Relationship between rhythmic discharge patterns of neurons in the central nucleus of the amygdala, blood pressure fluctuations and cortical activity

Journal of the

Autonomic Nervous System

ELSEVIER

Journal of the Autonomic Nervous System 57 (1996) 158 162

Relationship between rhythmic discharge patterns of neurons in the central nucleus of the amygdala, blood pressure fluctuations and cortical activity G. Schulz a.., M. Knuepfer b, M. Lambertz ~, P. Langhorst c, G. Stock a a Research Laboratories, Sehering AG, SIOI, 11 Floor. M~llerstr 170-178, D-13342 Berlin. Germany b Department of Pharmaeology, St. Louis University St. Louis, MO. USA c Institute of Physiology. The Free University of Berlin. Berlin, Germany

Abstract

In the discharge sequences recorded from single neurons in the central nucleus of the amygdala of chronically instrumented awake cats, rhythmical patterns with period durations of 5-12 s were observed. At the same time blood pressure and the degree of synchronisation of the EEG showed similar period fluctuations with positive correlation to neuronal activity. It is proposed that the central amygdaloid nucleus uses rhythmic patterns to coordinate somatomotor and vegetative systems. Keywords: Amygdala; Central nucleus; Blood pressure; Cortical activity; Correlation analysis

1. Introduction

The amygdaloid complex, a key element of the limbic system, is critically involved in the elaboration of complex patterns of psychomotor behaviour, in learning and memory formation [4,8,16]. Highly preprocessed information from all sensory modalities via cortical and subcortical structures but also from the dorsal vagal nucleus and the nucleus tractus solitarii (NTS) in the brain stem converge onto these neurons [4]. With many of these structures the amygdaloid complex is reciprocally connected. The neutones of the central subnucleus project to the hypothalamus which is essential for the expression of emotional behaviour and its cardiovascular components and further down to the medulla [1,6,10,12,14]. It also projects to the globus pallidus and the frontal premotor area possibly used to exert an influence upon forebrain motor mechanisms [12,19]. One of the most prominent features observed during stimulation of the central amygdaloid nucleus has been an arousal response, accompanied by cardiovascular adaptations [5,8,20,21]. In previous publications we have shown that in awake cats single neurons in the central nucleus respond to visual or auditory stimuli with an

* Corresponding author. 0165-1838/96/$15.00 © 1996 Elsevier Science B.V. All rights reserved SSDI 0 1 6 5 - 1 8 3 8 ( 9 5 ) 0 0 0 7 5 - 5

increase in discharge frequency clearly preceding changes in blood pressure, heart rate and also motor activity [18]. In the present investigation the relationship between the spontaneous discharge pattern and blood pressure on the one hand and cortical activity on the other hand were studied.

2. Materials and methods

In male cats under pentobarbitone anesthesia pairs of glass insulated stainless steel electrodes with an interpolar distance of 1.5 mm were implanted stereotactically into the left premotor cortex (with the coordinates frontal 17, lateral 5) from the atlas of Reinoso-Suarez [15] for chronic EEG recordings. A socket was chronically implanted (center: frontal 12; lateral 8.5) for attachment of a microdrive to record single unit activity in the right amygdala. For recording of muscle activity (EMG) stainless steel wires were sewed into neck muscles. A PE catheter was chronically implanted into the descending aorta and connected to a Statham transducer for continuous blood pressure recordings. After two weeks of recovery from surgery experiments were run twice a week. During recording sessions between 10 a.m. and 4 p.m. animals could move freely in a plexiglass chamber. Single unit activity was recorded with a tungsten electrode (tip diameter 7 - 1 2 /zm)

159

G. Schulz et al./ Journal of the Autonomic Nervous System 57 (1996) 158-162

3. Results

I iiiilili~H~fI]1~i| Ill~lJli~! Ji!iiP!iiIll|if! IJlli~!~lilllj I iidllii~

Sl ""'

ou-

'

los

,

Fig. l. Rhythmical modulation of discharge frequency of a single neuron

recorded in the central amygdala during wakefulness with period durations of 7 to 10 s. Cardiac frequency is modulated in the respiratory rhythm with slower variations superimposed. BP, blood pressure; HBI, heart beat interval; SI, standardized spikes; IF, integrated frequency.

lowered into the amygdala by a microdrive. Action potentials were continuously controlled for stability of amplitude and waveform with a delay circuit and a storage oscilloscope. Electrode positions were read from stereotactic data and verified histologically at the end of the experiments. All data were stored on magnetic tape for off-line analysis as described in detail elsewhere (cf., Ref. 33 in Ref. 11). Temporal relationships between neuronal events, blood pressure and electrocorticogram were analysed by means of computing auto-and cross-covariance histograms (ACH, CCH) and respective power spectra (PS) with analog signals appropriately filtered.

(l•rPr)

AUTOCOVARIANCE

The results presented were obtained in 4 cats. Spontaneous discharges of 142 neurones in the central nucleus of the amygdala were recorded during the experiments. The discharge rates of most neurones were below 1 spike/s. Spike trains of 20 neurones observed throughout at least one full cycle of sleep and wakefulness without obvious trends of variation in discharge frequency were chosen for detailed computer analysis. Discharge rates of these 20 neurones varied between 1.9 and 22 spikes/s. In slow wave sleep discharge rates (mean 8.2 spikes/s) were generally lower than in wakefulness (mean 10 spikes/s) and phases of REM sleep (mean 13.2 spikes/s). Besides tonic discharge patterns 17 out of 20 neurones showed rhythmically modulated discharge patterns of variable amplitude and period duration. (Fig. 1). These recordings were further analysed by computing ACHs and PS. Two examples of the results are given in Fig. 2. For 13 out of 17 neurones rhythmic modulation was observed in all states of sleep and wakefulness. Period durations determined by the main peak in the power spectra ranged from 5 to 12 s (mean 9 s). Parallel to rhythmic discharge patterns of the neurons fluctuations of blood pressure and heart rate were often observed. The temporal relations between neuronal discharge modulations and blood pres-

SNN(V)

AUTO SPECTRUM

0.7

°l " ( ~ N ('r)

o!2;: AUTOCOVARIANCE

SNN[v)

AUTO

SPECTRUM

0.4a

024

1~2 V'S

Fig. 2. Autocovariancehistograms (ACH) and power spectra (PS) of spontaneous discharge sequences of two different neurons recorded in the central amygdala during wakefulness. Maximal time lag in autocovariances:40 s; Bin width: A-r= 80 ms (in the time domain), 3v = 0.012 Hz (in the frequency range); Ordinate scaled in relative figures. Both ACHs show periodicities with durations of 8 to 9 s. Frequency analysis by PS shows maxima at approx. 0.11 Hz and 0.12 Hz.

G. Schulz et ul./Journal qf the Autonomic Nervous System 57 (1996) 158-162

160 "AN,,I

~AB "

AUTOCOVARIANCE

0.13

OTOCO

A",A"CE

005

CROSSCOVARIANCE

~~~'~o

s N ~l 0.16-

o~

AUTO SPECTRUM

AUTO SPECTRUM

09-

o 0048

2.4v-

"J2'.,,.s °o~

C

0.048

CROSS SPECTRUM

075-

~

"/'~2~.,,.s

Fig. 3. Temporal relation between rhythmic discharge pattern of a neuron from the central amygdala and blood pressure fluctuations in quiet wakefulness. Maximal time lag 20 s; Bin width A-r = 40 ms (time domain), Av = 0.024 (frequency range). ACHs of neuron (AN) and blood pressure (AB) show periodicities in a range of 8 s. The cross-covariance histogram (CCH) has periodic maxima right to the ordinate, i.e., blood pressure increase starts 2 s after an increase in neuronal discharge rate. Time lag to first maximum: 4 s. Both PS and the Cross Power Spectrum have a common maximum at 0.12 Hz. The PS of blood pressure has an additional maximum at 0.21 Hz, which is the respiratory frequency range.

sure fluctuations were investigated computing cross-covariances between spike trains and pulse waves (Fig. 3). In states of quiet wakefulness positive correlations were found in 12 out of 16 neurones and in 13 out of 13 in REM sleep. In these cases the CCHs had their maxima right to the ordinate, which indicates that an increase in discharge frequency is followed by blood pressure increase. The onset of blood pressure fluctuations followed that of neuronal discharge variations with time lags of 4 - 6 s, In quiet wakefulness either spontaneously or in response to acoustic or visual stimuli an increase in activity of the same neurones was followed by desynchronisation of the electrocorticogram (ECoG) of the sensorimotor cortex (Fig. 4). The temporal relationship between these events was further analysed by CCH. The result is shown in Fig. 5. The ACH of the discharge sequence of the

neuron indicates a modulation with a period duration of about 8 s. The ACH of the ECoG has a similar modulation and in addition contains a periodicity in a lower frequency range. The CCH for neuronal activity > rectified ECoG shows a synchronisation between neuronal activity and ECoG. In 10 neurones this relationship was seen in quiet wakefulness.

4. Discussion All neurons were recorded in the central nucleus of the amygdala which is reciprocally connected with cortical and subcortical areas as well as the hypothalamus and the lower brain stem [1,6,10]. In correspondence with other reports, spontaneous discharge rates were low and state

EC~

I "

NEURON i

2s

.: " ~

i ::

ii

' ,

4

;~

J!r

Fig. 4. Desynchronisation of the electrocorticogram (ECoG) during a spontaneous discharge burst of a single neuron in the central amygdala lasting for about 10 s.

G. Schulz et aL /Journal of the Autonomic Nervous System 57 (1996) 158-162 0.14-

q~NN(T)

~E('r)

AUTOCOVARIANCE NEURON

0.1.

CROSSCOVARIANCE NEURON > ECoG

011 ~E ECT)

161

AUTOCOVARIANCE

~

ECoG ( rectified)

I~1E(T) 014"I

CROSSCOVARIANCE I NEURON > J/ ECoG( rectified )

Fig. 5. Temporal relation between rhythmic discharge pattern of a single neuron in the central amygdala and the degree of synchronisation of the ECoG. ECoG was rectified and low pass filtered, cut-off frequency 1 Hz. Maximal time lag 20 s; Bin width A~-= 40 ms. Both ACHs indicate periodic variations of 8 s. The ACH of the ECoG contains also a periodicity in a lower frequency range. The CCH for neuron > rectified ECoG shows that both signals share the same periodicities. The deep trough left to the ordinate indicates that a high degree of synchronisation goes parallel with a low discharge frequency of the neuron. For the computation of the CCH (bottom left) the ECoG was unrectified and band pass filtered (1-15 Hz). It shows a maximum 160 ms fight to the ordinate.

dependent being higher in REM sleep and wakefulness than in slow wave sleep [7]. As described in a previous publication single neurons in the central amygdala respond to visual and acoustic stimuli with an increase of discharge frequency which precedes autonomic and somatomotor reactions [18]. The primary finding of the present study is that in phases of rhythmic discharge behaviour an ongoing relationship between neuronal activity and blood pressure can be demonstrated. To our knowledge this is the first time that rhythmic modulations of the discharge patterns of amygdaloid neurons with period durations between 5 and 12 s are described. Though respiration was not measured in these experiments it is highly unlikely that this modulation stems from an influence of the respiratory oscillator since respiratory cycle length in unanesthetized cats did not exceed 4 s [19]. Zhiang et al. described respiratory modulation of amygdaloid neurons with period durations around 2.5 s [23]. Rhythmic discharge patterns with period durations between 8 and 12 seconds were described in detail for postganglionic sympathetic nerve activity, for reticular neurons of the lower brain stem and in the midbrain of cats (cf., Ref. 11 and references therein). This suggests that the anatomically interconnected structures of amygdala, midbrain and lower brain stem share the property of rhythmic discharge behaviour which can be used to coordinate different functionally related systems. The underlying coordinating principles for different oscillators

were most elegantly elaborated by Koepchen, taking the phenomenon of blood pressure waves as an example [9]. In this context the effect of pressoreceptors as cause of the rhythmic discharge patterns of amygdaloid neurons has to be discussed especially against the background of the known anatomical connections between the NTS, the first relay station of baroreceptor afferents and the amygdala [6]. With pulse wave triggered histograms of amygdala unit activity Frysinger et al. showed an influence on the timing of discharges [2]. However, this effect was independent of blood pressure values. Furthermore, in all CCHs the maximum was located right to the ordinate which indicates that neuronal activity precedes blood pressure increase which makes it unlikely that the correlation is caused by a feedback via pressoreceptors. This is supported by an investigation showing a reduced effect of pressoreceptor activity on heart rate during stimulation of the amygdala [17]. The results presented give no indication for a constant tonic drive of the amygdala on basic sympathetic tone and thereby blood pressure. However, the amygdala could be a candidate for the supramedullary structures being necessary to maintain basic blood pressure levels [19]. This is supported and broadened in perspective by results from Galeno et al. demonstrating that destruction of the amygdala in developing spontaneously hypertensive rats attenuates the development of high blood pressure [3].

162

G. &'hulz et al./Journal of the Autonomic Nervous @stem 57 (1996) 158-162

Another important result is that, besides having a positive correlation with blood pressure fluctuations, the same neurons also show a positive correlation with cortical activity recorded from the premotor area. The demonstration of this functional connection is well in line with the anatomical connections of the amygdala to the premotor cortex and the results from stimulation experiments leading to an arousal reaction [5,22]. Since activity increase in the amygdala is followed by a desynchronisation of EEG in the premotor area this correlation is considered to be the expression of a timing effect to elicit somatomotor patterns. The time course of slow rhythmic changes in the amplitude of ECoG waves runs parallel to the rhythmic oscillations in the neuronal activities. Similar results were described by Oakson et al. for midbrain reticular neurons and anterior neocortical EEG waves [13]. The results presented add to the hypothesis that rhythmic patterns are used to coordinate different functional systems for a common task in response to stimulus combinations dependent on the complex situation. In this context the central amygdala receiving preprocessed information from all sensory modalities can be viewed as a key structure in the limbic system to integrate cardiovascular and somatomotor components of behaviour.

References [1] Amaral, D.G., Price, J.L., Pitkanen, A. and Carmichael. S.T.. Anatomical organization of the primate amygdaloid complex. In J.P. Aggleton (Ed.), The Amygdala. Wiley-Liss, New York, NY, 1992, pp. 1-66. [2] Frysinger, R.C., Zhang J. and Harper, R.M., Cardiovascular and respiratory relationships with neuronal discharge in the central nucleus of the amygdala during sleep-waking states, Sleep, I1 (1988) 317-332. [3] Galeno, T.M., Van Hoeen, G.W., Maixner, W., Johnson, A.K. and Brody, M.J., Contribution of the amygdala to the development of spontaneous hypertension, Brain Res. 246 (1982) 1-6. [4] Gloor, P., Temporal lobe epilepsy: its possible contribution to the understanding of the functional significance of the amygdala and of its interaction with neocortical-temporal mechanisms, Adv. Behav. Biol., 2 (1972) 423-457. [5] Heinemann, H., Stock, G. and SchMer, H., Temporal correlation of responses in blood pressure and motor relation under electrical stimulation of limbic structures in unanaesthetized, unrestrained cats, Pfliigers Arch.', 343 (1973) 27-40.

[6] Hopkins. D.A. and Holstege, G., Amygdaloid projections to the mesencephalon, pons, and medulla oblongata in the cat, Exp. Brain Res., 32, (1978) 529-547. [7] Jacobs, B.L. and McGinty, D.J., Amygdala activity during sleep and waking, Exp. Neurol., 33 (1971) 1-15. [8] Kaada, B.R., Stimulation and regional ablation of the amygdaloid complex with reference to functional respresentation, Adv. Behav. Biol.. 2 (1972) 205-281. [9] Koepchen, H.P., Die Blutdruckrhythmik. Steinkopff, Darmstadt, 1962. [1(/] Krettek, J.E. and Price, J.L., Amygdaloid projections to subcortical structures within the basal forebrain and brain stem in the rat and cat, J. Comp. Neurol., 178 (1978) 225-253. [11] Langhorst, P., Schulz, G. and Lambertz, M., Oscillating neuronal network of the "common brain stem system". In K. Miyakawa, H.P. Koepchen and C. Polosa (Eds.), Mechanisms of Blood Pressure Waves. Springer, Berlin, 1984, pp. 257-276. [121 Llamas, A, Avendano, C. and Reinoso Suarez, F., Amygdaloid projections to prefrontal and motor cortex, Science, 195, (1977) 794-796. [13] Oakson, G. and Steriade, M., Slow rhythmic oscillations of EEG slow-wave amplitudes and their relations to midbrain reticular discharge, Brain Res., 269 (1983)386-390. [14] Peiss. C.N., Concepts of cardiovascular regulation: past, present, future. In W.C. Randall (Ed.) Nervous Control of the Heart. Williams and Wilkins, Baltimore, MD, 1965, pp. 154-197. [15] Reinoso-Suarez, F., Topographischer Hirnatlas der Katze fiir experimental physiologische Untersuchungen. E. Merck, Darmstadt, 1961. [16] Sarter, M. and Markowitsch, H., Involvement of the amygdala in learning and memory: a critical review, with emphasis on anatomical relations, Behav. Neurosci., 99 (1985) 342-380. [17] Schl6r, K.H., Stumpf, H. and Stock, G., Baroreceptor reflex during arousal induced by electrical stimulation of the amygdala or by natural stimuli, J. Auton. Nerv. Syst., 10 (1984) 157-165. [18] Schulz, G., Lambertz, M., Stock, G. and Langhorst, P., Neuronal activity in the amygdala related to somatomotor and vegetative components of behavior in cats, J. Auton. Nerv. Syst. 17, (1986) 639-648 (Suppl.). [ 19] Sieck, G.L. and Harper, R.M., Pneumotactic area neuronal discharge during sleep and waking in the cat, Exp. Neurol., 67 (1980) 79-102. [20] Shinonaga, Y., Takada, M. and Mizuno, N., Direct projections from the central amygdaloid nucleus to the globus pallidus and substantia nigra in the cat, Neuroscience, 51 (1992) 691-703. [21] Stock, G., Schl6r, K.H., Heidt, H. and Buss, J., Psychomotor behaviour and cardiovascular patterns during stimulation of the amygdala, Pfli]gers Arch., 376 (1978) 177-184. [22] Stock, G., Rupprecht, K., Stumpf, H. and Schl6r, K.H., Cardiovascular changes during arousal elicited by stimulation of amygdala, hypothalamus and locus coeruleus, J. Auton. Nerv. Syst., 3 (1981) 503-510. [23] Zhang, J., Harper, R.M. and Frysinger, R.F., Respiratory modulation of neuronal discharge in the central nucleus of the amygdala during sleep and waking states, Exp. Neurol., 91 (1986) 193-207.