www.elsevier.com / locate / bres
Research report
Propagation of synchronous burst discharges from entorhinal cortex to
morphologically and electrophysiologically identified neurons of rat
lateral amygdala
a ,
*
a bMakoto Funahashi
, Ryuji Matsuo , Mark Stewart
a
Department of Physiology, Okayama University Dental School, 2-5-1 Shikata-cho, Okayama 700-8525, Japan b
Department of Physiology and Pharmacology, SUNY Health Science Center, Brooklyn, NY 11203, USA Accepted 15 August 2000
Abstract
Intracellular and field potential recordings were taken from the lateral nucleus of the amygdala in a rat horizontal brain slice preparation that included hippocampal formation. Pyramidal cells comprised the majority of labeled cells (77%). Electrophysiological classification based on hyperpolarizing or depolarizing afterpotentials subdivided both the pyramidal and non-pyramidal cell classes, although pyramidal cells tended to have hyperpolarizing afterpotentials (70%) and non-pyramidal cells tended to have depolarizing afterpotentials (63%). Synchronous population bursts were triggered with single extracellular stimuli in the deep layers of entorhinal cortex. These events propagated from deep layers of entorhinal cortex into the lateral nucleus of the amygdala. Latencies were consistent with a direct entorhinal to amygdala projection. Individual lateral nucleus neurons exhibited responses ranging from a long burst response that included an initial period of 200 Hz firing and a tail of gamma frequency firing lasting over 100 ms (grade 1) to an epsp with no firing (grade 4). Half of pyramidal cells responding to events initiated in entorhinal cortex were found to receive epsps strong enough to trigger firing. Only one stellate neuron fired in response to entorhinal stimulation. Excitatory postsynaptic responses included NMDA and non-NMDA receptor mediated components. We demonstrate that synchronous population events can propagate from entorhinal cortex to the lateral nucleus of the amygdala and that pyramidal neurons of the lateral nucleus are more common targets than stellate neurons. We conclude that other synchronous events such as sharp waves and interictal spikes can spread from entorhinal cortex to amygdala in the same manner. 2000 Elsevier Science B.V. All rights reserved.
Theme: Other systems of the CNS
Topic: Limbic system and hypothalamus
Keywords: Amygdala; Retrohippocampal area; Sharp wave; Rat
1. Introduction The amygdala has been subdivided into multiple nuclei and subnuclei with some variability in nomenclature The amygdala is a subcortical component of the limbic [11,23], but much is known about its organization [22]. In system that is critical for the expression of emotion. Basic particular, the lateral nucleus of the amygdala is considered neuroanatomical studies [1,11,24,30] and experiments with a principal target for the cortical and subcortical projec-classically conditioned fear responses [6,14,16] are being tions that carry information about a conditioning stimulus used to define the cortical and subcortical pathways into that animals learn to fear [6,14,16]. For example, direct the amygdala that mediate emotional learning and memory projections from the auditory thalamus and relayed
projec-[4]. tions from the auditory association cortex have been shown
to converge on the lateral nucleus of the amygdala with information about a tone that signals a foot shock.
*Corresponding author. Tel.: 181-86-235-6642; fax: 1
81-86-235-Another system of inputs to the amygdala may encode
6644.
E-mail address: [email protected] (M. Funahashi). information about the context within which an association
between conditioning and unconditioned stimuli is formed NaCl, 124; KCl, 5; CaCl , 2.0; MgCl , 1.6; NaHCO , 26;2 2 3 [12]. These arise from several parts of the hippocampal glucose, 10; pH was maintained at 7.4 when exposed to formation, including area CA1 (at the CA1-subiculum 95% O / 5% CO ). Thick slices of tissue (about 2–4 mm2 2 transition) and the entorhinal cortex. The entorhinal cortex thick), which included portions of the lateral amygdala, and perirhinal cortex (areas 35 and 36) have projections to hippocampus, entorhinal cortex and subicular complex, the lateral amygdala, whereas the other hippocampal were cut horizontally from the intact hemispheres. Thin afferents terminate in different amygdalar nuclei [11,30]. slices (about 350mm) were cut from such blocks using a Our interest in the entorhinal inputs to the amygdala tissue slicer (Microslicer DTK-3000, Dosaka, Japan) and derives in part from an interest in hippocampal formation maintained in a holding chamber at room temperature inputs that may converge on cells receiving thalamic or (23–258C) for at least 1 h. Single slices were transferred to association cortical inputs in the lateral nucleus of the a nylon mesh support in an interface recording chamber, amygdala. These entorhinal afferents may also mediate the where they were perfused with ACSF. The upper surfaces spread of synchronous population events that begin in the were exposed to a warmed, moistened atmosphere of 5% hippocampus or entorhinal cortex. Such events include CO in O . The temperature of the chamber was controlled2 2 ‘abnormal’ synchronous discharges that can arise in the at 3560.18C.
hippocampal formation such as interictal spikes [28], and
‘normal’ synchronous discharges such as sharp waves 2.2. Recording techniques [3,26]. Return projections from the amygdala to the
entorhinal cortex have been shown to carry another Extracellular recording electrodes were stainless steel synchronous population discharge, called a sharp sleep (acute conical tips; Roboz Microprobe, Rockville, MD) potential, from the basolateral amygdala to the entorhinal with tip impedances (at 1 kHz) of 0.9 to 1.1 MV. Signals,
cortex [21]. referred to the bath, were amplified (DPA-100D, Dia
One reason the hippocampal sharp wave has attracted Medical Systems, Tokyo, Japan), filtered (0.1 Hz to 10 attention is that the firing patterns of neurons participating kHz,26 dB / octave), and digitized (Digidata 1200, Axon in the generation of a sharp wave resemble patterns of Instruments, Foster City, CA) for off-line analysis. In-stimulation used to induce long term potentiation. This tracellular recording electrodes were pulled from 1 mm includes a relatively brief period of very high frequency diameter filament-containing glass capillary tubes and activity (200 Hz) at the beginning of the event and a filled with 3 M potassium acetate (tip resistances 80–120 trailing, longer duration period of gamma frequency (40– MV) or a solution containing 2% Neurobiotin tracer 100 Hz) activity. In the entorhinal cortex, distant sites can (Vector Laboratories, Burlingame, CA, USA) in 2 M be synchronous at zero phase lags during periods of potassium acetate (tip resistances 100–180 MV) [13]. gamma activity [8]. While it is unclear what information Intracellularly recorded signals were amplified by a high may be carried in a sharp wave, these events can lead to input impedance amplifier with facilities for current in-changes in synaptic efficacy themselves or they may jection using a bridge circuit and capacitance compensa-facilitate changes at other weaker inputs. tion (Neurodata IR-183, New York, USA). Extracellular Sharp waves are believed to originate in intact animal stimulating electrodes were parallel bipolar (150 mm brains in area CA3 of the hippocampus [3]. From here, diameter stainless steel, 0.5 mm tip exposure, 0.19 mm tip these events propagate through successive hippocampal separation; FHC, Brunswick, ME, USA). Stimulus pulses formation regions to the entorhinal cortex [5]. The spread were put through constant-current isolation units (Isolator-of activity beyond the entorhinal cortex has not been 10, Axon Instruments, Foster City, USA) at 0.25 Hz, 50ms defined. We sought to explore the propagation of syn- duration, 0.1–0.35 mA.
chronous population discharges from the entorhinal cortex All electrodes were placed under direct visual guidance to the lateral nucleus of the amygdala and to identify the with a dissecting microscope according to measurements cellular targets of this activity. These data from brain slices from observable landmarks (e.g. pial surface, angular may be a useful background for studies of sharp wave bundle, alveus, fimbria, ventricle). During experiments, the
propagation in vivo. locations for all recording and stimulating electrode
place-ments were drawn by hand using a dissecting microscope. Cell locations of labeled neurons were obtained by locating
2. Materials and methods cells directly in Nissl stained sections.
2.1. Slice preparation and maintenance 2.3. Pharmacological manipulations
against NMDA and non-NMDA receptor mediated gluta- stellate neurons from our dataset are shown in Fig. 1. matergic transmission, respectively. Picrotoxin (100 mM, Pyramidal and stellate neurons were subclassified based on Sigma Chemical Co., St. Louis, MO, USA) was used to their dendritic spine density (spiny, slightly or sparsely block GABA receptor-mediated inhibitory transmission.A spiny and aspiny). Spine density was actually a principal characteristic in the classification scheme of McDonald
2.4. Neurobiotin-labeling of single cells [17].
Separately, recorded cells were classified based on their For cells recorded with Neurobiotin-containing elec- electrophysiological properties. As in a previous study trodes, after electrophysiological recording, Neurobiotin- [25], our principal characteristic was the appearance of tracer was injected into these cells using 2–4 nA depolariz- afterpotentials following action potentials (Fig. 2). We ing rectangular current pulses (150 ms duration at 3.3 Hz distinguished two types of afterpotential that followed an for 20–30 min). Post-injection survival times ranged from action potential initiated by current injection. One was a 10 to 60 min. Slices with Neurobiotin-injected cells were hyperpolarizing afterpotential appearing in 63% of re-fixed in 4% paraformaldehyde and 0.2% picric acid in 0.1 corded cells. The other was a depolarizing afterpotential M phosphate buffer (pH 7.4) from overnight to 10 days. appearing in 37% of recorded cells.
Frozen sections (40–60 mm thick) were cut from the fixed Neither of the two basic electrophysiological classes tissue and kept in phosphate buffered saline (PBS, pH 7.4). corresponded to a single morphological cell class. The After rinses with PBS, these sections were treated with basic electrophysiological and morphological properties of 0.1% H O2 2 for 20 min and Triton-X100 (0.4–0.5% in the cell classes are given in Tables 1 and 2. Cells with PBS) for 2 to 3 h. They were rinsed in PBS and then hyperpolarizing afterpotentials were found to be both incubated in the Vectastain ABC Reagent (Vector Lab- pyramidal and stellate in shape, although most (19 / 22) oratories) in PBS for 2 to 3 h. After rinses with PBS, were pyramidal cells. Cells with depolarizing afterpoten-sections were reacted with diaminobenzidine (DAB) and tials were also found to be both pyramidal (8 / 13) and H O (0.003%) in PBS to visualize the injected cells. The2 2 stellate in shape (5 / 13), although the split was more sections, which included the successfully stained cells, equitable.
were counter stained by Nissl staining.
3.2. Intrinsic electrophysiological properties
3. Results Both electrophysiological classes of cells discharged several spikes at very short intervals at the onset of large Intracellular recordings were taken from 76 neurons in depolarizing currents applied via the recording electrode the lateral nucleus of amygdala in 72 horizontal slices from (Fig. 2A2, B2). Cells with a hyperpolarizing afterpotential rat brains. In some experiments, multiple cells were showed repetitive firing in response to depolarizing current recorded, but, in the vast majority of experiments, a single injection without adaptation of the firing frequency (Fig. neuron was recorded in each slice to permit unequivocal 2A2, A3, A4). Firing frequencies were stabilized to higher matching of morphological results with electrophysiologi- rates in cells with a hyperpolarizing afterpotential. Hy-cal results. All recorded cells had overshooting action perpolarizing current pulses revealed marked inward recti-potentials and resting membrane recti-potentials in the range of fication in this electrophysiological class of cells.
268 to278 mV that were stable for 20 min to 1 h. Field Cells with a depolarizing afterpotential showed slight potential recordings were taken from at least one location adaptation of firing frequency in response to depolarizing (deep layers of the retrohippocampal cortices or the lateral current injection, but overall, their firing rates during
nucleus of amygdala) in each slice. comparable current pulses were lower (Fig. 2B2, B3, B4).
Rectification of inward or outward currents was much less 3.1. Classification of cells in lateral nucleus for this electrophysiological class of cells. In three cells, the depolarizing afterpotential led to the generation of Of the Neurobiotin-filled neurons, 35 cells were success- spike doublets in response to current injection.
fully recovered to permit morphological classification.
Labeled cells were classified as pyramidal (27 / 35; 77%) or 3.3. Responses evoked by entorhinal stimulation stellate cells (8 / 35; 23%) based on the pattern of dendritic
branching. Pyramidal cells had a clear ‘apical’ dendrite Evoked responses were recorded in 40 lateral nucleus that was distinct from their multiple smaller ‘basal’ neurons to extracellular stimulation of the deep layers of dendrites. Stellate cells were characterized by multiple the entorhinal cortex. Twenty-three of these cells were also dendrites of uniform diameter and with a uniform dis- morphologically identified as pyramidal (N517) or stellate tribution around the soma. Similar classification schemes (N56) neurons.
Fig. 1. Photomicrographs of Neurobiotin-labeled neurons in the lateral nucleus of the rat amygdala. (A1) A low power image of a typical pyramidal cell with a thick primary (‘apical’) dendrite and ‘basal’ dendrites. This cell exhibited a hyperpolarizing afterpotential. (A2) Higher magnification image shows dendritic spines. (B) Low (B1) and high (B2) magnification images of a typical stellate neuron with several thin spiny dendrites uniformly distributed around the soma. This cell exhibited a depolarizing afterpotential. Calibration bar is 20mm.
exhibited by deep layer retrohippocampal neurons (Fig. 3), The second grade (2) of response was characterized by to epsps with no associated firing. We graded responses an event duration totaling about 50 ms. The maximum based on the pattern of evoked action potentials. Grade 1 firing frequency during grade 2 responses reached 100 Hz. responses resembled responses of deep layer entorhinal Epsp amplitudes were considerably smaller than those in neurons. These consisted of a brief (10–50 ms) period of grade 1 responses. Grade 2 responses were recorded in very high frequency (200 Hz) firing followed by a variable 8 / 40 cells, including two identified pyramidal and one period of sustained membrane depolarization without stellate neuron.
Table 1
a Membrane properties of Neurobiotin-labeled neurons in the lateral nucleus of amygdala
Afterpotential Cell shape n Resting Input Time Spike Spike
potential resistance constant amplitude half-width
(mV) (MV) (ms) (mV) (ms)
Hyperpolarization Pyramidal 19 270.665.0 44.1611.1 12.764.6 72.7612.6 1.3860.4 Stellate 3 271.564.9 25.561.1 12.561.9 67.468.5 1.5260.2 Depolarization Pyramidal 8 268.863.3 44.6611.6 12.764.52 71.9615.2 1.3560.2 Stellate 5 268.861.8 39.8610.1 10.362.71 72.066.3 1.4260.3 a
Values expressed as mean6S.D.
cells). Representative responses of cells from each grade in these cells. Epsps were seen in relation to each of the
are shown in Fig. 4. afterdischarges in field potential recordings (Fig. 6B).
Evoked responses of lateral nucleus neurons in all
grades were associated with similar field potential dis- 3.5. NMDA and non-NMDA components of epsps charges in the entorhinal cortex and lateral nucleus of the
amygdala. Synchronous activation of lateral nucleus neu- CPP and CNQX were used to examine epsps in three rons was also evidenced in pairwise recordings (Fig. 5). lateral nucleus cells responding to entorhinal stimulation. One example of a grade 4 cell is shown in Fig. 7. A long 3.4. Effects of GABA receptor blockadeA sustained epsp was evoked by stimulation to the deep layer of the entorhinal cortex. The amplitude and duration of To test the possibility that variations in the evoked epsps were gradually decreased during the first 10 min responses were due to differences in the intensity of following addition of CPP to the perfusate. The remaining concurrent inhibition that may be activated by propagating component, stable in amplitude and duration after 10 min population events, we tested four cells with grade 3 (epsp of CPP exposure was completely suppressed by addition of plus one action potential) and four (epsp only) evoked CNQX to the perfusate.
responses in the presence of 100 mM picrotoxin (Fig. 6). At resting membrane potential, cells that were classified as
grade-3-evoked responses developed progressively more 4. Discussion
intense responses that depended on the duration of
pic-rotoxin exposure. Within 5 min in each grade 3 neuron, Neurons of the lateral nucleus of the rat amygdala were responses were considered grade 2, with multiple dis- recorded in horizontal brain slices that included the charges at less than 100 Hz. Longer exposure led to entorhinal cortex and other retrohippocampal and hip-multiple repetitive burst discharges that resembled epi- pocampal areas. Lateral nucleus neurons were separated leptiform events in hippocampal and retrohippocampal into two groups based on the afterpotentials following cortices. Responses never resembled grade 1 events be- action potentials elicited by current injection. Cells were cause none exhibited the tail of gamma frequency activity. also classified based on their morphology as pyramidal or Concurrent field potential recordings showed these in- stellate neurons and subclassified based on dendritic tracellular discharges to coincide with population spikes in spininess. We found that morphological classification based
the field. on cell shape or on dendritic spininess did not correspond
Cells originally scored with grade 4 responses never to electrophysiological classification based on afterpoten-developed firing even after application of picrotoxin (100 tials.
mM) to the bath. The amplitude and duration of epsps We examined the responses of morphologically and increased, but firing was only seen when the cells were electrophysiologically identified cells to the spread of held depolarized from rest. Burst firing was never observed synchronous population events started in the entorhinal
Table 2
Characteristics of Neurobiotin-labeled cells in the lateral nucleus of amygdala a
Afterpotential Cell shape n Diameter Spines
(mm)
Aspiny Slight Spiny
Hyperpolarization Pyramidal 19 15.961.6 10 9
Stellate 3 17.161.2 1 2
Depolarization Pyramidal 8 17.761.6 1 3 4
Stellate 5 15.461.2 4 1
a
Fig. 3. Comparison of lateral nucleus neuron responses with deep layer entorhinal neuron responses to deep layer entorhinal and parasubicular stimulation. (A) Layer V entorhinal neuron response (upper trace) and field response (lower trace) to entorhinal stimulation. Note the initial high frequency firing coincident with the field spike and gamma frequency activity at the end of the intracellular response. Synchrony of this activity can appear as an oscillation in field recordings. Synchronous gamma activity is not seen in this response. (B) Lateral nucleus neuron response (upper trace) and layer V entorhinal field response (lower trace) to entorhinal stimulation. Latency difference is consistent with propagation from the entorhinal cortex to the lateral nucleus. (C) Different lateral nucleus neuron response (upper trace) and layer V entorhinal field response (lower trace) to parasubicular stimulation. Note the longer latency in a lateral nucleus neuron following parasubicular stimulation. Arrows below the field recordings mark the time of stimulation. Illustrations: drawing of horizontal slice showing electrode location. (d) Field potential electrode, in this and subsequent figures.
Fig. 4. Intracellular (left column) and simultaneous field (right column) responses in lateral nucleus to extracellular stimulation of deep layers of entorhinal cortex. Each row (A–D) is an example of a different ‘grade’ of evoked response. (A) Response resembling the responses typically seen in deep layer retrohippocampal neurons consisting of an initial burst of very high frequency activity followed by a variable duration depolarizing plateau without firing and a trailing period of gamma frequency activity. Note the long duration of the field response. The second grade of response (B) consisted of an evoked epsp with multiple action potentials. Cells responding with a single action potential even on very large epsps were labeled grade 3. Finally, some cells only exhibited epsps with no action potentials (grade 4). Arrows below traces in (D) mark the time of the stimulus for all traces to the deep entorhinal cortex.
4.1. Classification of lateral nucleus neurons combined electrophysiological / morphological studies of basolateral [29] and lateral [25] nucleus neurons.
Cells labeled with Neurobiotin were classified as Our findings most closely match the morphological pyramidal or stellate and further subclassified based on the categories of pyramidal cells and stellate cells proposed by relative spininess of their dendrites. Separately we classi- Millhouse and DeOlmos [19]. The proportions of each cell fied cells based on the appearance of the afterpotentials shape of the present study match theirs.
Fig. 5. Synchronous grade 1 discharges in a pair of lateral nucleus neurons to stimulation of the deep entorhinal cortex. Cross-correlogram below responses shows spike coincidence within 1 ms. The|8 Hz oscillation in the cross-correlogram results from the slow depolarizations and associated firing during the
initial event and the single afterdischarge. The subthreshold gamma frequency oscillations seen in the two cells following the early very high frequency discharges do not contribute significantly to the cross-correlogram shown.
Sparsely spiny neurons (class II) had various shapes and groups contained pyramidal and stellate neurons. Their dendritic patterns, but only comprised about 5% of cells. sample appears to have come from a more ventral location Rarer still were the neurogliaform cells (class III) and than the sample of the present study. Theirs was from the these were in basolateral nucleus. Only if we group level where the dorsal endopiriform nucleus and the lateral together the spiny and sparsely spiny cells can we approxi- ventricle define a wedge of lateral nucleus. We routinely mate the proportions in categories resembling those of used slices from more dorsal levels as evidenced by the McDonald. More likely is that variation in the proportions drawings of the slices in the figures. Differences in of pyramidal and stellate shaped class I cells changes sampling more dorsally or more ventrally in the lateral between basolateral and lateral nuclei, and even within nucleus were associated with differences in proportions of
lateral nucleus [18]. pyramidal and stellate shapes. Such differences in
sam-An electrophysiological classification of lateral nucleus pling may also be associated with differences in pro-neurons based on afterpotentials was proposed by Sugita et portions of electrophysiological cell classes.
al. [25]. They identified cells with depolarizing
afterpoten-tials and found that these had pyramidal cell shapes after 4.2. Spread of synchronous activity from the entorhinal biocytin labeling. They reported two other groups, both cortex to amygdala
with hyperpolarizing afterpotentials, that were found to
Fig. 6. Picrotoxin increases firing in cells that fired in normal media, but not in cells showing only epsps. (A) Example of a cell with a grade 3 response to entorhinal stimulation before and after 100mM picrotoxin was added to the perfusate. Note afterdischarges in the field recordings (also taken from the lateral nucleus) and the associated epsps early (5 min) and firing later (7 min). (B) Example of a cell with a grade 4 response to entorhinal stimulation (epsp only) before and more than 8 min after 100mM picrotoxin exposure. Only when the cell was held depolarized from rest with constant current injection was it possible to see single spikes in association with the epsps corresponding to the primary burst or afterdischarges. Resting potential was268 mV. Note the greater amplitude changes in the epsp associated with the primary field event as the membrane potential was varied from the resting level.
Fig. 7. NMDA and non-NMDA receptor mediated components of lateral nucleus neuron responses to entorhinal stimulation (0.32 mA / 4 s). Each trace is an average of 10 evoked responses under the conditions labeled. Application of the NMDA receptor antagonist CPP (10mM) gradually eliminated the late slow epsp leaving an early briefer epsp that was sensitive to CNQX (10mM), an antagonist of AMPA / kainate type glutamate receptors. Responses shown in a, c and e are superimposed at top right.
work focused on the sensory cortical and subcortical learning in some areas [20]. In the presence of GABA pathways that would be needed for associating a con- receptor blockade or low magnesium media, however,
ditioning tone with a foot shock. seizure activity can be seen in amygdala without attached
Clearly, paths exist for the spread of synchronous entorhinal cortex [9,10].
activity into the lateral nucleus of the amygdala [2]. These The interconnections of the entorhinal cortex and could well impact on other amygdala nuclei via intra- amygdala are clearly sufficient to mediate the spread of amygdala connections and ultimately a wide array of sharp wave-like events from the entorhinal cortex to the cortical and subcortical targets. We will not discuss these amygdala (as shown in this paper, also Ref. [2]) and sharp connections in any detail. We do note, however, that such sleep potentials from the amygdala to the entorhinal cortex projections, especially to subcortical targets in the hypo- [7,21]. Whereas the occurrence of sharp waves in any brain thalamus and brainstem may account for the viscerosen- region during fear conditioning has not been documented, sory or autonomic phenomena that occur during temporal the spread of events in our in vitro preparation suggests
lobe seizures [11]. that they may profoundly influence amygdala activity. This
Sharp waves are presumably normal synchronous dis- study of the cells and connections mediating the spread of charges that are more limited in duration than seizure in vitro synchronous events from the entorhinal cortex may events. They have been suggested to have a role in serve as a useful background for the study of the spread of learning [3]. As such, the spread of sharp wave activity in vivo events such as the sharp wave or interictal spikes. from the hippocampus to the amygdala may serve as a link
between the spatial context and emotional learning. It is interesting that we have seen grade 1 responses (i.e. events
that start with 200 Hz firing and end with gamma Acknowledgements
frequency activity) in lateral nucleus in horizontal slices,
[16] S. Maren, M.S. Fanselow, The amygdala and fear conditioning: has
References
the nut been cracked?, Neuron 16 (1996) 237–240.
[17] A.J. McDonald, Neurons of the lateral and basolateral amygdaloid [1] D.G. Amaral, J.L. Price, A. Pitkanen, S.T. Carmichael, Anatomical nuclei: a Golgi study in the rat, J. Comp. Neurol. 212 (1982)
organization of the primate amygdaloid complex, in: J.P. Aggleton 293–312.
(Ed.), The Amygdala, Neurobiological Aspects of Emotion, Mem- [18] A.J. McDonald, Neuronal organization of the lateral and basolateral ory, and Mental Dysfunction, Wiley-Liss, New York, 1992, pp. amygdaloid nuclei in the rat, J. Comp. Neurol. 222 (1984) 589–606. 1–66. [19] O.E. Millhouse, J.F. DeOlmos, Neuronal configurations in lateral [2] L.A. Brothers, D.M. Finch, Physiological evidence for an excitatory and basolateral amygdala, Neuroscience 10 (1983) 1269–1300.
pathway from entorhinal cortex to amygdala in the rat, Brain Res. [20] W.H. Miltner, C. Braun, M. Arnold, H. Witte, E. Taub, Coherence of 359 (1985) 10–20. gamma-band EEG activity as a basis for associative learning, Nature [3] G. Buzsaki, Hippocampal sharp waves: their origin and significance, 397 (1999) 434–436.
Brain Res. 398 (1986) 242–252. [21] D. Pare, J. Dong, H. Gaudreau, Amygdalo-entorhinal relations and [4] L. Cahill, N.M. Weinberger, B. Roozendaal, J.L. McGaugh, Is the their reflection in the hippocampal formation: generation of sharp
amygdala a locus of ‘conditioned fear’? Some questions and caveats, sleep potentials, J. Neurosci. 15 (1995) 2482–2503.
Neuron 23 (1999) 227–228. [22] A. Pitkanen, V. Savander, J.E. LeDoux, Organization of intra-[5] J.J. Chrobak, G. Buzsaki, Selective activation of deep layer (V–VI) amygdaloid circuitries in the rat: an emerging framework for retrohippocampal cortical neurons during hippocampal sharp waves understanding functions of the amygdala, Trends Neurosci. 20 in the behaving rat, J. Neurosci. 14 (1994) 6160–6170. (1997) 517–523.
[6] M. Davis, Neurobiology of fear responses: the role of the amygdala, [23] J.L. Price, Toward a consistent terminology for the amygdal J. Neuropsych. Clin. Neurosci. 9 (1997) 382–402. complex, in: The Amygdaloid Complex, INSERM Symposium 20, [7] D.M. Finch, E.E. Wong, E.L. Derian, X.H. Chen, F.N. Nowlin, L.A. Elsevier / North Holland Biomedical Press, Amsterdam, 1981, pp.
Brothers, Neurophysiology of limbic system pathways in the rat: 13–18.
projections from the amygdala to the entorhinal cortex, Brain Res. [24] J.L. Price, F.T. Russchen, D.G. Amaral, The limbic region. II. The 370 (1986) 273–284. amygdaloid complex, in: A. Bjorklund, T. Hokfelt, L.W. Swanson [8] M. Funahashi, M. Stewart, Properties of gamma-frequency oscilla- (Eds.), Integrated systems of the CNS, Part I, Handbook of tions initiated by propagating population bursts in retrohippocampal Chemical Neuroanatomy, Vol. 5, Elsevier Science, Amsterdam, regions of rat brain slices, J. Physiol. Lond. 510 (1998) 191–208. 1987, pp. 279–388.
[9] P.W. Gean, P. Shinnick-Gallagher, Picrotoxin induced epileptiform [25] S. Sugita, E. Tanaka, R.A. North, Membrane properties and synaptic activity in amygdaloid neurons, Neurosci. Lett. 73 (1987) 149–154. potentials of three types of neurone in rat lateral amygdala, J. [10] P.W. Gean, P. Shinnick-Gallagher, Epileptiform activity induced by Physiol. Lond. 460 (1993) 705–718.
magnesium-free solution in slices of rat amygdala: antagonism by [26] S.S. Suzuki, G.K. Smith, Spontaneous EEG spikes in the normal
N-methyl-D-aspartate receptor antagonists, Neuropharmacology 27 hippocampus. I. Behavioral correlates, laminar profiles and bilateral (1988) 557–562. synchrony, Electroencephalogr. Clin. Neurophysiol. 67 (1987) 348– [11] P. Gloor, The amygdaloid system, in: The Temporal Lobe and 359.
Limbic System, Oxford University Press, New York, 1997, pp. [27] L.W. Swanson, G.D. Petrovich, What is the amygdala?, Trends
591–721, Chapter 6. Neurosci. 21 (1998) 323–331.
[12] P.C. Holland, M.E. Bouton, Hippocampus and context in classical [28] R.D. Traub, R. Miles, in: Neuronal Networks of the Hippocampus, conditioning, Curr. Opin. Neurobiol. 9 (1999) 195–202. Cambridge University Press, New York, 1991, p. 281.
[13] H. Kita, W. Armstrong, A biotin-containing compound N-(2-amino- [29] M.S. Washburn, H.C. Moises, Electrophysiological and morphologi-ethyl)biotinamide for intracellular labeling and neuronal tracing cal properties of rat basolateral amygdaloid neurons in vitro, J. studies: comparison with biocytin, J. Neurosci. Methods 37 (1991) Neurosci. 12 (1992) 4066–4079.
141–150. [30] M.P. Witter, H.J. Groenewegen, F.H. Lopes da Silva, A.H. Lohman, [14] J.E. LeDoux, Emotion: clues from the brain, Annu. Rev. Psychol. 46 Functional organization of the extrinsic and intrinsic circuitry of the
(1995) 209–235. parahippocampal region, Prog. Neurobiol. 33 (1989) 161–253. [15] S. Maren, Long-term potentiation in the amygdala: a mechanism for
