Laboratorio de
Biología de Sueño, Departamento de
Biomedicina, Instituto de Ciencias de la Salud,
Universidad Veracruzana, México.
Abstract
Kuniomi Ishimori and Henri Piéron
were the first researchers to introduce the
concept and experimental evidence for a chemical
factor that would presumably accumulate in the
brain during waking and eventually induce sleep.
This substance was named hypnotoxin. Currently,
the variety of substances which have been shown
to alter sleep includes peptides, cytokines,
neurotransmitters and some substances of lipidic
nature, many of which are well known for their
involvement in other biological activities. In
this chapter, we describe the sleep-inducing
properties of the vasoactive intestinal peptide,
prolactin, adenosine and anandamide.
Intoduction
Henri
Piéron was the first researcher
to introduce the concept and experimental
evidence for a chemical factor that would
presumably accumulate in the brain during waking
and eventually induce sleep. This concept lay
dormant for many years, in absence of adequate
techniques and knowledge to pursue the
identification of such a hypnogenic factor. With
the development of biochemical techniques and
the description of various neurotransmitters
with specific anatomical localizations, it was
suggested that a subset of these endogenous
chemicals could be involved in the generation of
sleep. In the 1960s, Michel Jouvet championed
the concept that sleep and waking could be
respectively generated by specific chemical
neurotransmitters contained in specific neuronal
systems.
Recently, peptides and hormones have been
identified as substances capable of modulating
the sleep-wake cycle. Thus, chemicals of
different molecular sizes were suggested to
function as neurotransmitters, neuromodulators
or neurohormones, providing the possibility for
short-to-long acting molecules that could
participate collectively in the generation and
maintenance of the sleep-wake cycle. The variety
of substances which have been shown to alter
sleep includes peptides, cytokines,
neurotransmitters and some lipid-containing
substances, many of which are well known for
their involvement in other biological
activities. These substances, referred to as
sleep-inducing factors (SIFs), have been
classified with criteria by some authors in
order to be considered as SIFs. In this section,
we describe the evidence accumulated to date on
several SIFs and the changes that they produce
on sleep patterns.
The Origin of Sleep Inducing
Factors
First investigators to propose the existence
of a substance capable of inducing sleep were
René
Legendre and Henri Piéron (1913 in
France and Kuniomi Ishimori in Japan (1909).
Kubota
K. Kuniomi Ishimori and the first discovery
of sleep-inducing substances in the brain.
Neurosci Res. 1989;6(6):497-518.
In the first classic experiment,
Piéron removed cerebrospinal fluid (CSF)
from sleep-deprived dogs (for periods varying
from 30 to 505 hours, but in no case beyond the
point where they showed extreme sleepiness) and
transferred it to normal dogs. The recipient
animals were then observed to sleep longer than
usual. Sleep-deprived dogs showed significant
reductions of the muscle tone observed as a
visible decrease of the postural tone for the
head and flexion of the legs. Sleep-deprived
dogs demonstrated no significant changes in the
composition of the blood, blood pressure, heart
beat frequency, breathing frequency or corporal
temperature. Nevertheless, sleepdeprived dogs
had the following changes in the cytology of the
deep layers of the prefrontal cortex;
displacement of the neuronal nuclei,
vacuolization of the cytoplasm and disappearance
of Nissl grains [4]. The damage in the
cerebral cortex was proportional to the sleep
deprivation time. Interestingly, if
sleep-deprived dogs were allowed recovery sleep,
the cellular damage was not observed. These
results suggested that the accumulation of
sleep-promoting substances in CSF during
prolonged wake periods, collectively termed
"hypnotoxin", was sufficient to induce sleep in
normal animals. The hypnotic effects of
hypnotoxin disappeared when CSF was warmed to
65°C in 5 minutes, suggesting the
protein-like nature of the substance.
Nevertheless, it was not possible to isolate
that substance due to the technical limitations
of the epoch. On the other hand, Ishimori
performed experiments similar to those of
Piéron, but Ishimori named the
accumulated substance in the CSF of
sleep-deprived dogs a "hypnogenic" substance.
The works of Henri Piéron together with
Ishimori marked the beginning of the study of
humoral regulation of sleep.
Vasoactive Intestinal Peptide
(VIP)
VIP was originally isolated from the small
intestine and the lung as a peptide with potent
vasodilator effects and has recently been
suggested to play a major regulatory role in a
large number of acute and chronic diseases. The
primary structure of VIP is closely related to
pituitary adenylate cyclase activating
polypeptide. VIP contains 20 amino acids with a
molecular weight of 3329 Da. The amino acid
sequence of VIP in the rat, human and other
mammalians species is identical. VIP receptors
are widely distributed in the cerebral cortex,
amygdaloid nuclei, hippocampus, olfactory lobes,
thalamus, and suprachiasmatic nucleus.
Several studies have suggested a role for
VIP as a SIF. In the decade of 1980's Riou and
co-workers demonstrated that
intracerebroventricular (i.c.v.) administration
of 100 ng of VIP promotes rapid-eye movement
(REM) sleep in rats. Subsequent studies
indicated that i.c.v. injection of VIP also
promotes REM sleep in other species such as cats
and rabbits.
In 1986, Prospero-García and
co-workers reported that both the cerebral CSF
from cats sleep-deprived for 24 hours, as well
as VIP, were each capable of restoring REM sleep
in cats pretreated with para-chlorophenylalanine
(PCPA), a specific inhibitor of serotonin
synthesis which reduces sleep time. These
results suggested that the CSF of sleep-deprived
cats contains a VIP-like sleep factor possibly
involved in triggering REM sleep. Subsequent
studies demonstrated that i.c.v. administration
of CSF in cats incubated with VIP antibodies
blocked its REM sleepinducing effects
[18], and that VIP also induces REM
sleep recovery in insomniac forebrain-lesioned
cats.
Subsequently, it has been observed that
VIPimmunoreactivity in various areas of the rat
brain after 24 and 48 hours of REM sleep
deprivation did not change in any analyzed brain
structure, suggesting that the increase in VIP
peptide levels does not correlate with the REM
sleep homeostatic process. However, other
studies have shown that in both the
suprachiasmatic nucleus and periventricular
nucleus the concentration of VIPimmunoreactivity
increases during the dark period, when rodents
are mostly awake, and decreases during the
subsequent light period, when most of the time
is spent in sleep. VIP-immunoreactivity exhibits
daily variations in the locus coeruleus nucleus,
the periaqueductal gray matter and the
paraventricular nucleus. These daily variations
suggest that VIP may act as an endogenous
hypnogenic factor which progressively
accumulates during the waking period, until it
reaches a critical level at which it could
trigger REM sleep. Additionally, it has been
demonstrated that the density of VIP receptors
increases after 72 hours of REM sleep
deprivation. The increase of VIP receptors
occurred mainly in the laterodorsal tegmental
nucleus and posterodorsal tegmental nucleus,
brainstem areas involved in the generation and
maintenance of REM sleep.
PRL and Sleep Regulation
In 1986, Michel Jouvet demonstrated that the
systemic administration of PRL enhances the
total time of REM sleep in cats. Studies in
hypo-prolactinemic rats showed that REM sleep
duration decreased, the circadian rhythm of REM
sleep disappeared, while that of non-rapid eye
movement (NREM) sleep remains unchanged. These
studies were the first to provide evidence of
PRL-involvement in sleep regulation. Further
studies have shown that the administration of
exogenous PRL in additional animal species
including rabbits and rats induced REM
sleep.
Additionally, the systemic administration of
anti-PRL antibodies and anti-serum in rats
reduced REM sleep. Likewise, it has been
observed that the effect of PRL on REM sleep is
photoperiod-dependent, because the amount of REM
sleep decreased or increased when PRL was
administered during the dark or light period,
respectively. Furthermore, it has been reported
that prolactinreleasing peptide (PrRP) also
promotes REM sleep when it is administered to
rats. These results suggest that PRL stimulated
by i.c.v. injection of PrRP could contribute to
the induction of REM sleep.
The firing frequency of mesopontine
tegmental neurons, an area involved in REM sleep
generation, increases after PRL injection. This
result suggests that PRL is capable of inducing
REM sleep and modulating cholinergic activity,
but the mechanisms remain unclear. Additionally,
there are other studies that show that PRL
injection into the central nucleus of the
amygdala, an area containing high concentrations
of PRL immunoreactive fibers and receptors,
decreased NREM sleep. However, microinjections
of cholinergic agonist in the same nucleus
induced REM sleep.
PRL administered locally into the rat
dorsolateral hypothalamus, an area that contains
PRL immunoreactive neurons, either increases REM
sleep when it is given diurnally or decreases
REM sleep when given nocturnally. Also, it has
been demonstrated that in hypoprolactinemic rats
under light-dark conditions, the circadian
rhythms of NREM sleep and REM sleep display an
alteration of their phase relation. This result
suggests that the promoting effect of REM sleep
by PRL could be regulated by circadian
factors.
In addition, it has been shown that
relatively higher levels of PRL secretion occurs
during NREM sleep and that PRL secretion was
coupled to delta waves in humans, in contrast
alpha and beta bands frequencies were inversely
proportional to PRL secretion. It is well
established that plasma PRL concentrations
exhibit a sleep-dependent pattern, with high
levels occurring during sleep and low levels
during waking. Thus, it has been suggested that
endogenous PRL accumulation during NREM sleep
contributes to subsequent REM sleep.
[....]
Adénosine AD, the Sleep-Inducing
Purine
As discussed in the introduction, several
molecules modulate the steps leading to the
initiation of sleep. In addition to
neurotransmitters, such as gamma-aminobutyric
acid and peptide factors, including VIP and PRL,
there are endogenous compounds with
sleep-inducing properties. AD is a product of
natural cellular metabolism and has been
suggested to be an important homeostatic sleep
factor acting in neurons in basal forebrain and
preoptic areas through A1 and A2A receptors. In
this section we will review the role of AD on
sleep modulation.
Feldberg and Sherwood were the first to show
that AD induced sleep-like behavior in the cat.
Further research in rodents described that
systemic administration of AD, its analogs or
inhibitors of its metabolism increase NREM
sleep.
To date, several AD receptors have been
described. These receptors are
G-protein-coupled, and compromise the following
members: A1, A2A, A2B and A3. A1 (and A3)
inhibit adenylyl cyclase, while A2 displays
opposite properties. A1 are postsynaptic,
leading to a stimulation of K+ channels, as well
as a hyperpolarization and inhibition of neural
activity.
The case for AD as a sleep factor has been
suggested from multiple findings. Microdialysis
studies have confirmed that cholinergic cells in
the basal forebrain are presumably responsible
for the accumulation of this purine during
natural waking and during sleep deprivation.
Additionally, infusions of AD into the basal
forebrain of cats increase sleep, whereas the
extracellular levels of this purine have been
described to be about 20% higher during
wakefulness (W) compared with sleep
[80]. Importantly, when the period of
prolonged W was increased to 6 h using sleep
deprivation, a significant and robust increase
of AD (200%) was found, and this value decreased
during the recovery sleep period.
[....]
The sleep-inducing mechanism suggests that
AD acts on an AD1 autoreceptor placed in
cholinergic neurons in the basal forebrain. The
activation of the AD1 decreases the activity of
the cholinergic neurons blocking the inhibition
of gamma-aminobutyric acid-ergic neurons in the
ventrolateral preoptic area leading to a sleep
induction. It is widely accepted that the drive
to sleep is determined by the activity of the
basal forebrain cholinergic neurons, which in
turn release AD. It has been suggested that the
increase in the endogenous levels of AD are due
to the enhancement of the metabolic activity
associated with the neuronal discharge activity
during W. Thus, accumulating AD inhibits these
neurons so that sleep-active neurons can become
active. Importantly, this current hypothesis has
been strengthened with experimental evidence
showing that systemic or central administrations
of AD induce sleep and its extracellular levels
are increased during prolonged waking. Recently,
it was demonstrated that the basal forebrain
cholinergic neurons are central to the AD
regulation of sleep drive. In rats in which at
least 95% these neurons are successfully
lesioned via injections of the neurotoxin
192-IgG-saporin, the extracellular levels of AD
levels did not increase with 6 h of prolonged
waking. However, the lesioned rats had intact
sleep drive after 6 and 12 h of prolonged
waking. These results suggested that AD
accumulation in the basal forebrain is not
necessary for sleep drive. Additionally, it was
demonstrated that the basal forebrain
cholinergic neurons were not essential to sleep
induction as administration of the selective AD
A1 receptor agonist N6-cyclohexyladenosine into
basal forebrain of rats continued to induce
sleep. Thus, results suggest that neither the
activity of the cholinergic neurons placed in
the basal forebrain nor the accumulation of AD
in this brain area during wake is necessary for
sleep drive.
Anandamide (ANA)
During the 1960s several experiments were
carried out in order to evaluate the effects of
the cannabinoids on sleep. The main conclusion
of those experiments was that cannabinoids
modulate sleep by increasing NREM and REM
sleep.
Subsequently, transmembrane proteins in the
central nervous system (CNS) that recognize the
principal compound of marijuana, the
delta-9-tetrahydrocannabinol were described. The
receptors were classified as central or
peripheral, CB1 and CB2, respectively. The
presence of the CB1 receptor has been shown in
specific areas of the CNS such as cortex,
hippocampus, striatum, limbic system,
cerebellum, and brainstem. This protein
possesses 7 transmembrane domains and inhibits
cAMP formation through Gi alpha subunit
protein.
Once described as the receptors for the
exogenous cannabinoids, the search for
endogenous compounds for these cannabinoid
receptors started. To date, the family of
endogenous cannabinoids or endocannabinoids
comprises arachidonoylethanolamine,
2-araquidonylglycerol (2AG), virodhamine,
noladin-ether and N-arachidonyldopamine.
Cannabimimetic effects, including disruptions in
functions such as learning and memory, feeding,
pain perception and sleep have been reported
after systemic or central administrations of
these compounds.
ANA was the very first endocannabinoid
described, and it has been the most studied so
far. Pharmacological studies have shown that
injections of this lipid induce several
intracellular and behavioural changes including
sleep generation.
The distribution of ANA in the CNS includes
regions such as cortex, hippocampus, striatum,
cerebellum, and brainstem. Histological studies
suggest that ANA and 2-AG are well-positioned to
act upon CB1 in the CNS. Biochemically, the
release of ANA follows a different mechanism
from classical, synaptic vesicle-stored
neurotransmitters. It has been hypothesized that
ANA is released "on demand". This suggests that
there is a biological mechanism that involves
the activity of membrane phospholipids that are
the precursors of those compounds. In this
model, the biosynthesis of ANA is followed by
its immediate release. The mechanism of ANA
degradation involves two different pathways. In
the first, the compounds are transported to the
interior of the cell via specific transporters.
For ANA the putative transporter is the ANA
membrane transporter (AMT). Once the lipid
modulator is inside the cell, a hydrolysis
mechanism is initiated by the fatty acid amide
hydrolase (FAAH). Drugs targeting the AMT
[110] or FAAH have been shown to
modulate endogenous levels of ANA.
The physiological role of ANA on sleep is
under investigation. The first approach to
answer this question was carried out by Santucci
and co-workers in 1996. Systemically injected a
CB1 cannabinoid receptor antagonist, SR141716A,
to rats increased W and reduced NREM sleep. As a
conclusion they suggested that the wake-inducing
properties of SR141716A might be due to the
blocking of the CB1 cannabinoid receptor.
Later, our laboratory demonstrated that the
endocannabinoid ANA also modulated sleep. I.c. v
injections of ANA in rats induced an opposite
effect to that observed by Santucci and
colleagues. We found a significant decrease in W
and an enhancement in NREM and REM sleep. We
also observed that the effects caused by ANA on
sleep were more significant once injected into
the pedunculopontine tegmental nucleus, a brain
region suggested as a key element in sleep
promotion. If the changes observed in sleep
after the administration of ANA were due to the
activation of the CB1 cannabinoid receptor, we
hypothesized that the administration of the CB1
cannabinoid receptor antagonist SR141716A before
the injection of ANA might block the effects on
sleep. This experimental manipulation showed
that SR141716A efficiently blocked the effects
caused by ANA on sleep. Furthermore, it is known
that the CB1 cannabinoid receptor activates
intracellular elements such as phospholipase C.
Furthermore, we found that the injection of the
phospholipase C inhibitor U73122 before ANA
administration blocked the sleep-inducing
effects caused by the endocannabinoid. Finally,
we showed that systemic administration of ANA
induced an increase in sleep, and perhaps more
importantly, a considerable enhancement of the
levels of the sleep-inducing molecule AD.
[....]
CONCLUSIONS
To date, the SIFs differ in several aspects,
from their chemical nature and function per se
of each to the effect that they have on the
mechanisms of induction and/or maintenance of
sleep. These SIFs also work in concert with the
substances that regulate W. Within this
perspective it is difficult to define to a
single substance as one directly responsible for
the humoral regulation of sleep. Instead the
regulating effects of these substances are
dependent not only on the interactions among
them, but on their involvement with the cerebral
regions which are directly responsible for the
effects observed in the sleep pattern.
One hundred years after the discoveries of
Ishimori and Piéron, the number of sleep
regulating substances have grown substantially.
However, current data indicate that sleep is not
the result of the action of any one particular
substance on a specific brain site, but rather
that sleep is the result of a chain of
biochemical events, which lead to the cellular
processes necessary for the transitional states
required by the sleep-wake cycle.
