High frequency (200 Hz) oscillations and firing patterns in the basolateral amygdala and dorsal endopiriform nucleus of the behaving rat

Behavioural Brain Research 141 (2003) 123–129

Research report

High frequency (200 Hz) oscillations and firing patterns in the basolateral amygdala and dorsal endopiriform nucleus of the behaving rat Alexei A. Ponomarenko∗ , Tatiana M. Korotkova, Helmut L. Haas Department of Physiology, Heinrich-Heine University, D-40001 Düsseldorf, Germany Received 21 August 2002; received in revised form 2 October 2002; accepted 2 October 2002

Abstract The known repertoire of rhythms in the amygdala and paleocortex includes a range of oscillations from slow waves (<1 Hz) to fast gamma (40–100 Hz). In the present report, we show ∼200 Hz oscillations in the basolateral nucleus of the amygdala (BL) and the adjacent dorsal endopiriform nucleus (EPN) of the behaving rat. Microwire techniques were applied for recording single units and field activity from these structures and EEG from the dorsal or temporal CA1 subfields of the hippocampus. Units from both EPN and BL exhibited similar irregular firing patterns with bursts. The mean firing rates in EPN were <1 Hz, whereas units in the BL fired in a range of <1–17 Hz. Neuronal activity in both BL and EPN was phase-locked with high-frequency field oscillations (HFO, ∼200 Hz). Amygdaloid/EPN HFO displayed on average lower numbers of cycles and smaller amplitudes than hippocampal ripples. Neuronal firing and HFO in the BL and EPN were state dependent with a maximal occurrence during slow-wave sleep (SWS), being lower during waking and paradoxical sleep. Cross-correlation between hippocampal ripples and EPN or BL units and field HFO did not reveal any synchrony. These data suggest common principles of temporal coding in BL and EPN in certain behavioural states via short scale population synchrony though they convey signals of different modalities. © 2002 Elsevier Science B.V. All rights reserved. Keywords: High frequency; Ripple; Memory; Hippocampus; Mesial temporal lobe; LTP

1. Introduction Amygdala and paleocortex display various oscillatory patterns depending on behavioural state. During emotional arousal and paradoxical sleep theta rhythm is recorded from nuclei of the basolateral complex and perirhinal cortex [18,27,29]. Arousal states are also characterised by oscillations in beta/gamma bands in the pre-piriform cortex, whereas coherent slow delta waves in basolateral (BL), lateral (L) nuclei and rhinal cortices are typical for slow-wave sleep (SWS) [29]. Amygdaloid neurons fire at delta frequency during SWS (BL) and show gamma-related modulation of firing probability (L) [19]. Population bursts of BL neurons during SWS give rise to “sharp potentials” in the entorhinal cortex, layer II which targets the hippocampus [28]. The dorsal endopiriform nucleus (EPN) which lays lateral adjacent to the BL complex generates high-frequency bursts and population dis∗ Corresponding author. Tel.: +49-211-811-4422; fax: +49-211-811-4231. E-mail address: [email protected] (A.A. Ponomarenko).

charges in vitro [20,36]. This region shares the connectivity profile, projecting to enthorhinal cortex and hippocampus, and the cytoarchitectonic organisation with deep layers of the piriform cortex and the basolateral complex [2,3,34]. The EPN is critical for generation of epileptiform activity in the underlying cortex [21,23,24]. The theta-rhythm (6–12 Hz), with intermittent gamma patterns (40–100 Hz) is observed during exploratory behaviour and REM sleep in the hippocampus, whilst in a state of awake immobility or slow-wave sleep sharp wave-associated synchrony in the CA3 region triggers high-frequency oscillations (200 Hz, “ripples”, HFO) in CA1 [15]. These are believed to be critically involved in the consolidation of memory traces [25] and may be the natural stimuli to induce long-term potentiation [10–12]. Hippocampal ripples and prefrontal cortical sleep spindles occur in close temporal synchrony and hippocampal neurons fire before cortical ones during the ripple–spindle complex [33]. Phase-locked firing of interneurons and discharge of selected CA1 pyramidal cells within a narrow time window constitute the cellular basis of ripple oscillations in the hippocampus [39]. Ultra fast (up to 500 Hz) focal ripples, recorded from parahippocampal regions in models of

0166-4328/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 4 3 2 8 ( 0 2 ) 0 0 3 2 7 - 3

124

A.A. Ponomarenko et al. / Behavioural Brain Research 141 (2003) 123–129

temporal lobe epilepsy and epileptic patients are associated with burst-like firing of pyramidal neurons [5–8]. The basolateral complex of the amygdala and the hippocampus are the principal components of the medial temporal lobe memory system that mediates declarative learning and provides a complex representation for emotionally charged decisions and voluntary behaviour [16,22]. However, the network basis of functional integration between amygdala and hippocampus remains largely unknown [1]. In the present report, we show that in freely behaving rats the population discharge of neurons in the basolateral amygdaloid complex and the EPN is associated with field oscillations (∼200 Hz) in these structures. Furthermore, these patterns appear to be independent from ripple oscillations in the CA1 regions of both temporal and dorsal poles of the hippocampus.

2. Methods Three male Wistar rats (350–400 g) were housed at 22 ± 2 ◦ C with a 12 h light/dark period and supplied with food and water ad libitum. Rats were anaesthetised with a mixture of ketamine (100 mg/kg, Ketavet, Upjohn GmbH, Heppenheim, Germany) and 10 mg/kg xylazinhydrochloride (Rompun, Bayer AG, Leverkusen, Germany), by intra-peritoneal injection. The number of animals used was the minimal requirement for this study. All experiments were conducted in compliance with German law and with the approval of the Bezirksregierung Duesseldorf. All efforts were made to minimise the pain and discomfort of the experimental animals. Pain reflexes were regularly checked and if necessary additional anaesthetic was given. Microwire techniques were applied for recording unit and field activities [26]. Two single-barrel microdrive assemblies were implanted unilaterally 1 mm above EPN or BL and the temporal or dorsal CA1 area. Bundles of 3–5 microwires, formvar-insulated 25 ␮m stainless-steel wire for the amygdala or the EPN and 50 ␮m tungsten wire for the hippocampus (California Fine Wire Company, Grover Beach, USA) were inserted into the barrel of the microdrive cannula. EMG electrodes were inserted in the neck muscle. Neocortical EEG and indifferent electrodes were screws in the skull. Recording began after 5 days of recovery and adaptation of rats to a transparent box (30 cm in diameter, 30 cm height) situated in a sound-attenuated shielded recording chamber. First, CA1 electrode positions were adjusted for optimal sharp wave-associated ripple detection [13]. Electrodes were advanced in 5–20 ␮m steps until stable and well-isolated amygdaloid single- or multi-units were seen on an oscilloscope. After amplification (differential AC amplifier, A-M system, WA, USA) signals were wideband (1 Hz–10 kHz) digitised at 25 kHz using Power 1401 (CED, Cambridge, UK). EMG and neocortical EEG were band-pass filtered (1–120 Hz) and digitised at 1 kHz. Unit discrimination and further analysis were carried out off line using Spike2 software (CED, Cambridge, UK). A

template matching algorithm was applied for spike sorting in 500 Hz–10 kHz bandpass digitally (Finite Impulse Response) filtered recording, no more than 1 unit was extracted from one trace. The isolation of single units was verified by the absence of a peak in the auto-correlogram with a lag <1 ms, reflecting the refractory period. Recording epochs containing movement artefacts were rejected manually with a reference to EMG. High-frequency field oscillations were detected by a peak-detection algorithm in a low pass filtered (500 Hz), down-sampled (to 1 kHz) and bandpass re-filtered (140–300 Hz) signal from the same microwires that recorded the units. Derived events were used for cross-correlational analysis. An estimation of the mean number of cycles within a HFO was achieved by averaging HFO-containing waveforms and taking cycles with an amplitude exceeding 7 S.D. above mean of the HFO-free recording. Behavioural state was assessed from 1–120 Hz filtered neocortical and hippocampal EEG and EMG, sleep scoring was done with 10 s resolution according to standard criteria for waking, slow-wave sleep (SWS) and paradoxical sleep (PS). Correlation analysis was performed utilising a Spearman rank correlation. Following the completion of the experiments the animals were deeply anaesthetised, perfused intracardially with 10% formalin–sucrose solution and decapitated. Brains were fixed for 3 days in 30% formalin–sucrose, then frozen, cut in 40 ␮m slices and stained with cresyl violet. Recording sites were confirmed.

3. Results Sixty-two neurons were recorded, 38 from the basolateral amygdaloid complex and 24 from EPN. Units from these structures exhibited similar irregular firing patterns. In the majority of neurons (31 in the amygdala and 23 in the EPN) background firing was occasionally interrupted by bursts of 2–8 spikes with a firing frequency of 100–400 Hz. Seven neurons in the basolateral complex, six located in the lateral nucleus (L),which fired <0.01 Hz, are not been considered here due to a non-sufficient number of spikes for analysis in the individual recordings. Notably, mean firing rates in EPN were typically lower than 1 Hz, whereas units in the BL fired from <1 Hz (bursting units, n = 29) to 17 Hz (phasic units, n = 2; Table 1). Auto-correlograms of single units showed a sharp peak at 4–6 ms (with shorter lags for EPN units) and different distributions of longer interspike intervals (Figs. 1A, 2D and 4D). The EEG signal displayed high-frequency oscillations (HFO) mostly at the time of neuronal discharge in BL and EPN. The percentage of action potentials in BL and EPN participating in HFO was 24.3±5.1% and 41.9±11.8% of all spikes, respectively. Power spectra and averaged waveforms revealed a frequency of the field HFO of approximately 200 Hz (Figs. 1B and 2A). Firing of both bursting and phasic single units appeared synchronous with a positive HFO peak when relatively long intervals at cross-correlograms

A.A. Ponomarenko et al. / Behavioural Brain Research 141 (2003) 123–129

125

Table 1 Neuronal firing patterns in the EPN and BL in different vigilance states SWS EPN units (Hz) EPN units (low:burst firing ratio) BL bursting units (Hz) BL bursting units (low:burst firing ratio)

PS

W

0.59 ± 0.03 7.72 ± 2.30

0.41 ± 0.20 15.84 ± 9.73

0.43 ± 0.11 13.85 ± 2.09

0.74 ± 0.11 13.25 ± 5.19

0.65 ± 0.20 24.40 ± 4.66

0.38 ± 0.06 20.09 ± 5.31

(3 s) were considered. Cross-correlograms for EPN units were slightly sharper around the HFO peak than those for BL units, reflecting the slightly higher mean firing rates in the BL (Table 1). At a higher resolution, phase-locking of neuronal discharge with positive peaks of HFO was evident at least for three oscillatory cycles. The number of cycles in an individual HFO epoch tended to correlate with the number of spikes in a burst both in EPN (Fig. 1C) and BL. Single spikes were associated with a short (10 ms) field potential whilst trains of eight spikes corresponded to a similar number of oscillatory cycles (HFO duration up to 80 ms). Amygdaloid and EPN HFO had on average a lower number of cycles than hippocampal ripples (3–6 versus 11 cycles in averaged waveforms, as shown in

Section 2) and smaller amplitudes (Fig. 4C) but belonged to the same frequency band (∼140–200 Hz). In many cases HFO coincided with sharp potentials in a band 1–20 Hz lasting 15–100 ms depending on HFO duration (Figs. 1B and 2B). Auto-correlograms (up to 3 s) of HFO did not reveal any intrinsic slow rhythms in the occurrence of these events. Firing of both BL and EPN units exhibited a clear tendency to state dependency with maximal mean rates during SWS, intermediate ones during PS (for BL) and almost two times lower rates during W (significant for BL in cases of SWS/W and PS/W, P < 0.05; Table 1). Units of BL and EPN fired more bursts (instantaneous frequency >100 Hz) in relation to low frequency firing (<20 Hz) during SWS than

Fig. 1. HFO and neuronal firing in the EPN. (A) Auto-correlogram of a single unit (bin width 50 ms). Insets show higher resolution auto-correlogram and spike waveform (bin width 1 ms, scale 1 ms). Note the absence of interspike intervals shorter than 1 ms. (B) Cross-correlogram between field signal and neuronal firing (n = 9) triggered by a peak of HFO (upper trace). Higher resolution cross-correlogram showing phase relationships of unit firing and HFO peaks (middle). Each bin in the HFO waveform histogram has 1 ms width. Averaged waveform in 1–20 Hz band triggered by HFO peak (lower waveform). (C) Representative examples of correspondence of unit spiking (lower trace) and number of cycles in HFO (upper trace). The last sweeps show an example of an oscillatory epoch that was not accompanied by spiking. Bandpass filtered 500 Hz–10 kHz and 140–250 Hz signals, respectively.

126

A.A. Ponomarenko et al. / Behavioural Brain Research 141 (2003) 123–129

Fig. 2. HFO and neuronal firing in the BL. (A) Cross-correlogram between the field signal and neuronal firing (n = 8) triggered by a peak of HFO. Insets: higher resolution cross-correlogram showing phase relationships of unit firing and HFO peaks (upper) and HFO waveform histogram. Bin width 1 ms. (B) Averaged waveforms (n = 162) of HFO in 140–250 Hz band and an associated sharp potential in 1–20 Hz band triggered by HFO peak. (C) Representative example of firing pattern of a BL unit (instantaneous frequency). (D) Corresponding bandpass filtered 500 Hz–10 kHz (upper) and 140–250 Hz (lower) signals, respectively, showing occurrence of single spikes during short HFO and sometimes longer HFO as well. Auto-correlogram of the unit (bin width 50 ms) with insets of higher resolution (bin width 1 ms) and waveform.

Fig. 3. State dependence of EPN neuronal firing (EPN UNIT), occurrence of HFO in the EPN (EPN HFO) and ripple oscillations in the temporal CA1 (HIP RIP). The number of oscillatory epochs is expressed as events per 30 s. Hypnogram (HYPNO): 1, waking; 2, SWS; 3, PS.

A.A. Ponomarenko et al. / Behavioural Brain Research 141 (2003) 123–129

127

Fig. 4. Hippocampal correlates of HFO in the EPN/BL. Examples of cross-correlograms between individual BL unit (A), EPN unit (B) firing and temporal and dorsal CA1 ripple, respectively. Correlation was triggered by hippocampal ripple’s peak. (C) Non-synchronous occurrence of hippocampal ripples and HFO in BL. (D) Auto-correlogram of the unit with insets of higher resolution and waveform. Bin width in all correlograms 50 ms with exception of inset (1 ms).

during PS or waking (not significant). On the other hand, frequency of BL/EPN HFO occurrence varied markedly in individual recordings and across behavioural states. Consistently higher HFO numbers (up to 20 min−1 in the EPN and 23 min−1 in the BL) occurred during SWS than during PS or W (down to 1 min−1 ; Fig. 3). Hippocampal ripples occurred up to 20 min−1 during SWS, around 2 min−1 during waking with tonic EMG activity and were virtually absent during PS and waking with phasic EMG. More recordings with longer epochs of PS and immobile W are needed to fully elucidate the state dependence of HFO in the EPN and BL. Short- and long-scale correlation of amygdala and EPN activity with hippocampal ripples was also assessed both for HFO and single units. Averaged cross-correlograms were triggered by the positive peak of either temporal or dorsal hippocampal ripples for units (n = 9 for EPN, and n = 8 for BL) or field HFO from the same electrodes did not reveal a significant time-relation of any signal. Individual cross-correlograms for single units were more heterogeneous: some BL and EPN units tended to discharge well before (up to 2 s) the peak of hippocampal ripples but others fired at almost zero lag (Fig. 4A and B). Two neurons from BL displayed a right-shifted lag (several hundred milliseconds) from a hippocampal ripple’s positive peak. Hippocampal ripple triggered cross-correlograms for EPN and BL HFO peaks showed no correlation.

4. Discussion We provide here the first evidence of field high-frequency (200 Hz) oscillations phase-locked with neuronal discharge in BL and the adjacent EPN. These HFO are independent of the fast synchronous population patterns in the CA3–CA1 axis and most likely also from those in the parahippocampal cortices downstream from CA1 [9,14,17]. Thus, temporal CA1/subicular afferents to BL [37] are probably not responsible for synchronising BL neurons at the scale of tens of milliseconds. Another major finding is the similarity of neuronal firing and associated EEG events of EPN and BL that provide further support for the functional resemblance of these nuclei. The firing behaviour of a majority of the recorded units in the amygdala corresponds to the class of bursting projection neurons of BL [29,31,32,38]. Results on state dependent modulation of BL neuron firing complement previously published data in the behaving rats [4]. In the cat, activity of these BL cells is believed to initiate the generation of sharp potentials in the entorhinal cortex and does not correlate with population events beyond the dentate gyrus [28]. We found no reliable link between HFO in BL or EPN with CA1 ripples in the rat. The burst firing superimposed on field HFO is in keeping with in vitro findings of populational discharge in the EPN

128

A.A. Ponomarenko et al. / Behavioural Brain Research 141 (2003) 123–129

[36]. Bursting of principal cells in the EPN depends on low and high-voltage activated Ca2+ currents with fast inactivation kinetics [20]. In light of the extensive collateral connections between EPN principal neurons [3] this bursting can entrain a large number of cells in population synchrony. The presence of sharp field potentials superimposed on HFO suggests a mechanism analogous to the known interplay of sharp wave-associated dendritic depolarisation and oscillatory somatic inhibition in CA1 [39]. The basolateral amygdaloid circuitry (L) displays GABAB -receptor dependent attenuation of inter-neuronal IPSC in the principal cells under the conditions of paired pulse stimulation of putative cortical and thalamic afferents [35]. Furthermore, specific subsets of BL principal neurons generate repetitive or single spike bursts in response to prolonged depolarisation [30]. Short-scale synchronous firing of BL projection neurons receiving different inputs may organise precise timing of multi-modal information outflow to other amygdaloid nuclei and extra amygdaloid regions. It is assumed that population discharge of EPN neurons depolarises target pyramidal cells throughout the olfactory cortices [3]. The EPN and basolateral amygdaloid complex might constitute part of the subplate layer of the temporal and piriform cortices; these regions are among the (peri)amygdaloid nuclei that give rise to an excitatory output [34]. They use the common principles of efferent temporal coding via focal or global population synchrony though they convey signals of different modalities.

Acknowledgements The authors wish to thank Jian-Sheng Lin for methodological advice and Serguey Chepurnov for insightful discussions. We are grateful to Krister Eriksson and Claudia Wittrock for expert assistance in histology. Supported by Deutsche Forschungsgemeinschaft (DFG) Graduiertenkolleg G85/00.

References [1] Aggleton JP. The amygdala. Oxford: Oxford University Press, 2000. [2] Akmaev IG, Kalimullina LB. Amygdaloid brain complex: functional morphology and neuroendocrinology. Moscow: Nauka; 1993. p. 48–63. [3] Behan M, Haberly LB. Intrinsic and efferent connections of the endopiriform nucleus in rat. J Comp Neurol 1999;408:532–48. [4] Bordi F, LeDoux J, Clugnet MC, Pavlides C. Single-unit activity in the lateral nucleus of the amygdala and overlying areas of the striatum in freely behaving rats: rates, discharge patterns, and responses to acoustic stimuli. Behav Neurosci 1993;107:757–69. [5] Bragin A, Engel Jr J, Wilson CL, Fried I, Buzsaki G. High-frequency oscillations in human brain. Hippocampus 1999;9:137–42. [6] Bragin A, Engel Jr J, Wilson CL, Fried I, Mathern GW. Hippocampal and entorhinal cortex high-frequency oscillations (100–500 Hz) in human epileptic brain and in kainic acid–treated rats with chronic seizures. Epilepsia 1999;40:127–37.

[7] Bragin A, Mody I, Wilson CL, Engel Jr J. Local generation of fast ripples in epileptic brain. J Neurosci 2002;22:2012–21. [8] Bragin A, Wilson CL, Engel Jr J. Chronic epileptogenesis requires development of a network of pathologically interconnected neuron clusters: a hypothesis. Epilepsia 2000;41(Suppl 6):S144–52. [9] Buzsaki G. Hippocampal sharp waves: their origin and significance. Brain Res 1986;398:242–52. [10] Buzsaki G. The hippocampo-neocortical dialogue. Cereb Cortex 1996;6:81–92. [11] Buzsaki G. Memory consolidation during sleep: a neurophysiological perspective. J Sleep Res 1998;7(Suppl 1):17–23. [12] Buzsaki G, Haas HL, Anderson EG. Long-term potentiation induced by physiologically relevant stimulus patterns. Brain Res 1987;435:331–3. [13] Buzsaki G, Horvath Z, Urioste R, Hetke J, Wise K. High-frequency network oscillation in the hippocampus. Science 1992;256:1025–7. [14] Buzsaki G, Horvath Z, Urioste R, Hetke J, Wise K. High-frequency network oscillation in the hippocampus. Science 1992;256:1025–7. [15] Buzsaki G, Leung LW, Vanderwolf CH. Cellular bases of hippocampal EEG in the behaving rat. Brain Res 1983;287:139–71. [16] Chepurnov SA, Chepurnova NE. Amygdaloid complex of the brain. Moscow: Moscow University Press; 1981. p. 197–225. [17] Chrobak JJ, Buzsaki G. High-frequency oscillations in the output networks of the hippocampal-entorhinal axis of the freely behaving rat. J Neurosci 1996;16:3056–66. [18] Collins DR, Lang EJ, Pare D. Spontaneous activity of the perirhinal cortex in behaving cats. Neuroscience 1999;89:1025–39. [19] Collins DR, Pelletier JG, Pare D. Slow and fast (gamma) neuronal oscillations in the perirhinal cortex and lateral amygdala. J Neurophysiol 2001;85:1661–72. [20] de Curtis M, Pare D, Llinas RR. The electrophysiology of the olfactory–hippocampal circuit in the isolated and perfused adult mammalian brain in vitro. Hippocampus 1991;1:341–54. [21] Demir R, Haberly LB, Jackson MB. Voltage imaging of epileptiform activity in slices from rat piriform cortex: onset and propagation. J Neurophysiol 1998;80:2727–42. [22] Everitt BJ, Morris KA, O’Brien A, Robbins TW. The basolateral amygdala-ventral striatal system and conditioned place preference: further evidence of limbic–striatal interactions underlying reward-related processes. Neuroscience 1991;42:1–18. [23] Hoffman WH, Haberly LB. Bursting-induced epileptiform EPSPs in slices of piriform cortex are generated by deep cells. J Neurosci 1991;11:2021–31. [24] Hoffman WH, Haberly LB. Kindling-induced epileptiform potentials in piriform cortex slices originate in the underlying endopiriform nucleus. J Neurophysiol 1996;76:1430–8. [25] Kudrimoti HS, Barnes CA, McNaughton BL. Reactivation of hippocampal cell assemblies: effects of behavioural state, experience, and EEG dynamics. J Neurosci 1999;19:4090–101. [26] Lin JS, Sakai K, Jouvet M. Hypothalamo-preoptic histaminergic projections in sleep–wake control in the cat. Eur J Neurosci 1994;6:618– 25. [27] Pare D, Collins DR, Pelletier JG. Amygdala oscillations and the consolidation of emotional memories. Trends Cogn Sci 2002;6:306– 14. [28] Pare D, Dong J, Gaudreau H. Amygdalo–entorhinal relations and their reflection in the hippocampal formation: generation of sharp sleep potentials. J Neurosci 1995;15:2482–503. [29] Pare D, Gaudreau H. Projection cells and interneurons of the lateral and basolateral amygdala: distinct firing patterns and differential relation to theta and delta rhythms in conscious cats. J Neurosci 1996;16:3334–50. [30] Pare D, Pape HC, Dong J. Bursting and oscillating neurons of the cat basolateral amygdaloid complex in vivo: electrophysiological properties and morphological features. J Neurophysiol 1995;74:1179– 91.

A.A. Ponomarenko et al. / Behavioural Brain Research 141 (2003) 123–129 [31] Rainnie DG, Asprodini EK, Shinnick-Gallagher P. Intracellular recordings from morphologically identified neurons of the basolateral amygdala. J Neurophysiol 1993;69:1350–62. [32] Rosenkranz JA, Grace AA. Dopamine attenuates prefrontal cortical suppression of sensory inputs to the basolateral amygdala of rats. J Neurosci 2001;21:4090–103. [33] Siapas AG, Wilson MA. Coordinated interactions between hippocampal ripples and cortical spindles during slow-wave sleep. Neuron 1998;21:1123–8. [34] Swanson LW, Petrovich GD. What is the amygdala? Trends Neurosci 1998;21:323–31. [35] Szinyei C, Heinbockel T, Montagne J, Pape HC. Putative cortical and thalamic inputs elicit convergent excitation in a population of GABAergic interneurons of the lateral amygdala. J Neurosci 2000;20:8909–15.

129

[36] Tseng GF, Haberly LB. Deep neurons in piriform cortex. Part I. Morphology and synaptically evoked responses including a unique high-amplitude paired shock facilitation. J Neurophysiol 1989;62:369–85. [37] van Groen T, Wyss JM. Extrinsic projections from area CA1 of the rat hippocampus: olfactory, cortical, subcortical, and bilateral hippocampal formation projections. J Comp Neurol 1990;302:515– 28. [38] Washburn MS, Moises HC. Electrophysiological and morphological properties of rat basolateral amygdaloid neurons in vitro. J Neurosci 1992;12:4066–79. [39] Ylinen A, Bragin A, Nadasdy Z, Jando G, Szabo I, Sik A, et al. Sharp wave-associated high-frequency oscillation (200 Hz) in the intact hippocampus: network and intracellular mechanisms. J Neurosci 1995;15:30–46.