Sleep Medicine Reviews, Vol. 4, No.

5, pp 505–514
doi:10.1053/smrv.2000.0112,
available online at http://www.idealibrary.com on
SLEEP
MEDICINE
reviews
HISTORICAL NOTE

Frédéric Bremer 1892–1982: a pioneer in sleep

research
M. Kerkhofs1 and P. Lavie2
1
Sleep Laboratory, CHU A. Vésale, Université Libre de Bruxelles, Route de Gozée 706,
6110 Montigny-le-Tilleul, Belgium; 2 Sleep Laboratory, Faculty of Medicine, Technion, Israel
Institute of Technology, Haifa, Israel

Frédéric Bremer was one of the pioneer neurophysiologists who dedicated their career to understanding
the neural mechanisms involved in the control of sleep–wake regulation. This paper follows his career
and his major achievements. We found that Bremer’s interest in sleep resulted from his unexpected
observations after transecting the brain at the midcollicular level in the “encéphale isolé” preparation.
The sleep-like behaviour of the animal, accompanied by slow waves in the cortex, convinced him that
sleep resulted from cortical deafferentation. He was further convinced that deafferentation was the cause
of sleep when he found that transecting the brain at the medullary level did not much affect the sleep–wake
cycle. As we show, Bremer’s views that sleep is a passive phenomenon imposed on the brain because of
deafferentation was shared by most of his contemporaries. Years later Bremer admitted that he interpreted
his experimental findings wrongly. He continued to investigate sleep using his preparations and made
important contributions to understanding the relationships between the brainstem reticular formation
and the basal forebrain hypnogenic centres, as well as the importance of light on these relationships.
 2000 Harcourt Publishers Ltd

Key words: Bremer, encéphale isolé, cerveau isolé, sleep, deafferentation.

The way to sleep

Frédéric Bremer was born in Arlon, in the Belgian Ardennes in 1892. His family was
an intellectual one: both his father and mother were teachers at the Arlon Public
Secondary School. Frédéric Bremer himself was a bright student, who excelled in his
medical studies at the University of Brussels. However, these studies were interrupted
by the First World War, during which he served as military physician in a cavalry
regiment, then later as a medical auxiliary at the famous Ocean Ambulance. This was
a military hospital situated on the Belgian coast, in the city of La Panne, and was
frequently attended by Queen Elisabeth of Belgium who worked there as a volunteer.
In 1919, Bremer started his training in neurology, as the assistant of Professor Pierre
Marie at the Hospital de la Salpêtrière in Paris. Then, as a Fellow of the Belgian

Correspondence should be addressed to: Myriam Kerkhofs. Fax: +32 71 29 5226; E-mail
mkerkhof@ulb.ac.be

1087–0792/00/050505+10 $35.00/0  2000 Harcourt Publishers Ltd

506 M. Kerkhofs and P. Lavie

American Educational Foundation, he travelled to the USA, and spent a year at

Harvard with H. Cushing, followed by visits to the physiology laboratories of W. B.
Cannon and A. Forbes. Before finally returning to Belgium, he also spent some time
in the famous laboratory of Sir Charles Sherrington in Oxford. In 1924 Bremer became
an Aggregate (Fellow) of the University of Brussels and in 1932 started teaching “General
Pathology” at the Faculty of Medicine.
Bremer, who was a popular teacher, became Professor in 1934. His first publications
in 1920 and 1921 were clinical descriptions of aphasia and apraxia [1,2]. His first
scientific research paper, which was published in collaboration with P. Bailey from
Cushing’s laboratory, dealt with the neurogenic origin of diabetes insipidus [3]. Bremer,
together with Bailey, showed that a minimal lesion of the hypothalamus can induce
experimental diabetes insipidus. These lesions were also shown to induce sleepiness
and adiposity. In addition, their data suggested the existence of a “thirst” centre in
the hypothalamus.
When Bremer returned to Belgium and embarked on a research career of his own,
he turned his attention to the investigation of the physiology of the cerebellum [4–6],
and the neural control of muscular tone [7,8]. These studies eventually led him to
sleep research.

Cerveau isolé and encéphale isolé

We do not have any documented explanation why, during the early 1930s, Bremer
decided to redirect his research interests from the regulation of muscular tone to sleep.
One possible explanation is that his interest in sleep was a logical extension of his
interest in muscle tone regulation, since sleep is associated with a profound change
in muscle tone of which Bremer was very well aware (see, for instance, his discussion
of this subject in The Neurophysiological Problem of Sleep, 1966, p. 67) [9]. It could
also be that von Economo’s publication in 1929 [10], on the existence of wake and
sleep centres in the vicinity of the hypothalamus, based on his observations in
degenerative encephalitis lethargica, played a catalysing role. Bremer gave credit to
von Economo’s prophetic vision of sleep regulation mechanism in several of his papers,
but it is equally likely that this was only sheer scientific curiosity. As he himself
explained it: “my motivation (to perform the brain-stem transection studies) was an
unprejudiced curiosity” [11, p. 5]. And he continued: “Having adapted a technique of
decerebration in the cat that left the forebrain in situ after a mesencephalic transection
immediately caudal to the third nerve nuclei, I wished to know what the brain
structures thus separated from the rest of the neuraxis would become functionally, as
one likes to know what happens behind a curtain, the curtain being here the plane of
the intercollicular section”. Thus, although the aim of performing mesencephalic
transections may not have been directly planned to investigate sleep, the resulting
effects on sleep–wake behaviour were so dramatic that they profoundly affected
Bremer, thus influencing his future research career.
Bremer started his brain transection experiments at an opportune time. Neuro-
physiological techniques had greatly improved during the early 1930s with the in-
troduction of Alex Forbes’ electronic amplification. Electromagnetic oscillographs had
replaced the Lippman electrometer and the string galvanometer in electrophysiological
experiments. These newly introduced techniques allowed Lord Adrian and Matthews
 Frédéric Bremer 1892–1982 507

Figure 1. Frédéric Bremer in his Laboratory (Archives of the Université Libre de

Bruxelles).

to confirm Hans Bergers revolutionary discovery of alpha electroencephalographic

activity in humans.
Bremer’s mesencephalic transection produced rather dramatic results. As he himself
described them: “When I contemplated the ocular syndrome of the isolated forebrain
(i.e. the complete immobility of the eyes and the fissured pupils) and its elec-
trocorticogram of 6–10 Hz regular amplitude modulation, its highly synchronous
potential waves and its endless monotony, contrasting with the labile fast elec-
trocorticogram of the alert cat, I had the shock which one experiences in the face of a
new natural phenomenon”.
Bremer attributed the EEG characteristics of the “cerveau isolé cat” to the de-
afferentation of the telencephalon which deprived the brain from the flow of ascending
sensory impulses. This led him to formulate the hypothesis that the immediate cause
of sleep was a more or less complete deafferentation of the telencephalon [12]. Thus,
Bremer concluded this pioneering experiment by saying: “In summary, the complete
deafferentation (the olfactory and optic nerves excepted) of the brain in the cat, by a
transection of the brainstem behind the third nerve which leaves in place the te-
lencephalon, leads immediately to a functional state very similar, if not identical, to
natural and barbiturate sleep. This state, which persists indefinitely, is characterized
by an extreme myosis and the disappearance of olfactory and optic reflexes”. (Trans-
lation from M.K. CR Soc Biol, Cerveau “Isolé” et Physiologie du Sommeil, 1935, p. 1240)
[12].
In the encéphale isolé procedure, the telencephalon was isolated from all incoming
sensory input, except for those carried by the olfactory and visual nerves. These,
508 M. Kerkhofs and P. Lavie

Figure 2. The cerveau isolé experiment (reprinted from the Bull Acad Roy Med Belg
1937; 4: 68–86, with permission).

however were unable to maintain the telencephalon in a waking state. In order to

further evaluate his deafferentation “sleep hypothesis”, Bremer conducted a control
experiment in which he purposely left more sensory pathways intact. Here, he made
the transection at a much lower level—just above the union of the brainstem with the
spinal cord. He termed this procedure the “encéphale isolé” [13]. By performing this
transection, Bremer deprived the brain only of the sensory information arising from
the spinal cord, sparing the input of the sensory cranial nerves.
The impact of the lower brainstem transection on sleep–wake cycles was dramatically
different from that of the midbrain transection [13,14]. In contrast to the cerveau isolé
procedure, which led to a continuous sleep-like behaviour, the encéphale isolé cat
showed an alternation between sleep and wakefulness which did not differ from that
of an intact cat. This strengthened Bremer’s position regarding the cardinal role of
deafferentation of the telecephalon as the cause of sleep. Being unaware of the existence
of the brainstem reticular activating system, to be discovered some 14 years later by
Murozzi and Magoun, Bremer attributed the different impact of the mesencepahlic
and lower brainstem transections on sleep–wake cycles as resulting from their dif-
ferential effects on the flow of sensory input to the brain [15]. Thus, he concluded that
 Frédéric Bremer 1892–1982 509

Figure 3. The encéphale isolé experiment (reprinted from the Bull Acad Roy Med Belg
1937; 4: 68–86).

sleep in mammals is the manifestation of a decrease in cortical “tone”. “This tone”,

he continued, “is principally maintained by the continuous flux of sensory information
to the brain”. The decrease in cortical tone, resulting from the functional deafferentation
of the cerebral neural network, would in itself induce a decrease or suppression of
the continuous influx of neural transmissions from the cerebral cortex and from the
diencephalon, which would affect the state of vigilance by influencing posture and
reflexes (MK translation).
Bremer’s explanation of sleep as caused by a deafferentation of the brain, was very
much in line with that of almost all his predecessors, as well as most of his prominent
contemporaries. We find the roots of the view that sleep was a passive condition
created by the isolation of the brain from the rest of the body, in the writings of the
ancient Greeks. Alcmaeon, who lived in the sixth century BC, claimed that sleep was
caused by the receding of blood from the blood vessels in the skin to the interior parts
of the body. This caused immobility and lack of sensation. Aristotle, in his De Somno
et Vigilia (On Sleeping and Waking) further elaborated the isolation theory of sleep.
The first experimental evidence that supported the critical role of sensory stimuli in
maintaining vigilance was provided by Flourens in 1822. He reported to the Académie
Royale des Sciences his observations on the behaviour of a pigeon after bilateral
ablation of the cerebral hemispheres. Flourens summarized his results as follows: “Just
510 M. Kerkhofs and P. Lavie

imagine an animal which has been condemned to be permanently asleep, one that has
been devoid even of the ability to dream during sleep; this is more or less the situation
of the pigeon in which I had ablated the cerebral hemispheres” (quoted in Moruzzi,
1964, p. 21) [16]. Flourens’ experiments greatly influenced Purkinje (1846) who con-
ceived sleep as the consequence of a mechanical block in the neural conduction along
the internal capsule [17]. Even pre-1953 Kleitman, who should be credited more than
anyone else for the change from the passive to active theories of sleep, believed that
sleep was nothing but the cessation of wakefulness. This emanated from Kleitman’s
position that “it is perhaps not sleep that needs to be explained but wakefulness” [18,
p. 502] and then continued to explain: “Is it not just as correct to say that a person felt
the oncoming of an irresistible attack of sleep as to say that it was utterly impossible
for him to remain awake? Superficially the two expressions would seem to be equivalent,
but their implications are entirely different. The first implies an active onset of sleep,
while the other implies a cessation of an active condition of wakefulness” (18; p. 520).
Kleitman was well aware of Bremer’s experiments. Moreover, he saw Bremer’s ex-
perimental results as supporting his own sleep theory: “Bremer draws the inevitable
conclusion that in sleep there occurs functional deafferentation of the cerebral cortex,
thus giving support, from an entirely new quarter, to the theory that I, and others
before and after me, have proposed as the immediate cause of sleep” [18, p. 51].
Only 37 years after his original publication of the deafferentation theory of sleep
Bremer admitted that “If I had been more anatomically minded, I would have concluded
that between the planes of a spinal section at the level of the first cervical segment
and a mesencephalic transection, lies a neural structure that is necessary for the
maintenance of the waking condition of the diencephalon and the telencephalon.
Instead, I attributed the sleep-like ocular and electroencephalographic characteristics
of the isolated forebrain to its extensive deafferentation, depriving it of the minimal
flow of ascending sensory impulses apparently necessary for the central “tone” of its
neuronal population” [11].

The discovery of the reticular formation

The discovery of Moruzzi and Magoun of the reticular formation in 1949 [19] shed a
new light on Bremer’s data. Bremer himself admitted that “These facts can now be
easily incorporated into the activating reticular concept which Moruzzi and Magoun
deduced in 1949 with their discovery of the arousal effect of electrical stimulation of
the mesencephalic reticular tegmentum and of its similarity to sensory arousal, and
from the demonstration of the contrasting behavioral and electroencephalographic
consequences of selective lemniscal and brain-stem core sections performed by Magoun
and his associates”. There is no doubt however that he immediately grasped the
importance of the new discovery, and embarked on a series of studies to investigate
the relationship between the cortex and the newly discovered brainstem structure.
Together with Terzuolo, he showed that electrical stimulation of a specific area of the
cortex activated the reticular system, which led to an increase in “cortical tonus” and
in behavioural awakening [20–22]. These experiments were the first to show how the
cortex, when receiving a significant sensorial stimulation, could induce awakening of
the entire brain. Later [23], using the encéphale isolé procedure, Bremer showed
that electrical stimulation of the mesencephalic tegmentum evoked inhibition in
the hypnogenic area of the basal forebrain, while transection of the mesencephalic
 Frédéric Bremer 1892–1982 511

tegmentum resulted immediately in an intense and lasting activation of the hyponogenic

area which resembled the activation that accompanied the slow wave sleep of the
intact encéphale isolé. From these observations, Bremer suggested that the maintenance
of sleep “can be explained by an imbalance between the cerebral antagonistic influences
of the arousal and hypnogenic systems”. The arousal and hypnogenic systems re-
ciprocally inhibit each other.
Interestingly, the intense preoptic activation that followed the reticular formation
transection could be partialy inhibited by light. Here, again, Bremer being such a keen
observer made an important observation much ahead of his time, whose importance
was recognized many years later. He described it in the following words: “An
unexpected feature of the preoptic hyperactivity which follows the tegmental tran-
section was that a steady illlumination, even a faint one, resulted regularly in a marked
reduction of the intensity of the continuous spike discharge. . . . The strict limitation
of the photic response to the area on the side of the opposite to the illuminated eye
suggests that it is mediated by direct retino-preoptic fibers” [23, p. 73]. This unexpected
observation led him to conduct a more focused investigation on the effects of light on
the hypnogenic structure [24]. Again using the encéphale isolé preparation, Bremer
showed that photic inhibitory effect of light on the preoptic area was dependent on
activation of the suprachiamatic nucleus (SCN), which he defined as “a mammalian
neuronal structure in the very mesial and caudal portion of the basal preoptic area”.
SCN activation was stimulated by light [24, p. 146]. The reason he decided to investigate
the role of the SCN in mediating the inhibitory effects of light was the observations
that the SCN receives the projection of a predominantly crossed retino-preoptic
pathway, reported 2–3 years before his observations.
Being aware of the pioneering studies of Ibuka and Kawamura (1975) [25], who
showed that lesion of the SCN in rats resulted in a loss of the circadian rhythm
in sleep–wakefulness, he predicted that his observations on the involvement of the
SCN in mediating the photic inhibitory effects on the preoptic area would broaden
the functional significance of the SCN. Bremer was right: subsequent research has
proved that the SCN is the site of the biological clock regulating a variety of
circadian rhythms including that of melatonin secretion, which is intimately linked
with lighting conditions. In his last paper describing experimental results, published
in 1979 [26], Bremer reported that electrical stimulation of the optic nerves fibre
evoked in the mediobasal hypothalamus, slow negative potentials that could be
potentiated by diffuse light and were dependent on the integrity of the basal
preoptic area. In the last sentence of that paper, Bremer hypothesized that
the responsiveness of the hypothalamus to light has functional significance for
neuroendocrine regulations.
Bremer’s views on sleep regulation evolved alongside all the subsequent discoveries
in the field such as rapid eye movement (REM) sleep, the association of different
neurotransmitters with different sleep stages and the emerging complex picture of
neurophysiological control of sleep stages. He published several reviews in which his
conceptions are described [9,11,28–30]. In 1977, at the age of 85, his last famous review
in which he suggested that sleep could result from a “chain of disinhibitory and
inhibitory processes culminating in deactivation of the brainstem arousal system” was
published in the Annals of Neurology [27]. Being such an honest gentleman of science
he credited Sherrington, from whom he borrowed the concepts of inhibition and
disinhibition in the nervous system.
512 M. Kerkhofs and P. Lavie

Figure 4. Frédéric Bremer.

Epilogue

Frédéric Bremer made outstanding contributions to Sleep research and to Neuro-

physiological Sciences in general. His now famous preparations of the cerveau isolé and
encephale isolé greatly contributed to a better understanding of the neurophysiology of
sleep. But what impressed us most while reading his papers and interviewing people
who knew him intimately was his outstanding personality. Frédéric Bremer was a true
gentleman of science with a sense of irresistible curiosity in the search of the scientific
truth. He was an outstanding experimenter who performed all of his experiments by
himself, even after retiring from the University. At the same time he was a modest
person who did not hesitate to openly admit that he wrongly interpreted his ex-
perimental results. His papers should be obligatory readings to our own students as
a model of honest scientific writing.
 Frédéric Bremer 1892–1982 513

Acknowledgments

We are deeply grateful to Professor Antoine Bremer, Frédéric Bremer’s son, who
provided useful information and documents. We also want to thank Professors J. Reuse
and J.E. Desmedt, and the Department of Archives of the Université Libre de Bruxelles,
for their help.

References
1 Bremer F. Aphasie de Wernicke; unique séquelle d’une contusion cérébrale par contre-coup.
Rev Neurol 1920; 27: 133–136.
2 Bremer F. Global aphasia and bilateral apraxia due to an endothelioma compressing the
gyrus supramarginalis. Arch Neurol Psych 1921; 5: 663–669.
3 Bailey P, Bremer F. Experimental diabetes insipidus. Arch Int Med 1921; 28: 773–803.
4 Bremer F. Contribution à l’étude de la physiologie du cervelet: la fonction inhibitrice du
paléocerebellum. Arch Int Physiol 1922; 19: 189–226.
5 Bremer F. Recherches sur la physiologie du cervelet chez le pigeon. C R Soc Biol 1924; 90:
381–384.
6 Bremer F, Ley R. Recherches sur la physiologie du cervelet chez le pigeon. Arch Int Physiol
1927; 28: 58–95.
7 Bremer F. Tonus and contracture of skeletal muscles. Arch Surg 1929; 18: 1463–1490.
8 Bremer F. Contribution à l’étude de la transmission neuro-musculaire. Arch Int Pharmacol
1930; 38: 300–324.
9 Bremer F. The neurophysiological problem of sleep. In: von Grunebaum, Caillois (eds). The
Dream and Human Societies; University of California Press 1966; 53–75.
10 Von Economo C. Encephalitis lethargica. Wien Med Wochenschr 1923; 73: 777–782.
11 Bremer F. Historical development of ideas on sleep. Basic Sleep Mechanisms. New York:
Academic Press, Inc. 1974; 3–11.
12 Bremer F. Cerveau “isolé” et Physiologie du Sommeil. C R Soc Biol 1935; 118: 1235–1241.
13 Bremer F. Nouvelles recherches sur le mécanisme du sommeil. C R Soc Biol 1936; 122: 460–464.
14 Bremer F. Etude oscillographique des activités sensorielles du cortex cérébral. C R Soc Biol
1937; 124: 842–846.
15 Bremer F. L’activité cérébrale au cours du sommeil et de la narcose. Contribution à l’étude
du mécanisme du sommeil. Bull Acad Roy Med Belg 1937; 4: 68–86.
16 Moruzzi G. The historical development of the deafferentation hypothesis of sleep. Proc Am
Philosoph Soc 1964; 108: 19–28.
17 Purkinje JE. Wachen, Schlaf, Traum und verwandte Zustande. In Wagner (ed.). Handworterbuch
der Physiologie. Vol. 3. Braunschweig: Vieweg and Sohn 1846; 412–480.
18 Kleitman N. Sleep and Wakefulness. Chicago: The University of Chicago Press 1939.
19 Moruzzi G, Magoun HW. Brain stem reticular formation and activation of the EEG. EEG
Clin Neurophysiol 1949; 1: 455–473.
20 Bremer F, Terzuolo C. Rôle de l’écorce cérébrale dans le processus du réveil. Arch Internat
Physiol 1952; 60: 228–231.
21 Bremer F, Terzuolo C. Nouvelles recherches sur le processus physiologique du réveil. Arch
Int Physiol 1953; 61: 86–90.
22 Bremer F, Terzuolo C. Contribution à l’étude des mécanismes physiologiques du maintien
de l’activité vigile du cerveau. Interaction de la formation réticulée et de l’écorce cérébrale
dans le processus du réveil. Arch Int Physiol 1954; 62: 157–178.
23 Bremer F. Existence of a mutual tonic inhibitory interaction between the preoptic hypnogenic
structure and the midbrain reticular formation. Brain Res 1975; 96: 71–75.
24 Bremer F. Photic responses of the basal preoptic area in the cat. Brain Res 1976; 115: 145–149.
25 Ibuka N, Kawamura H. Loss of circadian rhythm in sleep-wakefulness cycle in the rat by
suprachiasmatic nucleus lesions. Brain Res 1975; 96: 76–81.
26 Bremer F. Hypothalamic potentials evoked by stimulation of the optic chiasma. Brain Res
1979; 177: 593–597.
514 M. Kerkhofs and P. Lavie

27 Bremer F. Cerebral hypnogenic centers. Ann Neurol 1977; 2: 1–6.

28 Bremer F. La formation réticulaire activatrice et le problème physiologique du sommeil. Une
évaluation. Acta Neurol belg 1972; 72: 73–84.
29 Bremer F. Analyse du mécanisme de l’action hypnogène d’une structure centrale. C R Acad
Sc Paris 1972; 274: 712–714.
30 Bremer F. Tonus cortical et afférences cérébrales. Arch Ital Biol 1973; 3: 462–467.