Synergistic actions of noradrenergic b- and a1-receptors 495

Neuroscience Vol. 99, No. 3, pp. 495–505, 2000

䉷 2000 IBRO. Published by Elsevier Science Ltd
Pergamon Printed in Great Britain. All rights reserved
PII: S0306-4522(00)00215-3 0306-4522/00 $20.00+0.00
www.elsevier.com/locate/neuroscience

SYNERGISTIC SEDATIVE EFFECTS OF NORADRENERGIC a1- AND b-RECEPTOR

BLOCKADE ON FOREBRAIN ELECTROENCEPHALOGRAPHIC AND
BEHAVIORAL INDICES
C. W. BERRIDGE* and R. A. ESPAÑA
Psychology Department, University of Wisconsin, Madison, WI 53706, USA

Abstract—The locus coeruleus–noradrenergic system exerts an activating influence on forebrain neuronal and behavioral activity
states. For example, in the anesthetized rat, unilateral locus coeruleus stimulation elicits bilateral activation of forebrain electro-
encephalographic activity. Pretreatment with a noradrenergic b-antagonist blocks this effect, suggesting that b-receptors play a
critical role in locus coeruleus-dependent activation of the forebrain. Consistent with this, stimulation of b-receptors located in
certain basal forebrain structures evokes sustained periods of alert waking in the unanesthetized rat. Similar forebrain and
behavioral activating effects are observed with a1-receptor stimulation within these basal forebrain regions.
To assess the extent to which a1- and b-receptors contribute to the maintenance of behavioral and forebrain activation, we
examined the electroencephalographic and behavioral effects of a1-, b- and combined a1/b-receptor blockade in the unanesthetized
rat. Rats were treated individually or in combination with either varying doses of the a1-antagonist, prazosin (intraperitoneally),
and/or the b-antagonist, timolol (intracerebroventricularly). Thirty minutes following treatment, animals were placed in a mildly-
arousing novel environment, which has been demonstrated previously to elicit activation of central noradrenergic systems and
sustained waking in vehicle-treated controls. Behavior and electroencephalographic activity were recorded and later scored.
Electroencephalographic activity was analysed using power spectrum analysis.
The following were observed: (i) b-receptor blockade alone does not alter behavioral or electroencephalographic indices of alert
waking; (ii) a1-receptor blockade alone increases high-voltage spindle activity in cortical electroencephalographic activity that was
associated with decreased behavioral activity; (iii) combined a1- and b-receptor blockade elicits a substantial increase in slow-wave
activity (0.33–2.0 Hz), also in association with decreased behavioral activity. All of these effects were dependent on the dose
administered and time following initiation of testing.
These results indicate that the combined actions of a1- and b-receptors exert distinct and synergistic actions on cortical neuronal
activity patterns that are essential elements of alert waking. 䉷 2000 IBRO. Published by Elsevier Science Ltd. All rights reserved.

Key words: Arousal, norepinephrine, locus coeruleus, waking, cortex, EEG.

Cortical and thalamic neurons display distinct activity
patterns across varying behavioral states. 34,40,42 These activity
patterns are reflected in electroencephalographic (EEG)
measures 41 and are associated with the ability of these
neurons to process sensory information efficiently. 30,39,41 For
example, during slow-wave sleep cortical neurons display a
burst mode of activity and cortical EEG (ECoG) is dominated
by large-amplitude, synchronized slow-waves (⬍2 Hz). In
contrast, during waking cortical neurons display a single-
spike activity mode and ECoG is characterized by low-
amplitude, mixed-frequency desynchronized activity. 35,37
The locus coeruleus (LC) is a small brainstem nucleus that
through an extensive efferent projection system provides the
major source of norepinephrine (NE) to the brain. 21,22
Substantial evidence indicates that the LC-noradrenergic
system modulates behavioral and forebrain neuronal activity
states. 6,9 For example, LC neurons display state-dependent
alterations in discharge rates across the sleep–wake
cycle. 2,3,20,25 Importantly, alterations in LC discharge rates
precede changes in behavioral and EEG indices of behavioral
*To whom correspondence should be addressed. Tel.: ⫹1-608-265-5938,
fax: ⫹1-608-262-4029.
E-mail address: berridge@facstaff.wisc.edu (C. W. Berridge).
Abbreviations: AECF, artificial extracellular fluid; EcoG, cortical EEG;
EEG, electroencephalographic; EMG, electromyographic; HVS, high-
voltage spindles; LC, locus coeruleus; MS, medial septal area; MPOA,
medial preoptic area; NE, norepinephrine; PSA, power spectrum analy-
sis; REM, rapid eye movement.
495
state, suggesting a causal relationship between LC neuronal
activity and behavioral state. 2,3,20,21,25 Supporting this hypoth-
esis, unilateral activation of LC neurons elicits a robust,
bilateral activation of forebrain EEG in the halothane-
anesthetized rat. 6 Conversely, a2-agonist-induced suppres-
sion of LC discharge activity and NE release potently
increases EEG and behavioral measures of sedation in both
anesthetized and unanesthetized rat. 7,16,18,19,46
Combined, these observations demonstrate an involvement
of central noradrenergic neurons in the maintenance of beha-
vioral and EEG indices of arousal. These behavioral state-
modulatory actions of NE appear to involve b-receptors.
For example, i.c.v. pretreatment with a b-antagonist blocks
forebrain activation observed following LC stimulation. 6
Further, b-agonist infusion into a region of the basal forebrain
encompassing the medial septal area (MS) increases EEG
indices of arousal in the halothane-anesthetized rat. 7 In the
unanesthetized rat, stimulation of b-receptors within either
MS or the medial preoptic area (MPOA) increases EEG
and/or behavioral indices of arousal/waking. 9,29
a1-Receptors also exert an activational influence on beha-
vioral and forebrain neuronal activity states. For example,
stimulation of a1-receptors within MPOA increases beha-
vioral and EEG indices of waking. 29 Similarly, in preliminary
studies, a1-agonist infusion into MS elicited EEG activation
in anesthetized rat (unpublished observations). In addition to
the behavioral state-modulatory actions of basal forebrain
noradrenergic receptors, b- and a1-receptors act directly on
496 C. W. Berridge and R. A. España
cortical, hippocampal, and thalamic neurons to modulate
activity patterns of these neurons. 15,31–33,36 Thus, both a1-
and b-receptors exert activational effects on forebrain
neuronal and behavioral activity states, via actions within
cortical, thalamic, and basal forebrain sites.
The extent to which the actions of b- and a1-receptors
contribute to the maintenance of EEG and behavioral
indices of alert waking is less clear. In the anesthetized rat,
b-antagonist treatment (both i.c.v. and bilateral infusion into
MS) decrease EEG indices of arousal. 6,8,9 However, in the
unanesthetized rat b-antagonist infusions into either MS or
MPOA do not alter EEG/behavioral indices of arousal. 12,26 In
fact, b-antagonists appear to increase cocaine-induced
increases in locomotor activity. 24 Finally, in both preliminary
and the current studies, i.c.v. b-antagonist administration
failed to alter either ECoG or behavior in unanesthetized
animals exposed to an arousal-increasing novel environment
when compared to vehicle-treated controls (see below).
In contrast to the minimal effects of b-receptor blockade on
EEG and behavioral indices of arousal, a1-receptor blockade
increases the incidence and duration of high-voltage spindles
(HVS), 14 a form of synchronized activity associated with
sleep and immobile waking. 1,14,42 However, this increase in
HVS does not resemble the a2-agonist-induced increases in
slow-wave EEG activity, typically associated with deep
sleep. 16,18
To summarize, stimulation of both b- and a1-receptors
elicit arousal-increasing actions in sleeping, unanesthetized
animals (e.ḡ. low arousal level baseline). In contrast, blockade
of either receptor subtype alone does not increase EEG and
behavioral indices of sedation comparable to that observed
with pharmacological suppression of noradrenergic neuronal
discharge activity and NE release. During waking, LC
neurons display elevated discharge rates which results in
the simultaneous activation of a1- and b-receptors. To date,
the contribution of combined a1- and b-receptor activation
to the maintenance of behavioral and forebrain neuronal
activity associated with waking has not been examined
systematically. To better understand the relative contribution
of b- and a1-receptors in the maintenance of EEG and beha-
vioral indices of arousal, the present studies examined the
effects of b- and a1-antagonist administration, separately
and in combination, on ECoG and behavior in animals
exposed to a mildly-arousing, novel environment. 5,10
EXPERIMENTAL PROCEDURES
Animals and surgery
Male Sprague–Dawley rats (280–320 g, Sasco, Oregon, WI, USA)
were housed in pairs for at least seven days prior to surgery with ad
libitum access to food and water on a 13 h/11 h light/dark cycle (lights
on 7.00 a.m.). On the day of surgery, animals were anesthetized using
halothane and then placed in a stereotaxic instrument with the incisor
bar set ⫺11.5 mm below ear bar zero. A 26-gauge guide cannula
(Plastics One, Roanoke, VA, USA) was implanted over the lateral
ventricle (⫺1.0 A, ^1.0L, ⫺1.7 V) for i.c.v. infusions. A bipolar
EEG electrode was implanted into the frontal cortex and bipolar
electromyographic (EMG) electrodes were implanted into neck muscle
(see below). The cannula, EEG, and EMG electrodes were cemented
into position, using acrylic cement (Plastics One, Roanoke, VA, USA).
A stainless-steel wire stylet was inserted and tightened into the cannula
via a plastic connector (Plastics One). Animals were allowed to recover
for seven to 10 days before testing. All efforts were made to minimize
animal suffering and to minimize the number of animals used and all
experiments were conducted in accordance with the Principles of
laboratory animal care (NIH publication no. 85-23, revised 1985).
Drugs/infusions
On the day of testing, animals were removed from their homecage,
weighed, and transferred to a Plexiglas chamber (46 × 46 × 21 cm).
The stylet was removed and a 33-gauge infusion needle containing
either artificial extracellular fluid (AECF; 147 mM NaCl, 1.3 mM
CaCl2, 0.9 mM MgCl2, 2.5 mM KCl, 5.0 mM NaH2PO4; pH 7.4) or
timolol (Sigma, St Louis, MO, USA) was inserted into the cannula.
The infusion needle was attached to a liquid swivel via PE20 tubing
filled with water. An approximately 50-nl air bubble was used to
separate the infusate from the water contained within the tubing. The
PE20 tubing and needle were housed within a stainless-steel coil
spring, the distal end of which was attached to the liquid swivel
(Instech Laboratories, Plymouth Meeting, PA, USA). The inlet of
the liquid swivel was connected to a 10-ml syringe via PE20 tubing
filled with water. The plunger of the syringe was advanced by a micro-
processor-controlled infusion pump (Harvard Apparatus, South
Natick, MA, USA). An air bubble was placed in the tubing above
the liquid swivel to allow visualization of fluid displacement during
the infusion.
Five minutes following i.c.v. needle insertion, a 2-ml infusion of
either vehicle (AECF) or varying doses of timolol (10, 50, 150 mg)
was made over a 2-min period. Five minutes later, the infusion needle
was removed and animals received an i.p. injection of either
0.5 × normal saline (0.45% NaCl) or varying doses of prazosin (250,
500, 1000 mg/kg; Sigma, St. Louis, MO, USA), dissolved in 0.5 ×
normal saline. In all cases the injection volume was 1.0 ml/kg body
weight. Prazosin was administered i.p. due to the poor water solubility
of the drug. Animals were then returned to their homecage. Thirty
minutes later, animals were transferred individually to novel Plexiglas
testing chambers (32 × 32 × 40 cm). The testing chambers were housed
within a wooden, sound-attenuated outer chamber containing a 15W
light bulb, a speaker through which white noise (80 dB) was played
and a 12-V fan run at a reduced speed to provide ventilation. The outer
chamber had two10-cm holes: one in the top panel of the chamber to
allow exit of EEG/EMG cables, and one in the front panel to permit
videotaping of behavior. Behavior and EEG/EMG were recorded for
30 min starting 5 min after introduction into the testing chamber. The
5-min delay was due to the often necessary adjustment of cables and/or
polygraph and VCR recording settings immediately following place-
ment of the animal into the testing chamber.
Electroencephalographic and electromyographic recording and
analyses
EEG and EMG recordings were performed as described previously. 9
Briefly, a bipolar surface-to-depth electrode was used to record ECoG
from frontal cortex (A ⫹ 3.0; L ^ 1.5). EMG was recorded using two
4-cm lengths of flexible, insulated wire (Cooner Wire, Chatsworth,
CA, USA) threaded into neck muscle and positioned such that an
approximately 3-mm portion of exposed wire was in direct contact
with tissue. A screw electrode was placed over the cerebellum and
served as ground. The ends of the EEG, EMG, and screw electrodes
were then inserted into a five-pin plastic connector which was cemen-
ted into position. EEG signals (0.3–100.0 Hz bandpass) and EMG
signals (1.0–30.0 Hz bandpass) were amplified, filtered, and recorded
on a polygraph and video recording tape using a modified VCR (Vetter
Instruments, Rebersburg, PA, USA). Five consecutive 5-min epochs
were analysed using power spectrum analysis (PSA). The first epoch
began 5 min after the animal was introduced into the testing chamber.
Each epoch was digitized at a sampling rate of 300 Hz and was tapered
at the ends using a cosine function and analysed using PSA. The mean
relative power (per cent of total power) was calculated for the follow-
ing frequency bands: 0.3–2.0, 2.0–4.0, 4.0–8.0, 8.0–12.0, 12.0–20.0,
20.0–30.0, 30.0–40.0 and 40.0–50.0 Hz. The incidence of HVS was
estimated from polygraph records. Determination of HVS was based
on: (i) peak amplitude greater than 2.5 × that of baseline; (ii) the
characteristic waxing/waning envelope pattern that persisted for a
minimum of 2 s.
Behavioral analyses
Behavioral analyses were as described previously. 9 Behavior was
videotaped using a black and white low-level illumination video
camera (Panasonic WV-BL2000). The output of the camera was sent
to a time and date imprinter, a video monitor and VCR. The first 30 min
of behavior was scored from videotape by a trained observer using a
 Synergistic actions of noradrenergic b- and a1-receptors 497

Table 1. Mean relative power for vehicle-treated animals (%) these studies, was based on previous studies conducted in
anesthetized animals that demonstrated a robust and sustained
Time (min) ( ⬇ 90–120 min) increase in slow-wave EEG activity follow-
Frequency (Hz) 5–10 10–15 15–20 20–25 25–30
ing timolol administration at this dose. 8 The dose of prazosin
used was based on previously published studies 14 and pilot
0.3–2.0 8 ^ 0.7 8 ^ 0.8 7 ^ 1.1 8 ^ 1.3 8 ^ 1.4 studies demonstrating EEG effects when administered in this
2.0–4.0 12 ^ 0.7 12 ^ 0.8 12 ^ 1.1 12 ^ 0.8 13 ^ 1.4 dose range in unanesthetized animals.
4.0–8.0 22 ^ 0.9 21 ^ 0.8 21 ^ 0.8 21 ^ 0.5 20 ^ 1.0 During testing, vehicle-treated rats displayed EEG and
8.0–12.0 13 ^ 0.3 13 ^ 0.3 14 ^ 0.7 13 ^ 0.9 15 ^ 1.0
12.0–20.0 15 ^ 0.4 16 ^ 0.5 17 ^ 1.0 16 ^ 0.8 16 ^ 1.0
behavioral indices of alert waking throughout the first 30
20.0–30.0 12 ^ 0.4 13 ^ 0.6 13 ^ 0.5 12 ^ 0.3 11 ^ 0.8 min of testing (Table 1). In these animals, ECoG activity
30.0–40.0 9 ^ 0.6 9 ^ 0.5 8 ^ 0.5 8 ^ 0.7 7 ^ 0.8 was characterized by high-frequency, low-amplitude activity
40.0–50.0 7 ^ 0.6 6 ^ 0.5 6 ^ 0.5 6 ^ 0.6 7 ^ 1.1 (Fig. 1). In vehicle-treated controls, HVS was observed infre-
quently during the 5–30-min testing epoch (Table 2). HVS
Mean (^ S.E.M.) ECoG relative power expressed as per cent for the speci-
fied frequency bands. Animals received i.c.v. and i.p. injections of vehi-
was never observed during the first 5 min of testing in
cle (n ˆ 7). PSA was conducted on 5-min EEG epochs collected at the vehicle-treated animals. b-Receptor blockade alone (timolol)
specified times following introduction into the testing chamber. Note that did not noticeably alter ECoG activity, compared to vehicle-
across the 30 min of testing, vehicle-treated animals displayed relatively treated controls (Fig. 1, Table 2). In contrast, a1-receptor
little slow-wave (0.3–2.0 Hz) activity. blockade alone (prazosin) elicited a robust increase in the
incidence of HVS (Fig. 1, Table 2). In all but one animal,
prazosin-induced HVS was observed within the first 5 min of
computer-based event recorder (Noldus Information, Wageningen, testing (range ˆ 0–29, mean ˆ 8 ^ 3.7). However, the magni-
The Netherlands). The following behaviors were scored: (i) asleep
(body resting on the floor, head resting on floor); (ii) quiet awake tude of the prazosin-induced HVS increased substantially
(body resting on floor, head raised off the floor); (iii) active awake over time. In the majority of cases, the robust increase in
(body off the floor with movement); (iv) inactive awake (body off HVS persisted throughout the entire recording session and
the floor, absence of movement, including sniffing and head move- was not replaced by slow-wave activity. However, in one
ments); (v) grooming; (vi) eating; (vii) drinking; (viii) quadrant entries; case HVS was observed for only approximately 5 min, at
(ix) rears. The frequency and duration of all behaviors, except quadrant
entries (frequency only) were scored for two consecutive 15-min which point large-amplitude, slow-waves predominated
epochs beginning 5 min after placement in the testing chamber. with HVS-like frequency intermixed.
Interestingly, blockade of b-receptors in the presence of
Statistical analyses a1-receptor blockade, produced qualitatively different
ECoG activity patterns than those observed with either
Drug effects were analysed using one-way (drug as factor) or two-
way mixed-design ANOVA with drug treatment as the between- treatment alone (Fig. 1). Thus, combined timolol/prazosin
subjects variable and time as the within-subjects variable. When statis- treatment elicited a substantial increase in large-amplitude
tical significance was indicated (P ⬍ 0.05), post hoc analyses were slow-wave activity in the absence of noticeable HVS (Fig.
conducted using Tukey’s honestly significant difference test. Dose- 1, Table 2). This effect was apparent within the first 5 min of
dependent effects were analysed using linear contrasts and post hoc
analyses were conducted using means comparison contrasts when
testing, reaching near continuous slow-wave activity within
statistical significance was indicated. the next 5–10 min, which persisted throughout the record-
ing session.
Histology and data selection criteria
After each experiment, rats were deeply anesthetized and received Cortical electroencephalographic effects of timolol, prazosin,
an i.c.v. infusion of 2 ml 0.5% Chicago Blue dye. Animals were then and combined timolol/prazosin: power spectrum analyses
perfused with 60 ml of 4% formaldehyde, and the brains removed and
placed in 20 ml of perfusion solution. After approximately 24–48 h, PSA was conducted on ECoG collected in the above-
brains were frozen on dry ice, and 50-mm sections were sliced and described studies to quantify the ECoG effects of a1- and/or
collected through the infusion needle track. The extent of dye diffusion b-receptor blockade (see Fig. 2). Analyses were conducted on
was noted. Sections were stained using Neutral Red dye and cover
slipped for examination of infusion needle location. Data were five 5-min epochs: (i) t ˆ 5–10 min; (ii) t ˆ 10–15 min; (iii)
included in the above analyses only when EEG recordings were elec- t ˆ 15–20 min; (iv) t ˆ 20–25 min; (v) t ˆ 25–30 min.
trically adequate, infusion needle placement was accurate, and dye was Analyses were not conducted on the first 5 min of recording
distributed throughout the ventricular system. due to the occasional need to adjust headstage connections
and/or EEG recording settings. The effects of drug treatment
RESULTS on ECoG relative power (per cent total power) are depicted in
Fig. 2. All results are presented in terms of percent vehicle-
Cortical electroencephalographic effects of timolol, prazosin,
controls (see Table 1).
and combined timolol/prazosin: qualitative observations
Consistent with the qualitative observations described
The EEG effects of b- and a1-receptor blockade was exam- above, b-receptor blockade alone did not alter ECoG relative
ined initially in animals receiving either i.p. administered a1- power at any of the examined time epochs. In contrast, PSA
antagonist (prazosin) or vehicle, and i.c.v. administered b- demonstrated significant effects of a1- and combined a1/b-
antagonist (timolol) or vehicle. ECoG was recorded in four receptor blockade on ECoG relative power. Prazosin admin-
groups of rats that received the following treatments: (i) i.c.v. istration alone increased relative power in the 4.0–8.0 and
vehicle (AEFC) ⫹ i.p. vehicle (0.5 N saline, n ˆ 7); (ii) i.c.v. 12.0–20.0 Hz frequency bands. These results are consistent
150 mg timolol ⫹ i.p. vehicle (n ˆ 7); (iii) i.c.v. vehicle ⫹ with previous observations, 1,42 indicating that HVS are
500 mg/kg i.p. prazosin (n ˆ 7); (iv) i.c.v. 150 mg timolol ⫹ comprised of two distinct frequencies of approximately 7
i.p. 500 mg/kg prazosin (n ˆ 8). The dose of timolol used in and 14 Hz (see Fig. 1B). These increases were statistically
498 C. W. Berridge and R. A. España

Fig. 1. Effects of a1-, b-, and combined a1/b-receptor blockade on ECoG activity. Thirty minutes prior to testing, animals were treated with either: (i) i.c.v.
vehicle ⫹ i.p. saline (VEH/SAL, n ˆ 7); (ii) 150 mg i.c.v. timolol ⫹ i.p. saline (TIM/SAL, n ˆ 7); (iii) i.c.v. vehicle ⫹ 500 mg/kg prazosin (VEH/PRAZ, n ˆ 7),
and; (iv) combined timolol/prazosin (TIM/PRAZ, n ˆ 8). (A) Shown are approximately 5-min ECoG traces (printed at 1 mm/s) from the second 5-min
epoch of exposure to the novel environment. Vehicle-treated controls displayed behavioral and ECoG indices of waking throughout much of the recording
session. This is reflected in sustained ECoG desynchronization (low-amplitude, high-frequency). Timolol treatment alone had no effect on ECoG activity.
Prazosin elicited substantial increases in the frequency, amplitude, and duration of HVS. In the presence of a1-receptor blockade (prazosin), b-receptor
blockade (timolol) elicited substantial increases in large-amplitude, slow-wave activity. (B) Shown are the same ECoG traces as shown in A, printed at a
faster chart speed (5 mm/s). Note that the presence of slow-wave activity with combined treatment of timolol/prazosin is not apparent with prazosin
treatment alone.

Table 2. High-voltage spindles during the 5–30-min testing epoch

Vehicle-controls Timolol Prazosin Timolol/prazosin

Incidence 6.2 ^ 6.0 3.0 ^ 1.9 68.6 ^ 23* 11.6 ^ 4.9

Incidence range 0 – 28 0 – 14 2 – 158 0 – 32

Mean (^ S.E.M.) incidence and range of HVS episodes during the 5–30-min testing period. HVS were
hand-scored from polygraph records. Vehicle-, timolol-, and timolol/prazosin-treated animals
displayed low levels of HVS throughout the testing period. In contrast, prazosin-treatment elicited a
substantial increase in the number of HVS episodes. *P ⬍ 0.01.
 Synergistic actions of noradrenergic b- and a1-receptors 499

Fig. 2. Effects of 150 mg timolol (Tim, n ˆ 7), 500 mg/kg prazosin (Praz, n ˆ 7), and combined timolol/prazosin (Tim/Praz, n ˆ 8) on ECoG relative power of
the identified 5-min epochs. Shown are means ^ S.E.M. for the three groups (see Fig. 1 for legend) expressed as percentage of vehicle-treated controls (n ˆ 7).
Prazosin-induced increases in HVS are reflected in increases in relative power of the 4.0–8.0 and 12.0–20.0 Hz frequency bands. Combined timolol/prazosin-
induced increase in slow-wave activity is reflected in an increase in relative power of the 0.3–2.0 Hz frequency band. ⫹P ⬍ 0.05; *P ⬍ 0.01 significantly
different from vehicle-treated controls.

significant in the 4.0–8.0 Hz band during the last four time
epochs. There was a substantial increase in this frequency
band during the 5–10-min time epoch, although this was
not statistically significant. The increase in relative power
of the 12.0–20.0 Hz frequency band was statistically sig-
nificant in the 5–10 and 10–15-min time epochs. How-
ever, the same general trend was observed in all analysed
time epochs. Additionally, prazosin reduced the relative
power of the 30.0–40.0 Hz (10–15 and 15–20 min) and
40.0–50.0 Hz (5–10, 10–15, 15–20 and 25–30 min)
frequency bands.
Combined administration of timolol and prazosin sig-
nificantly increased relative power of the 0.3–2.0 Hz
frequency band and significantly decreased relative power
of the 20.0–30.0, 30.0–40.0 and 40.0–50.0 Hz frequency
bands. The increase in relative power of the 0.3–2.0 Hz
frequency band and the decrease in the 30.0–40.0 and
40.0–50.0 Hz frequency bands were significant in all
epochs. The only exception to this was a non-significant
increase in the 0.3–2.0 Hz frequency band during the 25–
30-min epoch.
Dose-dependence of the prazosin- and timolol-induced
changes in cortical electroencephalographic activity
Additional studies were conducted to determine the dose-
dependent nature of the above described effects of a1- and
combined a1/b-receptor antagonist treatment. First, varying
doses of prazosin (250 and 1000 mg) were administered in
combination with i.c.v. vehicle infusions. These groups
were in addition to the 500-mg prazosin dose examined in
the first study (Figs 1, 2). As shown in Fig. 3, the prazosin-
induced increases in relative power of the 4.0–8.0 and 12.0–
20.0 Hz bands and the decreases in relative power of the
higher frequency bands were dependent on dose. At the high-
est dose tested, prazosin also tended to increase relative power
of the 8.0–12.0 Hz frequency band and decrease relative
power of the 0.3–2.0 Hz range.
To determine the dose-dependent nature of the slow-wave-
inducing effects of combined a1/b-receptor blockade, doses
of either timolol or prazosin, were varied while the dose of the
other drug was held constant. Animals were treated with
either the same dose of timolol used in the above described
500 C. W. Berridge and R. A. España

Fig. 3. Effects of varying doses of prazosin on ECoG relative power. Shown are means ^ S.E.M. expressed as percentage of vehicle-treated controls. Animals
were treated with either 250 (Praz 250, n ˆ 5), 500 (Praz 500, n ˆ 7), or 1000 (Praz 1000, n ˆ 5) mg/kg prazosin. Shown are means ^ S.E.M. expressed as
percentage of vehicle-treated controls. The vehicle-control and the Praz 500 groups are the same as that shown in Fig. 2. Prazosin elicited greater increases in
relative power of the 4.0–8.0 and 12.0–20.0 Hz frequency bands and greater decreases in relative power of the higher frequency bands at the higher doses than
those observed with the lowest dose. ⫹P ⬍ 0.05; *P ⬍ 0.01 significantly different from vehicle-treated controls.

studies (150 mg) and a lower dose of prazosin (250 mg/kg,
n ˆ 5) or the same dose of prazosin used above (500 mg/kg)
in combination with two lower doses of timolol (10, 50 mg,).
As shown in Fig. 4, in the presence of a constant dose of
timolol, prazosin elicited dose-dependent increases in slow-
wave activity. Thus, relative power of the 0.3–2.0 Hz
frequency band was greater at the 500-mg dose of prazosin,
when combined with 150-mg timolol, than that observed with
250-mg prazosin which, in turn, was greater than that
observed with timolol alone. Similarly, in the presence of a
constant dose of prazosin (500 mg), timolol elicited a dose-
dependent increase in slow-wave activity, with greater rela-
tive power observed in the 0.2–3.0 Hz frequency band with
increasing dose of timolol (Fig. 5).
Effects of timolol, prazosin, and combined timolol/prazosin on
behavioral activity
Behavioral analyses were conducted on animals included
in the PSA analyses described above. Analyses were
conducted on data collected within a 30-min period, begin-
ning 5 min after introduction to the testing chamber. As
shown in Fig. 6, the only significant effect of drug treatment
was a prazosin- and prazosin/timolol-induced increase in
time spent inactive (body up, immobile). There was a non-
significant trend for both treatments to decrease rears. In addi-
tion, time spent in quiet awake appeared lower in the timolol/
prazosin-treated group, possibly due to increased time spent
eating. However, neither of these effects were statistically
significant.
DISCUSSION
The current studies examined the degree to which nor-
adrenergic b- and a1-receptors contribute to the maintenance
of EEG and behavioral indices of alert waking. The results
obtained in these studies demonstrate qualitatively distinct
effects of b-, a1-, and combined b/a1-receptor blockade on
cortical neuronal activity state as measured by EEG. Thus,
b-receptor blockade alone had no effect on EEG measures
whereas, a1-receptor blockade increased the incidence of
HVS, with near-continuous HVS trains observed in some
cases. This increase in HVS is similar to that reported
previously with a1-receptor blockade. 14 However, the magni-
tude of the increase in HVS appeared larger in the current
studies than that observed previously, possibly due to the dif-
ference in testing conditions between the studies. In the cur-
rent studies, the animals were tested in a novel environment to
 Synergistic actions of noradrenergic b- and a1-receptors 501

Fig. 4. Effects of varying doses of prazosin in the presence of a constant dose of timolol (150 mg) on ECoG relative power. Shown are means ^ S.E.M.
expressed as percentage of vehicle-treated controls. Animals were treated with 150 mg i.c.v. timolol and either 0 (Tim, n ˆ 7), 250 mg/kg (Tim/Praz 250,
n ˆ 5), or 500 mg/kg (Tim/Praz 500, n ˆ 8) prazosin. Vehicle-controls and Tim 500 are those shown in Fig. 2. In the presence of a constant dose of timolol,
prazosin elicited a dose-dependent increase in the relative power of the 0.3–2.0 Hz frequency band. ⫹P ⬍ 0.05; *P ⬍ 0.01 significantly different from vehicle-
treated controls.

elicit sustained epochs of alert waking whereas, in the
previous studies animals were tested in the home cage. The
extent to which HVS was observed was dependent on time
spent in the testing apparatus. Thus, although prazosin-
induced HVS was usually observed during the first 5 min of
testing, the magnitude of HVS induction increased over the
subsequent testing period. Although HVS was generally
observed across the entire observation period, in one animal,
the characteristic waxing/waning envelope pattern of HVS
was replaced by slow-wave activity intermixed with HVS-
like frequency during the latter portions of testing.
Although b-receptor blockade alone did not alter ECoG
activity patterns, in the presence of a1-receptor blockade,
b-receptor blockade elicited a robust and dose-dependent
increase in ECoG slow-wave activity (0.3–2.0 Hz). In the
anesthetized rat, b-antagonists increase ECoG and hippo-
campal EEG indices of sedation 6,8 and block amphetamine-
induced ECoG activation. 11 The dose range over which timo-
lol increased slow-wave activity in the current study (in the
presence of a1-receptor blockade) is nearly identical to that
observed with timolol blockade of amphetamine-induced
EEG activation in anesthetized animals. 11 Thus, in the
unanesthetized animal, a1-receptor blockade more closely
mimics the sensitivity of forebrain EEG to noradrenergic
b-receptor blockade observed in anesthetized animals. It
remains to be determined why b-receptor blockade is suffi-
cient to prevent forebrain activation in the presence of
anesthesia but not in the absence of anesthesia. Possible
explanations include lower rates of neurotransmission at a1-
receptors during anesthesia compared to waking or lower
rates of neurotransmission within other activational systems
during anesthesia compared to waking. Similar to that
observed with prazosin-induced increases in HVS, the magni-
tude of the combined a1/b-receptor blockade-induced
increases in slow-wave activity was dependent on time of
testing: increases in slow-wave activity are observed over
time.
a1- and a1/b-receptor blockade also decreased measures of
502 C. W. Berridge and R. A. España

Fig. 5. Effects of varying doses of timolol in the presence of a constant dose of prazosin (500 mg/kg) on ECoG relative power. Shown are means ^ S.E.M.
expressed as percentage of vehicle-treated controls. Animals were treated with 500mg/kg prazosin and either 0 (Praz) 10 mg (Tim 10/Praz, n ˆ 5), 50 mg (Tim
50/Praz, n ˆ 9), or 150 mg (Tim 150/Praz, n ˆ 8) timolol. Praz and Tim 150/Praz groups are those shown in Fig. 2. In the presence of a constant dose of
prazosin, timolol elicited dose-dependent increases in the 0.3–2.0 Hz frequency bands. ⫹P ⬍ 0.05; *P ⬍ 0.01 significantly different from vehicle-treated
controls.

behavioral activation, consistent with the decrease in EEG
indices of forebrain activation. These treatments elicited a
large increase in time spent immobile with the body raised
above the floor, and tended to decrease locomotor activity.
Consistent with previous studies, 14 in the current study, HVS
was largely limited to periods of immobility. None of the drug
treatments produced animals that appeared to be asleep on the
basis of behavioral measures (prone, head down) during the
first 30 min of testing, despite the presence of near continuous
slow-wave ECoG activity in some of the timolol/prazosin-
treated animals (see Figs 1 and 6). In fact, combined timolol/
prazosin treatment appeared to decrease time spent in quiet
awake, possibly due to increases in time spent eating,
although neither of these effects were statistically significant.
The extent to which these treatments altered behavioral
indices of arousal is partially obscured by the fact that
vehicle-treated animals did not display substantial locomotor
activity under the current testing conditions.
Noradrenergic mechanisms of anesthesia
The above described results are consistent with previous
studies demonstrating profound sedative actions of a2-
agonists, which suppress LC discharge rates and rates of
NE release, when administered systemically, i.c.v., or within
or adjacent to LC. 17–19,23,46 The ability of a2-agonists to lower
the effective dose of general anesthetics make these drugs
useful adjuncts to surgical anesthesia. 13,27 The current obser-
vations suggest that the sedative actions of a2-agonists, may
derive from the simultaneous decrease in rates of neurotrans-
mission at a1- and b-receptors. Previous studies demonstrated
anesthesia-potentiating actions of b-receptor blockade,
suggesting that centrally acting b-antagonists might also be
useful adjuncts to general anesthesia. 11,8 The extent to which
a1-receptor blockade reduces the effective anesthetic dose
and the extent to which combined a1/b-receptor blockade
reduces effective anesthetic dose beyond that observed with
 Synergistic actions of noradrenergic b- and a1-receptors 503

Fig. 6. Effects of vehicle, timolol, prazosin, and combined timolol/prazosin treatment on behavioral responses to the novel environment. Shown are mean ^
S.E.M. time (s) spent engaged in the designated behaviors in vehicle (Veh)-, timolol (Tim)- (150 mg), prazosin (Praz)- (500 mg/kg), and combined timolol/
prazosin (Tim/Praz)-treated animals. Prazosin and combined timolol/prazosin increased time spent inactive (body up). Non-significant trends for decreased
rears were observed with all drug treatments. Non-significant trends for decreased quiet waking and increased eating were observed in timolol/prazosin-treated
animals. ⫹P ⬍ 0.05; *P ⬍ 0.01 significantly different from vehicle-treated controls.

either a1- or b-receptor blockade alone remains to be
determined.
Relevance to behavior
The current studies demonstrate potent actions of
combined inhibition of noradrenergic a1- and b-receptor
mediated neurotransmission on cortical neuronal activity
state. Available evidence indicates that cortical and thalamic
neuronal activity patterns associated with ECoG activation
are essential for the efficient transfer and processing of
sensory information from periphery to neocortex. 34,40,42
Consistent with this, ECoG activity state closely correlates
with performance in tests of vigilance. 3,20,30 Thus, it is
assumed that forebrain EEG activation is an essential con-
dition for optimal operation of various cognitive and
affective processes that guide behavior. The current observa-
tions, therefore, suggest that combined activation of a1- and
b-receptors, which occurs normally during the awake state, is
a necessary condition for optimal performance of specific
cognitive operations. Consistent with this, a correlation
between LC discharge activity and performance in a test of
vigilance has been described previously. 4,39 The functional
significance of HVS remains to be determined. However,
HVS are typically observed during transitions between EEG
activation and the appearance of large-amplitude, slow-wave
activity. Thus, at the level of EEG, a1-blockade alone
produces a decrease in forebrain activation that is intermedi-
ate to that observed with combined a1/b-receptor blockade.
Future studies will need to determine the extent to which a1-,
b-, and combined a1/b-receptors impact vigilance and other
cognitive and/or affective processes that guide interaction
with the environment.
It is important to note that despite the profound alterations
in cortical activity state elicited by a1- and a1/b-receptor
blockade, the magnitude of these effects were smaller earlier
in the observation period than that observed in the latter
portions of testing. Thus, during the initial exposure to the
environment, which is presumably associated with the
greatest levels of arousal/vigilance, blockade of a1- and
b-receptors had smaller effects on the general activity state
of the forebrain than observed later in testing. Whether this is
also observed following a2-agonist-induced suppression of
NE release, is currently unknown. Actions of other ascending
504 C. W. Berridge and R. A. España

activational systems likely contribute to this phenom-
enon. 28,35,38,40,41,44,45 None the less, it is important to note
that blockade of a1- and a1/b-receptors is sufficient to
decrease EEG indices of arousal, at time periods when vehicle-
treated animals continue to display prominent behavioral and
EEG indices of alert waking and when these other neurotrans-
mitter systems are presumably active. These latter observations
indicate a critical role of noradrenergic systems in the main-
tenance of an activated forebrain in alert waking, at least
under certain conditions. Similar sedative effects of cholin-
ergic and serotonergic receptor blockade are also observed at
the level of EEG (for review see Ref. 45). Thus, at least under
certain environmental/behavioral conditions, the actions of
multiple ascending activational are necessary for activation
of the forebrain associated with alert waking. Removal of any
of these systems results in an alteration in forebrain neuronal
activity state, and presumably in the ability of the forebrain to
support certain cognitive and/or affective processes.
The current observations are in contrast to that observed in
rapid eye movement (REM) sleep which is associated with
forebrain EEG activation 43 and the virtual absence of LC
neuronal discharge activity. 25 The precise physiological
differences between waking and REM sleep remain to be
characterized completely. Presumably REM sleep is asso-
ciated with actions of one or more neural systems that are
not present during waking and that are sufficient to elicit
forebrain activation in the absence of LC-noradrenergic
neurotransmission.
Potential sites of action
The sites within which a1- and b-receptors act to modulate
activity state of the forebrain remains to be determined. One
possibility is that NE acts directly within neocortical, hippo-
campal, and thalamic regions to alter the activity patterns of
these neurons. For example, previous studies demonstrate that
NE, through actions at both a1- and b-receptors, modulates
cortical and thalamic neuronal activity patterns. 31,34 Specifi-
cally, direct application of NE to cortical and thalamic
neurons induces the transition from a burst-type firing mode
associated with slow-wave sleep to a single spike firing mode
associated with waking. Alternatively, the basal forebrain
sites encompassing MS and MPOA are sites within which
a1- and b-receptors act to modulate both forebrain EEG and
behavioral indices of behavioral state. Thus, stimulation of
either b- or a1-receptors within MS or MPOA elicits robust
and sustained increases in behavioral and forebrain EEG
indices of waking. 8–10,29 Finally, noradrenergic efferents
may act, either directly or indirectly, to modulate activity of
other ascending activational systems implicated in the regula-
tion of behavioral state, including histaminergic, serotonergic
and cholinergic systems. 28,35,38,40,41,44,45
CONCLUSIONS
The current studies demonstrate synergistic actions of a1-
and b-receptor blockade on forebrain neuronal activity state,
as measured by EEG, within alert waking. Thus, b-receptor
blockade alone does not alter EEG activity patterns, a1-
receptor blockade increases HVS, and combined a1/b-
receptor blockade elicits substantial first increases in ECoG
slow-wave activity. These effects were observed beginning
shortly following initial exposure to the testing chamber, at
a time when vehicle-treated animals display prominent EEG
and behavioral evidence of alert waking. Thus, these results
demonstrate that via the combined actions of a1- and b-
receptors, noradrenergic systems are an integral component
of the neural architecture that supports forebrain activation
in alert waking. Forebrain neuronal activity patterns asso-
ciated with waking are necessary for optimal operation of
forebrain neural circuits. Thus, it is concluded that waking-
associated noradrenergic neurotransmission at a1- and b-
receptors is essential for optimal performance of cognitive
and affective processes which are dependent on forebrain
neuronal activity.
Acknowledgements—The authors are thankful for the comments and
insight of Dr Eric Stone, which formed a starting point for these
studies. The authors gratefully acknowledge the expert technical
assistance of Mr John O’Neill. This work was funded by PHS grant
DA10681 (CWB).

REFERENCES

1. Andersen P. and Andersson S. A. (1968) Physiological basis of the alpha rhythm, Appleton-Century-Crofts, New York.
2. Aston-Jones G. and Bloom F. E. (1981) Activity of norepinephrine-containing locus coeruleus neurons in behaving rats anticipates fluctuations in the
sleep–waking cycle. J. Neurosci. 1, 876–886.
3. Aston-Jones G. and Bloom F. E. (1981) Norepinephrine-containing locus coeruleus neurons in behaving rats exhibit pronounced responses to non-
noxious environmental stimuli. J. Neurosci. 1, 887–900.
4. Aston-Jones G., Rajkowski J., Kubiak P. and Alexinsky T. (1994) Locus coeruleus neurons in the monkey are selectively activated by attended stimuli in
a vigilance task. J. Neurosci. 14, 4467–4480.
5. Berridge C. W. and Dunn A. J. (1989) Restraint-stress-induced changes in exploratory behavior appear to be mediated by norepinephrine-stimulated
release of CRF. J. Neurosci. 9, 3513–3521.
6. Berridge C. W. and Foote S. L. (1991) Effects of locus coeruleus activation on electroencephalographic activity in neocortex and hippocampus.
J. Neurosci. 11, 3135–3145.
7. Berridge C. W., Page M. E., Valentino R. J. and Foote S. L. (1993) Effects of locus coeruleus inactivation on electroencephalographic activity in
neocortex and hippocampus. Neuroscience 55, 381–393.
8. Berridge C. W., Bolen S. J., Manley M. S. and Foote S. L. (1996) Modulation of forebrain electroencephalographic activity in halothane-anesthetized rat
via actions of noradrenergic b-receptors within the medial septal region. J. Neurosci. 16, 7010–7020.
9. Berridge C. W. and Foote S. L. (1996) Enhancement of behavioral and electroencephalographic indices of waking following stimulation of noradrenergic
b-receptors within the medial septal region of the basal forebrain. J. Neurosci. 16, 6999–7009.
10. Berridge C. W., Mitton E., Clark W. and Roth R. H. (1999) Engagement in a non-escape (displacement) behavior elicits a selective and lateralized
suppression of frontal cortical dopaminergic utilization in stress. Synapse 32, 187–197.
11. Berridge C. W. and Morris M. (2000) Noradrenergic b-receptor blockade prevents arousal-enhancing actions of amphetamine on forebrain EEG.
Psychopharmacology 148, 307–313.
12. Berridge C.W. and Wifler K. Contrasting effects of noradrenergic b-receptor blockade within the medial septal area on forebrain electroencephalographic
and behavioral activity state in anesthetized and unanesthetized rat. Neuroscience, 97, 543–552.
 Synergistic actions of noradrenergic b- and a1-receptors 505

13. Bloor B. D. and Flacke W. E. (1982) Reduction of halothane anesthetic requirements by clonidine, an alpha-2 adrenergic agonist. Anesth. Analg. (Cleve)
61, 741–745.
14. Buzsaki G., Kennedy B., Solt B. V. and Ziegler M. (1991) Noradrenergic control of thalamic oscillation: the role of a-2 receptors. Eur. J. Neurosci. 3,
222–229.
15. Chaulk P. C. and Harley C. W. (1998) Intracerebroventricular norepinephrine potentiation of the perforant path-evoked potential in dentate gyrus of
anesthetized and awake rats: a role for both a- and b-adrenoceptor activation. Brain Res. 787, 59–70.
16. Correa-Sales C., Rabin BC. and Maze M. (1992) A hypnotic response to dexmedetomidine, and a2-agonist, is mediated in the locus coeruleus in rats.
Anesthesiology 76, 948–952.
17. Danysz W., Dyr W., Plaznik A. and Kostowski W. (1989) The effect of microinjections of clonidine into the locus coeruleus on cortical EEG in rats. Pol.
J. Pharmac. 41, 45–50.
18. De Sarro G. B., Ascioti C., Froio F., Libri V. and Nistico G. (1987) Evidence that locus coeruleus is the site where clonidine and drugs acting at alpha-1
and alpha-2 adrenoreceptors affect sleep and arousal mechanisms. Br. J. Pharmac. 90, 675–685.
19. De Sarro G. B., Baggeta G., Libri V. and Nistico G. (1988) Microinfusion of clonidine and yohimbine into locus coeruleus alters EEG power spectrum:
effects of aging and reversal by phosphatidylserine. Br. J. Pharmac. 95, 1278–1286.
20. Foote S. L., Aston-Jones G. and Bloom F. E. (1980) Impulse activity of locus coeruleus neurons in awake rats and monkeys is a function of sensory
stimulation sand arousal. Proc. natn. Acad. Sci. USA 77, 3033–3037.
21. Foote S. L., Bloom F. E. and Aston-Jones G. (1983) The nucleus locus coeruleus: new evidence of anatomical and physiological specificity. Physiol. Rev.
63, 844–914.
22. Foote S. L. and Morrison J. H. (1987) Extrathalamic modulation of neocortical function. Ann. Rev. Neurosci. 10, 67–95.
23. Gatti P. J., Hill K. J, Da Silva A. M. T., Norman W. P. and Gillis R. A. (1988) Central nervous system site of action for the hypotensive effect of clonidine
in the cat. J. Pharmac. exp. Ther. 245, 373–380.
24. Harris G. C., Hedaya M. A., Pan W. J. and Kalivas P. (1996) Adrenergic antagonism alters the behavioral and neurochemical responses to cocaine.
Neuropsychopharmacology 14, 195–204.
25. Hobson J. A., McCarley R. W. and Wyzinski P. W. (1975) Sleep cycle oscillation: reciprocal discharge by two brainstem neuronal groups. Science 189,
55–58.
26. Kumar V. M., Datta S., Chhina G. S. and Singh B. (1986) Alpha adrenergic system in the medial preoptic area involved in the sleep–wakefulness in rats.
Brain Res. Bull. 16, 463–468.
27. Kaukinen S. and Pyyko K. (1979) The potentiation of halothane anesthesia by clonidine. Acta anesthesiol. scand. 23, 107–111.
28. Lin J. S., Hou U., Kazuya S. and Jouvet M. (1996) Histaminergic descending inputs to the mesopontine tegmentum and their role in the control of cortical
activation and wakefulness in the cat. J. Neurosci. 16, 1523–1537.
29. Mallick B. N. and Alam M. N. (1992) Different types of norepinephrinergic receptors are involved in the preoptic are mediated independent modulation
of sleep–wakefulness and body temperature. Brain Res. 591, 8–19.
30. Makeig S. and Inlow M. (1993) Lapses in alertness: coherence of fluctuations in performance and EEG spectrum. Electroencephalogr. clin. Neurophys.
86, 23–35.
31. McCormick D. A. (1989) Cholinergic and noradrenergic modulation of thalamocortical processing. Trends Neurosci. 12, 215–220.
32. McCormick D. A. and Pape H. C. (1990) Noradrenergic and serotonergic modulation of a hyperpolarization-activated cation current in thalamic relay
neurons. J. Physiol. 431, 319–342.
33. McCormick D. A., Pape H. C. and Williamson A. (1991) Actions of norepinephrine in the cerebral cortex and thalamus: implications for function of the
central noradrenergic system. Prog. Brain Res. 88, 293–305.
34. McCormick D. A. and Bal T. (1997) Thalamocortical mechanisms. A. Rev. Neurosci. 20, 185–215.
35. Metherate R., Cox C. L. and Ashe J. H. (1992) Cellular bases of neocortical activation: modulation of neural oscillations by the nucleus basalis and
endogenous acetylcholine. J. Neurosci. 12, 4701–4711.
36. Mouradian R. D., Seller F. M. and Waterhouse B. D. (1991) Noradrenergic potentiation of excitatory transmitter action in cerebrocortical slices: evidence
of mediation by an a1-receptor-linked second messenger pathway. Brain Res. 546, 83–95.
37. Pape H. C. and McCormick D. A. (1989) Noradrenaline and serotonin selectively modulate thalamic burst firing by enhancing a hyperpolarization-
activated cation current. Nature 340, 715–718.
38. Peck B. K. and Vanderwolf C. H. (1991) Effects of raphe stimulation on hippocampal and neocortical activity and behavior. Brain Res. 568, 244–252.
39. Rajkowski J., Kubiak P. and Aston-Jones G. (1994) Locus coeruleus activity in monkey: phasic and tonic changes are associated with altered vigilance.
Brain Res. Bull. 35, 607–616.
40. Steriade M. and Buzsaki G. (1990) Parallel activation of thalamic and cortical neurons by brainstem and basal forebrain cholinergic systems. In Brain
Cholinergic Systems (eds Steriade M. and Biesold D.). Oxford UP, Oxford.
41. Steriade M., Datta S., Pare D., Oakson G. and Dossi R. C. (1990) Neuronal activities in brain-stem cholinergic nuclei related to tonic activation processes
in thalamocortical systems. J. Neurosci. 10, 2541–2559.
42. Steriade M., McCormick D. A. and Sejnowski T. J. (1993) Thalamocortical oscillations in the sleeping and aroused brain. Science 262, 679–685.
43. Timo-Iaria C., Negrao N., Schmidek W. R., Hoshino K., Lobato de Menenzes C. E. and Da Rocha T. L. (1970) Phases and states of sleep in the rat.
Physiol. Behav. 5, 1057–1062.
44. Vanderwolf C. H. and Robinson T. E. (1981) Reticulo-cortical activity and behavior: a critique of the arousal theory and a new synthesis. Behav. Brain
Sci. 4, 459–514.
45. Vanderwolf C. H. (1988) Cerebral activity and behavior: control by central cholinergic and serotonergic systems. Int. Rev. Neurobiol. 30, 225–341.
46. Waterman A., Livingston A. and Bouchenafa O. (1988) Analgesic effects of intrathecally-applied a2-adrenoreceptor agonists in conscious, unrestrained
sheep. Neuropharmacology 27, 213–216.

(Accepted 2 May 2000)