimulation of proximal afferents (for about 20 ms in CA1, and for about 300 ms in CA3). With inverse combination of the stimuli the period of suppression was much shorter (3--8 ms). Tetanization of the mossy fibres was followed by prolonged (2--30 min) depression of the CA3 responses to the perforant path stimulation. No other reliable long-lasting posttetanic heterosynaptic effects were observed.     

Late post-learning effect of entorhinal cortex electrical stimulation persists despite destruction of the perforant path

In an appetitive learning task in mice, stimulation of the lateral entorhinal cortex (LEC) 30 min after training produced an improvement in retention 24 h later, as well as faster extinction of conditioning. This effect persisted in animals with bilateral lesions of the perforant path. In addition, the threshold for hippocampal after-discharges produced by LEC stimulation was raised significantly in perforant-path lesioned animals. The results indicate a functional dissociation between hippocampal and cortical mechanisms involved in memory consolidation.     

Abnormal responses to perforant path stimulation in the dentate gyrus of slices from rats with kainate-induced epilepsy and mossy fiber reorganization.

Previous electrophysiological studies have demonstrated that in a subset of hippocampal slices from tissue resected from patients with mesial temporal lobe epilepsy, perforant path stimulation can elicit prolonged negative field-potential shifts in the dentate granule cell layer (Masukawa et al., 1989. Brain Res. 493, 168-174; Isokawa and Fried, 1996. Neuroscience 72, 31-37). In this investigation, hippocampal slices were prepared from rats: (1) 2-4 days following kainate treatment, when little or no reorganization of the mossy fibers would be present and (2) 3-13 months after kainate treatment, when mossy fiber reorganization would have occurred. In saline-treated controls, perforant path stimulation typically evoked a single population spike. In contrast, perforant path stimulation could evoke 3-12 population spikes in nearly all slices from kainate-injected rats 2-4 days and 3-13 months after treatment. The majority of slices from kainate-injected rats 3-13 months after treatment had qualitatively similar responses to perforant path stimulation as that observed in slices from kainate-injected rats 2-4 days after treatment. However, in 17% of the slices from kainate-treated rats 3-13 months after treatment (29% of rats), the multiple population spikes were followed by a prolonged negative field-potential shift (duration: 140 ms-1.5 s) with variable superimposed population spike activity. This type of epileptiform activity was only observed in slices with robust Timm's staining in the inner molecular layer and similar responses could also be evoked in these slices with hilar stimulation. Furthermore, pharmacological depression of inhibition by adding the GABA(A) receptor antagonist bicuculline unmasked hilar-evoked prolonged negative field-potential shifts in most slices from kainate-treated rats 3-13 months following treatment, and these slices had robust Timm's staining in the inner molecular layer. Such events were not observed in slices from saline-treated controls or kainate-injected rats 2-4 days after treatment. In conclusion, the prolonged negative field-potential shifts evoked to perforant path stimulation in normal ACSF were associated with mossy fiber reorganization, but the relative contribution of altered inhibition, increased synaptic excitation, or even non-synaptic mechanisms is unknown.     

The Basomedial and Basolateral Amygdaloid Nuclei Contribute to the Induction of Long-term Potentiation in the Dentate Gyrus In Vivo

Recent behavioural studies have provided evidence that the amygdala modulates hippocampal-dependent memory. To test the possibility that the amygdala modulates hippocampal synaptic plasticity, we investigated the effects of surgical lesions of the amygdaloid nuclei on the induction of long-term potentiation (LTP) in the dentate gyrus of anaesthetized rats. Previously we reported that LTP in the dentate gyrus was attenuated by lesion of the basolateral amygdala, but was not affected by lesion of the central amygdala. In the present study, dentate gyrus LTP was significantly attenuated by basomedial amygdala lesion but not by medial amygdala lesion. These results suggest that, among the amygdaloid nuclei, the basomedial and basolateral nuclei are involved in the modulation of hippocampal plasticity. The roles of the basomedial and basolateral amygdala were further supported by experiments examining the effects of electrical stimulation of these nuclei. High-frequency stimulation of the basomedial amygdala alone did not induce dentate gyrus LTP, but when applied at the same time as tetanic stimulation of the perforant path increased the magnitude of the dentate gyrus LTP. Similarly, high-frequency stimulation of the basolateral amygdala enhanced LTP induced by tetanic stimulation of the perforant path. Furthermore, facilitation of dentate gyrus LTP by basomedial or basolateral amygdala stimulation was observed even in rats lesioned in either amygdala, suggesting that neurons in the basomedial and basolateral amygdala can modulate dentate gyrus LTP independently. Activity-dependent facilitation of hippocampal plasticity by the basomedial and basolateral amygdala may underlie memory processing associated with emotion.     

Permanently altered hippocampal structure, excitability, and inhibition after experimental status epilepticus in the rat: the "dormant basket cell" hypothesis and its possible relevance to temporal lobe epilepsy.

The relationship between an episode of status epilepticus, the resulting hippocampal pathology, and the subsequent development of pathophysiological changes possibly relevant to human epilepsy was explored using the experimental epilepsy model of perforant path stimulation in the rat. Granule cell hyperexcitability and decreased feedforward and feedback inhibition were evident immediately after 24 hours of intermittent perforant path stimulation and persisted relatively unchanged for more than 1 year. All of the pathophysiological changes induced by perforant path stimulation were replicated in normal animals by a subconvulsive dose of bicuculline, suggesting that the permanent "epileptiform" abnormalities produced by sustained perforant path stimulation may be due to decreased GABA-mediated inhibition. Granule cell pathophysiology was seen only in animals that exhibited a loss of adjacent dentate hilar mossy cells and hilar somatostatin/neuropeptide Y-immunoreactive neurons. GABA-immunoreactive dentate basket cells survived despite the extensive loss of adjacent hilar neurons. However, parvalbumin immunoreactivity, present normally in a subpopulation of GABA-immunoreactive dentate basket cells, was absent on the stimulated side. Whether this represents decreased parvalbumin synthesis in surviving basket cells or a loss of a specific subset of inhibitory cells is unclear. Hyperexcitability and decreased paired-pulse inhibition in response to ipsilateral perforant path stimulation were also present in the CA1 pyramidal cell layer on the previously stimulated side, despite minimal damage to CA1 pyramidal cells or interneurons. The possibility that CA1 inhibitory neurons were hypofunctional or "dormant" due to a loss of excitatory input to inhibitory cells from damaged CA3 pyramidal cells was tested by stimulating the contralateral perforant path in order to activate the same CA1 basket cells via different inputs. Contralateral stimulation evoked CA1 pyramidal cell paired-pulse inhibition immediately in the previously stimulated hippocampus. Thus, we propose the "dormant basket cell" hypothesis, which implies that despite malfunction, inhibitory systems remain intact in "epileptic" tissue and are capable of functioning if appropriately activated.     

Uptake ofd-[H]aspartic acid by hippocampal slices: the influence of low and high frequency activation of nerve endings

Schaffer collaterals (Sch. coll.) projection and perforant path (PP) fibers in the hippocampal slices were stimulated in the presence of 0.8 μMd-[³H]aspartic acid — a marker of glutamergic nerve terminals. The activation of these fibers evoked a frequency and sodium-dependent increase ofd-[³H]aspartate uptake. The increase ofd-aspartate uptake during Schaffer collaterals stimulation was much higher than during perforant path stimulation. The stimulation of Sch. coll. nerve endings, which was preceded by a high frequency stimulation of these fibers (potentiation), evoked a decrease ofd-aspartate uptake as compared to stimulated controls.     

Direct connection between perirhinal cortex and hippocampus is a major constituent of the lateral perforant path

Single-pulse stimulation of the perirhinal cortex (PRC) evoked field responses in the dorsal hippocampal CA1 region in urethane-anesthetized rats. In depth profiles conducted by moving the PRC stimulating electrode, the largest amplitude hippocampal potential was generated when the stimulating electrode was located within the perirhinal region. More dorsal (temporal cortex) or more ventral (lateral entorhinal cortex) stimulating sites elicited minimal hippocampal potentials. The hippocampal response was maintained during 100 Hz stimulation of the PRC, suggesting that it was monosynaptic, and high-frequency stimulation (400 Hz) of the PRC produced a significant potentiation of hippocampal CA1 field potentials (46.73 ± 4.14%). When the PRC and the lateral perforant path (LPP) were stimulated separately, the depth/amplitude profiles obtained from a roving recording electrode located within the dorsal hippocampus were similar. In order to determine if fibers from PRC project to the hippocampus via the LPP, the PRC-CA1 and LPP-CA1 potentials were recorded prior to and during procaine (20%, 0.5 μl) blockade of the LPP. A simultaneous loss of both potentials was observed immediately following procaine infusion, while a commissural control potential was unaffected. Both LPP and PRC potentials returned approximately 30–40 min later. Electrolytic lesions of PRC produced a significant decrease in the amplitude of LPP-hippocampal potentials when testing was conducted 4–5 days postlesion. Lesions of lateral entorhinal cortex or temporal cortex did not produce such effects. These data suggest that a direct pathway from perirhinal cortex to the dorsal hippocampal CA1 field can undergo long-term potentiation (LTP) and that this pathway makes a major contribution to the lateral perforant path. © 1996 Wiley-Liss, Inc.