Theme : Dreams
Differential Responses of Brain Stem Neurons during Spontaneous and
Stimulation-Induced Desynchronization of the Cortical EEG in Freely Moving
Cats
Birendra N. Mallick1, Stephen Thankachan1 and Fakhrul Islam2
1School of Life Sciences, Jawaharlal Nehru University, New Delhi 110 067,
India and 2Department of Medical Elementology and Toxicology, Jamia Hamdard,
Hamdard University, New Delhi 110 062, India
Abstract
The EEG is desynchronized during wakefulness and REM sleep. There are
awake and REM sleep-related neurons in the brain stem. This study was
carried out to investigate if the same neuron in the brain stem reticular
formation may be responsible for EEG desynchronization during wakefulness
and REM sleep. Single neuronal activity was recorded in chronically prepared
freely moving normal cats and their activities were correlated with EEG
desynchronization during spontaneous wakefulness, REM sleep, and during
wakefulness induced by stimulation of the brain stem reticular formation.
A majority of the neurons showed an increased firing associated with spontaneous
EEG desynchronization during wakefulness and REM sleep, however, about
55% of them showed a similar behavior during stimulation-induced desynchronization.
It was found that responses of a majority of the neurons during stimulation-induced
desynchronization were similar to that of their firing rate during EEG
desynchronization associated with spontaneous wakefulness irrespective
of their behavior during REM sleep; the REM-ON neurons were not affected
by the stimulation-induced desynchronization. A majority of the neurons
which showed an increased firing during spontaneous and stimulation-induced
EEG desynchronization received an excitatory input from the brain stem
reticular formation. The results of this study suggest that although some
neurons may be common, there is a strong possibility that the same neuron
in the brain stem reticular formation is not involved in EEG desynchronization
during wakefulness and REM sleep.
Current Claim: Separate groups of neurons are possibly involved in EEG
desynchronization during wakefulness and REM sleep.
Activate the ShortNotes by clicking on this link. Your notes will be stored
in this area and automatically retrieved upon your next visit.
The role of brain stem in EEG desynchronization is well established. The
EEG is synchronized in cerveau isolé preparation (brain stem transected
at mid-collicular level) while the synchronization alternated with desynchronization
in the encéphale isolé (transection was made at the junction
of brain stem and the spinal cord C1/C2 level) preparation (Moruzzi, 1972).
This was further confirmed by lesion and stimulation studies. The rostral
portion of the brain stem was reported to be responsible for wakefulness
and EEG desynchronization (Batini et al., 1959; Kasamatsu, 1970). Subsequently,
it has been shown in anesthetized as well as in freely moving animals
that although neurons related to wakefulness and EEG desynchronization
are scattered in the brain stem reticular formation, they are predominantly
located in the rostral brain stem (Huttenlocher, 1961; Manohar et al.,
1972). The brain stem area which induces EEG desynchronization has been
found to affect neurons present in other sleep-waking areas in the brain
(Mallick et al., 1986) and vice versa (Fenske et al., 1975; Szymusiak
and McGinty, 1989). Recently it has been suggested that brain stem neurons
induce EEG desynchronization by acting through the thalamic neurons (Paré
et al., 1988; Steriade et al., 1990, 1996).
In addition to wakefulness, the EEG is desynchronized during rapid eye
movement (REM) sleep as well, and the neurons related to REM sleep are
also located in the brain stem reticular formation (Jones, 1991). The
aminergic neurons are REM-OFF type (Chu and Bloom, 1974; McGinty and Harper,
1976; Jacobs, 1986), while REM-ON neurons may be cholinergic or non-cholinergic
(El Mansari et al., 1989; Kayama et al., 1992; Sakai and Koyama, 1996).
Although the mechanism of generation of REM sleep is not yet clear, neurons
in the dorsolateral pontine region ventral to the locus coeruleus are
attributed to muscle atonia (Sakai, 1985; Lai and Siegel, 1988) and those
in the parabrachial area to PGO waves (Datta, 1995) during REM sleep.
However, it is not known with certainty which neurons are responsible
for EEG desynchronization during REM sleep. Since wakefulness and REM
sleep are two distinct behaviors and EEG desynchronization is one of the
signs (apparently) common to them, at least two possibilities could exist.
One, that the same neurons in the brain stem may be involved in EEG desychronization
which switch from one neuronal circuitry responsible for behavioral wakefulness
to circuitry responsible for REM sleep or vice versa. Two, that there
may be two separate groups of neurons forming separate circuitry responsible
for EEG desynchronization during wakefulness and REM sleep behaviors.
It was hypothesized that if there were two separate groups of neurons
for EEG desynchronization during wakefulness and REM sleep, then neurons
related to the former only should respond similarly to brain stem reticular
formation stimulation-induced wakefulness and EEG desynchronization. Therefore,
the behavior of the same awake and REM sleep related neurons from the
reticular formation in freely moving cats was correlated with EEG desynchronization
during spontaneous wakefulness as well as REM sleep and also during brain
stem stimulation-induced EEG desynchronization. Nevertheless, considering
the complexity and redundancy of the nervous system, the other complex
possibility, that although there may be a predominance of separate groups
of neurons, it cannot be ruled out that a small portion of the neurons
may be involved in EEG desynchronization during both the behavioral states.
Activate the ShortNotes by clicking on this link. Your notes will be stored
in this area and automatically retrieved upon your next visit.
Experiments were performed on 9 adult cats (5 male and 4 female), weighing
between 2.5 and 3.4 kg. Under surgical anesthesia (sodium pentobarbitone,
35 mg/kg, i.v.) cats were prepared for chronic sleep-wakefulness (S-W)
recording (Mallick et al., 1989). One pair of small stainless steel screw
electrodes was fixed in the skull above the sensorimotor cortex to record
bipolar EEG and another pair over the orbital bone of the eye to record
electrooculogram (EOG). Two flexible insulated (except a small portion
at the end) wires were attached bilaterally to the dorsal neck muscles
to record bipolar electromyogram (EMG). Stainless steel insulated (except
at the tip) tripolar electrodes were bilaterally implanted at stereotaxic
coordinates (Berman, 1968) of A6, L10, H2 to record pontogeniculooccipital
(PGO) waves. One stainless steel tripolar stimulating electrode was stereotaxically
implanted in the midbrain reticular formation (MRF) at the coordinates
of A2, L3, H-1 (Mallick et al., 1986).
The microdrives consisted of two thin-walled stainless steel 19G outer
barrels (20 mm length) parallely placed 2 mm apart. Two 22G thin-walled
stainless steel inner barrels (40 mm length) were passed through the outer
barrel and arranged on a screw-nut assembly for smooth and easy vertical
movement (McGinty and Harper, 1976). A pair of microdrives was stereotaxically
introduced into the brain bilaterally at an angle of 30o through drill
holes made in the skull so as to reach coordinates of AP0-P3, L1.5-3.5.
The microdrives were introduced so that the tips of the inner cannulae
reached 6.0 mm above the target H-1. The outer cannulae of the microdrives
were cemented on the skull with dental acrylic so that the inner cannulae
could move freely dorsoventrally. Four microelectrode bundles each consisting
of nine 32 µm Formvar-coated stainless steel wires (California Fine
Wire Inc., U.S.A.) soldered to a 9-pin connector were lowered through
each of the inner guide cannulae so as to extend 5.0 mm beyond the inner
cannulae of the microdrives. Thus, the tips of the microelectrodes were
positioned 1.0 mm above the target. The microelectrodes were glued to
the inner cannulae so that when the inner cannulae were advanced manually
by means of the attached screw, the microelectrodes also advanced through
the brain. The microdrives could be moved dorsoventrally by a total length
of about 4 mm. Thus, with this arrangement in this study single neuronal
activities were recorded within the coordinates of AP0-3, L1.5-3.5, H-1
to -4.
About 2 weeks were allowed for the cats to recover from surgical trauma.
During this recovery period the cats were habituated to the recording
chamber. After recovery, electrophysiological signals viz., EEG, EOG,
EMG and PGO were recorded simultaneously on a Grass polygraph from unrestrained,
freely moving cats. Once a single unit activity (S:N > 3:1) was encountered,
it was amplified (A-M Systems Inc., U.S.A.) and fed to an adjustable window
discriminator (Fredrick Haer Brunnswick, ME). The raw signal from the
window discriminator was monitored on an oscilloscope (TDS 420 Tektronix
Inc., U.S.A.). The window-triggered output from the window discriminator
was fed into a pulse counter and the output was recorded simultaneously
on one of the channels of the polygraph along with other electrophysiological
signals in other channels. The signals were also recorded and processed
in a computer connected to a CED 1401plus intelligent interface. At least
two spontaneous S-W-REM sleep cycles were recorded and then the MRF was
stimulated (100 Hz, 300 µs, 200-300 µA) for 8-12 sec at the
background of quiet wakefulness and the effects on sleep-wakefulness as
well as on the unit activity were recorded simultaneously as long as the
effects lasted. Thereafter, in order to study if the neurons received
any direct projection from the MRF stimulation site, the response of the
neurons to 1 Hz (300 µs, 500-600 µA) stimulation of MRF was
also studied by overlapping 10-15 stimulus-bound responses on the oscilloscope.
Such 1 Hz stimulation of the MRF did not produce wakefulness, unlike the
effects on high frequency stimulation. Thus, the effect on the neurons
by 1 Hz stimulation was independent of changes in EEG desynchronization
and associated behavior.
After completion of the recording sessions the animals were deeply anesthetized
with sodium pentobarbitone (45 mg/kg) and a small electrolytic lesion
(50-100 µA for 10 sec) was made at the tip of the successful unit
recording site. The cat was then euthanized by intracardial perfusion
with 0.5 L of saline followed by 1 L of 4% paraformaldehyde and 2% potassium
ferrocyanide and preserved for histological identification of the lesioned
site and reconstruction of the recording sites. The brain stem was trimmed
and preserved in 30% sucrose before making 40 µm coronal sections
which were processed for staining with cresyl violet or hematoxylin and
eosin. Some of the sections were immunostained for tyrosine hydoxylase
(TH) by using standard methods (Reiner and Vincent, 1987). The former
two stained sections showed recording sites and the latter showed if the
REM-OFF neurons were recorded from the TH-positive sites. For immunostaining,
the sections were processed using primary antibody (rabbit anti-TH polyclonal,
1:1000, Eugene Tech) and incubated in secondary antibody (biotinylated
goat anti-rabbit immunoglobulin, 1:400, Vector Lab). This was followed
by treatment with avidin-biotin complex (ABC complex, 1:500, Vector Lab)
and the brown reaction product was visualized by incubating in DAB (Vector
Lab) and H202. The sections were mounted on subbed slides, dehydrated
and prepared for microscopic examination. Reconstruction of sections showing
unit recording sites and one histological TH-stained section showing microwire
track passing through TH-positive neurons are shown in Figure 1.
The sleep-wakefulness-REM sleep stages were classified as reported earlier
(Ursin and Sterman, 1981; Mallick et al., 1989). In brief, active awake
(AW), was characterized by EEG desynchronization accompanied by eye movement,
increased EMG tone and/or phasic motor activities; quiet awake (QW) by
EEG desynchronization with occasional (<20% time) synchronization and
spindling, reduced eye movements and reduced EMG tone (absence of phasic
motor activity). Sleep stage was defined by EEG synchronization and spindles,
no eye movements, reduced EMG tone and absence of phasic motor activity.
REM sleep had characteristic signs of EEG desynchronization, rapid eye
movement accompanied by muscle atonia and presence of PGO waves.
Mean neuronal discharge rate during each of the sleep-wakefulness-REM
sleep states was calculated by taking the average of five 60-second epochs
of respective states. Mean neuronal activity per second during different
states was compared statistically by applying analysis of variance (ANOVA)
coupled with Newman-Keul's test taking QW as the basal level to establish
the relationship of spontaneous neuronal firing rate to different states.
To study the effect of MRF stimulation-induced EEG desynchronization on
the unit activity, the firing rate of neuronal activity during post-stimulation
EEG desynchronization was also compared with that of the rate of firing
of the unit activity during pre-stimulation QW period. The firing rate
of the neurons for at least three 30 sec episodes each during pre-stimulation
QW and post-stimulation desynchronization were calculated and statistically
tested by applying ANOVA and Newman-Keul's test. As a rule, firing rate
during the period of MRF stimulation was not taken because occasionally
the stimulation artifact falling within the window levels could not be
avoided.
Activate the ShortNotes by clicking on this link. Your notes will be stored
in this area and automatically retrieved upon your next visit.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Eighty-four well isolated single units were recorded between caudal
midbrain to rostral pontine tegmental region of the brain stem from chronically
prepared freely moving normally behaving cats. The neuronal activities
(spikes per second) during each of the S-W-REM sleep states were compared
statistically against QW to establish the relationship of individual neuronal
firing rates to different states. It was found that 61 increased, 3 decreased
and 20 remained unaffected in relation to AW only; while 70 neurons increased,
7 decreased and 7 remained unaffected in relation to REM sleep only (Table
1). However, the neurons could be classified into 7 groups (Table 2) considering
their firing rate both during AW and REM sleep when the EEG during both
the behavioral states was desynchronized.
The mean firing rates for each of the neurons, in all the 7 groups so
classified, during different states are shown in Figure 2. About 62% (52
out of the 84) of the recorded neurons showed significantly increased
firing rates (Fig. 2A and B) during AW (p<0.01, F1,102=44.81) and during
REM sleep (p<0.01, F1,102=40.67). The mean firing rates of this group
of neurons during AW and REM sleep were 21.1 ± 2.1 spikes/sec and
21.6 ± 2.3 spikes/sec, respectively, while during QW the firing
rate was 5.6 ± 1.0 spikes/sec. The second largest group of neurons
(15 out of 84) showed a significantly (p<0.01, F1,28=8.1) increased
firing rate during REM sleep without being significantly affected during
AW and sleep states (Figure 2E). The mean firing rates of this group of
neurons were 18.4 ± 4.4 spikes/sec during REM sleep, 5.0 ±
1.8 spikes/sec during QW and 5.9 ± 2.0 spikes/sec during AW. There
were 4 neurons whose firing rate significantly (p<0.01, F1,6=1.33)
increased during AW but did not change significantly during REM sleep
(Figure 2D). The mean firing rate of this group of neurons during AW,
QW and REM sleep was 9.5 ± 3.5 spikes/sec, 4.8 ± 2.1 spikes/sec
and 5.6 ± 2.2 spikes/sec, respectively.
There were 7 neurons whose firing rates were significantly lower during
REM sleep (Figure 2C). Out of these, for 5 neurons the firing rate was
higher (p<0.01, F1,8=4) during AW and significantly lower (p<0.01,
F1,8=1.9) during REM sleep compared to QW and their mean firing rates
during AW, QW and REM sleep were 14.4 ± 4.4 spikes/sec, 4.7 ±
1.3 spikes/sec and 1.76 ± 1.02 spikes/sec, respectively. The other
two neurons decreased their firing rate exclusively during REM sleep (Figure
3) where the mean rate of firing was 1.6 ± 0.6 spikes/sec, 1.85
± 0.7 spikes/sec and 0.6 ± 0.02 spikes/sec during AW, QW
and REM sleep, respectively. These were possibly norepinephrinergic REM-OFF
neurons, since their spike shape and duration followed the criteria reported
earlier for such neurons (Koyama et al., 1994) and they were recorded
from TH-positive sites (when confirmed histologically). For 4 neurons
the firing rate during REM sleep was significantly higher (p<0.01,
F1,4=1.76) while during AW the rate was significantly lower (p<0.01,
F1,4=14.5) (Fig. 2F). The mean firing rates of this group of neurons during
REM sleep, QW and AW were 3.4 ± 1.9 spikes/sec, 0.9 ± 0.15
spikes/sec and 0.2 ± 0.1 spikes/sec, respectively. These neurons
may possibly be cholinergic or non-cholinergic REM-ON neurons because
the shape and duration of spikes from these neurons followed those of
the putative cholinergic neurons as reported by others (El Mansari et
al., 1989; Kayama et al., 1992; Sakai and Koyama, 1996). There were three
neurons whose firing rates remained unaffected during AW and REM sleep
as compared to QW when the mean firing rates were 8.7 ± 1.8, 10.2
± 8.3 and 9.8 ± 8.0 spikes/sec, respectively. Of these three
neurons, one showed increased firing (p<0.01, F1,2=30.5), the second
decreased firing (p<0.01, F1,6=1.49) and the third remained unaffected
during sleep compared to QW, when the firing rates were 0.24 ±
0.16, 3.58 ± 0.58 and 25.75 ± 2.5 spikes/sec, respectively
(Figure 2G).
Behavior of neurons (classified above) to high frequency (100 Hz) MRF
stimulation-induced desynchronization of the EEG
100 Hz stimulation of the MRF induced long-lasting EEG desynchronization
and behavioral arousal during and after cessation of the stimulation.
The stimulation strength was adjusted such that although the animal was
awake there was no significant active movement other than raising head
and being attentive. There was no sign of discomfort, fright or any other
unusual behavioral response.
Effect on group of neurons classified on the basis of their relationship
to spontaneous AW only: As shown in Table 3, out of the 61 neurons which
showed increased firing during spontaneous wakefulness, 35 (57%) showed
increased activity during MRF stimulation-induced wakefulness while the
other 26 remained unaffected. Two out of the 3 neurons which were inhibited
during spontaneous AW were inhibited on MRF stimulation-induced desynchronization
while the other remained unaffected and 19 out of 20 neurons which remained
unaffected during spontaneous AW remained unaffected on MRF stimulation-induced
desynchronization and wakefulness.
Effect on group of neurons classified on the basis of their relationship
to spontaneous REM sleep only: Out of the 70 neurons whose firing rate
increased during spontaneous REM sleep, 38 (54%) remained unaffected,
29 (41%) increased while 3 decreased their firing during stimulation-induced
desynchronization. Five of the 7 neurons which were inhibited during spontaneous
REM sleep showed increased firing during stimulation-induced desynchronization,
while 6 of the 7 neurons which remained unaffected during spontaneous
REM sleep showed a similar behavior during stimulation-induced desynchronization
(Table 4).
Effects of MRF stimulation-induced EEG desynchronization on groups of
neurons classified based on their firing rate during both spontaneous
AW and REM sleep: The effect of MRF stimulation-induced EEG desynchronization
was studied on 84 neurons and each of their firing rates have been shown
in Figure 2. There were 61 neurons which showed increased firing rates
during spontaneous EEG desynchronization irrespective of their behavior
during spontaneous REM sleep (when EEG was also desynchronized). A total
of 37 of these 61 neurons showed a significantly increased (p<0.01)
firing rate during EEG desychronization induced by MRF stimulation (Table
5). Twenty-nine out of 52 neurons (55%), which showed increased firing
rates during spontaneous AW and REM sleep, showed increased firing rates
during MRF stimulation-induced desynchronization (Figure 4). All the neurons
showing increased firing during AW and reduced/ceased firing during REM
sleep (the REM-OFF neurons) showed a significant increase in their firing
rate during MRF stimulation-induced wakefulness and EEG desynchronization
(Figure 5). Out of the 4 neurons which showed increased firing during
spontaneous AW without being affected during REM sleep, 3 increased while
one decreased firing during MRF stimulation-induced desynchronization
(Figure 2D). However, out of the three which increased, only one was statistically
significant while the other two remained insignificant, possibly because
of high standard deviation.
There were 17 neurons whose firing rates were affected during spontaneous
REM sleep but remained unaffected during spontaneous AW (EEG was desynchronized
during both the behaviors). Fifteen of them showed increased, while two
showed decreased, firing rates during REM sleep. All except one of these
neurons remained unaffected during MRF stimulation-induced EEG desynchronization
(Figure 6). Out of three classical REM-ON neurons, where the firing rate
decreased during spontaneous AW and increased during spontaneous REM sleep,
none showed any firing during or following stimulation-induced desychronization
(Figure 7). Two neurons, one REM-ON and the other REM-OFF, were recorded
simultaneously. The former increased while the latter decreased firing
during spontaneous REM sleep. The firing rate of the REM-OFF neuron increased
while the REM-ON neuron was not affected by MRF stimulation-induced desynchronization
(Figure 8). The interspike interval histograms (generated by SPIKE2, CED)
of these two neurons are shown in Figure 9. All the three neurons which
remained unaffected during spontaneous AW and REM sleep remained unaffected
during MRF stimulation-induced desynchronization of the EEG (Figure 2G).
Effect of 1 Hz stimulation of MRF on the brain stem neurons
Stimulation of MRF with 1 Hz did not affect EEG synchronization or desynchronization.
Such stimulation showed that MRF may influence neuronal activity irrespective
of induced alteration in the EEG. The effect was studied on 37 neurons
(Table 6). Out of the 32 neurons which increased firing during spontaneous
AW, irrespective of their behavior during REM sleep, 17 were excited,
14 remained unaffected while only one was inhibited by 1 Hz MRF stimulation.
A majority (17 out of 23) of the neurons which increased firing by high
frequency stimulation of the MRF were excited by 1 Hz stimulation (Table
7) with a latency of 5-10 msec (Figure 10A). Two of the 4 neurons which
decreased firing during spontaneous AW, increased during spontaneous REM
sleep and were not affected by 100 Hz stimulation, decreased (Figure 10B)
while the other two remained unaffected by 1 Hz MRF stimulation.
Classification of neurons into tonic and phasic types and their responses
to MRF stimulation
Depending on spontaneous firing pattern, the recorded neurons could be
grouped into tonic (N=63) and phasic (N=21) types as reported earlier
(El Mansari et al., 1989 and Steriade et al., 1990) (Figure 11). A majority
(N=48) of the tonic-type neurons showed a tonic increase in discharge
rate during spontaneous AW (p<0.01, F1,94=26.97) and during REM sleep
(p<0.01, F1,94=23.43) (these neurons would fall within the group shown
in Fig 2A and B), a characteristic of tonic-type I and 12 neurons showed
tonic increase (p<0.01, F1,22=1.45) of firing from QW to sleep and
highest during REM sleep (neurons within the group shown in Fig 2E and
F), a characteristic of tonic-type II. Out of the 21 phasic-type neurons,
13 showed burst firing during AW and REM sleep, while 8 showed burst firing
primarily during REM sleep.
On high frequency MRF stimulation-induced wakefulness and EEG desynchronization,
all the 24 affected neurons (out of 48 tonic-type I) and 9 out of 13 phasic
neurons (burst firing during AW and REM sleep) showed an increased firing.
On the other hand, all except 3 phasic-type neurons (10 out of 12 tonic-type
II and 7 out of 8 phasic type), remained unaffected by high frequency
MRF stimulation-induced wakefulness and EEG desynchronization; the 3 affected
neurons showed decreased firing rate. On 1 Hz MRF stimulation, when EEG
was not desynchronized, 14 out of 25 tonic-type I neurons and 4 out of
7 phasic-type neurons (burst firing during AW and REM sleep) were excited.
The effect of 1 Hz stimulation was studied on 2 tonic-type II and 3 phasic-type
neurons that showed firing only during REM sleep. The former were inhibited
while the latter showed no change.
Activate the ShortNotes by clicking on this link. Your notes will be stored
in this area and automatically retrieved upon your next visit.
In this study EEG, EOG, EMG, PGO and single neuronal activity were recorded
from brain stem reticular formation in freely moving normally behaving
chronically prepared cats. Based on their spontaneous firing rate during
wakefulness and REM sleep, when the EEG remained desynchronized, the neurons
were classified into seven groups. Although in earlier reports responses
of neurons have been characterized and correlated with spontaneous changes
in sleep, wakefulness and REM sleep, or specific correlates of REM sleep,
their behaviors during spontaneous as well as induced changes in those
behaviors or correlates were not studied. In order to study neuronal specificity
to a correlate of REM sleep, it is desirable to relate neuronal activity
with spontaneous as well as induced changes with the correlate under study.
It assumes significance for correlating neuronal activity with EEG desynchronization
which is present both during wakefulness as well as REM sleep. The advantage
of correlating unit activity with spontaneous and induced changes to a
sign is that the latter is induced in a relatively controlled condition.
Since this study attempted to identify neuronal specificity to EEG desynchronization
during wakefulness and REM sleep, the firing rate of the neurons during
quiet wakefulness was taken as baseline because there was absence of phasic
muscle activity during that phase although the EEG remained desynchronized.
The specificity of the neurons to spontaneous EEG desynchronization during
wakefulness was tested by studying and correlating their behavior during
MRF stimulation-induced desynchronization of the EEG. Since the aim of
this study was to investigate if the same neuron in the brain stem reticular
formation may be responsible for desynchronization of the EEG during both
wakefulness and REM sleep, the neuronal activities were recorded from
structures lying between caudal midbrain and rostral pons because awake
(Steriade and McCarley, 1990) and REM sleep-related (Siegel et al., 1984;
Sakai, 1985; Mallick et al., 1989, 1994) neurons are reported to be present
in these regions. The absolute firing rate of the neurons during different
states varied significantly and the neuronal spike shapes were mono-,
bi- or tri-phasic. The sites from where REM-OFF neurons were recorded
showed TH immunopositivity (Figure 1). The REM-ON and REM-OFF neurons,
as classified in this study, showed spike shape and duration as reported
earlier (El Mansari et al., 1989; Kayama et al., 1992; Koyama et al.,
1994; Sakai and Koyama, 1996).
It was found that a majority of the recorded neurons in this area altered
their firing rates whenever the EEG was desynchronized, irrespective of
whether the animal was in spontaneous awake or REM sleep state. Almost
95% of the neurons showed an increased firing rate correlated to EEG desynchronization
during both wakefulness and REM sleep and 61% of the neurons were common
and showed an increased firing rate during both the phases. Since this
area of the brain stem is attributed to EEG desynchronization (Moruzzi,
1972), it is reasonable that a majority of the neurons in this area increased
firing associated with EEG desynchronization. However, their differential
relationship, if any, to the two distinct behaviors wakefulness and REM
sleep, when EEG remains desynchronized, was not known. The observation
that a majority of the neurons increased firing during EEG desynchronization
irrespective of wakefulness or REM sleep behaviors tempts one to suggest
that the same neuron in the brain stem could be responsible for EEG desynchronization
during both the behavioral states. However, it may be an oversimplification
and will be discussed later. Earlier neurons recorded from brain stem
reticular formation were classified as tonic-type I, tonic-type II and
phasic-type (El Mansari et al., 1989; Steriade et al., 1990). In this
study, the number of neurons in those categories (when classified using
their criteria) were comparable (Figure 11) with minor differences which
could be due to differences in recording area and the techniques used.
It may be noted that there were more tonic-type neurons than phasic-type.
The former could be related to EEG desynchronization (a tonic phenomenon)
and the latter to movements, etc. (a phasic phenomenon).
There were some neurons which showed opposite response in firing rate
during REM sleep and wakefulness. These neurons may play a significant
role in modulating behavior associated to REM sleep and wakefulness, respectively.
For example, a varying degree of muscle tone is present during wakefulness,
while it is absent in REM sleep. Since the objective of this study was
to investigate if the same neuron may be responsible for EEG desynchronization
during both the behaviors (REM sleep and wakefulness), it was necessary
to study the behavior of the same neuron throughout during spontaneous
and stimulation-induced EEG desynchronization. Hence, after classifying
them on the basis of their responses to spontaneous behavior, the neurons
were studied if they maintained similar behavior during MRF stimulation-induced
EEG desynchronization. In addition, since the desynchronization and effect
on the units was induced by MRF stimulation, the responsiveness of the
neurons to inputs from the MRF without changes in the EEG was also studied.
This was achieved by overlapping 1 Hz stimulation bound responses of the
units on the oscilloscope. This showed if the recorded neurons received
input from the area of stimulation which functionally induced EEG desynchronization.
Relationship of neuronal activity to spontaneous and induced EEG desynchronization
and its physiological significance
Since EEG desynchronization is a common correlate during wakefulness
and REM sleep, it is possible that at least some of the neurons could
be common for induction of EEG desychronization during both the behaviors.
However, it was found that all except one of the neurons which remained
unaffected during spontaneous EEG desynchronization accompanied by wakefulness,
irrespective of their behavior during REM sleep, remained unaffected during
high frequency MRF stimulation-induced EEG desynchronization (Table 5).
Similarly, all except one of the REM-specific neurons (REM-ON and REM-OFF)
were affected in a manner similar to that of their behavior during spontaneous
wakefulness when EEG desynchronization was induced by high frequency stimulation
of the MRF. The REM-ON neurons were inhibited while the REM-OFF neurons
were either excited or not affected by high frequency MRF stimulation-induced
desynchronization of the EEG. Thus, these groups of neurons during stimulation-induced
desynchronization of the EEG behaved similarly to their behavior during
spontaneous EEG desynchronization accompanied by wakefulness and not as
during spontaneous REM sleep. Therefore, it is reasonable that the neurons
not related to EEG desynchronization during wakefulness were unaffected;
while the so called REM-OFF neurons excited (Figure 10A) and the REM-ON
neurons remained inhibited (Figure 10B) on 1 Hz stimulation of the MRF.
However, although a majority of the AW-related neurons were affected by
1 Hz MRF stimulation, some were not affected even though they were affected
during high frequency stimulation-induced EEG desynchronization. The former
could be a cause for EEG desynchronization and the latter an effect of
wakefulness and desynchronization of the EEG. The latter neurons may also
be related to muscle activity and movement associated to wakefulness.
The influence of MRF on the REM sleep-related neurons suggest that there
is possibly an interaction between the neurons related to REM sleep and
the wakefulness-generating area in the brain stem. The excitation and
inhibition observed in this study may be a direct excitatory or a disinhibitory
effect or an inhibitory or a disfacilitatory effect, respectively, which
cannot be confirmed from this study. However, either or both the possibilities
may be supported by the presence of excitatory and inhibitory neurons
and inputs to this area (Sakai, 1985).
One of the limitations of this study is that the neuronal responses to
specific muscle movement have been compromised. We are aware of the fact
that most of the brain stem neuronal activity could be correlated with
some muscle movement (Siegel, 1979; Siegel and Tomaszewki, 1983) during
wakefulness-associated desynchronization. In that case it is very difficult
to understand and interpret the cause and effect relationship if a neuronal
activity is related to movement or wakefulness or EEG desynchronization.
Since these phenomena are associated and are correlates of wakefulness
they are difficult to isolate behaviorally in in vivo studies. This is
an inherent deficiency of behavioral studies, and may be taken care of
if the response of the same neuron can be studied after induction of REM
sleep in addition to the studies being reported here (preliminary data
support this view [Thankachan et al., 1997]). Nevertheless, this limitation
has been minimized in this study by recording the neuronal activity during
comparable physical conditions, as far as possible, during spontaneous
and MRF stimulation-induced wakefulness.
Although a good proportion of brain stem neurons showed increased firing
rates when the EEG remained desynchronized during spontaneous wakefulness
as well as REM sleep, only a little over half of them showed a similar
behavior during MRF stimulation-induced desynchronization (Table 5). This
is probably because it is unlikely that all the neurons would be involved
in desynchronization of the EEG during those behaviors. The neurons which
were related to wakefulness and showed a similar behavior during MRF stimulation-induced
desynchronization may be responsible for EEG desynchronization, while
those which did not behave similarly to MRF stimulation-induced desynchronization
of the EEG may be involved in other physiological changes during wakefulness.
The neurons which increased firing during spontaneous as well as MRF stimulation-induced
EEG desynchronization are likely to be involved in desynchronization of
the EEG during wakefulness. The neurons which increased firing during
REM sleep only, but were not affected by MRF stimulation, may be involved
in EEG desynchronization during REM sleep. This may be supported by the
fact that 1 Hz stimulation of MRF significantly increased the firing rate
of 50% of the tonic-type and 80% of the phasic-type of neurons. The important
fact to be noted is that most of the REM sleep-related neurons (either
tonic or phasic category) remained unaffected by such stimulation (Figure
11). None of the studied neurons was antidromically activated by MRF stimulation.
This suggests that the recorded neurons neither projected to or passed
through MRF, the area of stimulation. Nevertheless, since a majority of
wake-active neurons was activated by 1 Hz stimulation of MRF at a latency
of 5-10 msec, it is likely that the MRF has a multisynaptic and/or unmyelinated
projection to the recorded neurons. Interestingly, among the neurons affected
by 1 Hz MRF stimulation, a majority of the neurons showing increased firing
during AW as well as during MRF stimulation-induced desynchronization,
irrespective of their behavior during REM sleep, received excitatory inputs
(Table 6). These MRF inputs to wake active neurons might provide a bias
to the reticular neurons for inducing EEG desynchronization during wakefulness.
Thus, it is reasonable that REM-OFF neurons were not inhibited (excited),
while the REM-ON neurons were either inhibited or remained unaffected
by 1 Hz stimulation of the MRF (Table 6). This view may be supported by
previous reports (El Mansari et al., 1989; Steriade et al., 1990) that
some of the tonic as well as phasic neurons project to the thalamus and
might be responsible for EEG desynchronization, possibly during wakefulness.
Although thalamic projection of the recorded neurons has not been investigated
in this study, it may be said that since the proportion of tonic and phasic
neurons are comparable to earlier studies, the tonic neurons in this study
are likely to project through the thalamus for modulation of the EEG as
hypothesized in those studies.
It may also be argued that those neurons which increased firing during
REM sleep but were not affected by MRF stimulation-induced desynchronization
of the EEG may be related to muscle atonia during REM sleep. Some of these
neurons may be responsible for inducing atonia and others for inducing
desynchronization of the EEG during REM sleep. The latter view may be
supported by earlier reports that depending on the anatomical location
of the neurons, some may be involved in atonia and others in EEG desynchronization
during REM sleep (Sakai, 1980, 1985). These neurons may be cholinergic
as well as non-cholinergic (Sakai and Koyama, 1996). Also, it has been
shown that neurons having non-NMDA receptors are related to atonia while
those with NMDA receptors are not (Lai and Siegel, 1991). Therefore, it
may be suggested that the latter neurons may be involved in EEG desynchronization
during REM sleep. The neurons which increased firing during wakefulness
as well as REM sleep may be involved in common behavior during wakefulness
and REM sleep, e.g., eye movement.
Since it has been found in this study that most of the neurons during
MRF stimulation-induced desynchronization behaved in a manner similar
to that of spontaneous desynchronization associated to wakefulness irrespective
of their behavior during REM sleep, it suggests that separate groups of
neurons are probably involved in EEG desynchronization during wakefulness
and REM sleep. However, the observation that neurons showing increased
firing during REM sleep and AW and also receiving inputs from MRF may
suggest that at least some neurons may be common for EEG desynchronization
during REM sleep and wakefulness. This is understandable in view of the
complexity of brain function and its redundancy hypothesis. Thus, although
the possibility of separate groups of neurons in the brain stem for EEG
desynchronization during wakefulness and REM sleep may be suggested, it
needs to be confirmed. The effect of MRF stimulation has been investigated
in this study; however, the role of induced REM sleep is under study as
well. Studying the same neuronal behavior during spontaneous and induced
wakefulness and REM sleep is likely to provide a clearer picture.
1. Batini C, Moruzzi G, Palestine M, Rossi GF, Zanchetti A. Effects of
complete pontine transections on the sleep-wakefulness rhythm the midpontine
pretrigeminal preparation. Arch Ital Biol 1959; 97: 1-12.
2. Berman AI. The Brain Stem of the Cat: A Cytoarchitectonic Atlas with
Stereotaxic Co-ordinates. Madison: University of Wisconsin Press, 1968.
3. Chu NS, Bloom FE. Activity patterns of catecholamine pontine neurons
in the dorso-lateral tegmentum of unrestrained cat. J Neurobiol 1974;
5: 527-44.
4. Datta S. Neuronal activity in the peribrachial area: Relationship
to behavioral state control. Neuroscience Biobehav Rev 1995; 19: 67-84.
5. El Mansari M, Sakai K, Jouvet M. Unitary characteristic of presumptive
cholinergic tegmental neurons during the sleep-waking cycle in the freely
moving cats. Exp Brain Res 1989; 76: 519-29.
6. Fenske M, Ellendorff F, Wuttke W. Response of medial preoptic neurons
to electrical stimulation of the mediobasal hypothalamus, amygdala and
mesencephalon in normal, serotonin or catecholamine deprived female rats.
Exp Brain Res 1975; 22: 495-507.
7. Huttenlocher PR. Evoked and spontaneous activity in single units of
medial brain stem during natural sleep and waking. J Neurophysiol 1961;
24: 451-68.
8. Jacobs BL. Single unit activity of locus coeruleus neurons in behaving
animals. Prog Neurobiol 1986; 27: 183-94.
9. Jones BE. Paradoxical sleep and its chemical/structural substrates
in the brain. Neuroscience 1991; 40: 637-56.
10. Kasamatsu T. Maintained and evoked unit activity in the mesencephalic
reticular formation of the freely behaving cat. Exp Neurol 1970; 28: 450-79.
11. Kayama Y, Ohta H, Jodo E. Firing of 'possibly' cholinergic neurons
in the rat laterodorsal tegmental nucleus during sleep and wakefulness.
Brain Res 1992; 569: 210-20.
12. Koyama Y, Jodo E, Kayama Y. Sensory responsiveness of "Broad-spike"
neurons in the laterodorsal nucleus, locus coeruleus and dorsal raphe
of awake rats: Implication for cholinergic and monoaminergic neuron specific
responses. Neuroscience 1994; 63: 1021-31.
13. Lai YY, Siegel JM. Medullary regions mediating atonia. J Neuroscience
1988; 8: 4790-6.
14. Lai YY, Siegel JM. Pontomedullary glutamate receptors mediating locomotion
and muscle tone suppression. J Neuroscience 1991; 11: 2931-3937.
15. Mallick BN, Kumar VM, Chhina GS, Singh B. Comparison of rostro-caudal
brain stem influence on preoptic neurons and cortical EEG. Brain Res Bull
1986; 16: 121-5.
16. Mallick BN, Siegel JM, Fahringer H. Changes in pontine unit activity
after REM sleep deprivation. Brain Res 1989; 515: 94-8.
17. Mallick BN, Thakkar M, Shiromani PJ, McCarley RW. REM sleep-selective
unit activity in the cholinergic pedunculopontine tegmental nucleus of
the freely moving cats. Sleep Res 1994; 23: 24.
18. Manohar S, Noda H, Adey RW. Behavior of mesencephalic reticular neurons
in sleep and wakefulness. Exp Neurol 1972; 34: 140-57.
19. McGinty DJ, Harper RM. Dorsal raphe neurons: depression of firing
during sleep in cats. Brain Res 1976; 10: 569-75.
20. Moruzzi G. The sleep waking cycle. Ergeb Physiol 1972; 64: 1-164.
21. Paré D, Smith Y, Parent A, Steriade M. Projections of brain
stem core cholinergic and non-cholinergic neurons of cat to intralaminar
and reticular thalamic nuclei. Neuroscience 1988; 25: 69-86.
22. Reiner PB, Vincent SR. Topographic relations of cholinergic and noradrenergic
neurons in the feline pontomesencephalic tegmentum: an immunohistochemical
study. Brain Res Bull 1987; 19: 705-14.
23. Sakai K. Some anatomical and physiological properties of ponto-mesencephalic
tegmental neurons with special reference to the PGO waves and postural
atonia during paradoxical sleep in the cat. In: Hobson JA, Brazier MAB,
eds. The Reticular Formation Revisited. New York: Raven Press, 1980, pp.
427-47.
24. Sakai K. Anatomical and physiological basis of paradoxical sleep.
In: McGinty DJ, Morrison A, Drucker-Colin R, Parmeggiani PL, eds. Brain
Mechanism of Sleep. New York: Raven Press, 1985, pp. 111-37.
25. Sakai K, Koyama Y. Are there cholinergic and non-cholinergic paradoxical
sleep-on neurons in the pons? NeuroReport 1996; 7: 2449-53.
26. Siegel JM. Behavioral functions of the reticular formation. Brain
Res Rev 1979; 1: 69-105.
27. Siegel JM, Nienhuis R, Tomaszewski KS. REM sleep signs rostral to
chronic transections at the pontomedullary junction. Neurosci Lett 1984;
45: 241-6.
28. Siegel JM, Tomaszewki KS. Behavioral organization of reticular formation:
studies in the unrestrained cat. I. Cells related to axial, limb, eye
and other movements. J Neurophysiol 1983; 50: 696-716.
29. Steriade M. Arousal: revisiting the reticular activating system.
Science 1996; 272: 225-6.
30. Steriade M, Datta S, Pare D, Oakson G, Curro-Dossi R. Neuronal activities
in brain-stem cholinergic nuclei related to tonic activation processes
in thalamocortical systems. J Neurosci 1990; 10: 2541-59.
31. Steriade M, McCarley RW. Brain Stem Control of Wakefulness and Sleep.
New York: Plenum Press, 1990.
32. Szymusiak R, McGinty D. Effects of basal forebrain stimulation on
the waking discharge of neurons in the midbrain reticular formation of
cats. Brain Res 1989; 498: 355-9.
33. Thankachan S, Islam F, Mallick BN. Behavior of brain stem neurons
during spontaneous and induced EEG desynchronization during wakefulness
and rapid eye movement sleep in free moving cats. Sleep Res 1997; 26:
54.
34. Ursin R, Sterman MB. A Manual for Standardized Scoring of Sleep and
Waking States in the Adult Cat. Brain Information Service/Brain Research
Institute, Los Angeles: University of California, 1981
Financial support to Birendra N. Mallick from the Department of Biotechnology
(India) is duly acknowledged. We are thankful to Prof. Jerry Siegel at
the University of California, Los Angeles (USA) for his constructive criticisms
and valuable comments while preparing this manuscript. We are also thankful
to Prof. Siegel for helping us with the TH-antibody.
Original address of this text :
http://www.sro.org/bin/article.dll?Paper&1379&0&0
Please copy this address to the address bar of your
internet browser and press the "enter" key.
(We prefer not to put actual links because
often page locations change and then our log files get cluttered with
error messages
- if the address does not work try to find it from the homepage of the
website in question).
|
