Actas De Fisiologia

Departamento de Fisiología Instituto de Biología Facultad de Medicina Facultad de Ciencias

Universidad de la República

ACTAS DE FISIOLOGIA

Ricardo A. Velluti, Editor Universidad de la República, Facultad de Medicina Departamento de Fisiología

Volumen 7 2001

Av. Gral. Flores 2125. 11800 Montevideo, Uruguay Fax: (5982) 924 8784 E-mail: [email protected] Versión electrónica: http://www.rau.edu.uy/universidad/medicina/

Montevideo, Uruguay Departamento de Fisiología Instituto de Biología Facultad de Medicina Facultad de Ciencias

Universidad de la República

Comité Editorial

Washington Buño, Madrid Roberto Caldeyro Barcia, Montevideo † Elio García-Austt, Montevideo Omar S. Macadar, Montevideo Héctor Mazzella, Montevideo Juan A. Roig, México José P. Segundo, Los Angeles Omar Trujillo-Cenóz, Montevideo

Comité Editorial Asociado

Luis Acerenza Federico Dajas Francisco R. Morales Cristina Arruti Máximo Drets Elia Núnes Luis Barbeito Ricardo Ehrlich Harol Núñez Lina Bettucci Gustavo Folle Eduardo Osinaga Michel Borde Julio Hernández Marisa Pedemonte Gustavo Brum Magela Laviña Gonzalo Pizarro Ruben Budelli Enrique Lessa Eugenio Prodanov Angel Caputi Gloria Martínez Rafael Radi Alberto Cirio Eduardo Mizraji Julio C. Velluti Jaime M. Monti NAVIGARE NECESSE World Federation of Sleep Research Societies (WFSRS)

Latin American Sleep Society (LASS)

“2001 Sleep Odyssey” Congress. Physiological Basis for Sleep Medicine

Ricardo A. Velluti - Chairman

WFSRS Executive Committee

Ennio Vivaldi – Vice President Adrian R. Morrison – Secretary General and Acting Treasurer Pier Luigi Parmeggiani – Chair of International Congress Program Committee Charles George – Membership Chairman Ron Grunstein – Chair of Education Committee Markku Partinen – Coordinating Secretary

LASS Committee Local Liaison Committee

Sérgio Tufik – President Marisa Pedemonte - Chairperson Rafael Salin-Pascual – Vice President Cecilia Orellana Luiz Menna Barreto – Secretary Cristina Scavone Lu Seabra – Secretary José L. Peña Monica Andersen – Secretary Daniel Lorenzo Francisco Hora José P. Arcos Sérgio Barros Vieira Julio E. Real Jaime M. Monti Sleep Research Society of Uruguay (SRSU)

Ricardo A. Velluti - President Jaime M. Monti – Secretary Cecilia Orellana - Treasurer José P. Arcos Alberto Rodríguez

10th A N N I V E R S A R Y of the WFSRS Cannes 1991 – Punta del Este 2001 Michael H. Chase - Founding President S Y M P O S I A Symposium 1

Computer based methods for sleep research

Chairperson: Thomas Penzel (Germany) Ennio Vivaldi, Universidad de Chile, Santiago, Chile Gerhard Klosch, University of Vienna, Austria Thomas Penzel, Hospital of Philipps-University, Germany Bob Kemp, Westeinde Hospital, The Netherland

Symposium 2

Functional significance of hypothalamic temperature changes during sleep

Chairperson: Pier Luigi Parmeggiani (Italy) Andrea Fuller, University of Witwatersrand, South Africa Pier Luigi Parmeggiani, University of Bologna, Italy Dennis McGinty, University of California Los Angeles, CA, USA Eus Van Someren, Netherlands I. for Brain Research, The Netherland Roberto Amici, University of Bologna, Italy 10 Actas de Fisiología 7, 2001

MODEL-BASED ANALYSIS OF THE NEURONAL SLOW-WAVE GENERATOR

Bob Kemp Sleep Centre, MCH - Westeinde Hospital, Den Haag, The Netherlands

Slow wave bursts do not reflect sleep depth bursts. The same holds for other EEG rhythms: they wax and wane while brain state appears to remain constant. This is confirmed by physiological models of the rhythm-generating neuronal systems: simula- tions by an electronic version of these models produce realistically looking EEG with waxing and waning (bursting) rhythms while all model parameters, including ‘sleep depth’, are kept constant. These findings strongly suggest that the amplitude of EEG rhythms is not only influenced by the underlying brain state. This is even more so since EEG amplitudes are strongly affected by anatomy (skull size and thickness). Still, sleep depth is commonly quantified using slow-wave amplitude. The physiological models contain a neuronal feedback loop that oscillates slow- wave rhythms. Slow-wave bursting is caused by random external activity because at one time, the random activity accidentally builds up an oscillating rhythm in the loop, while later it can accidentally counteract to it. So, amplitude variations reflect external- noise characteristics rather than sleep depth. The models also show which EEG characteristic is directly related to sleep depth. The neuronal feedback loop accumulates the present slow-wave and inserts it into the upcoming EEG. Physiological sleep depth is reflected by the strength of the feedback loop. Therefore, deeper sleep implies a stronger feedback, causing more of the present slow-wave to be inserted in the next EEG segment. So, sleep depth is directly reflected by the fraction (a %) of the current slow wave which is also seen in the next EEG segment. This % is exactly what our analyzer measures [1]. The analyzer output %, dubbed slow-wave microcontinuity, directly reflects physiological sleep depth. Because slow-wave shape is analyzed rather than amplitude, not- sleep-related parameters such as random external activity and skull anatomy have relatively little influence. The same is true for artifacts. Microcontinuity analysis duplicates the effects of temazepam and ultradian rhythmicity as shown by traditional delta (amplitude or power) plots. But these effects are clearly distinguished from the (large) effect of gender on slow wave amplitude: microcontinuity is not affected at all by gender [2]. Microcontinuity shows how temazepam attenuates slow waves while slow-wave sleep and subjective sleep quality remain positively correlated [3]. Thus, slow-wave microcontinuity (shape) more closely reflects physiological sleep depth.

1. Kemp B, Zwinderman AH, Tuk B, Kamphuisen HAC, Oberyé JJL. Analysis of a sleep- dependent neuronal feedback loop: the slow-wave microcontinuity of the EEG. IEEE-BME 47(9), 2000: 1185-1194 2. Mourtazaev MS, Kemp B, Zwinderman AH, Kamphuisen HAC. Age and gender affect different characteristics of slow-waves in the sleep EEG. Sleep 18, 1995:557-564. 3. Kemp B. Temazepam-induced subjective sleep quality, EEG slow waves and sleep spindles. J Sleep Res 9 (suppl 1), 2000: 98. Actas de Fisiología 7, 2001 11

THE POLYGRAPHIC AND CLINICAL DATABASE OF THE SIESTA PROJECT: A NEW APPROACH TO COMPUTER BASED SLEEP ANALYSIS

Gerhard Klösch(1), Bob Kemp(2), Thomas Penzel(3), Alois Schlögl(4), Ernst Trenker(5), Georg Gruber(6), Josef Zeitlhofer(7), Werner M. Herrmann(8), Sari Leena Himanen(9), Dieter Kunz(10), Manel J. Barbanoj(11), Joarchim Röschke(12) and Georg Dorffner(13)

1,7 Department of Neurology, University of Vienna, Austria; 2 Sleep Center, Westeinde Hospital, Den Haag, The Netherlands; 3 Medical Policlinic of Philipps- University Marburg, Germany; 4 Institute for Biomedical Engineering, University of Technology Graz, Austria; 5 Brain Research Institute, University of Vienna, Austria; 6 Department of Psychiatry, University of Vienna, Austria; 8 Laboratory of Clinical Psychophysiology, Department of Psychiatry, Benjamin Franklin Hospital, Free University of Berlin, Germany; 9 Department of Clinical Neurophysiology, Tampere University Hospital, Finland; 10 Department of Psychiatry and Psychotherapy, Lübeck, Germany, 11 Àrea d´Investigació Farmacologica, Institut de Recerca de l´Hospital de la Santa Creu i Sant Pau, UAB, Spain; 12 Department of Psychiatry, University of Mainz, Germany; 13 Austrian Research Institute for Artificial Intelligence, Vienna, Austria

For more than thirty years standard all- night sleep recordings have been based on visual interpretation of one electroencephalograhic (EEG), two eye movements (EOG) and one submental muscle activity (EMG) signal trace. However numerous sleep studies have concluded that sleep recordings cannot be limited to these traditional four channel recordings. A greater number of physiological signals and a variety of subjective and objective psychophysiological parameters including mood and cognitive performance have to be considered. In the SIESTA-project*, performed by a consortium of 16 clinical and bio-engineering centres spread all over Europe, multi-channel sleep recordings from 194 healthy subjects (15 males and 15 females out of each decade, aged 20 to 90 years) and from 98 patients with selected sleep disorders were collected. The standardised recording protocol included 14 days of activity monitoring and sleep diary, two nights of polysomnography and several psychometric tests to assess vigilance, sleepiness and performance. In addition to the standard EEG-, EOG- and EMG-channels, various physiological parameters including O2-saturation, ECG, excursion of chest and abdomen, nasal airflow as well as limb movements were recorded. All biosignals and standard (Rechtschaffen and Kales) sleep scorings were stored in European Data Format (EDF) and are available on CD-ROM. Based on this database new algorithms (e.g. continuous parameters NonREM-, REM-sleep and wakefulness) were developed to overcome the limitations of visual sleep scoring.

(*) The project was supported by a grant from the European Union Biomed-2 BMH4-CT97- 2040. 12 Actas de Fisiología 7, 2001

ANALYSIS OF RESPIRATION, HEART RATE AND OXIMETRY IN SLEEP STUDIES

Thomas Penzel, Ulrich Brandenburg, Sven Rostig and Jörg-Hermann Peter Department of Pneumology, Hospital of Philipps-University, Baldingerstr. 1, Marburg, Germany

Abstract: Computer based sleep analysis focuses on the automatic detection of sleep stages based on sleep EEG, EOG and EMG. Cardiovascular polysomnography includes the recording and analysis of non-EEG signals, such as respiratory and cardiovascular parameters. The required signals are defined by the disorders to be diagnosed, such as disordered breathing during sleep. The gold standard for recording of respiration is airflow with a pneumotachograph and respiratory effort with esophageal pressure - but due to patient discomfort this cannot be used under most clinical conditions. Standard respiratory signals include abdominal and chest wall movements, oronasal airflow, snoring sounds and oxygen saturation. The cardiovascular signals include heart rate and non-invasive blood pressure measurements. New signals such as pulse transit time (PTT) and peripheral arterial tonometry (PAT) can give additional insights for cardiovascular changes during sleep. Automatic analysis of respiratory signals during sleep depends very much on the type of signal under investigation. Movement artifacts and sudden changes of signal amplitude after positional changes during sleep can confound results of analysis. The sum of inductive plethysmography together with oxygen saturation provides reliable signals for subsequent automatic analysis. 93% of all manually detected respiratory events can be detected. The classification into apnea and hypopnea events is less reliable, but reaches sufficient sensitivity and specificity. Automatic analysis of oxygen saturation can always replace a manual counting of oxygen dips. The additional question arises to what extend an automatic analysis of oxygen desaturation may estimate a PSG derived apnea hypopnea index. Depending on the PSG definition of apneas and hypopneas, a sensitivity and specificity of 98% and 88% can be reached. The analysis of heart rate variability is used to investigate sleep stages and sleep disorders with characteristic patterns of heart rate, such as a cyclical variation of heart rate in sleep apnea. Many algorithms have been developed and success in detecting sleep stage and sleep apnea varies to a large extend. The results are promising but the preconditions are not completely clear. For new signals (PTT and PAT) algorithms are still being developed and validation studies are required once algorithms are available. Automatic analysis of respiration and oxygen saturation is reliable and can be used if the algorithms were validated and are disclosed to the user. Automatic analysis of heart rate still needs large validation studies which clarify the conditions for successful application.

Penzel T, Conradt R. Computer based sleep recording and analysis. Sleep Medicine Reviews 2000; 4:131-148. Actas de Fisiología 7, 2001 13

EXCHANGE OF RECORDINGS AND REMOTE INTERACTION VIA THE INTERNET IN SLEEP RESEARCH

Ennio A. Vivaldi, José M. Piquer, Juan Peirano, Jaime Acevedo and Adrián Ocampo-Garcés. Programa Disciplinario de Fisiología y Biofísica, I. C. B. M., Facultad de Medicina and Departamento de Ciencias de la Computación, Facultad de Ciencias Físicas y Matemáticas, Universidad de Chile

The advent of increasingly powerful hardware has enabled a steady progress in data storage and processing capabilities at affordable prices. Now, a technology inherited from the Internet is providing the means for a new concept of a Data Acquisition Lab- oratory based on the connection of instruments to an Intranet. Instruments, such as a polysomnograph, can act as measurement servers for clients anywhere. Usually, the Intranet is the campus or the hospital network using the TCP/IP protocol stack and it can be composed of several physical networks or subnets linked together through rout- ers. The Internet technology connects instrument servers to computers performing the data analysis and to web servers that provide access from outside the Intranet. The TCP/IP protocol stack is a family of software protocols that can run on any physical network enabling their interconnection. Web access, mail, ftp and all other well-known applications become available given a TCP/IP network. The interfacing for remote monitoring, control, storage and analysis can be based on standard web applications extended as needed by ad-hoc programming in Java language. The tasks of a sleep laboratory can be distributed among computers at different locations connected to a Local Area Network. The dedicated computer acting as a Data Acquisition server can be accessed by clients that read the data and store it for further processing. To enable monitoring and control from any computer connected to the Internet, even if it does not have much bandwidth available, a Web server acts as a gateway between the distant PC and the local data acquisition server. Results can be published in a Web page to enable monitoring from any machine connected to Internet by means of standard browsers. The computer that performs the data acquisition and controls the experimental set-up is also accessed through an Internet browser. The cli- ent can thus send commands to this server during a session and change acquisition parameters or experimental conditions. An example of such interaction is the remote implementation of a deprivation protocol.

Research funded by Grants from Fondecyt and Andes-Antorcha. 14 Actas de Fisiología 7, 2001

HYPOTHALAMIC TEMPERATURE CHANGES AND REGULATION OF REM SLEEP OCCURRENCE

Roberto Amici, Matteo Cerri, Rosa Domeniconi, Christine Ann Jones, Marco Luppi, Pier Luigi Parmeggiani, Emanuele Perez, Stefano Venturi and Giovanni Zamboni Department of Human and General Physiology, University of Bologna, Italy

In the rat, two types of REM sleep episode (RSE) have been separated on the basis of the bimodality of the frequency distribution of the duration of the interval (INT) between two consecutive RSEs (1). The occurrence of RSEs separated by short INTs (£ 180s; sequential RSEs), forming clusters of RSEs, have been shown to be much more influenced by changes in ambient conditions than that of RSEs separated by long INTs (> 180s; single RSEs) (1). This suggests the presence of a bimodal regulation of REM sleep occurrence, which could be characterized by different levels of autonomic activity. Since hypothalamic temperature (Thy) has been shown to oscillate as a consequence of state-dependent changes in autonomic activity (2), its time course has been analyzed during either short or long INTs. Sixteen male Sprague-Dawley rats (250g), acclimated to 23±0.5°C and to a 12:12h LD-cycle, were implanted, under deep anaesthesia, with surface electrodes and with a thermistor, placed over the hypothalamus, for EEG and Thy recording, respectively. In the figure, the time course of 0,06 A B b the average value of Thy dur- 0,03 c a ing the INT (L period) are °C shown for six classes of INT 0,00 duration (a-f), moving either d, e, f -0,03 INT duration (s) forward in time (A) from the a, b, c a: 31-60 end of a RSE or backward in -0,06 b: 61-120 f c: 121-180 time (B) from the onset of the d: 181-600 -0,09 d e: 601-960 next RSE. Thy values are ex- e f: > 961 pressed as the difference from -0,12 0 30 60 90 -90 -60 -30 0 those observed either at the end time after REM sleep (s) time before REM sleep (s) (A) of a RSE or at the onset (B) of the next RSE, taken as the reference. The results show that the decrease in Thy which follows the end of a RSE is lower during short (a-c) than during long (d-f) INTs. In the period which immediately precedes a RSE, Thy is higher during short (a-c) than during long (d-f) INTs. ANOVA, which has been carried out on the average Thy values of subsequent 30-s period, has shown that in both cases the difference between short (a- c) and long (d-f) INTs is statistically significant (p <0.01) over the 90-s period. In conclusion, the differeces in the time course of Thy during the INT would reflect the presence of different levels of autonomic activity in short and long INTs.

1) G. Zamboni et al., Arch. Ital. Biol., 1999. 2) P.L. Parmeggiani, Rev. Neurosc., 1995. Actas de Fisiología 7, 2001 15

BRAIN TEMPERATURE REGULATION IN MAMMALS WITH AND WITH- OUT A CAROTID RETE

Andrea Fuller1, Shane K. Maloney2, Claus Jessen3 and Duncan Mitchell1 Departments of Physiology, University of the Witwatersrand1, South Africa, University of Western Australia2, Perth, and University of Giessen, Germany3

Several groups of mammals, notably artiodactyls and felids, employ the carotid rete, a network of intertwining arteries at the base of the brain, as an intracranial heat ex- changer to lower hypothalamic temperature below arterial blood temperature. This phenomenon, termed selective brain cooling (SBC), is most conspicuous in tame, restrained laboratory animals with high body temperatures. However, in unrestrained mammals in which temperatures have been measured remotely by miniature data loggers, there is no fixed relationship between brain (or blood) temperature and the magnitude of SBC. Free-living springbok, eland, wildebeest and oryx employed SBC only sporadically, and never during intense exercise when their body temperatures were highest. In pigs, SBC was greatest during sleep when body temperatures were relatively low, and small- est, or even absent, during hyperthermia and natural fever. The suppression of SBC in these species appears to depend on sympathetically induced vasoconstriction in veins that otherwise carry cool venous blood from the nasal mucosa and respiratory passages to the carotid rete-cavernous sinus complex.

In mammals that do not possess a carotid rete, including humans, baboons, horses and rodents, whether, and how, selective brain cooling occurs remains controversial. In many studies surrogate measures of arterial blood temperature (for example rectal temperature), which usually overestimate arterial blood temperature, have been used. Brain and arterial blood temperatures have been measured simultaneously in horses and they exhibited SBC when exposed to extreme heat stress in the laboratory. However zebras, which are closely related to horses, never employed SBC when free-living in their natural habitat, even at brain temperatures of 41°C. There also is no clear consensus on whether humans exhibit SBC, because brain temperature (near the hypothalamus) and blood temperature never have been measured in a conscious subject. We recently have made these measurements in baboons in the laboratory. Brain temperature was always 0.2 - 0.5°C warmer than blood temperature, even when the baboons were exposed to globe temperatures exceeding 40°C and denied access to drinking water. Although local, direct cooling of the base of the brain may occur in some non-rete species, appre- ciable SBC has been observed only in species possessing a carotid rete. 16 Actas de Fisiología 7, 2001

ROLE OF PREOPTIC HYPOTHALAMIC TEMPERATURE-SENSITIVE NEURONS IN NREM SLEEP CONTROL

Dennis McGinty, Hui Gong, Noor Alam, Natalia Suntsova, and Ron Szymusiak Neurophysiology Research, VAGLAHS, and Depts. of Psychology and Medicine, UCLA, Los Angeles,USA

Local preoptic area (POA) warming increases both concurrent and subsequent NREM sleep. We hypothesized that hypothalamic preoptic area (POA) warm-sensitive and cold-sensitive neurons (WSNs, CSNs) control spontaneous as well as warming-induced sleep. These neuronal subgroups are identified by responses to local warming and cooling. In support of this hypothesis we have found that most WSNs exhibit increased discharge and CSNs exhibit decreased discharge prior to and during spontaneous NREM sleep. POA cooling, which inhibits WSNs and activates CSNs, suppresses spontaneous sleep. We found that activation of WSNs inhibits several wake-promoting neuronal groups in basal forebrain, posterior hypothalamus and midbrain, providing circuitry for regulation of sleep-wake states. Our current research is focused on further delineation of POA sleep-active and warm-sensitive neuronal populations. Electro- physiological studies1 showed that the c-Fos immunostaining method could be used to identify sleep-active neurons. We have confirmed and extended the report2 of a segregated group of sleep-active neurons in the ventrolateral POA (VLPO). We found an additional large segregated group of sleep-active neurons identified by c-Fos immunostaining in the median preoptic nucleus (MnPN)3. Electrophysiological studies have confirmed the presence of sleep-active neurons in MnPN. Using double-labeling methods, we found that more than 80% of sleep-active neurons in MnPN and VLPO co- localize the synthetic enzyme, GAD, of the inhibitory neurotransmitter, GABA. We hypothesized that sites containing sleep-active neurons would also contain sleep-active neurons. We found that sustained moderate ambient heat exposure during waking, which increases subsequent NREM sleep, induced high numbers of c-Fos immunostained neurons in both VLPO and MnPN. A stress control procedure did not have this effect. Thus, WSNs and sleep-active neurons identified by c-Fos are co-localized in the same sites. However, only 24-37% of WSNs co-localized GAD. Thus, neurons activated by sustained warming within VLPO and MnPN appear to be a separate population from sleep-active neurons. Further studies are needed to identify possible interactions of WSNs and sleep-active neurons.

1. Szymusiak et al, Brain Research, 803: 178-188, 1998 2. Sherin et al, Science, 271:216-219,1996 3. Gong et al, Amer. J. Physiol.279:R2079-R2088, 2000. Actas de Fisiología 7, 2001 17

A COMPARATIVE STUDY OF COUNTERCURRENT AND CONDUCTIVE SELECTIVE HYPOTHALAMIC COOLING DURING SLEEP

Pier Luigi Parmeggiani, Adele Azzaroni and Marcella Calasso Department of Human and General Physiology University of Bologna, Italy

Hypothalamic temperature decreases during NREM sleep with respect to quiet waking (QW) and increases during REM sleep with respect to NREM sleep, regardless of exposure to a wide range of ambient temperatures above and below ambient thermal neutrality (cf. 1). Primarily, the mechanisms for hypothalamic cooling underlie such state-dependent temperature variations, since metabolic heat production is coupled to heat clearance by blood flow. The arterial blood supplying the hypothalamus is cooled systemically through the venous blood return from systemic heat exchangers (upper airway mucosa, ear pinna, horn, glabrous skin, tail) to the heart (systemic cooling). In some species (e.g., cat, dog. sheep, goat) the carotid blood supply to the hypothalamus is, moreover, cooled selectively by countercurrent heat exchange between the carotid rete and cranial venous plexuses receiving cool venous blood from the systemic heat exchangers of the head (2). In species lacking the carotid rete (e.g., rabbit, rat) selective hypothalamic cooling is provided by conductive heat exchange between the basal portion of the brain and the cranial venous lakes which drain cool venous blood from the systemic heat exchangers of the head (3). Vertebral artery blood is not thermally conditioned by a selective heat exchange mechanism. Hence, the difference between pontine temperature (systemic cooling only) and hypothalamic temperature (both systemic and selective cooling) is a quantitative indicator of the intensity of selective hypothalamic cooling (1). From QW to NREM sleep, state-dependent changes in tonic vasoconstric- tor sympathetic outflow and the head posture, influencing hydrostatically transmural pressure in the vessels of heat exchangers, affect systemic and selective heat loss. Thus, the latter underlies a thermal feedback influencing selectively hypothalamic thermosensitive neurons during the homeothermic behavioral states of QW and NREM sleep. At REM sleep onset, hypothalamic temperature rises rapidly as a result of a marked decrease in the supply of cooler carotid blood to the circle of Willis and an autoregulatory increase in the supply of warmer vertebral blood to the circle of Willis. Anatomical characteristics of the two species, the cat with a carotid rete and a small ear pinna and the rabbit lacking a carotid rete and with a large ear pinna, underlie different effects of REM sleep-dependent hemodynamic changes on systemic and selective cooling. Namely, systemic cooling is less depressed than selective cooling in the cat whereas the reverse applies to the rabbit.

1) Parmeggiani, P. L. Rev. Neurosci. 6: 353-363, 1995. 2) Hayward,J. N. Baker, M. A. Brain Res. 16: 417-440, 1969. 3) Caputa, M., Kadziela, W. Narebski, J. Acta Neurobiol. Exp. 36: 625-638;1976. 18 Actas de Fisiología 7, 2001

INTERACTION BETWEEN THE CIRCADIAN REGULATION OF TEMPERATURE AND SLEEP

Eus JW Van Someren Netherlands Institute for Brain Research, Amsterdam, The Netherlands

Thermoregulation and sleep are closely related. In the anterior hypothalamic preoptic area (POAH) and other brain areas, subpopulations of warm sensitive neurons (WSNs) increase their firing rate at sleep onset. Experimental warming of the POAH induces a similar accelerated firing and facilitates sleep onset. It has been proposed that brain temperature may be involved in the physiological regulation of sleep1. However, contrary to the experimental findings of increased sleep onset probability with elevated brain temperature, the likeliness of sleep onset in natural conditions is actually minimal at the time when the circadian rhythm in brain temperature peaks. In fact, sleep onset probability increases on the falling limb of the circadian core temperature rhythm. How can this discrepancy of increased sleep with well-controlled experimental brain

34.8 Þ C warming be reconciled with increased sleep with brain 36.9 Þ C core cooling under natural conditions? A likely explana-

34.3 36.7 tion is that many of the locally thermosensitive neu-

proximal skin 36.5 rons also respond, in a similar way, to changes in skin 33.8 temperature. Thus, the very changes in cell membrane 36.3 properties leading to sleep-related alterations in fir- 33.3 mean 36.1 skin ing rate as induced by local warming in experimental

distal skin 35.9 conditions, may be induced by warming of the skin 32.8 35.7 under natural conditions. The importance of skin temperature, especially of the extremities, is supported 32.3 35.5 by a number of observations. First, skin warming in- 0:00 4:00 8:00 0:00 4:00 8:00 0:00

12:00 16:00 20:00 12:00 16:00 20:00 duces a stronger increase in POAH WSN firing than local warming and overrules the effect of local POAH 2,3 temperature . Second, sleep appetitive behavior like lying down is associated with a redistribution of warm hour of the day blood to the extremities, thus increasing their temperature. Third, skin temperature changes predict sleep onset4. The increase of sleep onset probability that occurs on the falling limb of the circadian core temperature rhythm can now be understood. Since the circadian drop in core temperature is mainly due to increased dissipation of body heat, skin temperature is actually elevated when core temperature is falling. When human constant routine temperature data5 are introduced into a model6 on thermosensitive brain areas involved in sleep regulation, a ‘sleep state probabiliy’ curve results that accurately describes habitual sleep onset and offset, the ‘post-lunch’ dip and the ‘forbidden zone’ (see figure).

1. McGinty D & Szymusiak, R (1990) TINS 13: 480-7; 2. Boulant JA et al. (1973)) Am J Physiol 225:1371-4; 3. Boulant JA et al. (1974) J Physiol 240:639-660; 4. Kräuchi K et al. (1999) Nature 401:36-7; 5. Kräuchi K et al. (1994) Am J Physiol 267:R819-29; 6. Van Some- ren EJW (2000) Chron. Int 17:313-54 Symposium 3

The pathophysiology of insomnia

Chairpersons: Hartmut Schulz (Germany) & Michael H. Bonnet (USA) Michael H. Bonnet, Wright State University, OH, USA Irma Gvilia, Beristashvili I. of Physiology, Tbilisi, Georgia Jaime M. Monti, Universidad de la República, Montevideo, Uruguay Hartmut Schulz, Klinikum Erfurt, Germany

Symposium 4

Sleep states and memory processes: Physiological and biochemical mechanisms

Chairpersons: Carlyle Smith (Canada) & Subimal Datta (USA) Carlyle Smith, Trent Univesrsity, Ontario, Canada Subimal Datta, Boston University, MA, USA Gina Poe, Washinton State University, WA, USA Ted Abel, University of Pennsylvania, PA, USA 20 Actas de Fisiología 7, 2001

PHYSIOLOGICAL CHANGES IN PATIENTS WITH INSOMNIA*

Michael H. Bonnet and Donna L. Arand Dayton Department of Veterans Affairs Medical Center, Wright State University, and Kettering Medical Center, Dayton, USA

In recent years, several studies have shown that patients with primary (psychophysiological) insomnia have physiological differences from matched control subjects. Those differences include decreased objective sleepiness (MSLT), increased 24-hour metabolic rate (VO2), increased heart rate, and increased heart rate variability indices consistent with increased sympathetic nervous system tone and decreased parasympathetic nervous system tone. These results are consistent with the hypothesis that patients with insomnia suffer from a primary disorder of hyperarousal. Other studies from our lab have examined the impact of disturbed sleep, with and without a hyperarousal component, upon daytime alertness and other common symptoms in insomnia patients. One central question in patients is whether their poor sleep produces daytime dysphoria and hyperarousal or whether hyperarousal produces poor sleep and daytime dysphoria. In a yoke-control experiment, we recorded sleep in a group of insomnia patients and produced the same sleep patterns in a group of normal sleepers (by induced awakenings and arousals) for one week. This ‘insomnia’ sleep pattern in normals produced symptoms of mild partial sleep deprivation but no indication of the typical hyperarousal, mood change, or continuing sleep difficulty seen in patients. In a separate study, we induced a state of chronic hyperarousal in normals by administering caffeine 400 mg TID for one week. These normal sleepers did display hyperarousal (i.e., increased metabolic rate), mood and personality change, and increased latencies on MSLT in addition to poor sleep at night – all typically seen in patients with primary insomnia. These data imply that hyperarousal, independent of history or personality, produces the symptoms seen in patients with chronic insomnia but that poor sleep alone does not produce such symptoms. This conclusion was further supported by the finding that patients with sleep state misperception, despite normal sleep time, had elevated metabolic rate. In a final study, patients with primary insomnia were experimentally given much worse sleep than normal for a week. This study tested the hypothesis that if sleep were the primary factor, the exacerbation of poor sleep would also exacerbate other symptoms. Despite significantly reduced sleep, the patients had decreased metabolic rate during the day, normalized sleep latency on the MSLT (11.1 min), and no change in body temperature, anxiety or personality measures. Again, it was concluded that sleep amounts, even in insomnia patients, were not directly related to secondary symptoms.

* Supported by a Merit Review Grant from the Department of Veterans Affairs and the Sleep- Wake Disorders Research Institute Actas de Fisiología 7, 2001 21

THE GUINEA PIG AS A NATURAL MODEL OF INSOMNIA

Irma Gvilia, Nato Darchia and Tengiz Oniani Dept. of Neurobiology of Sleep-Wakefulness Cycle, I. Beritashvili Institute of Physiology, Tbilisi, Georgia

Markedly unstable cyclic organization of sleep-wakefulness cycle (SWC), prolonged latency of sleep, its insignificant amount, and strongly fragmented nature due to a high index of arousal, inability to quickly habituate to a new environment, dissociation between EEG and behavior sleep parameters, anxiety and high responsiveness to external stimuli within 24h are the main characteristics of guinea pig’s regular SWC. The peculiar organization of SWC in this species has attracted our particular attention in the aspect that the typical characteristics of SWC in guinea pig are well-correlated with those objective indices, which are used in a clinical practice when diagnosing insomnia. Features of guinea pig’s SWC organization, effects of the animals’ adaptation to the experimental set, effects of total sleep deprivation and selective deprivation of paradoxical sleep, effectiveness of benzodiazepines, as “traditional therapy” for insomnia has been studied. According to data obtained both non-pharmacological and pharmacological interventions were successful in poor sleep of guinea pig and resulted in significant increase of total sleep duration, its intensification (in case of non-pharmacological influences), stabilization of cyclic organization of the animals’ SWC on the whole: consecutive alternation of different phases in the SWC acquired more regular and ordered character. In a course of the animals’ adaptation to the experimental set their sleep rendered more quiet by all behavioral parameters up to appearance of a sleep ritual. Evaluating the neurobiology of guinea pig’s SWC it is important to consider ecolo- gy of the species: this is an animal of prey and its safety in natural conditions is not secured by reliable shelter. Most of the time of day and night the animal has to be in a state of readiness to flee. So, the system of wakefulness (WS) in guinea pig has to be characterized by a strong instinctive control over sleep system (SS) partly suppressing its functioning. This factor and the obtained data suggest that effectiveness both non- pharmacological and pharmacological influences on poor sleep of guinea pig within the experiments seems to be caused by reducing the activity of the WS. As a result of partial releasing from under the WS control, the SS starts to work in optimal mode recovering balance between the two systems. Generalization of the aforesaid makes us think that guinea pig can be considered as a natural model for studying chronic insomnia. Besides, the animal’s SWC, standardized according to the measured indices of sleep and wakefulness can be used for evaluating of sleep-inducing effects of hypnotic drugs. It may contribute to the establishment of a safer and more effective treatment for chronic insomnia. 22 Actas de Fisiología 7, 2001

THE BRAINSTEM, SEROTONIN AND SLEEP - THE ROLE OF 5-HT1A RECEPTORS

1Jaime M. Monti and 2Daniel Monti 1Department of Pharmacology and Therapeutics, Clinics Hospital, Montevideo, Uruguay and 2Western Psychiatric Institute, University of Pittsburgh Medical Center. Pittsburgh, USA

The reciprocal interaction hypothesis of REM sleep (REMS) generation, is the one that best corresponds to the experimental evidence concerning REMS. It identifies cholinergic neurons in the laterodorsal (LDT) and the pedunculopontine (PPT) tegmental nuclei as promoting REMS, and their inhibition by serotonergic (5-HT) afferents from the dorsal raphe nucleus (DRN) and by noradrenergic afferents from the locus coeruleus (LC). It is currently accepted that the REMS induction region of the medial pontine reticular formation (mPRF) predominantly includes glutamatergic neurons, which are in turn activated by acetylcholine. In addition to cholinergic projections from the PPT and the LDT nuclei, the REMS induction zone receives serotonergic, noradrenergic and histaminergic inputs. In general, binding densities levels of serotonin (5-HT1-2) receptors in the LDT/PPT nuclei and the mPRF vary from low to medium.

The 5-HT1A receptor is located on the soma and dendrites (somatodendritic autoreceptor) of 5-HT neurons, and at postsynaptic sites. Stimulation of the somatodendritic

5-HT1A receptor inhibits the firing rate of serotonergic neurons, whereas activation of the postsynaptic receptor induces inhibitory responses on target structures. Currently, a number of ligands are available that show high affinity for the 5-HT1A-receptor binding site in rat and cat brain. They include 8-OHDPAT and flesinoxan, which are full agonists at pre- and postsynaptic sites, and WAY 100635 which acts as an antagonist at these receptors. Systemic injection of 8-OHDPAT or flesinoxan consistently reduces slow wave sleep (SWS) and REMS, and increases waking (W) in the rat. On the other hand, direct microinjection of 8-OHDPAT or flesinoxan into the dorsal raphe nucleus (DRN), or microdialysis perfusion of 8-OHDPAT into the same structure, significantly increases REMS. In contrast, administration of WAY 100635 into the DRN markedly reduces REMS occurrence. It has been determined that extracellular levels of serotonin in the PPT and the LDT are highest during W, and progressively lower during SWS and REMS. In this context, it has been found that microinjection of either serotonin, 8-OHDPAT or flesinoxan into the LDT reduces REMS in the cat and/or the rat. On the other hand, direct infusion of WAY 100635 into the LDT significantly increases REMS in the rat. Thus, presently available evidence strongly supports a regulatory role for serotonin in REMS occurrence: 1) activation of the somatodendritic 5-HT1A receptor in the DRN results in a reduction of serotonin neuron firing rate, and an increase of REMS; 2) activation of the postsynaptic receptor in the PPT/LDT nuclei induces the opposite effects.

- J.M. Monti and D. Monti. Role of dorsal raphe nucleus serotonin 5-HT1A receptor in the regulation of REM sleep. Life Sciences 66: 1999-2012 (2000). Actas de Fisiología 7, 2001 23

PSYCHOBIOLOGIC FACTORS: CAUSES OR CONSEQUENCES OF INSOMNIA?

Hartmut Schulz Helios Klinikum Erfurt, Clinic of Neurology, Erfurt, Germany

Insufficient sleep during the night and the relative inability to nap during daytime suggest that the propensity to sleep is reduced in insomniac patients throughout 24 hours. Based on the two-process model of sleep [1] one would predict that the monotonic increase of factor S during daytime is attenuated in insomnia. We have modelled this situation with a variant of the two-process model [2], with the assumption that sleep propensity is the direct product of the drive for slow wave activity (factor S; renewal function), and REM sleep (factor R; circadian function). In normal, undisturbed conditions, the multiplication of the two factors S and R results in a sleep propensity function (SPF) with (a) a maximum during the night, (b) a secondary peak in the afternoon, (c) a morning nadir after the end of sleep, and (d) an evening nadir. All these typical aspects of the SPF are experimentally substantiated [3]. In the simulation we assumed that in insomniac patients the drive for S and R is reduced. As a consequence the SPF is attenuated with decreased sleep propensity during bed rest at night and a marked attenuation of the secondary peak in the afternoon. The model predictions correspond to the finding of increased sleep latencies in MSLTs of insomniac patients. Yet, the daytime part of the SPF in insomniacs has not been determined in systematic nap studies. According to the simulation, based on the SxR model, the clinical features of insomnia with difficulty falling asleep, repeated awakenings, shallow sleep and early-morning awakening would be a consequence of reduced drives of factors S and R. Predisposing conditions for such a disturbance could be genetic, organic and environmental factors. There are convincing data showing that insomniacs as a group display signs of increased physiological and cognitive activation (Dr. Bonnet’s contribution to this symposium). Whether a deficit of slow wave sleep, as suggested by Gaillard [4] is another essential factors needs to be investigated in longitudinal studies. Likewise, data on the circadian distribution of REM sleep propensity in insomnia are lacking. The deficit in knowledge on basic aspects of sleep regulation in insomnia has different consequences: 1. For many features of insomnia it is unclear whether they are causally linked to the disease process, or whether they represent mere consequences of the disorder. 2. The development of effective behavioral and pharmacological treatments for insomnia depends essentially on an adequate understanding of the psychobiological basis of the disorder.

References: [1] Borbély AA. A two process model of sleep regulation. Hum Neurobiol 1982; 1:195-204. [2] Schulz H, Bes FW, Jobert M. Modelling sleep propensity and sleep disturbances. In Meier-Ewert K, Okawa, M. (eds.) Sleep-Wake Disorders. Plenum Press, New York, 1998, pp 11-26. [3] Broughton R. Important underemphasized aspects of sleep onset. In Ogilvie RD, Harsh JR (eds) Sleep Onset. Normal and abnormal Processes. American Psychological Association. Washington DC, 1994, pp19-35. [4] Gaillard J-M. Is insomnia a disease of slow- wave sleep? Eur Neurol 1976; 14:473-484. 24 Actas de Fisiología 7, 2001

SLEEP AND MEMORY: A MOLECULAR AND GENETIC ANALYSIS

Ted Abel University of Pennsylvania, Dept. Biology, Philadelphia, USA

The goal of our research is to elucidate the molecular machinery underlying sleep and to define the role of sleep in the consolidation of long-term memory. Sleep is involved in many physiological processes, and disturbances in sleep are a hallmark of many diseases, including neurodegenerative disorders such as Alzheimer’s disease, in which memory deficits are prominent. We are interested in bridging past behavioral studies, which have shown that sleep is needed for the consolidation of memory, with current knowledge about the molecular mechanisms underlying memory storage. Our recent work has used genetic and pharmacological approaches in mice to demonstrate that protein kinase A (PKA) activity is critically important for long-term memory storage. There are striking parallels between the effects of sleep deprivation and the effects of pharmacological manipulation of the PKA signaling pathway on the consolidation of memory. Both sleep deprivation and inhibition of PKA disrupt the consolidation of memory only at discrete times following training and these times vary depending on the strength of the training protocol. Based on this, we are examining the role of the PKA signaling pathway in the behavioral effects of sleep deprivation on hippocampus- dependent tasks in mice using genetic, pharmacological, electrophysiological and behavioral approaches. Actas de Fisiología 7, 2001 25

ANATOMICAL, PHYSIOLOGICAL AND BEHAVIORAL EVIDENCE TO SUPPORT THE HYPOTHESIS THAT BRAINSTEM PHASIC P-WAVE GENERATING CELLS ARE INVOLVED IN SLEEP-DEPENDENT LEARNING AND MEMORY PROCESSING

Subimal Datta Sleep Research Laboratory, Department of Psychiatry, Boston University School of Medicine, Boston, USA

There is now considerable evidence that sleep stages are involved with memory processing and improvement of learning in mammals. To account for this well-documented sleep-dependent memory processing and learning, the following hypothesis has been suggested. During wakefulness, acquired information is stored temporarily in the neocortex. During subsequent slow-wave sleep, this temporarily stored information is downloaded into the hippocampus, amygdala, and some parahippocampal areas. During tS-R (transition between slow-wave sleep and REM sleep) and REM sleep, activation of phasic P-wave generating cells reactivates the hippocampus, amygdala, and some parahippocampal areas to reprocess the temporarily stored information that was downloaded from the neocortex. During REM sleep, this reprocessed information in the hippocampus, amygdala, and parahippocampal areas is sent back to the neocortex for permanent storage. The present study provides anatomical, physiological, and behavioral evidence that the activation of phasic pontine P-wave generating cells is critical for memory processing and improvement of learning behavior in mammals. A series of investigations has yielded support for the above hypothesis. First, we recorded single cell activity patterns of P-wave generating cells in behaving cats (1). During tS-R and REM sleep, P-wave generating cells fired high-frequency bursts of 3-5 spikes (intra-burst frequency >500 Hz) along with tonically increased single spikes. Next, we mapped anatomical efferent connections of the functionally identified P-wave generator of the rat (2). These studies demonstrated that the P-wave generating cells send anatomical projections to the hippocampus, amygdala, parahippocampal areas, and other forebrain structures known to be involved in memory processing. Third, we have chemically (carbachol) stimulated the P-wave generator in the cat (3). Chemical microstimulation of the P-wave generator increased P-wave density and total amount of REM sleep which both remained elevated for 7-10 days. If long-term potentiation of the synaptic efficacy is known to be a physiological correlate of learning and memory, the combined results of the single cell activity, anatomical pathway tracing, and behavioral long-term potentiation studies together support the hypothesis that P- wave generating cells are involved in memory processing. Finally, recent behavioral studies in the rat have demonstrated that the level of improvement of learning in the retrial session depends positively on the percentage of the P-wave density increase in between the first and third episodes of REM sleep immediately after the first learning session of learning trials (4). These behavioral results together with anatomical and physiological results support the hypothesis that the activation of the P-wave generating cells is critical for the reactivation of the hippocampus, amygdala, and parahippocampal areas for memory processing and long-term storage of learned information.

1. Datta, S. and Hobson, J.A. (1994). Neuronal activity in the caudolateral peribrachial pons: Relationship to PGO waves and rapid eye movements. J. Neurophysiol. 71:95-109. 2. Datta, S., Siwek, D.F., Patterson, E.H. and Cipolloni, P.B. (1998). Localization of pontine PGO wave generation sites and their anatomical projections in the rat. Synapse 30:409-423. 3. Datta, S., Calvo, J.M., Quattrochi, J.J., and Hobson, J.A. (1992). Cholinergic microstimulation of the peribrachial nucleus in the cat. I. Immediate and prolonged increases in PGO waves. Arch. Ital. Biol. 130:263-284. 4. Datta, S. (2000). Avoidance task training potentiates phasic P-wave density in the rat: A mechanism for sleep-dependent plasticity. J. Neurosci. 15:8607-8613. Research was supported by NIH research grants NS34004 and MH59839. 26 Actas de Fisiología 7, 2001

HIPPOCAMPAL NEURAL ACTIVITY DURING REM SLEEP FOR LEARNING AND MEMORY REFINEMENT: SHORT AND LONG TERM EFFECTS OF PARTIAL REPEATED REM DEPRIVATION.

Gina R. Poe, Carena M. Thompson, Brett T. Riley, and Michael J. Tysor Washington State University College of Veterinary Medicine, Pullman, WA, USA

The hippocampus is a central assembly site in the brain for associative memories, for example, those formed when learning new environments. Recordings from the hippocampus reveal neural activity patterns consistent with learning (long term potentiation of synapses) during both waking learning sessions and subsequent recorded REM sleep periods. Neural activity patterning during REM sleep after learning a novel place is also theta (rhythmical slow activity, 5-10 Hz) – associated and consistent with strength- ening synapses, as is nearly identical to that of prior waking learning session. The similarities in neural firing pattern during waking and REM sleep implies that, for the hippocampus, the benefit of REM sleep for memory enhancement may be replaced by equal time in waking rehearsal. Indeed, the behavioral effects of REM sleep deprivation on performance of memory-dependent tasks, though significant, are short-lived given repeated learning sessions. However, physiologic and now behavioral data also support the possibility that REM sleep may be the only state in which the refinement of synapses occur, a process that in the long term may prove essential to efficient learning in a structure of limited synaptic, and thus memory-encoding capacity such as the hippocampus. With repeated experience the learning-consistent activity pattern in REM sleep reverses. After 5 or 6 days, hippocampal neural activity during REM sleep begins to shift in relation to theta to a pattern consistent with pruning away (depotentiating) the synapses of cells encoding familiar environments. This time course is consistent with the consolidation of place memory to long-term stores outside the hippocampus – after which the hippocampus is no longer necessary for the recognition of place. This spontaneously occurring depotentiation pattern has not been described in any other behavioral state. Depotentiaion is theoretically necessary to prevent an overabundance of potentiated synapses, the saturation of which would obscure the differences between stimuli and prevent further learning. Depotentiation may also be necessary for refining learning to not only encode what is, but also to prune back what is not. This type of knowledge would enable an animal to, for example, cease checking sites never or no longer baited with food, thus making performance of food searching tasks most efficient. The results of two different behavioral studies suggest that the effect of REM sleep deprivation over many days subtly but significantly reduces performance efficiency relative to normally sleeping animals. Thus, although additional waking practice sessions initially level the playing field between REM sleep deprived and normally sleeping animals, the increase in efficiency of performance after several days of a learning task is absent in animals who are REM sleep deprived for as little as 4 hours per day. Thus, depotentiation, and it’s role in pruning synapses, may constitute a unique function of REM sleep for memory and learning. Actas de Fisiología 7, 2001 27

CHANGES IN DENSITY OF REM SLEEP FOLLOWING ACQUISITION OF COGNITIVE PROCEDURAL TASKS IN HUMANS

Carlyle T Smith Dept. of Psychology, Trent University, Peterborough, Ontario K9J 7B8, Canada

It has been reported that increases in REM sleep parameters follow the successful acquisition of cognitive procedural tasks in humans. These increases have been reported to take the form of either an increase in number of minutes of REM sleep or an increase in the density of REM sleep with no changes in the length of the REM sleep periods. Unfortunately, some earlier studies, using human subjects did not report on changes in REM densities (1). In animal studies, with one exception, no actual rapid eye movement densities have been measured (2). Our earliest studies indicated that while the REM densities increased following learning, the task was a set of exams and measures of sleep were taken 3-5 days after the last exam was written (3). The present experiments examine the effects of two cognitive procedural tasks on the changes in sleep parameters on the same night as task acquisition. After an acclimatization and baseline night of sleep, subjects were asked to learn the Tower of Hanoi and Mirror Trace tasks. The subsequent night of sleep showed marked increases in the REM densities of the test subjects compared to the non-learning controls. These increases in number of rapid eye movements suggest that the brain stem mechanisms involved in generating PGO waves in animals are also at work in humans and appear to be involved with the processing or consolidation of newly learned material into long term memory.

References: 1) Smith, C. (2001). Sleep states and memory processes in humans: procedural vs. declarative memory systems. Sleep Med. Rev. (In press). 2) Smith, C. and Lapp, L. (1986). Prolonged increases in both PS and number of REMs following a shuttle avoidance task. Physiol. Behav. 36, 1053 -1057. 3) Smith, C. and Lapp, L. (1993). Increases in number of REMs and REM density in humans following an intensive learning period. Sleep, 14, 325 - 330.

Research Supported by Natural Sciences and Engineering Research Council (NSERC) and Canadian Institutes of Health Research (CIHR). Symposium 5

The role of orexin in the physiology of natural sleep & wakefulness control

Chairpersons: Robert McCarley (USA) & Robert E. Strecker (USA) Louis de Lecea, Scripps Institute, La Jolla, CA, USA Neil Upton, GlaxoSmithKline, Essex, UK Jerome Siegel, University of California Los Angeles, CA, USA Mahesh Thakkar, Harvard University, Boston, MA, USA Masashi Yanagisawa, UTSW, Japan

Symposium 6

Melatonin

Chairperson: Daniel P. Cardinali (Argentina) José Cipolla-Neto, Universidad de Sao Paulo, SP, Brazil Jaime M. Monti, Universidad de la República, Montevideo, Uruguay M. de Lourdes Ventura Seabra, Universidad de Sao Paulo, SP, Brazil Daniel P. Cardinali, Universidad de Buenos Aires, Argentina 30 Actas de Fisiología 7, 2001

ACTIVATION OF HYPOCRETIN NEURONS AND SLEEP

Luis de Lecea and Veronique Fabre Dept. of Molecular Biology, MB-10. The Scripps Research Institute. 10550 N. Torrey Pines Rd., La Jolla CA, USA

The hypocretins (hcrts), also known as orexins, are peptides derived from the same precursor and have recently been associated with the regulation of REM sleep and the pathophysiology of narcolepsy. In particular, hcrt-deficient mice display narcolepsy- like attacks 1, and human narcoleptic patients have dramatically reduced levels of hcrt- producing neurons 2,3. Hypocretin producing cells are restricted to ~3000 cell bodies in the lateral hypothalamus but their projections are widespread throughout the brain. One of the main hcrt projection areas is the locus coeruleus (LC) 4, the main source of noradrenergic neurons in the brain. Local administration of hcrt1 into the LC area produced significant changes in the amounts of REM sleep (F(4,24)=21.8, p<0.001), SWS2 (F(4,24)=8.4, p<0.001) and wakefulness (F(4,24)=16.9, p<0.001). Injection of hcrt1 (25pmol) induced profound changes characterized by a decrease in REM sleep amounts (-95 % of control levels; p<0.05; Fisher (LSD) and SWS2 amounts (-33 %; p<0.05) associated with an increase in wakefulness (+62%; p<0.05) during the 0-4h period post-injection as compared to baseline. The effect was likely mediated by hcrtr1 receptors in noradrenergic cells of the LC. These results suggest that hcrts activate noradrenergic neurons that modulate the boundaries between sleep states. To determine the effects of general activation of hcrt neurons in vivo we have generated transgenic mice that express the A subunit of cholera toxin under the control of a mouse hcrt promoter. Expression of cholera toxin in neurons constitutively activates adenylyl cyclase and results in increased levels of cAMP, mimicking constitutive ligand- induced activation of hcrt cells. A fragment containing 7 Kb of the mouse hcrt promoter directs accurate cell-specific expression of the transgene. The consequences of constitutive activation of hcrt cells on the noradrenergic system and the regulation of REM sleep will be discussed.

1 Chemelli, R. M., J. T. Willie, et al. (1999). ¿Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation? Cell 98(4): 437-51. 2 Peyron, C., J. Faraco, et al. (2000). ¿A mutation in a case of early onset narcolepsy and a generalized absence of hypocretin peptides in human narcoleptic brains? Nat Med 6(9): 991- 997. 3 Thannickal, T. C., R. Y. Moore, et al. (2000). ¿Reduced number of hypocretin neurons in human narcolepsy? Neuron 27(3): 469-74. Actas de Fisiología 7, 2001 31

HYPOCRETIN (OREXIN) AND MOTOR ACTIVITY

J.M. Siegel, M.F. Wu, J. John and N. Maidment VAGLAHS and UCLA Dept. of Psychiatry

Hypocretin (orexin) is a recently discovered hypothalamic neuropeptide that has been implicated in the etiology of narcolepsy. The normal behavioral function of this peptide is unclear, although a number of studies have suggested a role in food intake regulation. We measured hypocretin-1 (Hcrt-1) / orexin-A in the cerebrospinal fluid (CSF) of dogs during a variety of behaviors. We found that forty-eight hr of food deprivation produced no significant change in Hcrt-1 levels. Feeding after the deprivation procedure also produced no significant change in Hcrt-1 levels. In contrast, 24 hours of sleep deprivation produced a >75 % increase in the CSF hypocretin level in both normal and genetically narcoleptic dogs. However, the amount of increase was correlated with the level of motor activity during the sleep deprivation procedure, not the amount of sleep loss. A 30- to 120-minute period of motor activity in the same dogs produced a 30-50% increase in Hcrt-1 levels, with the magnitude of the increase being highly correlated with the level of motor activity. These results demonstrate that Hcrt release is closely linked to motor activity. The strong correlation between motor activity and Hcrt-1 release may explain many of the previously reported behavioral and physiological relations ascribed to the Hcrt system. 32 Actas de Fisiología 7, 2001

OREXINS: ACTIVITY REGULATORS

Mahesh M Thakkar and Robert E Strecker Psychiatry (151C), Harvard Medical School/VA Medical Center, Brockton MA, USA

The orexins/hypocretins are recently discovered neuropeptides implicated in the regulation of sleep-wakefulness. Cell bodies containing orexins are exclusively localized in and around the perifornical region in the lateral hypothalamus with dense projections to distinct forebrain and brainstem sites that are implicated in the control of sleep- wakefulness, including the cholinergic basal forebrain, the noradrenergic locus coeruleus, the serotonergic dorsal raphe nucleus , the cholinergic mesopontine tegmentum and the pontine reticular formation (PRF). Inherited canine narcolepsy is caused by an abnormal oexin (type II) receptor gene whereas targeted disruption of the orexin gene in mice produced narcolepsy/cataplexy like symptoms. Intracerebroventricular administration of orexin increased wakefulness and locomotor activity in rodents. These combined reports suggested that endogenous orexins suppressed REM sleep and induced locomotor activity , whereas the absence, or reduction, of orexin peptide or orexin receptors enhanced REM sleep and REM sleep-related phenomena including muscle atonia. Where do orexins act to control REM sleep and wakefulness. We propose that orexin may regulates REM sleep and muscle atonia via it selective action on brainstem site, whereas it may regulates wakefulness via it action on the cholinergic neurons of the basal forebrain. Using rat as our animal model, we conducted two studies to test our hypothesis. In the first study, we perfused antisense oligodeoxynucleotide against mRNA of orexin type II receptor, locally in the subcoerulean PRF, using the microdialysis probe. We chose orexin type II receptor because canine narcolepsy/cataplexy involves abnormalities of the orexin II receptor. Our choice of target site was based on the fact that the subcoerulean PRF is the key regulator of REM sleep and muscle atonia. In the second study, we perfused orexin-A in the cholinergic basal forebrain and determined its effect on sleep-wake cycle during the light cycle (inactive period). The cholinergic basal forebrain is a known regulator of wakefulness and cortical activation. The result of our studies revealed that 1) microdialysis of antisense oligonucleotide in the subcoerulean PRF induced cataplexy and increased REM sleep. 2) Mi- crodialysis perfusion of orexin-A in the basal forebrain produced a dose dependent increase in wakefulness and a dose dependent decrease in nonREM sleep. The data from our studies support the hypotheses that 1) orexins regulate REM sleep and muscle atonia via selective action on brainstem neurons, primarily those in the subcoerulean PRF with mediation by the orexin II receptor; and that orexins influences wakefulness through the cholinergic basal forebrain possibly with mediation by the orexin II receptor. Thus, orexins regulate waking and locomotion. Actas de Fisiología 7, 2001 33

ROLE OF THE OREXIN LIGAND-RECEPTOR SYSTEM IN MODULATING AROUSAL AND BEHAVIOUR: A PHARMACOLOGICAL CHARACTERISATION

Neil Upton Neurology CEDD, GlaxoSmithKline Pharmaceuticals, New Frontiers Science Park, Third Avenue, Harlow, Essex, CM19 5AW, UK

Orexin-A and orexin-B are novel neuropeptides initially isolated from hypothalamic extracts but now known to be present in fibres projecting to sites throughout the brain and spinal cord. The orexins bind with high affinity and activate two G-protein coupled receptors, orexin-1 (OX1) and orexin-2 (OX2) [Sakurai et al 1998 Cell 92:573- 585], which are also broadly distributed in the CNS thereby suggesting that the physiological role of the orexin ligand-receptor system is likely to be complex. Indeed, centrally administered orexins have been shown to evoke a number of behavioural changes including feeding, grooming, locomotion and arousal [Hagan et al 1999 PNAS 96:10911-10916]. In particular, there is growing evidence that the orexinergic system is an important modulator of the sleep-wake cycle of animals and man. For example, orexin-A (1, 10 or 30 ug/rat intracerebroventricularly) produces a dose-dependent increase in the time rats spend awake, largely at the expense of a marked reduction in paradoxical sleep (PS) and also deep slow wave sleep at the highest dose [Piper et al 2000 Eur. J. Neurosci. 12:726-730]. In addition, prepro-orexin knockout mice exhibit a narcolepsy phenotype [Chemelli et al 1999 Cell 98:437-451] and the sleep disorder canine narcolepsy is caused by a mutation in the OX2 receptor gene [Lin et al 1999 Cell 98:365-376]. Even more strikingly, researchers have now shown a dramatic 85- 95% loss of orexins in the hypothalamus and associated projection areas of narcoleptic subjects (Peyron et al 2000 Nature Medicine 6:991-997; Thannickal et al 2000 Neuron 27:469-474). Taken as a whole, available data suggest that deficient orexin neurotransmission is a key factor leading to narcolepsy. Although there are now numerous reports of diverse central actions of the orexins, the pharmacology of the orexin receptor subtype(s) which underlie these effects is still in its infancy. However, studies in vivo with the recently identified selective OX1 receptor antagonists SB-334867 (OX1 pKb = 7.4; OX2 pKb = 5.7) and SB-408124 (OX1 pKb = 7.8; OX2 pKb = 5.8), have just begun to clarify the situation. Thus, orexin-A evoked grooming (Duxon et al 2001 Psychopharmacology 153:203-209) and orexin-B induced hyperactivity (Jones et al 2001 Psychopharmacology 153:210-218) in the rat both appear to be primarily mediated by OX1 receptors, with a downstream 5- HT2C receptor involvement in the former response. Furthermore, antagonism of the delay in PS induction produced by orexin-A in the rat by SB-334867 also implicates OX1 receptors in the modulation of sleep. 34 Actas de Fisiología 7, 2001

MELATONIN AS A NEUROPROTECTIVE CHRONOBIOTIC AGENT

Daniel P. Cardinali, Luis I. Brusco and Analía M. Furio. Departament of Physiology, Faculty of Medicine, University of Buenos Aires, Argentina.

Melatonin is synthesized and secreted during the dark period of the light/dark cycle. The rhythmic nocturnal melatonin secretion is directly generated by the circadian clock, located in mammals in the suprachiasmatic nuclei (SCN), and is entrained to a 24-hour period by the light-dark cycle. The periodic secretion of melatonin may be used as a circadian mediator to any system than can “read” the message. In addition, direct effects of the hormone on the SCN could explain some of melatonin effects on the circadian system. Duration of melatonin nocturnal secretion is directly proportional to the length of the night and it has experimentally been demonstrated to be the critical parameter for photoperiod integration. Melatonin administration synchronizes the sleep- wake cycle in blind people and in individuals suffering from jet lag or delayed sleep phase syndrome. In elderly insomniacs, melatonin administration decreased sleep latency and increased sleep efficiency. The effect of melatonin on sleep is probably the consequence of increasing sleep propensity (by inducing a fall in body temperature) and of a synchronizing effect on the circadian clock (chronobiotic effect). Melatonin secretion correlates with sleepiness in sighted and blind people. The urinary levels of the main melatonin metabolite 6-sulphatoxymelatonin decrease with age and in chronic diseases like coronary heart disease. Data will be shown indicating that patients with coronary disease had a low melatonin production rate, with higher decreases in those with higher risk of cardiac infarction. Administration of melatonin (3 mg p.o.) for up to 6 months did not affect circulating prolactin, FSH, TSH or estradiol in elderly insomniac females. In this group of patients melatonin augmented sleep quality and duration, and decreased sleep latency and the number of awakening episodes. Estimates of next- day function also improved significantly. Melatonin has a demonstrable therapeutic effect on sleep and circadian disorders in Alzheimer disease (AD). Data will be shown derived from 56 patients, aged 77 ± 8 years, who received 3-9 mg gelatin melatonin capsules p.o. daily at bed time for 4 to 35 months. Clinically, the patients exhibited improvement of sleep and delay of cognitive decay. The results indicate that melatonin can be useful for treatment of sleep disorder in AD and that it may delay cognitive deterioration, either as a consequence of sleep improvement or by interfering with AD physiopathology (e.g., b-amyloid toxicity). Melatonin is a potent antioxidant prevent- ing a number of diseases linked to augmented oxidative stress. To what extent the results fit the concept of the free radical flux theory of sleep, that holds that cerebral free radical production increases during wakefulness and is lower during sleep, will be discussed.

The studies were supported by the University of Buenos Aires, CONICET, ANPCYT and Elisium S.A., Argentina. Actas de Fisiología 7, 2001 35

WHAT DOES MELATONIN DO ON SLEEP IN NORMAL SUBJECTS AND PATIENTS WITH INSOMNIA?

Jaime M. Monti Department of Pharmacology and Therapeutics, Clinics Hospital, Montevideo, Uruguay

Initial studies addressing the effect of melatonin on sleep made use of the intravenous or the intranasal route or administered very large doses of the methoxyindole by the oral route. From these early studies it was concluded that melatonin reduces sleep latency and induces sleepiness and fatigue. More recently, the effects of lower pharmacologic or physiologic doses of melatonin have been examined in different laboratories. These studies included normal subjects and patients with chronic insomnia, as well as dementia patients exhibiting sundowning syndrome (Monti and Cardinali, 2000). The effects of melatonin on normal subjects during daytime were assessed using polysomnography, multiple sleep latency tests (MSLTs), wrist-worn actigraphs, and performance tasks. Melatonin was administered in doses ranging from 0.1 to 40.0 mg, 60-120 min prior to the beginning of the evaluation procedure. Melatonin administered during daytime reduced sleep latency, increased total sleep time, induced feelings of sleepiness and fatigue and reduced correct responses on vigilance tasks. Moreover, in the study by Zhdanova et al. (1996) 0.3-1.0 mg melatonin given during nighttime significantly reduced sleep latency and increased sleep efficiency, as assessed polysomnographically. The effect of melatonin on the polysomnographic sleep of insomniac patients was determined in three studies. James et al. (1990) administered 1.0-5.0 mg melatonin 15 min before bedtime in one night to a group of young insomniacs; variables related to sleep induction and maintenance were not significantly modified. In the study by Hughes et al. (1998), middle-aged and elderly insomniacs made use of immediate-release and controlled-release melatonin (0.5 mg) that were taken 30 min before bedtime. Melato- nin significantly shortened latencies to persistent sleep. Monti et al. (1999) administered 3.0 mg of melatonin during 14 nights to a group of elderly patients with chronic primary insomnia. The hormone was given immediately before turning off the lights in the sleep laboratory. Wake time after sleep onset was significantly reduced whereas total sleep time and sleep efficiency were increased in 4 out of the 10 patients treated with melatonin. Administration of immediate-release or sustained-release melatonin to independently living or institutionalized elderly insomniacs improved sleep as monitored by wrist actigraphy or structured clinical interviews and sleep logs completed by the patients (Haimov and Lavie, 1995; Fainstein et al., 1997). In the retrospective study by Brusco et al. (1999) Alzheimer’s disease (AD) patients received 9 mg of melatonin daily for 22-25 months. A significant improvement of sleep quality was found in all cases. In addition, sundowning diagnosed clinically in all 14 patients examined, was no longer detectable in 12 of them. In conclusion, melatonin was more effective when given to elderly insomniacs, or AD patients, although sleep improvement was not strictly correlated with prior levels of the hormone. Symposium 7

Sleep as a biological rhythm

Chairperson: Luiz Menna-Barreto (Brazil) Ennio Vivaldi, Universidad de Chile, Santiago, Chile Marisa Pedemonte, Universidad de la República, Montevideo, Uruguay Claudia de Castro Moreno, Universidad de Sao Paulo, SP, Brazil Luiz Menna-Barreto, (Universidad de Sao Paulo, SP, Brazil

Symposium 8

Memory processing during sleep

Chairperson: Antonio Giuditta (Italy) Gyorgy Buzsaki, State University of New Jersey, NJ, USA Antonio Giuditta, University of Naples “Federico II”, Italy Pierre Maquet, University College London, UK Carlyle Smith, Trent University, ON, Canada Marcos Frank, University of California San Francisco, CA, USA 38 Actas de Fisiología 7, 2001

SLEEP-WAKE CYCLE CHANGES ALONG ONTOGENESIS - SOME LANDMARKS

Luiz Menna-Barreto Universidade de São Paulo, Brazil

Stability of the sleep-wake cycle is often linked to good sleep quality although cycle parameters do change importantly along the life span of individuals. Attention will be drawn to such changes with examples from field studies of healthy populations conducted by our laboratory at the Universidade de São Paulo, Brazil. Infants show important inter-individual differences in the composition and maturation of the circadian component of the sleep-wake cycle; children in the first years of school may be sleep- deprived; city-dwelling adolescents tend to phase delay the cycle more than their country-dwelling pairs; the reduction in amplitude of the cycle in the elderly may be an artifact. We conclude with recommendations of protocols where monitoring of the cycle play decisive roles in the comprehension of its stability. Actas de Fisiología 7, 2001 39

IRREGULAR WORKING TIME AND SLEEP: THE CASE OF TRUCK DRIVERS

Claudia Roberta de Castro Moreno Departamento de Saúde Ambiental- Faculdade de Saúde Pública-USP-Brasil

Truck drivers submitted to irregular working hours including night work change sleep- wake time arrangements, which may cause excessive sleepiness during working time and might affect the ability to drive safely. Based on the evidence of the problems generated by the inversion of the sleep-wake cycle and consequent sleep deprivation, it can be said that individual strategies related to sleep habits are crucial for the adaptation to the work schedule. A study with the purpose to investigate such strategies was conducted in a truck drivers population. Our results have shown shorter sleep duration and greater sleep irregularity in truck drivers with irregular working time. Moreover, many truck drivers working in an irregular time intake more caffeine and others substances that may help to keep them awake during working time. The main impression from this study is that the sleep need, the subjetive sleep quality and ratings of sleepiness can be influenced by the working hours. It should be emphasized, however, that the consumption of stimulants could be a masking factor on sleep and it must be object of further studies with truck drivers.

Support: Fapesp 98/13053-3. 40 Actas de Fisiología 7, 2001

SENSORY PROCESSING DURING SLEEP AND WAKING, INTERACTION WITH ULTRADIAN HIPPOCAMPAL THETA RHYTHM

Marisa Pedemonte Neurofisiología. Departamento de Fisiología. Facultad de Medicina. Universidad de la República. Montevideo, Uruguay.

Several points of interaction between sensory inputs and biological rhythms are postulated. Sensory information is continuously processed by the central nervous system during wakefulness and during sleep, shown by reported data demonstrating shifts in the evoked activity as well as in the unitary firing. Recently, an ultradian biological rhythm –the hippocampal theta– is proposed as a time giver for sensory neuronal processing. Changes in the firing rate and discharge pattern of visual and auditory neurons were present in sleeping animals. All auditory loci studied –from the cochlear nucleus up to the auditory cortex– exhibited neuronal populations that increased, decreased or maintained their discharge rate during sleep (3). Auditory and visual neurons have shown phase-locking to the hippocampal theta rhythm. This temporal correlation appears during both spontaneous and evoked activity. The phase-locking was demonstrated during wakefulness and in sleep phases. The phase correlation could be shown following a novel stimulus and during changes related to attention or behavioral states. Theta rhythm is postulated to act during sleep-waking behavior as a low frequency temporal organizer in both auditory and visual systems, adding a temporal dimension to the sensory processing. It was proposed (1, 2, 3) the neuronal activity in the sensory systems as a changing phenomenon which depends on the interaction of, at least, three sets of signals: a) the specific sensory incoming information, b) the actual state of the brain –W, SWS or PS– and c) the hippocampal theta rhythm.

1. Pedemonte M., Peña J.L. & Velluti R. Exp Brain Res, (1996) 112: 41-46. 2. Pedemonte M., Pérez-Perera L., Peña J.L. & Velluti R.A. (2001) Sleep Res Online, 4(2): 51- 57. 3. Velluti R.A., Peña J.L. & Pedemonte M. Biol Signals Recept, . (2000) 9: 297-308. Actas de Fisiología 7, 2001 41

TEMPORAL ORGANIZATION OF SLEEP AND SHIFTWORK

Ennio A. Vivaldi, Emilio Fernández and Rodrigo Catalán Programa Disciplinario de Fisiología y Biofísica, Instituto de Ciencias Biomédicas, Facultad de Medicina, Universidad de Chile

The proper management of rotating schedules and environmental conditions can improve Shiftwork adaptation. Mining activities at high altitude add further physiological burdens to shiftwork organization problems. A reliable knowledge of workers actual sleep-wake organization throughout the shiftwork schedule is a necessary point of departure for assessing and improving alternative programs. The present study was conducted in a copper mine at a high altitude with workers on a specific 12-days rotating schedule: EMEMEMNNNFFF (E = 1400-2200, M = 0600-1400, N = 2200-0600, F = free). Time of day and place allocated for sleep changed according to the rotation. The mine is located at an altitude of 3,400 m. with a 2-hour average travel time. An actigraphic study, a technique that allows long-term monitoring of sleep wake distribution, was conducted for 24 days in 36 workers. Sleep behavior was assessed in term of its total amount, its distribution and its latency. Data on labor accidents, seasonal effects, sleep while traveling and workers tasks were also considered. Data was also collected on cognitive tests and the Standard Shiftwork Index. Intermitent exposure to high altitude (including sleep in altitude) was analysed in the context of its interaction with homeostatic and circadian regulatory mechanisms. 42 Actas de Fisiología 7, 2001

CORTICAL PLASTICITY ENHANCED BY SLEEP

M. G. Frank, N. P. Issa and M. P. Stryker W. M. Keck Foundation Center for Integrative Neuroscience Dept. of Physiology, University of California San Francisco, USA

During a critical period of brain development, occluding the vision of one eye causes a rapid remodeling of the visual cortex and its inputs. These changes in cortical circuitry form a well-established model of cortical plasticity in vivo. Sleep has been linked to other processes dependent on synaptic remodeling, but a role for sleep in this form of cortical plasticity has not been demonstrated. We examined the effects of sleep on the plasticity evoked by monocular deprivation (MD) during the cat critical period for visual development. We first determined the effects of 6 hours of MD on cortical responses to visual stimuli (n=5). We then determined if this remodeling was increased immediately after a 6-hour period of sleep (n=6) in complete darkness. Finally, we determined if these changes in remodeling also occurred if cats were instead kept awake for 6 hours (n=7) in complete darkness. EEG and EMG recordings were used to determine sleep/wake states and to ensure that all cats were alert during the MD period. We used microelectrode recording and optical imaging of intrinsic signals in primary visual cortex to measure changes in synaptic remodeling. We found that the effects of MD on cortical responses were greatly enhanced by sleep (Mann-Whitney U, p<0.05). We also found that increases in cortical plasticity were highly correlated with nonREM sleep amounts (Pearson: r=0.95) indicating that these changes in cortical circuitry specifically occurred during nonREM sleep. Sleep deprivation during this 6-hour period completely blocked the enhancement of cortical plasticity observed after sleep (Mann- Whitney U, p<0.05). These findings demonstrate that sleep and sleep loss modify experience-dependent cortical plasticity in vivo. They suggest that sleep in early life may play a crucial role in brain development.

NIH Grant EY02874 and National Research Service Awards EY06831 and EY06880 supported this research. Actas de Fisiología 7, 2001 43

LEARNING AND SLEEP: THE SEQUENTIAL HYPOTHESIS

Antonio Giuditta, Stefania Vescia, Paola Mandile and Stefania Piscopo Department of General and Environmental Physiology, University of Naples “Federico II”, Naples, Italy

According to the sequential hypothesis of the function of sleep, memory processing during sleep is assumed to require two sequential sets of brain operations, respectively occurring during slow wave sleep (SS) and paradoxical sleep (PS). Supporting evidence is largely consisting in the observations that, in rats learning a two-way active avoidance task (fast learning or FL rats), post-training SS episodes significantly increase their average duration in comparison to baseline values and to the values of control rats. Several lines of evidence confirm that these effects are not due to contingent factors, such as stress or sleep loss (for a review, see 1). These data should be viewed in the context of the literature demonstrating the involvement of PS in memory processing. In more recent work we have shown that, in baseline sleep, variables of SS episodes initiating sleep sequences that contain an intervening bout of transition sleep (TS) strongly correlate with the level of performance determined the following day. On the other hand, variables of baseline sequences lacking TS do not display comparable correlations (2). In addition, the observation that baseline sleep sequences are clustered in long strings separated by waking episodes longer than one minute (trains) (3) has further established that trains including SS->TS->W sequences prevail in FL rats in comparison to slow learning (SL) and non learning (NL) rats, and that most of their variables strongly correlate with the level of performance determined the following day. On the other hand, trains lacking the above sequence do not display comparable features (4). The data support the view that TS plays an essential role in memory processing and that it markedly modifies the functional significance of the preceding SS episodes (for a review, see 5). In separate analyses of the waking EEG power spectra obtained by telemetry during the training session, we have shown that, in comparison with baseline, a selective increase occurs in the delta region of FL rats (but not in SL or NL rats) in concomitance with their early dramatic improvement of performance. On the other hand, comparable increments are present in SL and NL rats in the region of 7-9 Hz (6). The latter data suggest that quantitative variations in selected bands of the waking EEG do occur during training in FL rats, as well as in SL and NL rats.

1. Giuditta A. et al., Behav. Brain Res. 69:157-166, 1995. 2. Vescia S. et al., Physiol. Behav. 60:1513-25, 1996. 3. Piscopo S. et al., Behav. Brain Res. 119:93-101, 2001. 4. Piscopo S. et al., Behav. Brain Res., 120:13-21, 2001. 5. Ambrosini M.V., Giuditta A., Sleep Med. Rev, in press, 2001. 6. Mandile P. et al., Abstr. Ital. Soc. Neurosci., I 12, 1993. 44 Actas de Fisiología 7, 2001

THE REM SLEEP WINDOW AND ACETYLCHOLINE

Carlyle T Smith Dept. of Psychology, Trent University, Peterborough, Ontario K9J 7B8, Canada

It has been observed in animal studies that REM sleep deprivation impairs the memory for a variety of aversive, appetitive and spatial memory tasks. The amount of REM sleep deprivation need only be 3 - 4 hours in length, but it must be applied at the right post training time. These short vulnerable time periods have been named REM Sleep Windows (REMSW) (1). It would logically seem that some important memory processing is occurring during these REMSW to maximize memory efficiency. The nature of this process is unknown, but earlier work has shown that acetylcholine (ACh) is present in relatively high quantities during the time when the REMSW is active (2). In more recent work, using the Morris water maze, we have demonstrated that increases in ACh in rat wholebrain are present one week after the end of training compared to non-trained controls. More interestingly, if rats are subjected to 4 hours of post training REM sleep deprivation at the REMSW, the levels of ACh in wholebrain one week later are suppressed. REM sleep deprivation outside the REM sleep deprivation window does not suppress levels of ACh. REM sleep deprivation during the REMSW appears to interfere with memory mechanisms which utilize ACh as their transmitter (3). In our most recent work, we have injected Scopolamine, an ACh antagonist, directly into the amygdala of rats to coincide with the 9 -12 hour REMSW of a Conditioned Cue Preference task (CCP). Acquisition of the CCP task has been shown to require an intact amygdala (4). It has been observed that acquisition of the CCP can be blocked by the Scopolamine injected to coincide with the REMSW. Saline control injection rats learn the CCP task normally.

References: 1) Smith, C.T. (1985). Sleep states and learning: a review of the animal literature. Neurosci. Biobehav. Rev., 9, 157-168. 2) Smith, C., Tenn, C. and Annett, R. (1991). Some biochemical and behavioural aspects of the paradoxical sleep window. Can J. Psychol., 45, 115-124. 3) Smith, C., DeButte, M. and Annett, R. (1998). Effects of paradoxical sleep deprivation on levels of acetylcholine and memory for a spatial task. Sleep, 21, (suppl.), 7. 4) Vallance, K., McDonald, R.J. and Smith, C. (1999). Effects of paradoxical sleep deprivation on the memory for a conditioned cue preference task. Sleep, 22 (suppl.), S243.

Research Supported by the Natural Sciences and Engineering Research Council of Canada (NSERC). Symposium 9

Basal forebrain: new perspectives on the role of the wakefulness- promoting neurons in the control of behavioral state

Chairpersons: Robert E. Strecker (USA) & Radhika Basheer (USA)

Barbara E. Jones, McGill University, Montreal, Canada Robert McCarley, Harvard University, Boston, MA. USA Kazue Semba, Dalhousie University, USA Ronald Szymusiak, VA Medical Center, North Hills, CA, USA

Symposium 10

The physiological basis for restless legs syndrome

Chairperson: Richard P. Allen (USA) Wayne Hening, Robert Wood Johnson Medical School, New York, USA Christopher J. Earley Marco Túlio de Mello, Sao Paulo, SP, Brazil Richard P. Allen, Johns Hopkins University, USA 46 Actas de Fisiología 7, 2001

WAKE-PROMOTING NEURONS IN THE BASAL FOREBRAIN

Barbara E Jones Montreal Neurological Institute, McGill University, Montreal, Canada

It has long been known that the basal forebrain serves as an important relay for the brainstem reticular activating system in the stimulation and maintenance of cortical activation during waking. However, the specific neurons implicated in that relay and the manner in which they influence the cortex has only in more recent years come to light. Cortically projecting neurons of the basal forebrain comprise cholinergic, GABAergic and most recently identified putative glutamatergic neurons [1, 2]. As originally evident from in vitro recording [3], each of these cell groups likely has particular discharge properties and responds in a different manner to the neurotransmitters contained within the brainstem afferents [4]. Local microinjections of neurotransmitters that excite cholinergic neurons stimulate cortical activation, including both high frequency gamma and low frequency rhythmic theta activity [5, 6]. Using juxtacellular recording of neurons in anesthetized animals, we have most recently been able to identify cells groups in vivo and characterize the relationship of their discharge to EEG activity [7]. All cholinergic neurons discharge at the highest rate during evoked cortical activation and the majority discharge in bursts in association with the theta-like, rhythmic slow activity that accompanies that activation. Of the GABAergic neurons, a small sub-group also discharges in a rhythmic manner in association with rhythmic slow activity, but in a more cluster-like discharge pattern [8]. A third group of neurons that composes a major contingent of rhythmically discharging neurons are non-cholinergic, non-GABAergic neurons that contain phosphate-activated-glutaminase, the synthetic enzyme for the neurotransmitter pool of glutamate [9]. Together, these three groups of cortically projecting neurons have the capacity to stimulate cortical activation at high, gamma and beta, and low, theta-like, frequencies. By such complementary rhythmic modulation of cortical activity, these cell groups can collectively promote cortical activation and moreover promote coherent activity at high and low frequencies which is critical for integrative functioning within the cerebral cortex during active and attentive waking.

1. Gritti, I., L. Mainville, M. Mancia, and B.E. Jones. J. Comp. Neurol., 1997. 383: 163-177.; 2. Manns, I., L. Mainville, and B.E. Jones. Neuroscience, In press.; 3. Alonso, A., A. Khateb, P. Fort, B.E. Jones, and M. Muhlethaler. Eur. J. Neurosci., 1996. 8: 169-182.; 4. Jones, B.E. and M. Muhlethaler, in Handbook of Behavioral State Control: Cellular and Molecular Mecha- nisms, 1999, 213-233.; 5. Cape, E.G., I.D. Manns, A. Alonso, A. Beaudet, and B.E. Jones. J. Neurosci., 2000. 20: 8452-8461.; 6. Cape, E.G. and B.E. Jones. J. Neurosci., 1998. 18: 2653- 2666.; 7. Manns, I.D., A. Alonso, and B.E. Jones. J. Neurosci., 2000. 20: 1505-1518.; 8. Manns, I.D., A. Alonso, and B.E. Jones. J. Neurosci., 2000. 20: 9252-9263.; 9. Manns, I.D., A. Alonso, and B.E. Jones. Soc. Neurosci. Abst., 2000. 26: 1514. Actas de Fisiología 7, 2001 47

WAKEFULNESS AND AROUSAL: SHORT- AND LONG-TERM EFFECTS OF ADENOSINE IN THE BASAL FOREBRAIN

Robert McCarley Harvard Medical School, Boston VA Healthcare System, Brockton MA USA

The magnocellular regions of the basal forebrain (BF) are known to be important regulators of wakefulness and cortical activation. A majority of neurons in the BF are wakefulness-promoting, based on lesion and stimulation data, and BF neurons’ showing their highest discharge activity during wakefulness and a marked reduction in activity as the animal becomes drowsy and goes to sleep. What might be the factors leading to reduction of discharge rate? Adenosine is one of the prime candidates. We suggest that, as adenosine is a breakdown product of ATP, adenosine builds up during wakefulness when the neural metabolism is high, and then acts on adenosine A1 receptors to reduce the discharge activity of wake promoting neurons to cause the transition from wakefulness to sleep. We have conducted a series of studies in support of this hypothesis. In vitro studies have demonstrated that cholinergic neurons of the basal forebrain and the mesopontine tegmentum nuclei are under the tonic inhibitory control of adenosine. In vivo microdialysis perfusion of adenosine in the cholinergic BF and mesopontine cholinergic nuclei reduced wakefulness in freely behaving cats. Combined single unit recording and microdialysis have shown that adenosine acts via A1 receptor to inhibit the discharge activity of BF wake promoting neurons. Microdialysis perfusion of adenosine A1 receptor antisense induces wakefulness and reduces recovery sleep. In addition, microdialysis perfusion of an adenosine transport inhibitor (NBTI) not only doubles extracellular adenosine levels in the basal forebrain but also produced a significant decrease in waking. Extracellular concentrations of adenosine in the BF were highest during spontaneous wakefulness and the adenosine levels in the BF exhibited progressive increases during sustained, prolonged wakefulness followed by a slow decline during recovery sleep. Though all the above mentioned studies suggest that adenosinergic inhibition of the basal forebrain arousal system is a key element in the transition from wakefulness to sleep, a new question arise. Does accumulation of adenosine during prolonged wakefulness have any long term effects, such as might be involved in mediation of sleep debt? To answer this question, we began a search to identify the intracellular cascades and transcriptional mediators of the long-term effects of adenosine. RT-PCR studies demonstrated an increase in A1 receptor mRNA level after 6 hr of sleep deprivation. Calcium imaging studies showed that A1 receptor is the major mediator of an increase in intracellular Ca via the inositol trisphosphate receptor on the endoplasmic reticulum. Further, sleep deprivation and A1 agonists induces the nuclear translocation of NF-kB exclusively in cholinergic neurons of the basal forebrain. NF- kB is known to bind to the promotor region of the A1 receptor, and may mediate this effect. In summary, we believe there is strong evidence to suggest that adenosine is a key mediator of sleepiness and of the long-term effects of sleepiness. 48 Actas de Fisiología 7, 2001

MULTIPLE OUTPUT PATHWAYS OF THE BASAL FOREBRAIN: ROLES IN WAKEFULNESS AND LINKS WITH THE CIRCADIAN SYSTEM

Kazue Semba Departments of Anatomy & Neurobiology and Psychology, Dalhousie University, Halifax, Nova Scotia B3H 4H7 Canada

The basal forebrain (BF) has received much attention over the last two decades because of its cholinergic neurons and their role in learning and memory, cortical activation, wakefulness, and involvement in the pathophysiology of Alzheimer’s disease. These cholinergic neurons project to the cerebral cortex, providing the main source of cortical acetylcholine, as well as to the amygdala and olfactory bulb. However, it has become evident that the BF also contains non-cholinergic projection neurons. Some of these neurons are GABAergic and project to the cortex. Furthermore, as already noted in 1975 by Divac, the BF is a source of multiple descending projections. Descending BF axons are now known to innervate the thalamus, hypothalamus and brainstem, and they are predominantly non-cholinergic. The heterogeneity of BF neurons is seen not only in their anatomical features, but also in their neurophysiological properties, as revealed by both in vitro and in vivo electrophysiological investigation. The increasing knowledge about the complexity of the organization and function of the BF would require reconsideration of some of the results in earlier behavioural studies that employed non- selective lesions, or stimulation or drug injections that would affect all BF neurons indiscriminately. The first part of this presentation is intended to provide a brief summary of the current knowledge of the organization of efferent pathways of the BF, with emphasis on neurochemical heterogeneity, and to discuss possible roles of these multiple pathways in the control of vigilance. It is proposed that the BF plays a key role in orchestrating both cognitive and non-cognitive aspects of vigilance through its ascending and descending projections. Behavioural state is modulated by the circadian system, such that sleep is consolidated into either the dark or the light phase depending on the species. The main circadian clock is located in the suprachiasmatic nucleus (SCN), and after SCN lesions, the distribution of sleep becomes random with respect to time of day. The last part of the presentation will discuss anatomical and functional evidence for direct and indirect links between the BF and the SCN. In conclusion, the BF may play a key role in controlling a number of efferent systems involved in different functional aspects of vigilance during wakefulness, and serve as a link with the circadian system to pattern daily rhythms of sleep and waking. Actas de Fisiología 7, 2001 49

ACTIVITY OF NEURONS IN CHOLINERGIC REGIONS OF THE BASAL FOREBRAIN DURING WAKING AND SLEEP

Ronald Szymusiak, Md. Noor Alam, Hui Gong, and Dennis McGinty Research Service, V.A.GLAHS, North Hills CA, USA and Departments of Medicine and Psychology, University of California, Los Angeles, CA USA

Neurons located in cholinergic regions of the magnocellular basal forebrain (mBF) have been implicated in the regulation of behavioral and electrographic indices of arousal. We have examined extracellularly-recorded neuronal activity in the mBF of cats and rats during natural sleep and waking. A commonly encountered cell type in both species is one that exhibits elevated discharge rates during waking and REM sleep, with significantly reduced discharge rates during nonREM sleep. During waking, discharge of mBF neurons is strongly movement-related, with significantly reduced activity during quiet waking compared to active waking. In cats, orthodromic excitatory responses are evoked in mBF neurons in response to electrical stimulation of the pontine and midbrain reticular formation, suggesting that ascending input from the brainstem is a potential source of excitatory drive during waking and REM sleep. Multiple neurochemical factors appear to participate in the inhibitory modulation of mBF neurons across arousal states. Administration of adenosinergic agonists and antagonists via a microdialysis probe located adjacent to recorded neurons, demonstrates the presence of tonic adenosine-mediated inhibitory tone on mBF neurons during both waking and sleep. Similar studies using gamma-aminobuytryic acid (GABA) antagonists reveal that in some mBF neurons, increased GABAergic inhibition contributes to nonREM sleep-related reductions in discharge rate. Anatomical studies in progress in our laboratory suggest that GABAergic neurons located in the median preoptic nucleus may be a source of nonREM sleep-related inhibitory modulation of mBF neurons.

Supported by V.A. Medical Research Service and HL60296. Symposium 11

Sleep promoting neurons in the basal forebrain and preoptic region

Chairperson; Barbara E. Jones (Canada) Barbara E. Jones, McGill University, Montreal, Canada Clifford B. Saper, Harvard University, Boston, MA, USA Michel Muhlethaler, Geneva, Switzerland Ronald Szymusiak, VA Medical Center, North Hills, CA, USA

Symposium 12

Consideration of possible pathophysiologies of movement disorders in sleep

Chairperson: Adrian R. Morrison (USA) Adrian R. Morrison, University of Pennsylvania, PA, USA Mark Mahowald, University of Minneapolis, MN, USA Jack Yamuy, University of California Los Angeles, CA, USA David B. Rye, Emory University, Atlanta, GA, USA 52 Actas de Fisiología 7, 2001

SLEEP-PROMOTING NEURONS IN THE BASAL FOREBRAIN

Barbara E Jones Montreal Neurological Institute, McGill University, Montreal, Canada

Evidence from lesion, stimulation and unit recording studies suggests that the basal forebrain is importantly involved in promoting sleep. Lesions have resulted in decreases in slow wave sleep, and stimulation has elicited cortical slow wave activity and sleep. Single unit recording studies have identified neurons that are most active during slow wave sleep. However, the basal forebrain is also known to be important for cortical activation and waking, since lesions decrease and stimulation enhance cortical activation. It also contains many neurons that are active in association with cortical activation and waking. The basal forebrain thus appears to play a dual role in cortical modulation and sleep-wake states that must be fulfilled by distinct cell groups endowed with distinct properties, neurotransmitters and/or receptors. Corroborating evidence from in vitro studies in the basal forebrain, multiple cell types have been identified by in vivo recording and juxtacellular labeling that have distinctive profiles and patterns of discharge in relation to EEG activity. Of these, all cholinergic neurons are most active and tend to burst in association with cortical activation [1]. On the other hand, most GABAergic neurons are most active and discharge in distinct patterns in association with cortical slow wave activity similar to slow wave sleep in anesthetized animals [2]. In vitro pharmacological studies have indicated that cholinergic and non-cholinergic basal forebrain neurons respond differentially to neurotransmitters of the activating system, including importantly noradrenaline (NA), which excites cholinergic neurons and inhibits a group of non-cholinergic neurons [3]. The identified GABAergic neurons that are most active in association with cortical slow wave activity in vivo are endowed with alpha 2 adrenergic receptors [4]. Local microinjections of NA into the basal forebrain suppress slow wave activity and stimulate fast cortical activity most likely through the selective actions of NA on GABAergic sleep- active in addition to cholinergic wake-active neurons [5]. We conclude that specific GABAergic basal forebrain neurons endowed with alpha 2 adrenergic receptors are inhibited during waking, when noradrenergic locus coeruleus neurons are active, and disinhibited, when these cease, to become active and thus promote slow wave cortical activity and sleep.

1. Manns, I.D., A. Alonso, and B.E. Jones. J. Neurosci., 2000. 20: 1505-1518. 2. Manns, I.D., A. Alonso, and B.E. Jones. J. Neurosci., 2000. 20: 9252-9263. 3. Fort, P., A. Khateb, M. Serafin, M. Muhlethaler, and B.E. Jones. NeuroReport, 1998. 9: 1-5. 4. Manns, I.D., Hou, Y.P. and Jones, B.E. World Federation of Sleep Research Societies Ab- stracts, 2001. 5. Cape, E.G. and B.E. Jones. J. Neurosci., 1998. 18: 2653-2666. Actas de Fisiología 7, 2001 53

DIFFERENTIAL EFFECTS OF VENTROLATERAL PREOPTIC (VLPO) CLUSTER VS. EXTENDED VLPO ON NREM VS REM SLEEP

Clifford B. Saper and Jun Lu Department of Neurology and Program in Neuroscience Harvard Medical School and Beth Israel Deaconess Medical Center Boston, MA 02215, USA

The VLPO was originally defined as a cluster of neurons just dorsolateral to the optic chiasm that project to the ascending arousal system. These neurons are most active during sleep, as demonstrated both by Fos expression and by single unit recording studies. VLPO neurons with these properties have been found to contain the inhibitory neurotransmitters galanin and GABA. We have hypothesized that VLPO neurons may serve to inhibit the ascending arousal system during sleep and thus may be a necessary part of brain circuitry for causing sleep. In all of these studies, some neurons that have similar properties to those in the VLPO cluster have been identified extending dorsally and medially from the densest cell aggregation. Recent studies show that these neurons in the dorsal and medial extended VLPO have slightly differrent projections, targeting the noradrenergic locus coeruleus and the serotoninergic dorsal and medial raphe nuclei more than the VLPO cluster, which innervates the histaminergic tuberomammillary nucleus most heavily. These observations suggested that the extended VLPO may have different effects on sleep from the VLPO cluster. We targeted lesions with the excitatory amino acid, ibotenic acid, at the VLPO cell cluster. Animals with greater than 70% cell loss in the VLPO experienced more than 55% loss of total sleep time. The loss of neurons in the VLPO cluster in these experiments correlated closely with the loss of NREM sleep time. However, although the same animals had a similar degree of loss of REM sleep, the amount of REM sleep loss did not correlate with the injury to the VLPO cluster. Interestingly, the loss of REM, but not NREM sleep, did correlate with loss of sleep-active (Fos positive) neurons in the extended VLPO. To test the role of the extended VLPO in REM sleep, animals were treated with darkness during the first three hours of the normal light cycle. These animals had approximately three times as much REM sleep during this period as did animals that slept in the light. The animals who slept in the dark also had differential Fos activation of neurons in the extended VLPO. These studies demonstrate that the extended VLPO may have a more important role in the regulation of REM sleep, while the VLPO cluster may be more important in controlling NREM sleep. 54 Actas de Fisiología 7, 2001

SLEEP-PROMOTING NEURONS IN THE MEDIAN PREOPTIC NUCLEUS: ELECTROPHYSIOLOGY, NEUROCHEMISTRY AND EFFERENT PROJECTIONS

Ronald Szymusiak, Hui Gong, Natalia Suntsova and Dennis McGinty Research Service, V.A.GLAHS, North Hills CA, USA and Departments of Medicine and Psychology, University of California, Los Angeles, CA USA

Neurons in the rat median preoptic nucleus (MnPN) exhibit c-fos protein immunoreactivity (IR) in response to sustained sleep (Gong et al., Am. J. Physiol. 279: R2079- R0288, 2000). MnPN neurons appear to be responsive to temperature as well as to arousal state, as the number of Fos-IR neurons is increased in animals sleeping in a warm environment compared to animals sleeping at normal ambient temperatures. Chronic microwire recordings of extracellular unit activity reveal a high concentration of neurons with sleep-related discharge patterns in the MnPN. Many MnPN neurons exhibit increased activity during both nonREM and REM sleep compared to waking, similar to what we have previously described for ventrolateral preoptic area neurons (Szymusiak et al., Brain Res. 803: 178-188, 1998). Injections of the neuronal tracer, biotinylated dextran amine (BD), into the MnPn yield anterograde labeling in hypothalamic and forebrain regions implicated regulation of behavioral and electrographic arousal, including cholinergic regions of the magnocellular basal forebrain, and the hypocretin neuronal field in the perifornical lateral hypothalamus. Double-labeling for c-fos protein and a marker of GABAergic neurons, glutamic acid decarboxylase (GAD), reveals that 80-85% of MnPN neurons exhibiting sleep-related fos-IR are also GAD- positive. In conclusion, a subpopulation of MnPN GABAergic neurons is activated during sleep and may constitute a source of sleep-related inhibitory modulation of arousal mechanisms in the basal forebrain and lateral hypothalamus.

Supported by V.A. Medical Research Service and HL60296. Actas de Fisiología 7, 2001 55

REM SLEEP WITHOUT ATONIA: HISTORY OF AN ANIMAL MODEL

Adrian R. Morrison Department of Animal Biology, University of Pennsylvania, Philadelphia, USA

Jouvet and Delorme (1965) first recognized REM sleep without atonia (REM-A) for what it is although Jouvet (1962) actually described the phenomenon earlier as a hallucinatory-like state. Bilateral electrolytic lesions in the pontine tegmentum induce REM- A. Henley and Morrison (1969, 1974) later demonstrated that REM-A could appear within very few days post lesion, that the locus coeruleus was not involved and that the phenomenon was stable over months. The last contrasts with the recent puzzling finding that rats only exhibit behavior during REM for a few days before recovering atonia (Sanford et al. 2001). Hendricks et al. (1982) then reported that the behavior observed during episodes of REM-A depended on the loci of damage within the pontine tegmentum, arguing against the view (Sastre and Jouvet 1978) that cats were demonstrating oneiric behavior simply because motoneuronal inhibition had been removed. Indeed, the appellation, REM- A, has proved misleading in that regard: behaviors observed are the consequence of disturbances in a complex of systems. Nevertheless, the behaviors that occur are those typical of an active, very alert cat. Nuchal atonia can be eliminated without release of behavior by small, asymmetric or even immense lesions that presumably block systems organizing behavior. During wakefulness, all cats with REM-A demonstrated significant increase in exploratory locomotion (Morrison et al. 1981). This observation led to a model that has been used to explain REM-A, orienting in wakefulness and the abnormality, cataplexy (Morrison 1979).

1. Jouvet, M. (1962). Arch. Ital. Biol., 100, 125-206. 2. Jouvet, M. and Delorme, F. (1965). C.R.Soc.Biol., 159, 895-899. 3. Hendricks, J.C., Morrison, A.R. and Mann, G.L. (1982). Brain Res. 239, 81-105. 4. Henley, K. and Morrison, A.R. (1969). Psychophysiol., 6, 245. 5. Henley, K. and Morrison, A.R. (1974). Acta Neurobiol. Exp., 34, 215-232. 6. Morrison, A.R. (1979). Prog. Psychobiol. Physiol. Psychol., 8, 91-131. 7. Morrison, A.R., Mann, G.L. and Hendricks, J.C. (1982). Sleep, 4, 247-257. 8. Sastre, J. and Jouvet, M. (1978). Physiol. Behav., 22, 979-989. 9. Sanford, L.D., Cheng, C.S., Silvestri, A.J., Tang, X., Mann, G.L., Ross, R. J. and Morrison, A.R. (2001). Sleep Res. Online, 4, 1-5. http://www.sro.org/2001/Sanford/1 56 Actas de Fisiología 7, 2001

NEURAL SUBSTRATES GOVERNING NOCTURNAL MOVEMENTS: INSIGHTS FROM ANIMAL MODELS OF DESEASE

David Rye Sleep Disorders Laboratory, Emory University School of Medicine, Atlanta, USA

The subceruleal region of the dorsolateral pons figures prominently as the ultimate mediator of REM-sleep behavior disorder (RBD) since neurons here exhibit REM-sleep specific increases in discharge and descending pathways to ventromedial medullary regions essential in maintaining REM-atonia. When these pathways are interrupted, REM- sleep specific motor behaviors can be seen in humans. The pathophysiology of RBD must also account for the fact that it predates or accompanies diseases which share in common a parkinsonian phenotype. Its prevalence in idiopathic Parkinson’s disease may reach 15% and it is also not uncommon in narcolepsy/cataplexy. Some investigators have suggested that the pathophysiological basis of RBD may lie in loss of noradrenergic subcoeruleal or cholinergic pedunculopontine region (PPN) neurons given their frequent involvement by the primary pathology of many degenerative conditions. Such neuropathological assessments and correlations are generally over simplistic as they do not take into account the integrity of less conspicuous neural populations in this heterogeneous region, pathologies in other brain regions, and are inconsistent with experimental work. Loss of noradrenergic function, for example, might be expected to promote atonia rather than eliminate it, whereas loss of cholinergic PPN neurons would be expected to result in loss rather than enhancement of phasic motor elements of REM-sleep. Alternatively, abnormal afferent signals impinging upon the subcoeruleal region may alter neural responsiveness thereby releasing RBD. As the subceruleal region exhibits a wide array of forebrain afferents, there are numerous potential structures in the hypothalamus, amygdala or basal ganglia that might modify RBD. Of these afferent sources, clinical and experimental evidence argues that the basal ganglia nuclei are likely to be most relevant to the pathophysiology of “idiopathic” RBD. Each of the neurodegenerative conditions in which RBD is commonly observed, for example, share in common loss of nigrostriatal dopamine (DA) pathways, and imaging studies in “idiopathic” RBD demonstrate loss of DA transporter in nigrostriatal axons but no detectable alterations of D2 receptor density. Rats and non-human primates depleted of striatal DA lack overt RBD, but do exhibit heightened phasic and tonic somatomotor activity in REM-sleep that predominates in limb versus axial musculature. Enhancement of phasic motor phenomena of REM-sleep requires transient hyperpolarization of the subpopulation of glutamatergic and cholinergic PPN region neurons that exhibit low-threshold calcium spikes as might be afforded by pathologically elevated phasic bursting of GABA-ergic basal ganglia output nuclei known to occur in PD. Enhancement of tonic electromyo- graphic activity in REM-sleep may reflect inhibition of PPN region neurons whose REM- sleep specific activition is othwerwise necessary for engaging REM-atonia pre-motor elements in the ventromedial medulla. Thus, removal of pathologic pallidal influences upon the upper brainstem by pallidotomy can dampen REM-sleep elevations in somatomotor activity present in some PD patients. Pharmacologic dampening of pathologic pallidal influences with low doses of levodopa-carbidopa (sinemet®) likely underlies the reported benefits of this agent in treating RBD. Actas de Fisiología 7, 2001 57

EFFECTS OF HYPOCRETIN (OREXIN) IN PONTINE NUCLEI INVOLVED IN ACTIVE SLEEP

Jack Yamuy, Mingchu Xi, Simon J. Fung, Pablo Torterolo, Jian-hua Zhang, Manuel J. Rojas, Francisco R. Morales and Michael H. Chase Department of Physiology and The Brain Research Institute, UCLA School of Medicine, Los Angeles, CA 90095

A cholinoceptive region within the nucleus pontis oralis (NPO), that receives cholinergic projections from the latero-dorsal and pedunculo-pontine tegmental (LDT-PPT) nuclei, is thought to be involved in the control of wakefulness and active (REM) sleep. The newly discovered hypothalamic hypocretinergic system, which has been implicated in these behaviors, projects both to the LDT-PPT and NPO. We have recently reported that a high percentage of hypocretinergic neurons express Fos during carbachol-induced active sleep. Based upon these data, we sought to examine if hypocretin application into the LDT-PPT and NPO induces changes in the occurrence of active sleep and in the activity of NPO neurons recorded intracellularly. Two different in vivo preparations were used in this study. First, cats were prepared for the chronic monitoring of behavioral states and the microapplication of drugs. Second, a-chloralose anesthetized cats were prepared for intracellular recording of NPO neurons prior to and following the juxtacellular ejection of hypocretin. In the chronic cat, we found that a) the microinjection of hypocretin 1 into the LDT- PPT increased the percentage of time spent in wakefulness and decreased the time spent in active sleep, and b) the microinjection of hypocretin 1 and 2 into the NPO produced, with a short latency, either a behavioral state that was indistinguishable from natural active sleep or a dissociated state which consisted of atonia and others, but not all of the components of active sleep. In the acute preparation, we found that a) hypocretin 1 produced depolarization, an increase in the rate of spontaneous discharge and the mean amplitude of spontaneous EPSPs, and a decrease in rheobase in NPO neurons, b) electrical stimulation of the hypocretinergic area in the hypothalamus produced EPSPs in NPO neurons, and c) a long-latency (presumably cholinergic) EPSP induced in NPO neurons by the electrical stimulation of the LDT increased in amplitude following the juxtacellular application of hypocretin 1. The present data suggest that hypocretin acts as an excitatory neurotransmitter vis- à-vis neurons that are involved in active sleep in the NPO. Disruption of hypocretinergic mechanisms in the LDT-PPT and NPO may be important in producing the alterations in active sleep and wakefulness that occur in narcolepsy.

This work was supported by USPHS Grants NS23426 NS09999, MH43362, HL60296, and AG04307. Symposium 13

Central mechanisms of apnea

Chairpersons: David W. Carley (USA) & Misha Radulovacki (USA)

Clifford B. Saper, Harvard University, MA, USA Donald R. McCrimmon, Northwestern University, MI, USA Leszek K. Kubin, University of Pennsylvania, PA, USA David W. Carley, University of Illinois, IL, USA Misha Radulovacki, University of Illinois, IL, USA

Symposium 14

Arousal and awakening processes in early ontogenesis

Chairperson: Lilia Curzi-Descalova (France) Lilia Curzi-Descalova, Hopital Robert Debré, Paris, France Igino Fagioli, University of Florence, Italy José Groswasser, University Hospital Reine Fabiola, Brussels, Belgium Rosmary Horne, Monash University, Melbourne, Australia 60 Actas de Fisiología 7, 2001

THE PONTINE RESPIRATORY GROUP AND APNEIC REFLEXES

Clifford B. Saper and Nancy Chamberlin Department of Neurology and Program in Neuroscience Harvard Medical School and Beth Israel Deaconess Medical Center Boston, MA 02215, USA

The traditional description of the brainstem cell groups involved in respiratory control includes three interconnected sets of neurons: the dorsal (DRG) and ventral (VRG) respiratory groups in the medulla and the pontine respiratory group. The DRG consists of respiratory-related, mainly inspiratory neurons in the nucleus of the solitary tract, mainly constituting its ventrolateral, interstitial, and intermediate subnuclei. The VRG includes the respiratory-related neurons in several populations in the ventrolateral medulla, including expiratory neurons in the Bötzinger complex and caudal ventrolateral medulla, inspiratory neurons in the intermediate part of the ventrolateral medulla, and neurons that have a pacemaker-like function in the pre-Bötzinger complex, just caudal to the Bötzinger complex. The pontine respiratory group includes the parabrachial complex, including the Köl- liker-Fuse nucleus. In mapping studies involving threshold injections of glutamate, we found that sites in the lateral parabrachial nucleus produced hyperpnea with complete expiratory cycles, whereas injections into the rostral, dorsal part of the Kölliker-Fuse nucleus produced hyperpnea with incomplete expiration (similar to apneusis). We attributed these responses to neurons in the lateral crescent subnucleus and Kölliker-Fuse nucleus that project to the ventrolateral medulla, and to Kölliker-Fuse projections back to the intermediate NTS, where injections of glutamate cause apneic responses. We found only occasional true apneas, however, and all of these were obtained from the ventral edge of the Kölliker-Fuse nucleus and the pontine tegmentum ventral to the parabrachial nucleus. Subsequently, Deutschmann and Herbert reported that injections of glutamate into the Kölliker-Fuse nucleus could cause apneic responses. We therefore reinvestigated the source of these responses by extending our glutamate stimulation studies further into the more ventral components of the pons, beyond what had been classically considered to constitute the pontine respiratory group. We found that injection sites into the white matter area between the motor and principal sensory trigeminal nuclei consistently produced low threshold apneic responses. These responses were never seen within the Kölliker-Fuse nucleus core, but could occasionally be elicited from its ventral edge, suggesting that this intertrigeminal region (ITR) was the source of the apneic responses in all of the previous studies. The connections of the ITR were studied by injecting it with both anterograde and retrograde tracers. Inputs to the ITR were discovered from each of the sites that processes sensory inputs that produce apneas, including the nucleus of the solitary tract and the ventral part of the spinal trigeminal nucleus. Projections from the ITR were found to the VRG, suggesting that this site may play a key role in mediating sensory-reflex apneic responses. These observations also extend the concept of the pontine respiratory group to include a previously unrecognized but critical component of respiratory circuitry. Actas de Fisiología 7, 2001 61

AROUSAL DURING SLEEP : NEUROPHYSIOLOGICAL BASIS AND DEVELOPMENT IN EARLY HUMAN ONTOGENESIS

Lilia Curzi-Dascalova1, Ronal L. Ariagno2 and Majid Mirmiran2 INSERM, E3599, Hôpital Robert Debré, Paris, France1;Stanford University, USA2

The phenomena of arousal and awakening, particularly in infants during the first year of life, has become an important research area since there is general agreement that understanding these mechanisms is essential to understand the adaptive physiology of infants during development and illness. Most of the previous data which has been reported concern babies who have been recorded during the first months of age. It has been demonstrated that arousability is affected by birth weight, gestational and post-natal age, by sleeping position (i.e., prone versus supine), by prenatal exposure to drugs and smoking). The criteria used for arousal definition are highly variable. In a recent preliminary collaborative investigation we examined changes in various physiological variables following spontaneous body movements as part of the definition of arousal in healthy premature and full-term newborns. We tried to evaluate the use of the arousal related parameters recently proposed for older infants (Consensus of the Pediatric European Wake-Up Club, in preparation; also see Arousal and Awakening in Human Development Symposium summary, WFSRS meeting, Dresden, 1999). We found that from 31w post-conceptional age (PCA), changes in EEG amplitude and frequency could be observed during arousals. In both active (AS) and quiet sleep (QS) EEG modifications were usually seen after 34w PCA. EEG modifications mainly consisted of amplitude decrease and frequency increase. Heart rate changes related to arousals were the most consistent finding. Heart rate acceleration was frequently observed and the amplitude of this acceleration increased with PCA. Heart rate decelerations were rarely seen. Our data showed that preterm infants from 31 w PCA and term neonates from the moment of birth do respond to spontaneous generalized body movements. The amplitude of response increases with PCA and was similar for preterm infants at term PCA when compared to term infants. Furthermore preterm infants at 37-40 weeks PCA have more arousals and greater heart rate variability in supine than in prone sleep position. Finally, our results showed that the physiological response to spontaneous generalized body movement during the neonatal period is a change in EEG, respiration and heart rate seen during or shortly after arousals. 62 Actas de Fisiología 7, 2001

PSYCHOPHYSIOLOGICAL STUDIES ON INFANTS AWAKENINGS

Igino Fagioli1), Gianluca Ficca2), Fiorenza Giganti1) and Piero Salzarulo1) 1) Department of Psychology, University of Florence; 2) Department of Psychology, Second University of Naples (Italy)

The modalities of the awakening during the early development and their age-related changes could help to clarify the mechanisms underlying sleep termination which are involved in the transition from the infant polyphasic rhythm (characterised by a frequent alternate occurrence of waking and sleeping states) to the adult monophasic one. This transition is mainly accounted for by a decrease in the number and duration of awakenings, and in a change in their distribution, which determine both the nocturnal sleep “consolidation”, and the diurnal wakefulness consolidation. Studies based on polygraphic recordings showed that the number of awakenings decreases in the first year of life mainly during the night-time (Ficca et al., 1999). During the whole first year of life, and particularly in the first six months of life, awakenings occur more frequently during REM, like in young adults. Older infants show an increased propensity to wake up in NREM sleep, before the occurrence of REM sleep, after daytime naps (Salzarulo & Fagioli, 1992). During early development both awakenings and sleep onsets occur through REM sleep, which plays a “gating” role in both directions (Salzarulo et al., 2000). With ageing the major change regards prevalent state for sleep onset (REM sleep onset early in infancy, NREM sleep onset in adults); similar changes in the modality of awakening (with the only exception of naps) do not emerge. Data on the EEG background activity dynamics suggest that differences in homeostatic regulation of sleep could account for the frequently observed occurrence of one long nocturnal awakening which separates two sustained sleep bouts (biphasic nocturnal sleep). The developmental changes in the awakenings are at least partly related to the development of the homeostatic regulation of sleep, which is indexed by the characteristics of the EEG background activity.

References: Ficca, G., Fagioli, I., Giganti F., Salzarulo P. Spontaneous awakenings from sleep in across the first year of age. Early human development, 1999, 55: 219-228. Salzarulo, P, Fagioli, I.. Post-natal development of sleep organization in humans: speculations on the emergence of the “S process”. Neurophysiol. Clin., 1992, 22: 107-115. Salzarulo, P., Giganti, F., Ficca, G., Fagioli, I., Toselli M. Gates to awakening in early development. Clinical Neurophysiology, 2000, 53 : 352-354. Symposium 15

Dopamine in sleep mechanisms and pathologies

Chairpersons: Birgit Högl (Austria) & Jaime M. Monti (Uruguay) David B. Rye, Emory University, GA, USA Jaime M. Monti, Universidad de la República, Montevideo, Uruguay Birgit Högl, University of Innsbruck, Austria Jacques Montplaisir, Université de Montréal, Québec, Canada

Symposium 16

GABA or glycine: which one is the most important for REM sleep

Chairperson: Pierre H. Luppi (France) Barbara E. Jones, McGill University, Montreal, Canada Jerome Siege,University of California Los Angeles, CA, USA Jack Yamuy, University of California Los Angeles, CA, USA Pierre H. Luppi, INSERM, Lyon, France 64 Actas de Fisiología 7, 2001

EFFECTS OF DOPAMINERGIC SUBSTANCES ON NIGHT SLEEP AND DAYTIME SLEEPINESS IN PATIENTS WITH PARKINSON´S DISEASE

Birgit Högl Department of Neurology, University of Innsbruck, Austria

This contribution will review the results of older and more recent studies regarding effects of levodopa and various dopamine agonists on night sleep. Sleep onset and yawning induced by levodopa, and various dopamine agonists were first described years ago. However, only the report of sudden sleep attacks occurring without warning in Parkinson´s disease patients taking pramipexole and ropinirole, has led to increased awareness of these side effects. Subsequently, drowsniness and sleep attacks were also found with other, ergoline, dopamine agonists. The most recent studies try to establish and compare frequencies of sleep attacks with different dopamine agonists. According to some reports, the induction of sleepiness by dopamine agonists in patients with Parkinson´s disease is a dose dependent phenomenon, which appears to be more frequent with higher doses, and diminishes with dose reduction. This is in contrast with the biphasic effect of dopamine agonists on sleep and wakefulness seen in animals, where low doses of dopamine agonists seem to induce sleep and sleepiness and higher doses increase wakefulness (Lagos P 1998, Poewe and Högl 2001). Modafinil is a wake promoting drug used in the treatment of narcolepsy. In high doses modafinil diminishes parkinsonian symptoms in MPTP marmosets. Dopamine reuptake inhibition has recently been shown to be modafinil´s mode of action (Wisor 2001). Modafinil improved daytime sleepiness in Parkinson´s disease patients (see Högl et al, this issue).

References: Lagos P, Scorza C, Monti JM, et al: Eur Neuropsychopharmacol 1998;8:113-120 Poewe W, Högl B. Curr Opin Neurol 2000; 13:423-6. Wisor J, Hishino S, Sora I, Uhl G, Mignot E, Edgar D. J Neurosci 2001;21:1787-1994 Actas de Fisiología 7, 2001 65

THE ROLE OF DOPAMINE IN THE REGULATION OF SLEEP AND WAKING

Jaime M. Monti and Héctor Jantos Department of Pharmacology and Therapeutics. Clinics Hospital. Montevideo, Uruguay

Hypothesis regarding the role of dopamine (DA) on sleep and waking (W) have gone through several phases. In the 1970s, based on lesion studies, DA was hypothesized to be responsible for behavioral arousal. (Jones et al., 1973). Based on numerous pharmacologic studies, a hypothesis evolved in the early 1980s proposing that DA facilitates W and inhibits slow wave sleep (SWS) and REMS (Gillin, 1981; Monti 1982; Wauqui- er, 1985). Two different categories of DA receptors, D1 and D2, were initially identified in the CNS on neuronal processes postsynaptic to DA axons. Presynaptic receptors that regulate the release of DA have also been described, and their pharmacological properties resemble these of the D2 receptor. The application of molecular cloning techniques during the last few years led to the identification of novel DA receptors. They include the D5 receptor which displays structural and pharmacological similarities to the D1 receptor, and the D3 and D4 receptors which show molecular and pharmacological ho- mology with the D2 receptor and are included in the D2-like subfamily. The D3 receptor is expressed as postsynaptic receptor primarily in the olfactory tubercle, the islands of

Calleja, the nucleus accumbens and the bed nucleus of the stria terminalis. The D3 receptor is also expressed in the substantia nigra (pars compacta) and the ventral tegmental area, which indicates that it may serve a presynaptic function.

Studies where the predominantly D2 agonists apomorphine, bromocriptine, and quinpirole had been given to laboratory animals over a wide range of doses showed biphasic effects shuch that low doses decreased W and increased SWS and REMS, whereas relatively large doses induced opposite effects. Pretreatment with haloperidol, sulpiri- de or YM-09151-2 reversed the effects of both small and large doses of the predomi- natly D2 agonists on sleep variables. The D1 receptor agonist SKF 38393 significantly increased W and suppressed REMS in the rat. On the other hand, the selective D1 antagonist SCH 23390 increased SWS and REMS, and suppressed W. Pretreatment with compound SCH 23390 prevented the effects of the D1 agonist.

The D3 preferring DA agonist pramipexole induced dose-related effects on sleep and W. A low dose increased SWS and REMS and reduced W, and a large dose induced opposite effects. Compound YM-09151-2 effectively antagonized the increase of W and reduction of SWS induced by the large dose of the preferring D3 agonist, thus suggesting the involvement of postsynaptic D2 receptors. In conclusion, pharmacological studies support the proposal that DA plays a role in the regulation of sleep and W. Reduction of DA activity following stimulation of presynaptic D2 or D3 receptors or blockade of postsynaptic D1 or D2 receptors favors the occurrence of SWS and REMS, whereas the increase of DA activity following activation of postsynaptic D1 or D2 receptors or blockade of presynaptic D2 receptors favours the occurrence of W. Nevertheless, increased W after stimulation of postsynaptic D2 receptors could be partly related to the higher incidence of motor activity. 66 Actas de Fisiología 7, 2001

GABA PLAYS A CRITICAL ROLE IN REM SLEEP

Barbara E Jones Montreal Neurological Institute, McGill University, Montreal, Canada

GABAergic neurons are distributed through the brainstem reticular formation, from where they innervate surrounding neurons and in some cases project to forebrain or spinal cord targets. They accordingly have the capacity to inhibit or disinhibit large populations of brainstem neurons, including cholinergic, monoaminergic and other reticular neurons that are critically important in the generation of sleep-wake states and the ascending or descending components of those states. To date, it has not been possible to identify recorded neurons in the brainstem as GABAergic. We have used expression of c-Fos, the immediate early gene, as a reflection of neural activity in association with dual-immunostaining for the synthetic enzyme for GABA, glutamic acid decarboxylase (GAD) in order to examine potential changes of activity in GABAergic neurons in association with REM or paradoxical sleep (PS) [1-3]. In order to provoke extreme activity changes, we examined c-Fos expression in three groups: one submitted to ~48 hours PS deprivation (PSD: 0% PS), one submitted to ~48 hours PS deprivation followed by ~3 hours PS recovery (PSR: average ~28% PS), and the third maintained in the same recording condition for ~48 hours as control (PSC: average ~14% PS). PS recovery was associated with an increase in the number of c-Fos+/GAD+ neurons in multiple areas of the central gray region, raphe, and reticular formation. GABAer- gic neurons surrounding the noradrenergic locus coeruleus neurons and serotonergic dorsal raphe neurons increased in number expressing c-Fos, while the monoaminergic neurons decreased in number in association with PS. These local GABAergic neurons thus appear to be responsible for the important inhibition of adjacent noradrenergic and serotonergic neurons during the state of PS. In the medulla, GABAergic neurons increased in number expressing c-Fos, while serotonergic neurons decreased in number in association with decreasing EMG amplitude reflecting muscle atonia of PS. The GABAergic neurons within the medullary raphe nuclei might inhibit local serotonergic neurons, to produce a disfacilitation of spinal motor neurons, yet also as larger projection neurons produce a direct inhibition of the motor neurons. Finally, through the reticular formation, in only one region did GABAergic neurons expressing c-Fos decrease in number in association with PS, that is within the reticular nucleus pontis oralis. In this region alone, there was also a significant increase in the number of large reticular neurons expressing c-Fos in association with PS. GABAergic neurons may thus play a critical role in the disinhibition of particular reticular neurons that generate the ascending and descending components of the state of PS.

1. Maloney, K.J., L. Mainville, and B.E. Jones. J. Neurosci., 1999. 19: 3057-3072. 2. Maloney, K.J., L. Mainville, and B.E. Jones. J. Neurosci., 2000. 20: 4669-4679. 3. Maloney, K., L. Mainville, and B.E. Jones. Submitted. Actas de Fisiología 7, 2001 67

LOCATION OF THE GABAERGIC NEURONS GATING PARADOXICAL SLEEP ONSET: A GAD AND CTB DOUBLE LABELING STUDY

Pierre-Hervé Luppi, Romuald Boissard, Damien Gervasoni, Bruno Barbagli and Patrice Fort INSERM U480, Lyon Cedex 08, France

Iontophoretic ejection of bicuculline (Bic) in the rat pontine dorsal subcoeruleus nucleus (SubCD) induces a long-lasting paradoxical sleep-like state (PS-like, Boissard et al., 1999). To localize the neurons responsible for the tonic GABAergic inhibition of the SubCD PS-on neurons, the retrograde and anterograde tracer, cholera toxin B subunit (CTb) was ejected in the same site that Bic. Then, CTb and GAD or ChAT immunohistochemistry were combined. After CTb ejections in the SubCD, numerous retrogradely-labeled neurons were distributed in the lateral preoptic area, the nucleus of the central amygdala, the posterior and lateral hypothalamic area, the deep mesencephalic nucleus, the ventro-lateral periaqueductal gray (PAG) and the nuclei pontis oralis and caudalis. A substantial number of neurons were observed in the lateral bed nucleus of the stria terminalis, the zona incerta, the substantia nigra reticulata, the lateral paragi- gantocellular nucleus, the parvocellular, gigantocellular, gigantocellular alpha and ventral reticular nuclei. On sections double-stained with CTb and choline acetyltransferase, only a few cholinergic neurons were retrogradely-labeled in the laterodorsal tegmental and pedunculopontine nuclei. In contrast, a substantial number of GAD/CTb double- stained neurons were observed in the dorsal part of the deep mesencephalic reticular nucleus (DPMe), the rostral pontis oralis (rPnO) and the ventral part of the caudal pontis caudalis (vPnC). A small number of double-labeled neurons were observed in the lateral hypothalamic area and the zona incerta, the substantia nigra reticulata and in the medullary reticular nuclei. These results suggest that GABAergic neurons located in the DPMe, rPnO and/or vPnC could tonically inhibit during waking and slow wave sleep the PS-on neurons located in the SubCD. Supporting the hypothesis that the GABAergic neurons in the DPMe are playing this role, it has recently been shown in cats that pressure ejections of muscimol (a GABAA agonist) in the dorsal DPMe (Sas- tre et al.2, 1996) induced a strong PS hypersomnia. To test this hypothesis in rats, we made in head-restrained animals iontophoretic applications of muscimol in the DPMe (10mM, 100nA). Muscimol ejections during 20 min in the DPMe in rats induced a strong PS hypersomnia. To label the pathways involved, CTb or PHAL were ejected in the effective sites through a barrel of the micropipette. On CTb- and PHAL- stained sections, anterogradely-labeled varicose fibers were intermingled with CTb retrogradely- labeled neurons in the SubCD, confirming the existence of a reciprocal connection between the vlPAG and the SubCD. In addition, Mus was ejected in one rat during 1h30 to induce Fos staining. This rat showed 60% of PS during the first hour of Mus ejection and Fos positive neurons were seen in the SubCD. From these results, we propose that the onset and maintenance of PS is controlled by an inhibitory projection from a population of GABAergic neurons localized in the DPMe to the SubCD and other reticular regions containing PS-executive neurons. Symposium 17

Molecular and cellular correlates of sleep, wakefulness and sleep deprivation

Chairperson: Chiara Cirelli (USA) Tarja Porkka-Heiskanen,, University of Helsinki, Finland Chiara Cirelli, University of Wisconsin, MA, USA Raymond Cespuglio, Claude Bernard University, Lyon, France Louis De Lecea, Scripps Institute, San Diego, CA, USA

Symposium 18

Adult Sleepwalking

Chairperson: Claudio Bassetti (Switzerland) Claudio Bassetti,Universitätskliniken, Zürich, Switzeland Mark Mahowald,University of Minneapolis, MN, USA Jacques Montplaisir, Université de Montréal, Québec, Canada Geert Mayer,University of Marburg, Germany Mehdi Tafti, Neuropsychiatrie, Chene-Bourg, Switzerland 70 Actas de Fisiología 7, 2001

ENERGETIC PROCESSES UNDERLYING THE SLEEP-WAKE CYCLE

Raymond Cespuglio, Larissa Netchiporouk, Nathalyia Shram Unit INSERM 480, C . Bernard University, 8 av. Rockefeller 69373 Lyon – France

To date, despite the great variety of studies related to glucose and lactate metabolism, only a limited number of them deal with the changes occurring in the extracellular glucose or lactate levels throughout the sleep-wake cycle. The present work was thus undertaken in order to define how glucose consumption and lactate production fluctuate in relation with the sleep-wake states. Glucose and lactate sensors were made as described 1, 2 using a 30 ?m-diameter carbon fiber glued into a pulled glass micropipette. Substrate- specific enzymes glucose oxidase or lactate oxidase was immobilized on the active part of the sensor (carbon fiber) and covered with an acetate cellulose membrane. In vitro and in vivo, differential normal pulse voltammetry (DNPV) was used to detect H2O2 produced by the enzyme-catalyzed oxidation of either glucose or lactate. In vivo, the biosensors were inserted with a microlift into the somatosensory cortex of freely moving rats (male Sprague-Dawley) also equipped with polygraphic electrodes. For statistics an ANOVA followed by a LSD post-hoc test was used. The biosensors linearity range (glucose: 0.3-6.5 mM; lactate: 0.1-2.0 mM) was adequate for monitoring the cortical level of the analytes. The basal extracellular glucose concentration in the cortex of conscious rat was checked at 0.59 ± 0.3 mM during the light period. Spontaneous significant variations occur in the mean glucose level during slow-wave sleep (SWS = +13 %) and paradoxical sleep (PS = -11 %) as compared to the waking state (W = 100 %). It is to be noticed that during long periods of active waking (aW), glucose level tended towards a decrease which became significant after 15 min. (aW = -32%). On the contrary, during long episodes of SWS, it tended towards an increase which became significant after 12 min. (SWS = +28 %). According to these data, PS is pointed out as a state dependent on the energy availability and SWS as a period of energy saving. Mean concentration of the extracellular lactate in the cortex of the conscious rat was estimated at 0.4 ± 0.17 (n = 12, light period). Spontaneous changes occurred during SWS (-16.2%) versus W referenced at 100% and PS (+8.5%) versus SWS referenced at 100%. Finally, during the active waking (aW) triggered by a water puff stress, lactate level rose significantly in accordance with the animal activity (+53%,) versus W referenced at 100%. These results suggest that during W the astrocytes, coupling neuronal activity to energy metabolism, allow the transformation of the blood-borne glucose into lactate. After uptake, this substrate can fuel the neurons through the oxidative phosphorylation. The lactate increase observed during PS, after the quiescent situation of SWS, indicates that the lactate pathway becomes active again. Neuronal uptake of this lactate may fulfill the oxidative phosphorylation in order to supply the important ATP need of PS.

References: 1. Netchiporouk LI Shram NF Jaffrezic-Renault N Martelet C Cespuglio R (1996) In vivo brain glucose measurements: differential normal pulse voltammetry with enzyme-modified carbon fibber electrodes. Anal. Chem. 68, 4358-4364. 2. Shram NF Netchiporouk LI Martelet C Jaffrezic-Renault N Bonnet C Cespuglio R (1998) In vivo voltammetric detection of rat brain lactate with carbon fiber electrodes coated with lactate oxidase. Anal. Chem. 70, 2618-2622. Actas de Fisiología 7, 2001 71

CHANGES IN GENE EXPRESSION ACROSS THE SLEEP-WAKING CYCLE

Chiara Cirelli Department of Psychiatry, University of Wisconsin/Madison, USA

Distinctive categories of genes change their expression in the brain in response to the transition from sleep to wakefulness, as well as after sleep deprivation. Thus, sleep and wakefulness differ not only in terms of behavior, metabolism, and neuronal activity, but also in terms of brain gene expression. Until now, almost all the genes expressed at higher levels during sleep do not correspond to known sequences. On the other hand, most of the genes that are upregulated after periods of spontaneous wakefulness and sleep deprivation are known and can be grouped into several functional categories: immediate early genes/transcription factors, genes related to energy metabolism, growth factors and adhesion molecules, chaperones and heat shock proteins, vesicle and synapse-related genes, neurotransmitter receptors, transporters, enzymes, and others. This suggests that several basic cellular functions are affected by the arousal state of the animal. The increase in the expression of genes regulating mitochondrial activity and glucose transport may underlie a compensatory response of the brain to the increased metabolic demand of wakefulness. The high expression during wakefulness of genes related to neurotransmission and synaptic activity can also be part of a compensatory response. Moreover, many of the genes upregulated during wakefulness and sleep deprivation relative to sleep are involved in neural plasticity, suggesting that plastic changes, in as much as they require the induction of genes, occur during wakefulness rather than during sleep. Several heat shock proteins and chaperones involved in protein fold- ing and endoplasmic reticulum functions also show higher expression in wakefulness than in sleep. The analysis of the genes induced by long-term sleep deprivation is still in progress, but one transcript whose expression increases in proportion to the duration of wakefulness has already been identified. This transcript codes for the enzyme aryl sulfotransferase involved in the catabolism of catecholamines, and its induction may signify that an important function of sleep is to prevent the uninterrupted activity of the noradrenergic system. A connection between sleep and catabolism of catecholamines is also suggested by recent results obtained in Drosophila. Rest deprivation in Droso- phila is associated with an increased expression of arylalkylamine N-acetyltransferase (aaNAT1), an enzyme implicated in the catabolism of monoamines and functionally related to aryl sulfotransferase. Moreover, a Drosophila mutant in which the transcriptional level and activity of the aaNAT1Dat enzyme is deficient (Datlo) shows a greater and more prolonged rest rebound after rest deprivation relative to wild-type flies. Thus, the accumulation of monoamines in the brain may trigger sleep homeostasis. 72 Actas de Fisiología 7, 2001

ADULT SLEEPWALKING: A STUDY OF 74 PATIENTS

Claudio Bassetti, Dario Vadilonga, Mehdi Tafti** Neurologische Universitätskliniken, Zürich and Bern, IUPG-Division, Neuropsychiatrie, Chêne-Bourg**, Switzerland

Background Sleepwalking affects 1-4% of the adult population and may arise from the activation of thalamo-cingulate pathways and persisting deactivation (typical of sleep) of other thalamo-cortical pathways.1 The clinico-neurophysiological spectrum of adult sleepwalking (ASW) is poorly known. Patients: from 1969 to 2000 the diagnosis of ASW was made in 74 pts (males: females=49:25; mean age=41 years, range: 16-81). Methods Clinical data: Patients (pts) charts were reviewed. We assess 43 of them by means of personal interview, standard questionnaire on sleep habits/characteristics (108 questions), and ASW questionnaire (29 questions). Data were compared with those of a normal control population (n=40). An awake EEG was recorded in 43 pts, a long-term EEG in 18 pts, and a conventional polysomnography (PSG) in 53 pts. A brain-MRI was obtained in 15 pts. HLA-Typing was performed in 16 pts. Results: Clinical data: presenting complaints were ASW (59%), sleep-disordered breathing complaints (12%), hypersomnia (9%), insomnia (9%), and others (11%). Final main diagnoses included NREM parasomnia (77%), REM parasomnia (5%), overlap parasomnia (3%), epilepsy (1%), and others (14%). Associated neurological disorder were found in 18%, psychopathological factors in 23%, and alcohol abuse in 7%. Sleepwalking in childhood was reported by 58% (absent in 22%, unknown in 20%). A positive family history for sleepwalking was present in 24% (absent in 53%, unknown in 23%). In association with ASW awareness during deambulatory activity was reported (sometimes/often or always) by 35%, injuries by 19%, violence by 32%, and eating by 37%. Bruxism (33%), sleeptalking (72%), dreaming (91%), nightmares (56%), screaming at night (42%), hypnagogic/-pompic sleep paralysis (9%/16%) and hallucinations (26%/21%) were more commonly reported than in our normal control population. The wake EEG was abnormal in 15/43 pts, and the long-term EEG monitoring in 7/18 pts. On PSG an apnea-hypopnea index >10 was found in 11/53 pts. In the first 30 PSG reviewed in detail, episodes of sleepwalking were documented in one patient, confusional arousals in 50%, and increased muscle tonus or phasic activity during REM in 25% of pts. The brain MRI was abnormal in 6/11 pts. More data on PSG and genetic analysis will be presented at the meeting. Conclusion 1) Clinical features of ASW differ from those of childhood sleepwalking; 2) Our data suggest the existence of an overlap (and occasionally co-existence) between ASW and REM sleep behavior disorder.

Reference: 1) Bassetti al., Lancet 2000; 356:484-5. Actas de Fisiología 7, 2001 73

ADULT SLEEPWALKING: DIFFERENTIAL DIAGNOSIS

Mark W. Mahowald MN Regional Sleep Disorders Center, Hennepin County Medical Center Dept. of Neurology, University of MN Medical School, Minneapolis, MN, USA

Complex motor behavior arising from the sleep state was formerly thought to be a unitary phenomenon, due to sleepwalking, and was commonly attributed to psychiatric or psychological causes. Recent clinical and polygraphic analysis has revealed that such complex behaviors are, in fact, the result of a large number of very different conditions, most of which are diagnosable and treatable. Most, in fact, are not the manifes- tation of psychiatric disorders and are far more prevalent than previously suspected (for instance, 10% of the adult population occasionally experiences sleepwalking episodes) (1, 2). The most common complex motor parasomnias are: Disorders of arousal (confusional arousals, sleepwalking, sleep terrors) REM sleep behavior disorder Nocturnal seizures Psychogenic dissociative disorders Clinical experience has clearly indicated that these conditions may perfectly mimic one another, and that they cannot be accurately diagnosed by history alone. Extensive polygraphic and video monitoring are mandatory to arrive at a correct diagnosis, which should lead to effective treatment. The fact that these disorders may be very bothersome, and potentially injurious or violent to the patient or others underscores the importance of precise diagnosis (3). Further study of these fascinating conditions, particularly disorders of arousal and REM sleep behavior disorder, which are due to state dissociation, will undoubtedly lead to greater understanding of sleep and brain function (4).

References:

1) Mahowald MW, Schenck CH. Kryger MH, Roth T, Dement WC, editors. Principles and Practice of Sleep Medicine. 3rd ed. Philadelphia: W. B. Saunders; 2000. p. 724-741. 2) Broughton RJ. Kryger MH, Roth T, Dement WC, editors. Principles and Practice of Sleep Medicine. 3rd ed. Philadelphia: W. B. Saunders; 2000. p. 693-706. 3) Mahowald MW, Schenck CH..Kryger MH, Roth T, Dement WC, editors. Principles and Practice of Sleep Medicine. 3rd ed. Philadelphia: W. B. Saunders; 2000. p. 786-795.) Ma- howald MW, Schenck CH. Archives Italiennes de Biologie 2001;139:269-300. 74 Actas de Fisiología 7, 2001

ADULT SLEEPWALKING: PATHOPHYSIOLOGY AND TREATMENT

Geert Mayer, Eduard Leonhardt, Thomas Penzel Hephata Klinik, Schwalmstadt-Treysa, Germany University of Marburg, Department of Internal Medicine, Marburg, Germany

Introduction: Patients with frequent sleepwalking (SW) rarely do so under sleep laboratory conditions. Sleep deprivation, alcohol consumption and fever are some of the conditions that may cause SW episodes. The dynamics of slow wave sleep play a pivotal role in pathophysiology and treatment of SW. Sleepwalkers display high voltage delta activity prior to SW. Slow wave activity SWA is higher during slow wave sleep SWS prior to SW and increases significantly immediately prior to SW, arousal index, wake after sleep onset and SWS fragmentation are significantly higher than in controls (Espa 2000). 65% of our sleepwalkers report occasional dream recall prior to SW raising the question if SW represents a dissociation from NREM sleep. We report data on pathophysiology of SW from sleep deprivation studies and polysomnographic studies (PSG) relating to behavior, power spectra and REM-sleep. Based on a review of literature on treatment of SW we present own therapeutic data. Methods: A PSG study of 10 sleepwalkers and controls was performed before and after SD. PSG of 13 sleepwalkers before and after SW were investigated for no. of arousals (10 min before and after SW), no. of stage shifts, amount of SWS, time from REM to SW and from SW to next REM and compared to controls. Movements in both studies were classified by videometry as regular movements during SWS and subclinical or clinical SW. Patients were either treated nonpharmacologically with „self-hypnosis“ or clonazepam. Results: After SD there is significant increase of arousal length and of SW. Increase of arousal during SWS, SWS and sleep fragmentation is nonsignificant. SW is preceeded by significant longer SWS comprising higher delta power than regular movements in the 10 minutes prior to SW. The duration from SW to the next REM period is significantly longer than for regular movements. Number of arousals before and after SW, time from sleep onset and REM to SW does not differ from controls. Alpha, sigma and delta power of patients with and without SW did not differ throughout sleep-cycles and total sleep time. Sigma power was reduced prior to SW. SWA during total sleep time in sleepwalkers is lower than in controls. SW during SWS is preceeded by significantly more SWA than regular behavior. Treatment results are based on a questionnaire of our complete database of SW patients. About 40% of them required treatment, either due to frequent SW with consecutive nonrestorative sleep and daytime sleepiness or self-injury. 90% of the patients treated with “self-hypnosis“ succeeded in suppressing SW, whereas only few patients required clonazepam treatment, for which they had low compliance and low response. Conclusion: SW occurs in close vicinity to REM. PSG data shows increased pressure for SWS caused by an all-night reduction of SWA in sleepwalkers. Due to this SWA pressure its increase prior to SW episodes suppresses sigma power and REM-sleep, resulting in the dissociation of REM-sleep, which may explain the occurance of dreams. In contrast to Espa et al. increase of arousals during total sleep time and SWS could not be found. Only few sleepwalkers desire therapeutical interventions. Therapies meeting the criteria of evidence based medicine treatment with clonazepam and “self-hypnosis“ were as favourable in our patient population as reported in the literature. Reference: 1) Espa et al., Clin Neurophysiol 2000;111:929-39 Actas de Fisiología 7, 2001 75

SLEEPWALKING: ELECTROPHYSIOLOGICAL FINDINGS AND EFFECTS OF SLEEP DEPRIVATION

Jacques Montplaisir, Steve Joncas Steve, Antonio Zadra Centre d’étude du sommeil et des rythmes biologiques Hôpital du Sacré-Cœur de Montréal, Université de Montréal, Québec, Canada

Sleepwalking occurs in 1% to 15% of the general population and is more frequent in children and young adolescents (4% to 17%) than in adults (1% to 4%). Analyses of sleep macrostructure (e.g., sleep architecture, cyclic patterns of sleep stages) show no significant differences between adult somnambulistic patients and control subjects, except for a greater number of arousals out of SWS in sleepwalkers compared to controls.1,2,3 Spectral analysis of the EEG is a reliable method of assessing the dynamics of different sleep EEG frequency bands across the night. Since somnambulistic patients have repeated awakenings from early NREM sleep cycles they have significantly lower SWA power than control subjects during the first NREM cycle.1,2 Our results suggest that spectral analyses of SWA across NREM cycles can contribute to our understanding of the disorders of arousals. Delineating patterns of SWA associated with various parasomnias could help clarify their underling pathophysiology. Sleepwalking rarely occurs in the sleep laboratory and the fact that the incidence of somnambulistic episode is much lower in the laboratory than in the patient’s normal environment constitutes a potential obstacle in establishing a polysomnographically based diagnosis. Data exist to support the idea that sleep deprivation could be a good diagnostic tool for adult somnambulism.3,4 We recently studied 10 sleep walkers and 10 normal controls.4 The 38 hours of sleep deprivation resulted in a significant increase in the frequency of somnambulistic episodes during recovery sleep, but also in the complexity of those behavioral manifestations. The data therefore indicate that sleep deprivation results in an increased probability of obtaining a polysomnographically- based diagnosis.

References: 1) Gaudreau H, Joncas S, Zadra A, Montplaisir J. Dynamics of slow-wave activity during the NREM sleep of sleepwalkers and control subjects. Sleep 2000;23:755-760. 2) Espa F, Ondzé B, Deglise P, Billiard M, Besset A. Sleep architecture, slow wave activity, and sleep spindles in adult patients with sleepwalking and slep terrors. Clinical Neurophysiology 2000;111:929-939. 3) Mayer G, Neissner V, Schwarzmayr P, Meier-Ewert K. Sleep deprivation in somnambulism. Effect of arousal, deep sleep and selep stage changes. Nervenarzt 1998;69(6):495-501. Symposium 19

REM behavior disorder: sleep medicine bridge to neurophysiology, neurochemistry and psychology

Chairpersons: Rosalia Silvestri (USA) & Mark Mahowald (USA) Adrian R. Morrison, University of Pennsylvania, PA, USA Mark Mahowald, University of Minneapolis, MN, USA Rosalia Silvestri, Harvard University, Boston, MA, USA Ilonka Eisensehr, University of Munich, Germany

Symposium 20

Basal ganglia during sleep

Chairperson: Jean Askenasy (Israel) Jean Askenazy Oscar Gershanik, Hospital Francés, Buenos Aires, Argentina David B. Rye, Emory University, GA, USA Gregor Wenning Seiji Nishino, Stanford University, Palo Alto, CA, USA

Symposium 21

The process of transition into REM sleep

Chairperson: Giovani Zamboni (Italy) José M. Calvo, Instituto Nacional de Psiquiatría, México DF, México Ennio Vivaldi, Universidad de Chile, Santiago, Chile Giovani Zamboni, University of Bologna, Italy J.Allan Hobson, Harvard University, Boston, USA 78 Actas de Fisiología 7, 2001

INCREASED MUSCLE TONE DURING REM SLEEP CORRELATES WITH STRIATAL PRESYNAPTIC DOPAMINERGIC DYSFUNCTION: COMPARISON OF IPT AND IBZM SPECT INVESTIGATIONS IN SUBCLINICAL AND CLINICAL IDIOPATHIC REM SLEEP BEHAVIOR DISORDER WITH PARKINSON’S DISEASE AND CONTROLS

Ilonka Eisensehr, Rainer Linke, Klaus Tatsch, Bita Kharraz, Huberta von Lindeiner, Franz J Gildehaus, Thomas C Wetter, Claudia Trenkwalder, Johannes Schwarz and Sohey Noachtar Department of Neurology, University of Munich, Marchioninistr. 15, 81377 Munich, Germany

Introduction: REM sleep behavior disorder (RBD) is characterized by complex behavior during REM sleep. The etiology of this disorder is still unknown, but a recent study showed an association between RBD and Parkinson’s disease (PD) and we found reduced striatal dopamine transporters in idiopathic clinically manifested RBD. Methods: To further investigate the the role of striatal dopaminergic function in controling muscle tone during REM sleep, we studied striatal postsynaptic dopamine D2-receptor density with 123I-IBZM and the striatal presynaptic dopamine transporter with 123I-IPT using single-photon emission computed tomography (SPECT) in the following groups: 1. 8 patients without a history of movement disorders during sleep but with polysomnographically confirmed continuously increased muscle tone of more than 15% of their REM sleep without associated movements (subclinical RBD), 2. 8 patients with polysomnographically confirmed idiopathic clinically manifested RBD (clinical RBD), 3. 8 controls without a history of movement disordrs during sleep (continuously increased muscle tone of less than 15% of their REM sleep, no movements during REM sleep), and 4. 8 patients with PD stage Hoehn & Yahr I (IBZM SPECT was not performed in PD patients). Results: Controls and patient groups were similar in age and sex. Patients with clinically manifested RBD had significantly reduced striatal IPT binding compared with the controls (clinical RBD: right = 3.18±0.43, left =3.2±0.43, vs controls: right=4.07±0.29, left=4.07±0.30, p<0.0001) but significantly higher striatal IPT binding compared with the striatum contralateral to the symptomatic body side of the PD patients (PD: ipsilateral=3.25±0.35, p=0.798, contralateral= 2.51±0.28, p=0.003). SPECT results of patients with subclinical RBD suggest that their IPT uptake is intermediate: IPT uptake is lower compared with controls, but higher than in patients with clinical RBD (subclinical RBD: right=3.56±0.21, p(clinical RBD)=0.038, left=3.55±0.25, p(clinical RBD) =0.083), p(controls)<0.002). IBZM uptake was not significantly diffeent between the groups (controls, subclinical and clinical RBD). Con- clusion: This study demonstrates that reduced striatal dopamine transporters are one pathophysiologic mechanism of continuously increased muscle tone during REM sleep. Results are given as mean±SD Actas de Fisiología 7, 2001 79

THE ANIMAL BIOLOGY OF RBD: TOWARD AN UNDERSTANDING OF HUMAN PATHOLOGY

Adrian R Morrison Department of Animal Biology, School of Veterinary Medicine, University of Penn- sylvania, Philadelphia, USA

Abstract: Animal models take various forms: naturally occurring diseases that assume essentially identical form in humans and animals, conditions created in animals with the hope of mimicking a known human condition and experimental preparations that have led to the diagnosis of a problem in humans. These latter two preparations may differ considerably from the human condition, but they in some way lead to an insight that results in diagnosis. REM without atonia (REM-A) is an example of the last category, for awareness of this phenomenon led to the recognition of REM Behavior Disor- der (RBD) (Mahowald and Schenck 2000). Because of animal-rightist attacks on the validity of animal models, it is important that scientists understand the strengths and weaknesses of various animal models. Unlike RBD of the idiopathic type, specific lesions of the pontine tegmentum lead to REM-A. However, the various types of behavior that occur in REM-A, which depend on lesion site (Hendricks et al 1982), are observed in patients expressing RBD. Thus, the cells and fibers destroyed by the experimental lesions may give a clue to what structures of the human brain must degenerate or be altered in some way for RBD to appear. The possibilities will be reviewed. These include, at a minimum, pontine cholinergic neurons, pontine tegmental neurons projecting caudally to the medullary inhibitory area and afferents from the superior colliculus, interstitial nucleus of Cajal, basal ganglia and the amygdala (Morrison 1988, Rye 1997, Zagrodzka et al. 1998).

1. Hendricks, J.C, Morrison, A.R. and Mann, G.L. (1982). Brain Res., 239, 81-105. 2. Mahowald, M. and Schenck, C.H. (2000). In Principles and Practice of Sleep Medicine 3rd ed., pp. 724-741, M.H. Kryger, T. Roth and W. C. Dement (eds.), New York: Saunders. 3. Morrison, A.R. (1988). Arch. Ital. Biol., 126, 275-289. 4. Rye, D.B. (1997). Sleep, 20, 757-788. 5. Zagrodzka, J., Hedberg, C.E., Mann, G.L. and Morrison, A.R. (1998). Behav. Neurosci., 112, 589-602. 80 Actas de Fisiología 7, 2001

RBD INDUCTION AND SUPPRESSION BY DRUGS. WHAT DO THEY TELL US ABOUT THE NEUROCHEMISTRY OF THIS DISORDER?

Rosalia Silvestri Brigham and Women’s Hospital, Department of Neurology Harvard Medical School Boston, MA USA

The drugs which have so far proved helpful in the treatment of RBD as well as those that have instead been noticed to cause RBD-like side effects allow us to formulate some hypotheses about the neurotransmitters involved in this disorder and their pharmacology. A crucial role seems to be played by the dopamine system, as documented by the high incidence of RBD in dopamine related neurological disorders (eg. Parkin- son’s disease, Lewi body dementia, and OPCD) and by the effectiveness of L-dopa and dopa-agonists in reducing RBD symptoms. That the serotonergic system plays a major role is documented by the fact that SSRI and tricyclic antidepressants may cause RBD- like symptoms while drugs such as imipramine and clonazepam may reduce RBD symptoms. The cholinergic system also appears to be involved as indicated by biperiden induced RBD and by the adverse effects of the new ACH esterase inhibitors such as donepezil. All of these drugs presumably act through the brainstem systems responsible for aminergic inhibition and cholinergic excitation of phasic activity during REM. The drug effects which are observed in any given patient at any given time depend upon the activity and excitability levels of these intrinsic neuromodulatory systems. Actas de Fisiología 7, 2001 81

SLEEP INDUCED VERSUS SLEEP DEPRIVED MOTOR BENEFIT IN PARKINSON’S DISEASE, A PARADOX?

Oscar S. Gershanik Sección Enfermedades Extrapiramidales Centro Neurológico, Hospital Francés, BUENOS AIRES

Whether motor performance depends on sleep in patients with Parkinson´s disease is not known. An early hypothesis suggests increased dopamine storage during sleep in remaining dopaminergic nerve terminals. On the other hand, there is some evidence that sleep benefit in Parkinson´s disease, the experience of improved motor performance in the morning after a night´s sleep, is independent of sleep quality. Sleep benefit is even associated with more awakenings during the night and a tendency towards more abnormal polygraphic sleep measures compared to findings in patients without sleep benefit. Animal research provides plenty of evidence, that not only dopamine interferes in sleep wakefulness regulation, but that selective rapid eye movement (REM) sleep deprivation influences the dopamine system. Selective REM sleep deprivation increases behavioral and motor responses to dopamine agonists such as apomorphine, an effect possibly mediated by supersensitivity of postsynaptic dopamine receptors. In parkinsonian rat models, a significant increase in ambulatory behavior and rearing in animals with bilateral lesions of the nigrostriatal tract was seen after 3 days of selective REM sleep deprivation. REM sleep deprivation also induces a clear increase in apomorphine-induced turning behavior in rats with unilateral 6-OHDA nigro-striatal lesions, suggesting increased supersensitivity of dopamine receptors after REM sleep deprivation. Links between sleep deprivation and the dopamine system or motor function have also been evidenced by clinical studies: The therapeutic effect of sleep deprivation as an established treatment modality in patients with major depression has been related to dopaminergic mechanisms. The involvement of the dopamine system in the antidepressant effects of sleep deprivation has also been demonstrated by IBZM spect. Interestingly, several studies in depressed patients have noted, that sleep deprivation not only improved symptoms of depression, but also influenced motor performance: Increased motor output following sleep deprivation was found both in depressed responders and nonresponders and motor symptoms of depression were improved. The existence of a positive influence of sleep deprivation on motor function in Parkinson´s disease is at present controversial. Reports on the results of sleep deprivation in PD range from those showing significant improvement to no improvement at all. There is perhaps a subgroup of patients that in a manner similar to what has been observed with the phenomenon of sleep benefit, show improved motor scores after sleep deprivation, possibly indicating the existence of different response types to both sleep deprivation and sleep benefit. 82 Actas de Fisiología 7, 2001

THE DOPAMINERGIC SYSTEM, SLEEP AND SLEEP-RELATED INVOLUNTARY MOVEMENTS

Seiji Nishino Center for Narcolepsy, Stanford University School of Medicine, Palo Alto, USA

Using the canine model of narcolepsy (hypocretin receptor 2-mutated), we have demonstrated that the dopaminergic (DA) systems, especially dopamine uptake, release and D2/3 autoreceptor systems, are critically involved in the regulation of sleep and cataplexy [1-4]. With regard to the regulation of sleep, enhancement of DA transmission at DA terminals is likely to be one of the most important mechanisms for the pharmacological control of vigilance [3,4]. This is demonstrated by a series of pharmacological experiments: (1) DA uptake inhibitors potently increase EEG arousal and their in vivo potency on EEG arousal correlates well with their in vitro affinity to the dopamine transporter (DAT) [1]. (2) Am- phetamine derivatives (d-, l- and l-methyl amphetamines) have different potencies with respect to EEG arousal and their in vivo potency is correlated with their effects on increase in extracellular DA levels at the DA terminals [2]. (3) Furthermore, wake-promoting effects of amphetamine, GBR 12909 (a DA uptake inhibitor) and modafinil are completely diminished in DAT knockout mice [5]. In contrast, DAT uptake inhibitors have little effect on cataplexy. Systemic administration or local perfusion of D2/3 autoreceptor agonists into the substantia nigra, ventral tegmental area and A11, however, significantly aggravate cataplexy dogs [4, 6, 7]. Thus it is likely that modulation of midbrain/diencephalon dopamine is required to modulate cataplexy, while enhancement of DA transmission at the terminal is sufficient for mediation of the wake promoting effects of DA compounds. Considering the fact that narcoleptic dogs are very sensitive to DA D2/3 agonist effects (which induce cataplexy/sleep) when compared to control dogs [4]. DA/hypocretin interaction is likely to be underlying in the pathophysiology of canine narcolepsy. Narcolepsy is often associated with periodic leg movements during sleep (PLMS) [8]. Involvement of the DA system in PLMS is well established, and DA agonists have been widely used for the treatment of PLMS in humans. Interestingly, we also observed PLMS-like “leg movements” during sleep in our narcoleptic dogs [9]. Furthermore, we found that DA compounds used for the treatment of PLMS in humans (especially D2/3 agonists) aggravate cataplexy and enhance drowsy state, but reduce PLMS-like movements of these animals. We therefore hypothesized that D2/D3 mechanisms involved in the control of muscle tone during cataplexy may also be involved in the occurrences of PLMS: Dysregulation of the dopaminergic system may exist in canine narcolepsy, and this may be specifically involved in sleep-related motor inhibition (i.e. cataplexy) and activation (i.e. PLMS) and abnormal sleep tendency. The canine narcolepsy model may therefore be an invaluable resource for research on PLMS and other dopamine-related sleep/movement disorders. “The DA system, sleep and sleep-related involuntary movements” will be further discussed.

[1] Nishino, S., et al., Sleep Research Online, 1998. 1: p. 49-61. [2] Kanbayashi, T., et al.Neuroscience, 2000. 99(4): p. 651-659. [3]Nishino, S., et al., Sleep Res, 1996. 25: p. 317. [4] Reid, M.S., et al., Brain Res, 1996. 733: p. 83-100. [5] Wisor, J.P., et al., J Neurosci, 2001. 21(5): p. 1787-94. [6] Okura, M., et al., Sleep, 1999. 22S: p. S1. [7] Honda, K., et al., Neuroreport, 1999. 10(14): p. 3111- 3118. [8] Wittig, R., et al., Clin Electroencephalogr, 1983. 14(3): p. 130-4. [9] Okura, M., et al., Psychiatry Clin Neurosci, 2001. 55(3): p. 243-4. Actas de Fisiología 7, 2001 83

THE ROLE OF TEMPORAL LOBE AMYGDALA IN THE ONSET OF REM SLEEP

José M Calvo, Karina Simón-Arceo and Ignacio Ramírez-Salado Instituto Nacional de Psiquiatría, México D.F., México

Rapid eye movement sleep (REM) is a distinctive sleep stage that alternates with episodes of slow wave sleep (SWS) and is always preceded and accompanied by the occurrence of ponto-geniculo-occipital (PGO) waves. In humans, REM sleep is also accompanied by visual and auditory hallucinatory phenomena and by emotional and autonomic changes all related to rapid eye movements (REMs) and correlated with PGO wave activity. A widely distributed pontine neuronal generator network has been postulated for REM sleep and the pontine tegmental nuclei which play a major role in PGO wave generation have been considered as a major component of the neuronal network for REM sleep induction. However, the forebrain structures possibly involved in mediating hallucinatory, emotional and autonomic phenomena have not yet been completely localized. Furthermore, the precise role of such forebrain structures in the onset mechanisms of REM sleep is still to be well elucidated. We have found that PGO-propagated field potentials can also be recorded in structures of the limbic system such as the cingulate gyrus, hippocampus and the amygdala. Among these limbic structures, the amygdala is anatomically and physiologically linked to the brainstem nuclei responsible for PGO wave generation. We have found that electrical stimulation of the amygdala during REM sleep provokes a significant enhancement of PGO wave density, and that an electrolytic lesion of the central amygdaloid nucleus decreases PGO wave density. The amygdaloid PGO-propagated potentials and the facilitatory influence of the amygdala upon PGO wave activity strongly suggest a facilitatory interaction between the amygdala and the PBL region. On the other hand, we found that microinjetions of carbachol (8 µg / 0.5 µl) as well as vasoactive intestinal peptide (VIP)(0.33 µg / 1 µl) into the central amygdaloid nucleus, induced a prolonged (6 days) enhancement of both REM sleep and its preceding slow wave sleep episodes with PGO waves (sommeil phasique a ondes lenses, SPHOL). PGO wave density was also progressively and significantly enhanced during the six days following carbachol or VIP microinjections. These findings strongly suggest that amygdala constitutes a component of the forebrain neuronal network for REM sleep induction. 84 Actas de Fisiología 7, 2001

FROM SYNAPSE TO GENE PRODUCT: PROLONGED FOS EXPRESSION FOLLOWING A SINGLE MICROINJECTION OF CARBACHOL.

J. Allan Hobson Laboratory of Neurophysiology, Department of Psychiatry, Harvard Medical School, Boston, USA.

How the brain modifies its regulatory systems in response to the application of a drug is unknown especially over the long term of weeks and months. We have developed a model system approach to this question by manipulating the cholinergic cell groups of the laterodorsal and pedunculopontine tegmental (LDT/PPT) nuclei in the pontomesencephalic tegmentum (PMT) which are known to be actively involved in the timing and quantity of rapid eye movement (REM) sleep. In a freely moving feline model, a single point source microinjection into the caudolateral PMT of the cholinergic agonist carbachol conjugated to latex nanospheres elicits a long-term enhancement of one distinguishing phasic event of REM sleep, ponto-geniculo-occipital (PGO) waves, lasting 5 days but without any significant change in REM sleep or any other behavioral state. To determine potential regions involved in this long-term response, Fos immunohistochemistry was used to identify neurons activated during this long-term PGO enhancement (LTPE) effect. The number of Fos-immunoreactive (Fos-IR) neurons surrounding the caudolateral PMT injection site decreased sharply by postcarbachol day 3, while the number of Fos-IR neurons at remote sites in the more rostral LDT/PPT increased >30-fold and remained at a high level following the course of LTPE. These results suggest that carbachol’s acute effects induce LTPE via cholinergic receptors with subsequent transsynaptic activation of the LDT/PPT driving the LTPE effect. These experiments are the first demonstration of a sustained expression of Fos in response to pharmacological stimulation of the brain. Actas de Fisiología 7, 2001 85

SHORT-TERM HOMEOSTASIS OF REM SLEEP AND THE TIMING OF THE TRIGGERING OF A REM EPISODE

Ennio A. Vivaldi and Adrián Ocampo-Garcés Programa Disciplinario de Fisiología y Biofísica, Instituto de Ciencias Biomédicas, Facultad de Medicina, Universidad de Chile

A relationship between the length of an individual REM sleep episode and the length of the interval that immediately follows it has been demonstrated in long-term recordings of spontaneous sleep-wake cycling in the rat and other species. Consequently, the amount of REM sleep in a given episode permissively allows for the time that the animal will stay without entering a new episode. It follows that one mechanism that participates in regulating the timing in which a new transition into REM sleep is due or allowed, must somehow be related to the amount of REM sleep occurring in a given episode. REM sleep expression is markedly modulated throughout a 12:12 L:D schedule. Short-term REM sleep homeostasis is present at all times, even at the phases when REM sleep expression is minimal. The parameters of the regression curve of interval length on episode length can be assessed at different phases of the L:D schedule. Impli- cations for the interaction between circadian and homeostatic mechanisms of REM regulation can be discussed. Another issue to be discussed is whether this relationship has to do only with the REM state, i.e., REM presence vs. REM absence, or whether there is a differential regulatory relationship between REM sleep and the expression of NREM sleep within the interval. Further support for the existence of a short-term homeostatic mechanism can be found in the demonstration of a carry-over effect from one episode-interval cycle to the next. The fact that a given interval turns out to be shorter or longer than expected for its preceding episode is reflected in a tendency for the next cycle to show an error in the opposite, i.e., corrective, direction.

Research funded by Fondecyt Grants. 86 Actas de Fisiología 7, 2001

AMBIENT MODULATION OF REM SLEEP OCCURRENCE

On a physiological ground, the wake-sleep cycle may be viewed as a succession of periods during which physiological homeostatic regulations are either fully operant (wakefulness and NREM sleep) or impaired (REM sleep, REMS). The distribution of the periods separating subsequent REMS episodes (REMS intervals) is bimodal and this has allowed to distinguish, in the Rat, REMS episodes occurring at long intervals (>3min, single episodes) from those occuring at short intervals (£3min, sequential episodes) (1). The occurrence of REMS in the form of sequential episodes is reduced in ambient conditions characterized by an increase in the intensity of physiological homeostatic regulations (1, 2). For example, the decrease of ambient temperature below the lower thermoneutral value, proportionally reduces the occurrence of sequential REMS episodes, but not that of single REMS episodes. These results suggest that the dual classification of REMS episodes has a functional relevance. Such an hypothesis has been initially explored in the frequency domain of EEG activity, by analyzing the time course of the change in the average relative power of the delta band (0.75-4.0 Hz) in the 90s preceding the onset of a REMS episode (Fig. 1).

200 180 d, e, f 160 Fig.1. Time course of the average ) 140 relative power of the EEG delta band 120 (0.75-4.0 Hz) for the following REMS c 100 intervals (s): 80 a (30-60); b (61-120); c (121-180); b a 60 d (181-600); e (601-960); f (961-1800). elative delta power (% r 40 The onset of REMS is indicated by the 20 zero value on the x-axis. 0 -90 -60 -30 0 time before REMS (s)

The results show that the average relative delta power for REMS intervals £3min (Fig. 1, a-c), that define sequential REMS episodes: a) changes in proportion to the interval duration; b) is lower than that observed for the intervals >3min (Fig. 1, d-f).

1) R. Amici et al., J.Sleep Res. (1994) 3, 250-256. 2) R. Amici et al., Brain Res. (1998) 781, 252-258. FOCUS GROUPS Actas de Fisiología 7, 2001 89

Focus Group 1

What should be asked to a theory on the function of the sleep

Chairperson: Ruben V. Rial (Spain) Pier Luigi Parmeggiani, University of Bologna, Italy Jerome Siegel, University of California Los Angeles, USA Dennis McGinty, University of California Los Angeles, USA Antonio Guiditta, University of Naples, Italy R. J. Berger, University of California Santa Cruz, USA K. Honda, Tokyo Medical and Dental University, Japan Giulio Tononi, University of Wisconsin, MA, USA Ruben V. Rial, University Illes Balears, Spain C. Heller, Stanford University, USA T. Wehr, NIH Bethesda, USA P. Maquet, University of Liege, Belgium

Focus Group 2

Treatments for the Restless Legs Syndrome: Does one initial treatment fit all or should some types of patients get different treatments?

Chairpersons: Richard P. Allen (USA) & Wayne Hening (USA) Richard P. Allen, Johns Hopkins University, USA Phillip Becker Christopher J. Earley Claudia Trenkwalder, Max Planck Institute, Germany Marco Túlio de Mello, Sao Paulo, SP, Brazil Ronald D. Chervin, University Hospital, MI, USA Wayne Hening, Robert Wood Johnson Medical School, New York, USA Jacques Montplaisir, Université de Montréal, Québec, Canada) Arthur Walters, Robert Wood Johnson Medical School, New York, USA 90 Actas de Fisiología 7, 2001

WHAT SHOULD BE ASKED TO A HYPOTHESIS ON THE FUNCTION OF SLEEP

Rubén V Rial

Objectives

1) To clarify the properties that should be meet by a hypothesis on the function of sleep 2) Perhaps, to reduce the number of proposed theories, which is, evidently excessive. Researchers do not even agree on whether the main function of the sleep could be unique or multiple

Why?

The function of the sleep is perhaps the most searched and elusive goal proposed to sleep researchers. This Focus Group aims at clarifying the question. The function of the sleep can be understood as the function of mere rest in front of activity, but also can be understood in a restricted sense, for instance as the function of the slow wave mammalian sleep or that the REM. It can be even restricted to small traits, for instance, to the function of dreams or that of rapid eye movements. From other viewpoint, over a dozen of functions for the sleep have been proposed and it seems convenient to analyze why any one of them seems not to be acceptable to main community of sleep researchers. From this viewpoint, the arguments of necessari- ness and sufficiency as well as their constraints should be examined in relation to every one of the proposed functions. Finally, a point often neglected is that a well sustained hypothesis, in addition to its experimental support, should explain why the proposed function might only be achieved during sleep and not during other behavioral states. Focus Group 3

Sleep, stress, and the HPA axis

Chairpersons: Joëlle Adrien (France) & Veronique Viot-Blanc (France)

Joëlle Adrien, INSERM, Paris, France Francesco Benedetti, University of Milan, Italy Goron Hajak, University of Gottingen, Germany A. Vgontzas, Penn State University, PA, USA Mark Opp, University of Texas, TX, USA Deborah Suchecki, UNIFEST-EPM, Sao Paulo, Brazil Véronique Viot-Blanc, Hopital Lariboisière, Paris, France

Focus Group 4

Adrenergic inputs of the preoptic/basal forebrain in sleep/wakefulness regulation

Chairperson: V. Mohan Kumar (India) Subimal Datta, Boston University, MA, USA Noor Alam, Los Angeles, USA C. E. Berridge, Wisconsin, USA Barbara E. Jones, McGill University, Montreal, Canada Vijay Ramesh, Harvard Medical School. Brockton, MA, USA Thannickal C. Thomas, Los Angeles, USA Hrudananda N. Mallick, New Delhi, India T. Osaka, Osaka, Japan Hitoshi Matsumura, Osaka, Japan V. Mohan Kumar, New Delhi, India 92 Actas de Fisiología 7, 2001

SLEEP AND STRESS RESPONSE IN VARIOUS MOUSE MODELS OF DEPRESSION AND ANXIETY

Benjamin Boutrel, Daniela Popa, Michel Hamon and Joëlle Adrien INSERM U288, CHU Pitié-Salpêtrière, 91 Bd de l’Hôpital, 75013 Paris - France

In humans, sleep impairments and disturbances of hypothalamic-pituitary-adrenocorti- cal (HPA) function are observed in endogenous depression and anxiety disorders (1, 2). In animals, stress-elicited activation of the HPA axis triggers a REM sleep rebound (3) as well as adaptive changes at serotoninergic receptors (4). On the other hand, treatment with antidepressants that inhibit serotonin reuptake (SSRIs) induces an enhancement of ACTH secretion (5) but a reduction of REM sleep amounts (6). We have examined whether a link actually exists between sleep regulation and stress response in various strains of mice which exhibit depression- or anxiety-like behav- iours. For this purpose, we chose: -1. transgenic mice which express reduced levels of glucocorticoid receptors specifically in the brain (GRi) (7), and mice selectively bred to diverge in their immobility response in the tail suspension test (helpless versus non- helpless) (8); 2- knock-out mice that do not express either the 5-HT1A (5-HT1A-/-) or the 5-HT1B (5-HT1b-/-) type of serotoninergic receptors; whereas 5-HT1A-/- mutants exhibit increased levels of anxiety in a number of tests (9), 5-HT1B-/- mutants are, in contrast, less anxious than paired wild-type mice (10). Under baseline conditions, enhanced amounts of REM sleep were observed in helpless, transgenic and knock-out mice as compared with their respective controls (see also ref. 11). In addition, helpless mice also exhibited increased levels of light SWS (at the expense of wakefulness), and 5-HT1B-/- mice had decreased amounts of SWS. After an immobilization stress of 90 minutes, a REM sleep rebound was observed (for up to 36 consecutive hours) in control mice, but not in any of the experimental groups. These results indicate that impairments of serotoninergic or HPA function by different means in mice are associated with sleep-wakefulness alterations that do not correlate with the behavioural phenotype observed (depression- or anxiety-like). The absence of REM sleep rebound in these animals after acute stress might reflect a lack of adaptive capacities at the level of the serotoninergic and/or HPA systems.

1- Barden N, Reul JM, & Holsboer F. T.I.N.S., 1995, 18:6-11. 2- Benca RM, et al.. Arch. Gen. Psychiatry 1992, 49: 651-668. 3- Houdouin F, et al. Exp. Brain Res. 1991, 85: 153-162. 4- Young AH, et al. Neuropharmacology 1992, 31: 433-438. 5- Gilbert F, et a. Eur. J. Pharmacol. 1988, 147: 431-439. 6- Maudhuit C, et al. J. Neuropharmacology 1994, 33: 1017-1025. 7- Pepin MC, Pothier F, & Barden N. Mol. Pharmacol. 1992, 42: 991-995. 8- Vaugeois JM, et al. J. Eur. J. Pharmacol., 1996, 316: R1-R2. 9- Parks CL, et al. Proc. Natl. Acad. Sci., 1998, 95: 10734-10739. 10- Brunner D, et al. Behav. Neurosci., 1999, 113: 587-601. 11- Boutrel B, et al. J. Neurosci., 1999, 19: 3204-3212. Actas de Fisiología 7, 2001 93

CORTICOTROPIN-RELEASING HORMONE (CRH) AS A REGULATOR / MODULATOR OF WAKING

Mark R Opp Department of Anesthesiology, University of Michigan, Ann Arbor MI, USA

An essential and common feature of all living organisms is their ability to maintain a relatively stable internal physiologic environment in the face of constant and often threatening changes in their external environment. Acute stressors often induce a “fight or flight” response that is dependent on activation of the autonomic nervous system. This rapid autonomic activation is followed within minutes by increased glucocorticoid production, which mobilizes energy stores that may be required for “fight or flight”. Because CRH is a regulator of both the HPA axis and the autonomic nervous system, this peptide provides the organism with a response system capable of reacting to stressors of multiple modalities and on different time scales. For example, CRH and the autonomic nervous system respond rapidly to stimuli such as perceived threat, a cir- cumstance under which a delayed response could be detrimental. However, other types of stressors, such as infection, also induce CRH and HPA axis activity; these responses occur on a much longer time scale than responses to stimuli such as perceived threat. In response to immune challenge, one function of elevated glucocorticoids within the CNS is to moderate actions of immunomodulators before they become detrimental. Therefore, CRH, the HPA axis, and the autonomic nervous system constitute a system that allows the animal to respond appropriately to stressors of many different modalities. Studies of CRH have historically been “stress” studies, and indeed CRH is now considered the major (although not only) mediator of behavioral, autonomic, and endocrine responses to stressors. In this Focus Group we will discuss our research that indicates CRH is involved in the regulation / modulation of spontaneous waking in the absence of overt stressors. For example, central administration into rats of specific CRH receptor antagonists or antisense oligodeoxynucleotides reduces spontaneous waking and increase NREM sleep. We have recently demonstrated that HPA axis feedback mechanisms in brain may provide an important functional link between waking and sleep and responses to immune challenge. Interfering with CRH actions in brain reduces circulating corticosteroids, which in turn releases inhibitory tone on the synthesis of the cytokine interleukin (IL)-1. As a result, mRNA expression for IL-1 increases. IL-1 is well-known as an enhancer of NREM sleep; the increase in NREM sleep after central administration of CRH receptor antagonists is abolished when animals are pretreated with anti-IL-1 antibodies. Thus, CRH via multiple mechanisms appears to regulate / modulate spontaneous waking, and to mediate many of the effects of stressors on waking and sleep.

Supported in part by NIMH grants MH52275, MH54976, MH64843 94 Actas de Fisiología 7, 2001

STRESS AND INSOMNIA IN ANXIOUS AND DEPRESSED PATIENTS

Véronique Viot-Blanca and Chrystèle Deschauxb aHôpital Lariboisière, Paris, bSleep laboratory, Grenoble - France

It has been postulated that the clinical course of depression could be stressful life-events, then anxious disorder and finally depression, insomnia being present as a symptom at all three stages. Thus we have examined whether there is any clinical feature in insomniac patients that supports such a continuum. 250 GP were asked to complete 4 clinical questionnaires for insomnia, either chronic or not, together with a Hospital Anxiety Depression Scale. Among 976 patients, 770 completed the HAD: 27.8% were purely anxious (A group: A³10, D<10), 26% were anxious and depressed (AD group: A³10, D³10) 5,1% were purely depressed (D group: A<10, D³10) and 41.2% were non-psychiatric (N group: A<10, D<10). A, AD and N groups were compared using either Student T-test or Mann-Whitney (non-parametric) for quantitative variables and Chi-2 for qualitative comparisons. Anxiety, alone or coexisting with depression was more prevalent among women (N=59%, A=76%, AD=68%-N vs A: p<0.001, N vs AD: p<0.05) and was associated with specific clinical features: more awakenings during sleep (N=56%, A=70%, AD=80% - N vs A: p<0.001, N vs AD: p<0.01), more premature waking (N=32.5%, A=41%, AD=41.5% - N vs A: p<0.05, N vs AD: p = 0.05) more non-restorative sleep (N=33%, A=59.5%, AD=50% - N vs A: p<0.001, N vs AD: p <0.001). There was however no difference between groups in the difficulty to fall asleep. As a consequence of a poor night sleep, daytime sleepiness was more prevalent in the anxious groups (N=11%, A=18%, AD=20.5% - N vs A: p<0.05, N vs AD: p = 0.05) and the patients were also more tense and irritate (N=36%, A=54%, AD=64.5%-N vs A : p<0.001, N vs AD : p<0.001) while there was no difference regarding daytime fatigue, frequency of poor nights, sleep habits or possible medical causes of insomnia. When depression was associated with anxiety (comparison between group A and group AD), the difficulty to fall asleep was more prevalent (A=74%, AD=82.5% - p<0.05), as were sleep awakenings (A=70%, AD=80.5%- p=0.01) and non- restorative sleep (A=50%, AD=59% - p=0.05) but unexpectedly not premature waking. The classical clinical presentation e.g. difficulty falling asleep being related to anxiety and premature waking to depression, was not relevant in the overall population. In the depressive subgroup, chronic insomnia (N=55%, A=63%, AD=70% - N vs A: ns, N vs AD: p<0.001) and frequent poor sleep nights i.e. 4 or more during a week (N=28%, A=36%, AD=57% - N vs A: ns, N vs AD: p< 0.001, A vs AD: p<0.05) were more often present. Daytime consequences were more severe: they felt more often tense (A vs AD p<0.05), more often gloomy (N=23%, A=25%, AD=41% - N vs A: ns, N vs AD: p<0.001, A vs AD p<0.001). Anxiety alone was not related to a longer duration or a higher severity of insomnia. However, when depression was present, insomnia was more severe regarding frequency, duration and daytime consequences. The absence of other clinical difference is supporting the hypothesis of common mechanisms for insomnia in both anxiety and depression, therefore the hypothesis of a continuum between stress, anxiety and depression. Some depres- sions could be related to repetitive chronic stresses associated with an over-reactivity of the HPA axis followed by a progressive desensitization. In this context insomnia could be a marker of stress. Actas de Fisiología 7, 2001 95

ROLE OF ALPHA-NORADRENERGIC SYSTEM IN THE MEDIAL PREOPTIC AREA IN THE REGULATION OF SLEEP AND BODY TEMPERATURE

Subimal Datta1 and Velayudhan Mohan Kumar2

1Sleep Research Laboratory, Department of Psychiatry, Boston University School of Medicine, Massachusetts, Boston, USA 2Department of Physiology, All India Institute of Medical Sciences, New Delhi, India

The rostral half of the hypothalamus, especially the preoptic region, has been shown by lesion studies to play an important role in the rat capacity to sleep. Stimulation studies have provided evidence for the involvement of this area in the rat arousal or alerting mechanism. Nerve terminals containing norepinephrine (NE) were shown to be present in the medial preoptic area (mPOA). On the basis of chemical stimulation studies, it was proposed that the NE mechanism in the mPOA had a functional role in the alerting response. Incidental observations made in some thermoregulatory studies provided some clues regarding the role of the mPOA NE system in the regulation of sleep-wakefulness. It was suggested that the release of NE in the rostral hypothalamus was associated with the control of body temperature. A study was undertaken to investigate the possible involvement of mPOA noradrenergic mechanisms in the modulation of sleep-wakefulness and thermoregulation in rats. In this study, saline, NE, phenoxybenzamine (PBZ, an noradrenergic alpha receptor antagonist) and propranolol (PROP, a beta noradrenergic receptor antagonist) were injected in the mPOA, and in some control areas adjoining the mPOA, in different groups of male rats during the day and night. Arousal was produced by NE, and sleep by PBZ when applied in the mPOA. None of the other procedures, including application of NE and PBZ in the control areas and PROP in the mPOA, produced alterations in sleep-wakefulness. NE injected in the mPOA, produced a fall in body temperature, whereas PBZ produced an increase. Saline and PROP produced no alteration. These findings provide support for the alpha noradrenergic system in the mPOA playing a physiological role in arousal and thermoregulation, and furthermore, identify a specific area of action of this system.

Research was supported by NIH research grant MH59839 and All India Institute of Medical Sciences. 96 Actas de Fisiología 7, 2001

ADRENERGIC INPUTS OF THE MEDIAL PREOPTIC / BASAL FOREBRAIN AREA IN SLEEP-WAKEFULNESS REGULATION

V. Mohan Kumar Department of Physiology, All India Institute of Medical Sciences, New Delhi, India

Experimental studies have confirmed the role of the medial preoptic area (mPOA) in the is of sleep and control of sleep-wakefulness (S-W) rhythm. One of the afferent inputs to this area, connected with these functions, comes from the noradrenergic (NE) neurons in the brain stem. The role of these NE afferents in the regulation of S-W was studied in free-moving rats with chronically implanted electrodes for recording EEG, EMG and EOG. Changes in S-W in these animals were studied after local application of adrenergic agents in the mPOA, and destruction of NE fibres innervating this area. Application of noradrenaline (which predominantly acts on alpha receptors), clonidine (alpha-2 agonist) and isoproterenol (a beta agonist), at the mPOA, produced arousal. The beta antagonist propranolol, on the other hand, did not produce any change in S-W, whereas the alpha antagonists, phenoxybenzamine and phentolamine and alpha-2 antagonist, yohimbine produced sleep in free-moving rats. This indicated the involvement of alpha receptors, and not the beta receptors, in the tonic regulation of S- W. If it is assumed that these adrenergic blockers phenoxybenzamine and phentolamine act preferentially on alpha-1 receptors, it would suggest that the arousal is normally maintained by tonic activity of the NE system. But, similar results obtained on local application of alpha-2 antagonist, yohimbine, contradict this assumption. So, it is possible that the externally applied alpha agonists and antagonists might have also acted on the alpha-2 receptors, which are predominantly present on the presynaptic terminal, and inhibited the release of norepinephrine from NE terminals. When noradrenaline was locally applied at the mPOA after eliminating most of the alpha-2 receptors (by destroying the NE terminals and presynaptic receptors), it produced sleep. Application of beta agonist, after elimination of NE terminals did not produce any change. These findings support the role of postsynaptic alpha receptors in sleep, and indicate that the NE fibres might have a hypnogenic function at the level of the mPOA through its action on the alpha-1 receptors. The role of NE fibres in the induction of sleep was further confirmed by the observation that the destruction of these fibres in the mPOA produces an increase in quiet wakefulness. Actas de Fisiología 7, 2001 97

ROLE OF THE CATECHOLAMINERGIC FIBRES OF THE PREOPTIC AREA IN THERMAL PREFERENCE OF RATS: POSSIBLE IMPLICATION IN SLEEP REGULATION

H.N Mallick Department of Physiology, AIIMS, New Delhi, India

The preoptic area (POA) regulates not only the sleep-wakefulness but also the body temperature. The sleep and thermoregulation are modified by the catecholaminergic (CA) inputs to the POA. POA lesions as well as the lesions of CA fibres of the POA, modify sleep and the body temperature. It has been suggested that the alteration in sleep after the lesion is an indirect reflection of the thermoregulatory ability of the animal. There is a shift in the temperature preferred by the animal after destruction of the POA neurons. This may contribute towards the changes in sleep. Studies on the thermal preference of the rats after CA fibres lesion of the POA have not been undertaken. Animals have the ability to regulate their body temperature by both behavioral and non-behavioral mechanisms. This study deals with the thermal preference of rats when the POA catecholaminergic terminals are destroyed with 6-OHDA. The normal rats prefer an ambient temperature of 27 0C when allowed to choose between 24, 27 and 30 0C. After lesion of the CA terminals of the POA, the rats preferred 24 0C with resultant hyperthermia. Rise in rectal temperature by destruction of the POA catecholaminergic terminals is brought down by behavioral adjustment by the animal by seeking a lower ambient temperature. Although the integrity of the POA neurons is essential for behavioral thermoregual- rory mechanism, this study shows that CA input to the POA is important for non- behavioral thermoregulation. Thus the changes in sleep after CA fibre lesion may be related to the changes in thermoregulatory ability of the animal. 98 Actas de Fisiología 7, 2001

ALTERATIONS IN MONOAMINE NEUROTRANSMITTERS AT THE MEDIAL PREOPTIC AREA AFTER SLEEP DEPRIVATION

Vijay Ramesh1 and Velayudhan Mohan Kumar2 1Department of Psychiatry, Harvard Medical School/ VA Medical Center, West Roxbury, MA 02132, USA 2Department of Physiology, All India Institute of Medical Sciences, New Delhi, India

Intracerebral microinjection, unit recording, stimulation and lesioning, have demonstrated the role of the medial preoptic area (mPOA) in the regulation of sleep. The mPOA receives noradrenergic, dopaminergic and serotonergic afferent inputs. The role played by the monoaminergic fibres in the regulation of sleep-wakefulness (S- W) is still debated. It has been reported that the deprivation of sleep produced alterations in NA, DA and 5-HT levels in various regions of the brain including the hypothalamus. If some of these monoamines are involved in the regulation of sleep, at the level of the mPOA, their levels are likely to be altered in this region after sleep deprivation. Morphological study of dendritic spines could also give information about the altered transmitter activity during sleep deprivation. Excessive synaptic activity has been shown to increase the spine density in the neocortex and mPOA. If there is any alteration in the release of some of these monoaminergic transmitters, during sleep deprivation, it is likely to produce an alteration in the spine density at the mPOA. So, in this study, in addition to the levels of monoamines, the spine density of the neurons in the mPOA was studied after total sleep deprivation (TSD). The experiments were conducted on 24 adult male Wistar rats to find out the alteration in the levels of monoamines and dendritic spine densities in the medial preoptic area and cortex after total sleep deprivation. Noradrenaline was reduced in the medial preoptic area, though there was no significant change in the cortex. Dopam- ine and serotonin were decreased both in the medial preoptic area and in the cortex. Dendritic spine counts in the medial preoptic area and the motor cortex were increased after total sleep deprivation. Enhanced release of the monoamines and their subsequent breakdown during sleep deprivation could be responsible for the decreased levels of the transmitters. Increase in synaptic activity, resulting in the enhanced release of the transmitters, might be responsible for the increased spine density. Focus Group 5

Pathophysiology of Idiopathic Hypersomnia

Chairperson: Michel Billiard (France) Michel Billiard, Hopital G. De Chauliac, Montpellier, France Claudio Bassetti, Zurich, Switzerland Sona Nevsimalova, Charles University, Prague, Czech Republic Alain Besset, Montpellier, France Jens Nilsson, Stockholm, Sweden

Focus Group 6

Genetic manipulations for understanding sleep systems: constitutive knock outs vs. inducible knock outs vs. antisense technology

Chairpersons: Robert E. Strecker (USA) & Robert McCarley (USA) Joelle Adrien, INSERM, Paris, France Michael Chase, University of California Los Angeles, USA Chiara Cirelli, University of Wisconsin, MA, USA Luis de Lecea, Scripps Institute, La Jolla, CA, USA Bertil Fredholm, Karolinska Institute, Sweden Robert Greene, Harvard University, Boston, MA, USA Yoshihiro Urade, Osaka Biosciences Institute, Japan Joan Hendricks, University of Pennsylvania, PA, USA Tarja Porkka-Heiskanen, University of Helsinki, Finland Paul Shaw, NSI Mehdi Tafti, University of Geneva, Switzerland Mahesh Thakkar, Harvard Universty, Boston, MA, USA Fred Turek, Northwestern University, IL. USA Masashi Yanagisawa, UTSW 100 Actas de Fisiología 7, 2001

GENETIC MANIPULATIONS OF THE SEROTONINERGIC SYSTEM FOR UNDERSTANDING SLEEP REGULATIONS

Benjamin Boutrel, Chloé Alexandre, Véronique Fabre, Michel Hamon and Joëlle Adrien INSERM U288, CHU Pitié-Salpêtrière, 91 Bd de l’Hôpital, 75013 Paris - France

Genetic manipulation of the serotoninergic (5-HT) system leads to alterations of 5-HT neurotransmission and provides new opportunities to investigate the role of serotonin in the regulation of sleep. Indeed, it represents an alternative to the use of pharmacological tools and, to some extent, of localized lesions of the 5-HT system, which have been until recently the main approaches to investigate this question. The development of DNA sense and antisense technology in the rat (1), and of homologous recombinaison for the knock- out of genes coding for various proteins involved in 5-HT neurotransmission in the mouse, has allowed us to explore further the role of the serotonin transporter (5-HTT) and of the

5-HT1A and 5-HT1B receptors (2, 3) in the regulation of sleep and wakefulness. 1- The 5-HT transporter: In the rat, a recombinant plasmid containing either the entire coding sequence or a partial antisense sequence of the 5-HTT gene, together with the strong and universal promoter CMV, was complexed with the cationic polymer, polyethyl- enimine, to be injected into the dorsal raphe nucleus. Down-regulation of the 5-HTT (@30%) caused by the antisense plasmid was associated essentially with a reduction of the amplitude of the sleep-wakefulness rhythm, but no modifications of sleep or wakefulness duration over 24 hours. In contrast, knock-out mice which do not express the 5-HTT exhibited, as compared with their wild-type counterparts, increased amounts of REM sleep but no change in the amplitude of sleep-wakefulness rhythm. In addition, 5-HT1A receptor-mediated sleep regulations were decreased in these mice, probably because of the occurrence of a down- regulation of this receptor type in some brain regions.

2- In knock-out mice which do not express the 5-HT1A or 5-HT1B receptors, an enhancement of REM sleep amounts was observed. These modifications could be mimicked in wild-type controls by pharmacological blockade of these receptors. These data indicate that both receptor types exert a tonic inhibitory influence on REM sleep. In addition, 5-

HT1A and 5-HT1B-/- mutants exhibited modifications in the sensitivity of other serotoninergic receptor types (notably the 5-HT1A receptors in 5-HT1B-/- mice and vice versa), which suggests that adaptive changes at 5-HT neurotransmission develop as a consequence of the knock-out. To conclude, with regard to the serotonin transporter, the antisense plasmid approach and the constitutive knock-out did not generate equivalent results, which might be accounted for by the difference in species. In addition, it should be emphasized that neurotransmission impairments have different time courses, one being transient (DNA antisense) and the other definitive (knock-out). Furthermore, constitutive knock-outs undergo several adaptive processes involving other proteins than those coded by the invalidated gene, which render the interpretation of the corresponding sleep phenotype rather difficult. Inducible knock-outs will probably help to overcome this difficulty. Finally, we believe that association of genetic manipulations with relevant pharmacological ones should allow further progress in the understanding of sleep mechanisms.

1- Fabre V, Boutrel B, Hanoun N, et al. J. Neurosci. 2000, 20: 5065-5075. 2- Boutrel B, Franc B, Hen R, Hamon M, & Adrien J. J. Neurosci. 1999, 19: 3204-3212. 3- Monaca C, Boutrel B, Hen R, Hamon M, & Adrien J. J. Sleep Res. 2000, 9 S1: 133. Actas de Fisiología 7, 2001 101

CORTISTATIN AND OTHER GENES REGULATED BY SLEEP DEPRIVATION

Luis de Lecea and Marta Calbet Dept of Molecular Biology, MB-10. The Scripps Research Institute. 10550 N. Torrey Pines, La Jolla CA, USA

To determine differences in gene expression in the rat cerebral cortex after sleep deprivation, we have analyzed the gene expression profile of 24 hours-sleep-deprived rats using 2 complementary approaches: cDNA microarrays containing 5200 cDNAs and subtractive hybridization which allows us to study differences in the concentration of low abundant mRNAs. For both experiments we have obtained mRNA of rats that have been sleep deprived for 24 hours by gentle handling and mRNA of rats that have been deprived of sleep and allowed to recover with eight hours of sleep. We obtained a group of 64 clones that showed enriched expression in the cortex after forced waking, some of which were identified as novel species as they do not match any known sequence. The cDNA clones of known sequence encode proteins functionally involved in basic cellular functions. Our results suggest that cortical neurons adapt to increased neuronal activity by increasing the turnover of proteins involved in signal transduction, intracellular transport, RNA and protein synthesis machinery, peptide processing, metabolism, neural plasticity and neurotransmission. 102 Actas de Fisiología 7, 2001

QUANTITATIVE GENETIC DISSECTION OF SLEEP IN MICE

Mehdi Tafti1 and Paul Franken2 1Biochemistry & Neurophysiology, Department of Psychiatry, University of Geneva, Switzerland 2Department of Biological Sciences, Stanford University, USA

Sleep as other complex traits is under the control of multiple genes that each may decrease or increase one or several of its phenotypic manifestations resulting in a normal distribution. Although single gene manipulation can provide valuable information, the major limitation is the number of candidate genes to be analyzed. Mutagenesis can theoretically overcome this limitation by inducing mutations in any single gene. How- ever, genetic screens in this approach are for fully penetrant dominant and recessive mutations and therefore cannot identify small effect mutations that may turn out to be essential for some aspects of the phenotype. In addition, with currently available technologies, recording and analyzing sleep of thousands of mice in a mutant screen does not seem feasible, at least in a single academic laboratory. Quantitative trait loci analysis (QTL) has been proposed as a powerful approach in the genetic dissection of complex traits. Although, natural allelic variation of genes with small effect can be mapped through QTL analysis, the final identification of sequence variants (quantitative trait nucleotides or QTNs) in the QTL region with biologically significant effects on the phenotype, may represent prohibitive efforts in terms of both phenotyping and genotyping. Therefore a combination of several approaches is necessary for mapping and candidate gene analysis. High-resolution QTL mapping in conjunction with future developments in high-throughput phenotyping and genotyping, and the availability of whole genome sequences of several major mouse strains should identify candidate genes to be investigated. All those candidate genes can then be mutated either by classical homologous recombination (KO) or better by the newly developed serial nested chromosomal deletions (Cre-loxP). Systematic crosses between increasing and decreasing QTL allele and the complete set of mutants should physically identify the functional QTN. Since most QTNs will probably be involved in gene regulation rather than being mutations affecting the protein function, further gene expression profiling with microarray or TaqMan technologies and gene translation analyses should be used to uncover the molecular mechanisms involved. Meanwhile, since little is known about the molecular basis of sleep, we believe that any single approach may reveal interesting candidate pathways to follow by other genetic and genomic means. Actas de Fisiología 7, 2001 103

ANTISENSE TECHNOLOGY: A SIMPLE TOOL TO MANIPULATE GENE EXPRESSION

Mahesh M Thakkar and Robert McCarley Psychiatry, Harvard Medical School, VA Medical Center, Brockton, USA

Pharmacological techniques for determining effects of specific receptor types on behavior are limited by the specificity of the agent and ability to reach the receptor of interest. Molecular techniques of constitutive receptor “knock-outs” have proven powerful but are limited by epistatic (developmental) compensations and interactions with other genes, as well as lack of specificity to brain areas of interest. Inducible knockouts avoid some of the developmental constraints of the constitutive knock-outs but may be somewhat difficult to localize and likely will not be reversible -in addition to being quite time- and labor-intensive to make and thus expensive. It occurred to us that, if capable of being successfully applied to our sleep systems, antisense technology utilized to mimic a “localized, reversible receptor knock-out” might be useful. The main principle behind the antisense technology is to block or inhibit the transfer of information from the gene to the protein by interfering with the function of the mRNA. Despite the availability of wide range of potential mechanisms to modulate RNA function, antisense oligonucleotide mediated cleav- age of mRNA by RNAse H enzyme is among the most widely used technique. The antisense oligonucleotide forms a hybrid with the mRNA and this oligonucleotide:RNA hybrid is subject to degradation by RNAase H. The most widely used antisense oligonucleotide are phosphorothioate deoxyoligonucleotide. The replacement of one non-bridge oxygen atom with sulphur make these modified phosphorothioates oligonucleotides extremely resistant to endo- and exo-nucleases and prevents degradation. Further, these modified antisense oligonucleotides are water soluble, penetrate the cells efficiently, readily form antisense: RNA hybrids and are potential substrates for RNAase. The antisense technology, in most cases, has been used to manipulate neurotransmitter receptors and almost all the studies to date have used the microinjection technique to deliver the antisense locally at the site of choice in the brain. Using rat as our animal model, we conducted two experiments using the antisense technology. We used the microdialysis perfusion instead of microinjection for antisense delivery to specific brain area. Microdialysis perfusion allows a precise control to regulate both the concentration and the time of delivery. In the first study, we perfused antisense oligodeoxynucleotide against mRNA of orexin type II receptor, locally in the subcoerulean reticular formation. We chose orexin type II receptor because canine narcolepsy/cataplexy involves abnormalities of the orexin II receptor. Our choice of target site was based on the fact that the subcoerulean PRF is the key regulator of REM sleep and muscle atonia. In the second study, we perfused antisense oligodeoxynucleotide against mRNA of adenosine A1 receptor in the cholinergic basal forebrain. Adenosine A1 receptor in the cholinergic basal forebrain has been implicated in the control of sleep-wakefulnes. The result of our studies revealed that 1) microdialysis of antisense oligonucleotide against the mRNA of orexin type II receptor in the subcoerulean PRF induced cataplexy and increased REM sleep. 2) Microdialysis perfusion of adenosine receptor antisense in the basal forebrain induced wakefulness and reduced recovery sleep following six hours of sleep deprivation. The data from our studies clearly demonstrate that the antisense technology, combined with an efficient delivery system like microdialysis perfusion, can be successfully used to elucidate the molecular mechanisms in sleep research. 104 Actas de Fisiología 7, 2001

SLEEP BEHAVIORS OF GENE-KNOCKOUT MICE FOR PROSTAGLANDIN D2 SYNTHASE AND RECEPTOR

Yoshihiro Urade and Osamu Hayaishi Department of Molecualr Behavioral Biology, Osaka Bioscience Institute, Suita, Japan

Prostaglandin (PG) D2 is the most potent endogenous sleep-promoting substance. We recently reported that human PGD synthase (PGDS)-overexpressing transgenic mice exhibited excessive amounts of non-REM sleep in response to the noxious stimulus by tail clipping, coupled with significant increase in PGD2 production in the brain. To further investigate the involvement of PGD2 in sleep regulation, we generated gene knockout (KO) mice for PGDS and determined the sleep/awake pattern of the PGDS- KO mice and also KO mice for PGD2 receptor, the D type of prostanoid (DP) receptor. Infusion of PGD2 into the lateral ventricle of the brain induced non-REM sleep in wild-type (WT) mice but not in the DP-KO mice, indicating that PGD2 induced non- REM sleep in mice via the DP receptor. However, no difference was observed in the sleep/wake patterns between these KO and WT mice, suggesting that the effects of the deficiency of the PGD2-mediated system on the sleep regulation are efficiently compensated by other system(s). We found abnormalities of the sleep regulation in these KO mice after sleep deprivation. Both PGDS-KO and DP-KO mice showed less non- REM rebound sleep after sleep deprivation for 6 hr than the WT mice. On the other hand, the amount of REM rebound sleep after sleep deprivation was almost the same among PGDS-KO, DP-KO, and WT mice. These results indicated that PGD2, PGDS, DP receptor, and their genes are involved in the homeostatic regulation of non-REM sleep. Focus Group 7

What is the clinical significance of periodic leg movements during sleep?

Chairperson: Ronald D. Chervin (USA) Ronald D. Chervin, University Hospital, Ann Arbor, MI, USA Arthur Walters Wayne Hening, New York, USA Flavia Consens, University Hospital, Ann Arbor, MI, USA Richard Allen, MD., USA Mario-Giovanni Terzano, University of Parma, Italy Christian Guilleminault, Stanford University, Palo Alto, CA, USA Mark Mahowald, University of Minneapolis, MN, USA

Focus Group 8

Sleep and Anesthesia; commonalities and differences at cellular and systemic levels

Chairperson: Leszek Kubin (USA) Gyorgy Buzsaki, Rutgers State University, NJ, USA Barbara E. Jones, McGill University, Montreal, Canada Yoshimasa Koyama, Fukushima Medical University, Japan Igor Timofeev, University Laval, Canada Robert P. Vertes, Florida Atlantic University, USA 106 Actas de Fisiología 7, 2001

MORNING HYPOTENSION AND UARS

C Guilleminault, RA Stoohs and JL Faul Stanford University Sleep Disorders Center

Orthostatic intolerance (OI) and orthostatic hypotension (OH) cause a diminished quality of life and an increased incidence of falls. This study examines the presence of OI, OH, and low blood pressure in 4409 subjects referred for sleep studies (overnight 18- channel polysomnography). A low resting arterial blood pressure (systolic blood pressure less than 105 mmHg, diastolic blood pressure less than 65 mmHg) was present in 101 subjects (2.3 %). Low blood pressure was observed commonly in subjects with UARS [(23 %)], but was uncommon in subjects with OSAS [(0.06%)], parasomnia [1/ 127, (0.7%)], restless leg syndrome [(0.9%)], and psychological insomnia [(0.9%)]. In order to investigate blood pressure homeostasis, 15 subjects with OI and the upper airway resistance syndrome (UARS), 15 subjects with OSAS (matched for age and BMI), 5 insomniac subjects with low blood pressure, and 15 control subjects underwent tilt-table testing. Fifteen subjects with UARS and OI and fifteen control subjects also underwent 24-hour ambulatory blood pressure monitoring. Subjects with OI and UARS had lower mean daytime systolic blood pressure (119 ± 28 mmHg) and lower mean daytime diastolic blood pressure (75 ± 18 mmHg) than control subjects (131 ± 35 mmHg and 86 ± 19 mmHg, respectively) (p < 0.05). During tilt-table testing, subjects with UARS and a history of OI had a greater drop in blood pressure (27 ± 3 mmHg), than control subjects (7.5 ± 1.6 mmHg), subjects with OSAS (7.5 ± 1.6 mmHg) (p < 0.01), and insomniacs (9.4 ± 1.5 mm Hg) (p<0.01). We conclude that a fifth of subjects with UARS have low blood pressure and complain of OI in the morning. Morning tilt-table testing of these subjects demonstrates resting tachycardia and OH consistent with hypovolemia. Low early morning blood pressure may increase the risk of stroke in older patients. Actas de Fisiología 7, 2001 107

SLEEP AND ANESTHESIA; COMMONALITIES AND DIFFERENCES AT CELLULAR AND SYSTEMIC LEVELS

Leszek Kubin1, Gyorgy Buzsaki2, Barbara E. Jones3, Yoshimasa Koyama4, Igor Timofeev5 and Robert P. Vertes6 1Dept. of Animal Biology and Ctr. for Sleep & Respiratory Neurobiol., Univ. of Pennsylvania, Philadelphia, PA, USA; 2Ctr. for Neuroscience, Rutgers State Univ. of New Jersey, Newark, NJ, USA; 3Dept. Neurol. & Neurosurgery, McGill Univ., Montreal, Canada; 4Dept. of Physiology, Fukushima Medical University, Japan; 5Lab. Neurophysiol., Fac. Medicine, Univ. Laval, Quebec, Canada; 6Florida Atlantic Univ., Ctr. for Complex Systems, Boca Raton, FL, USA.

While sleep and anesthesia are different states, they have common traits and may share common mechanisms. Based on this notion, anesthetized animals have been used to study sleep mechanisms at levels spanning from a single cell to large and distributed neuronal networks. In addition, complex sleep- and arousal-like neural events may occur spontaneously, or can be elicited pharmacologically (e.g. by carbachol) and by sensory stimulation under (or despite) general anesthesia. The questions then arise: What is common, and what is different, in the mechanisms of sleep and anesthesia?; To what extent can sleep/wake-like cellular behaviors and interactions observed in anesthetized animals help us understand the mechanisms of natural sleep?; Is there sleep under anesthesia? The expert participants of this Focus Group investigated cellular mechanisms of sleep in both naturally behaving and anesthetized animals. They will present results, address the above questions and share their thoughts on the relevance and limitations of sleep studies under anesthesia based on cellular recordings from cortico-thalamic, basal forebrain, hippocampal and brainstem neurons. 108 Actas de Fisiología 7, 2001

FIRING PATTERNS OF HIPPOCAMPAL NEURONS ARE PRESERVED DUR- ING SUBSEQUENT SLEEP EPISODES AND MODIFIED BY NOVEL AWAKE EXPERIENCE

Gyorgy Buzsaki Center for Neuroscience, Rutgers University, Newark, NJ, USA

What determines the firing rate of cortical neurons in the absence of external sensory input or motor behavior, such as during sleep? Here we report that, in a familiar environment, the discharge frequency of simultaneously recorded individual CA1 pyramidal neurons and the co-activation of cell pairs remain highly correlated across sleep- wake-sleep sequences. However, both measures were affected when new sets of neurons were activated in a novel environment. Nevertheless, the mean firing rate of the pyramidal cell population remained constant across behavioral states and testing conditions. The findings suggest that long-term firing patterns of single cells can be modified by experience. Increased firing rates of recently used neurons are associated with a concomitant decrease in the discharge activity of the remaining population. Actas de Fisiología 7, 2001 109

INHIBITION BY CLONIDINE OF NORADRENERGIC (NA) A5 AND LOCUS COERULEUS NEURONS ELICITS REM SLEEP-LIKE EFFECTS IN URETHA- NE-ANESTHETIZED RATS

Victor Fenik, Vitaliy Marchenko, Richard O. Davies and Leszek Kubin Department of Animal Biology and Center for Sleep and Respiratory Neurobiology, University of Pennsylvania, Philadelphia, PA 19104, USA

REM sleep is characterized by increased activity of mesopontine cholinergic neurons and quiescence of NA cells. Consistent with the current concepts that cholinergic neurons play an active role in the generation of REM sleep, whereas silencing of aminergic cells has only a “permissive” role, cholinergic stimulation [1] or antagonism of adrenergic receptors within the dorsal pontine tegmentum [2] facilitate REM sleep, whereas activation of adrenoceptors in the dorsomedial pons [3] or stimulation of locus coeruleus (LC) cells [4] impede generation of REM sleep in behaving animals. In urethane-anesthetized, paralyzed and artificially ventilated rats, one can reliably and re- peatedly elicit REM sleep-like episodes (comprising cortical activation, hippocampal theta rhythm, silencing of pontine NA neurons, and suppression of hypoglossal (XII) nerve activity) by microinjecting a cholinergic agonist, carbachol (10 nl, 10 mM), into a discrete site within the dorsal pontine tegmentum [5-7]. We tested whether, in this preparation, suppression of pontine NA cell activity is sufficient to produce electrocor- tical, motor and respiratory changes similar to the REM sleep-like effects of carbachol. To silence NA cells, we injected an alpha-2 adrenergic agonist, clonidine (10-20 nl, 0.75 mM), into the left and then the right A5 region and the left and right LC in 4 rats. In all animals, 70-160 s following the last injection, the EEG was activated, hippocampal theta rhythm appeared, XII nerve activity was suppressed, and the respiratory rate reduced. In one rat, bilateral clonidine injections into A5 alone elicited the same effect. The episode durations (2.3-4.6 min), magnitude of the suppression of XII nerve activity (by 87%±6(SE)), and pattern of EEG changes resembled the REM sleep- like effects produced by pontine carbachol under the same conditions. The clonidine effect was of an all-or-none type; injections that did not elicit the entire response had no effect on any of the recorded signals. Thus, at least in this model, complex REM sleep- like changes can be elicited consequent to bilateral inactivation of pontine NA neurons, including the A5 group whose contribution to REM sleep mechanisms has received little attention.

References: 1. Bourgin,P., et al. 1995. NeuroReport, 6: 532-536. 2. Bier,M.J., & McCarley,R.W. 1994. Brain Res., 634: 333-338. 3. Tononi,G., et al. 1991. Pflügers Arch., 418: 512-518. 4. Bourgin,P., et al. 2000. J. Neurosci., 20: 7760-7765. 5. Fenik,V., et al. 1999. Soc. Neurosci. Abstr. 25: 2144. 6. Fenik,V., et al. 1999. Sleep Res.Online 2 (Suppl. 1): 30. 7. Kubin,L., 2001. Arch. Ital. Biol., 139: 147-168. 110 Actas de Fisiología 7, 2001

INTRINSIC AND SYNAPTIC FACTORS IN ACTIVITY OF NEOCORTICAL NEURONS DURING NATURAL WAKING-SLEEP CYCLE: AN INTRACELLULAR STUDY

Igor Timofeev, François Grenier and Mircea Steriade Dept. Anatomy and Physiology, Laval University, Quebec, Canada

Three major currents have an impact on the membrane potential of cortical neurons in normal conditions. These are leak current, voltage activated intrinsic currents and synaptic currents. All these compound currents are affected by activities of neuromodulatory systems and thus differently expressed in sleep and wake states. Here, in in vivo experiments during the natural sleep-waking cycle, we identified some currents that are implicated in the maintenance of the membrane potential of neocortical neurons during slow-wave sleep and quiet wakefulness. Slow-wave sleep was characterized by the presence of two distinct modes of the membrane potential. The down state was characterized by a great reduction of synaptic activity, linear voltage-current relations, sensitivity to intracellular Cs+; without intracellular injections of DC currents, the membrane potential rounded from 70 to 90 mV suggesting its domination by a leak conduc- tance. The mean membrane potential in the up state was 62 mV and it was composed of synaptic events. Voltage-current relations were linear in ranges above and below 65 mV, with a change in the slop of the voltage-current curve around 65 mV, indicating activation of a persistent Na+ current at depolarized levels. Waking was characterized by a unimodal distribution of the membrane potential with mean voltage around 62 mV (range from 53 mV to 72 mV). Some regular spiking neurons stopped or significantly reduced their firing during the transition from slow-wave sleep to quiet wakefulness, but still displayed synaptically generated fluctuations of the membrane potential. The membrane potential during waking was not sensitive to intracellular Cs+. Voltage-current relations were similar to the up state, but with notably steeper slope indicating higher input resistance. Antidromic invasion was potentiated during the tonic activation processes in waking, compared to slow-wave sleep and REM sleep. The facilitation may be due, in part, to an increase of neuronal input resistance during wake. Tha- lamically-evoked EPSPs increased their amplitudes and decreased their latencies during waking and REM sleep, compared to slow-wave sleep. Altogether our data indicate that one of the major difference between slow-wave sleep and wake is a great reduction of Cs+-sensitive potassium currents in waking, which is associated with an increased synaptic responsiveness. This reduction might occur via depression of leak current by neuromodulators.

Supported by Canadian Institute of Health Research, Fonds de la Recherche en Santé du Québec and Human Frontier Science Program. Focus Group 9

Pros and cons of the platform technique for REM sleep deprivation.

Chairperson: Anton Coenen (The Netherland) Anton Coenen, Psychologisch Laboratorium, Nijmegen, The Netherland Joëlle Adrien INSERM, Paris, France Dag Stenberg, University of Helsinki, Finland Vladimir Kovalzon Sergio Tufik, Universidad de Sao Paulo, SP, Brazil Tarja Porkka Heiskanen, University of Helsinki, Finland Deborah Hipolide, Universidad de Sao Paulo, SP, Brazil Kim Horne Birendra Mallick Mahesh Thakkar, Brockton VAMC, MA, USA Carlyle Smith, Trent University, Ontario, Canada Gregory Rose 112 Actas de Fisiología 7, 2001

THE PLATFORM TECHNIQUE FOR SELECTIVE REM SLEEP DEPRIVATION

Joëlle Adrien INSERM U288, CHU Pitié-Salpêtrière, 91 Bd de l’Hôpital, 75013 Paris - France

The platform technique for selective REM sleep deprivation has been used for almost four decades, firstly in the cat (1) and secondly in other species such as the rat (2, 3) and the mouse (4, 5). In most instances, the water surrounding the platform was deep enough (11 cm) to induce, besides REM sleep deprivation, severe restriction of locomotor activity and intense stress. In our hands, like in some others (4), the water was only 1 (for mice) or 2 (for rats) cm deep, and the animals could walk around in the deprivation setup. Furthermore, they were firstly trained to stand on a large platform, while polygraphic monitoring allowed to verify that sleep-wakefulness amounts were comparable to those under baseline conditions, i.e., in a standard recording cage. Secondly, animals were habituated to the small platform for 1 to 3 hours daily during several days prior to the deprivation experiment. Finally, at the beginning of the deprivation period, a platform of not minimal size was used and sleep-wakefulness states were observed continuously. If REM sleep attempts occurred, a smaller platform was substituted to the former one. Accord- ing to this technique, each animal was provided with a platform adapted to its capacities of developing SWS in the REM sleep deprivation set-up. In addition, several times a day, animals were transfered to a normal recording cage where they stayed until REM sleep attempts were observed polygraphically. At that time, they were returned to the deprivation cage. Recording the animals during the whole periods of both the control (large platform) and deprivation (small platform) conditions allowed not only to verify that animals were actually deprived of REM sleep, but also to make sure that the amounts of SWS were comparable to those under baseline conditions. This latter verification, together with assays of plasmatic corticosterone, is relevant to estimate the level of stress induced by the procedure. To conclude, we think that the size of the platform is not a standard which can be reliable across research groups or even across strains within the same species. Thus, the platform technique should always be validated by simultaneous monitoring of the sleep-wakefulness cycles, in order to verify that the conditions in a given laboratory correspond actually to selective REM sleep deprivation with minimal impairment of SWS. Furthermore, during the control and the deprivation procedures, animals should be returned to a standard recording cage at regular intervals, in order to limit the stress induced by the platform set-up.

1- Vimont-Vicary P, Jouvet-Mounier D, & Delorme F. EEG Clin. Neurophysiol. 1966, 20: 439- 449. 2- Morden B, Mitchell G, & Dement W. Brain Res. 1967, 5: 339-349. 3- Adrien J, & Dugovic C. Eur. J. Pharmacol. 1984, 100: 224-226. 4- Kitahama K, & Valatx JL. Neuropharmacology 1980, 19: 529-535. 5- Boutrel B, Franc B, Hen R, Hamon M, & Adrien J. J. Neurosci. 1999, 19: 3204-3212. Actas de Fisiología 7, 2001 113

PROS AND CONS OF THE PLATFORM TECHNIQUE FOR REM SLEEP DEPRIVATION IN RATS

Anton M. L. Coenen Nijmegen,P.O. Box 9104 6500 HE, The Netherlands

The platform (watertank or pedestal) technique to deprive animals of REM sleep was originally designed by D. Jouvet and colleagues in 1964. Animals, commonly rats, are placed on a small platform surrounded by water. Usually an inverted flower pot is placed in a tank filled with water. The rat can sleep but REM sleep is prevented to occur by the restricted size of the platform, providing not enough room for complete muscular relaxation of the rat. Due to the paralysis at the onset of REM sleep, the rats make physically contact with the water and are aroused. The technique is attractive for its simplicity and cheapness. Debates about the reliability and validity of the platform technique are as old as the technique itself. It is clear that the technique deprives an animal from REM sleep, given the REM rebounds after treatment. A problem, however, is that the multitude on effects induced by the technique are difficult to interpret. There is no theory in which the various behavioural/cognitive/neurochemical effects and consequences can be combined and united. One can argue that sooner or later such an all-embracing theory will be developed, but an even more serious problem is whether this mulititude on effects can be ascribed to deprivation of REM sleep perse. Every technique has intented (specific) and non-intended (non-specific) effects and two factors determine the reliability of the technique. The ratio between the specific and non-specific effects, and the possibilities to control for the non-specific effects. There are arguments to defend the position that the platform technique includes a number of non-specific effects for which it is hard to control for. Side-effects of the technique, such as stress and temperature, may play a major role in the induction of effects, while the control for these effects is not completely adequate. Moreover, when alternative techniques for REM sleep deprivation are applied, the effects obtained with the platform technique can mostly not be replicated. It is necessary to weight the merits of this technique against the criticisms. This gives perhaps an answer to the question whether application of this technique is justi- fied and under which circumstances and, furthermore, which scientific questions can be answered with this technique. The important question is whether the scientific output of this technique justifies the exposure of numerous animals to the platform treatment. 114 Actas de Fisiología 7, 2001

PARADOXICAL SLEEP DEPRIVATION REDUCES PLASMA HOMOCYSTEINE LEVELS IN RATS

Debora C Hipolide, Allan C Oliveira, Vania D’Almeida, Jose N Nobrega and Sergio Tufik

Homocysteine is an intermediate amino acid in the methionine-cysteine metabolic pathway, as it represents a branching point in which it can be remethylated to methionine or converted to cysteine. The importance of this metabolic pathway relates to its great number of physiological functions, such as providing the methyl donor (S-Adenosyl-methionine) for virtually all methylation reactions in the organism1 and the limiting substrate (cysteine) to the synthesis of glutathione2. Hyperhomocysteinemia has been associated with a growing number of pathological and stressful conditions. As sleep deprivation (SD) is associated with disruption of many physiological processes we hypothesized that it would also be associated with increases in total plasma homocysteine (tHcy) levels. Further, since we had previously found evidence of oxidative stress in brain following SD, we also searched for evidence of systemic oxidative stress by measuring total glutathione (tGSH) and thiobar- bituric-acid reactive substances (TBARS) levels in plasma. Thirty-eight male Wistar rats were sleep deprived for 96 hours using the classic platform technique. Ten animals were killed just after this period and another 18 sleep-deprived animals were allowed to undergo sleep rebound (SR) for 24 or 48 hours. Ten home cage animals composed the control group. All animals were killed by decapitation and blood was collected for biochemical analysis. tHcy levels were determined by HPLC, tGSH and TBARS levels were determined by colorimetric detection. Contrary to expectation, tHcy levels were reduced by SD (4.38  0.67 M) as compared to the control group (6.86  1.03 M) and did not recover to normal levels after 24 (4.96  0.84 M) or 48 (4.87  0.57 M) hours of SR. We observed a tendency towards decreased tGSH levels and increased TBARS levels in SD rats as compared to controls. The decrease in tHcy can be theoretically explained by the catabolic state related to sleep deprivation or by consumption in response to a higher demand for cysteine in order to maintain GSH status. As we observed a significantly higher TBARS/tGSH rate in SD rats, we believe sleep deprivation is associated to systemic oxidative stress, representing a situation of higher requirement for GSH and, consequently, for cysteine. In our case, the preferential use of homocysteine for cysteine synthesis may be biochemically explained by inactivation of both methionine synthase and S-adenosylmethionine synthetase and activation of cystathionine synthase by oxidation. The absence of significant differences in tGSH levels in the four groups studied may represent this metabolic effort to maintain tGSH levels and could also explain the occurrence of just a tendency to increase on TBARS levels on SD rats. If this is the case, it is still necessary to establish why tHcy levels remain decreased yet after 48 hours of sleep rebound and which different mechanisms causes increases on homocysteine levels in other stressful conditions.

Supported in part by funds from CNPq /PIBIC, FAPESP and AFIP (Brazil)

1. Chiang PK, Gordon RK, Tal J, et al. S-Adenosylmethionine and methylation. FASEB J 1996; 10: 471-80. 2. Anderson ME. Glutathione: an overview of biosynthesis and modulation. Chemico-Biological Interactions 1998;111-112: 1-14. P O S T E R S Actas de Fisiología 7, 2001 117

LEARNING-RELATED NEURAL CHANGES ARE ACCESSIBLE DURING REM SLEEP

Mercedes Atienza, Jose L. Cantero, Elena Dominguez-Marin and Rosa M. Salas Laboratory of Neurophysiology, Department of Psychiatry, Harvard Medical School, Massachusetts Mental Health Center, 74 Fenwood Road, Boston, United States

Perceptual learning is thought to be the result of neural changes that take place over a period of several hours or days, allowing information to be stored in long-term memory. Evidence suggests that contents of long-term memory may improve attentive and pre-attentive sensory processing. Therefore, it is plausible to hypothesize that learning- induced neural changes that develop during wakefulness could improve automatic information processing during human REM sleep. The MMN, an objective measure of the automatic change detection in auditory cortex, was used to evaluate long-term learning effects on pre-attentive processing during wakefulness and REM sleep. When subjects learned to discriminate two complex auditory patterns in wakefulness, an increase in the MMN amplitude was obtained in wake and REM states. The automatic detection of the infrequent auditory pattern was improved in both brain states presumably by reactivating information from long-term memory. These findings suggest that long- term learning-related neural changes are accessible during REM sleep as well. 118 Actas de Fisiología 7, 2001

NAPPING: A NATURAL WAY TO SPEED UP CONSOLIDATION OF VISUAL PERCEPTUAL LEARNING

Mercedes Atienza, Sarah Mednick, Jose L. Cantero, Ken Nakayama and Robert Stickgold Laboratory of Neurophysiology, Department of Psychiatry, Harvard Medical School, Massachusetts Mental Health Center, 74 Fenwood Road, Boston, United States

Within-day over-exposure to a perceptual learning task, the texture discrimination task (TDT), leads to impaired performance. Previous research has shown that learning- related neural changes that underlie the consolidation process of perceptual learning partially occur during nocturnal sleep. We hypothesized that detrimental effects on performance from within day, consecutive testing may be reversed by a daytime nap. All subjects were tested on the TDT in four sessions separated by 2-hour intervals. The experimental group took a nap of either 60 min. or 30 min. between the 2nd and 3rd session. We found that 60 min. naps reversed the detrimental effects on performance observed in the control group, whereas the benefit of half hour naps was more ephem- eral. Further, in the hour nap condition, improvement was correlated with the amount of slow wave sleep (SWS; r=.68, p=.05). These results, however, do not distinguish between whether the nap induced stimulus-specific neural changes or if experimental subjects were just generally more rested. Improvement in the TDT has been shown to be retinotopically specific, in which case the learning-related neural changes and the performance decrement should also be retinotopic. We also tested this hypothesis. Retinal specificity was evaluated by switch- ing the target stimulus to the opposite lower quadrant for the last of four within-day testing sessions. The no-switch group evidenced the aforementioned decline in performance across sessions, whereas the switch group’s performance returned to baseline during the 4th session. Therefore, the performance decline caused by over-exposure to the TDT seems to be specific to neural changes in early visual cortex and not due to generalized tiredness. These results suggest that benefits of daytime naps may stem from a speeding-up of learning-related, local neural changes in early sensory cortex during the SWS periods. Actas de Fisiología 7, 2001 119

DAYTIME SLEEPINESS, MOOD, AND PERFORMANCE FOLLOWING EX- PERIMENTALLY-INDUCED SLEEP FRAGMENTATION

Kimberly A. Cote, Catherine E. Milner, Karen D. Baxter, Laura B. Ray, and Stephanie L. Osip Psychology Department, Brock University, Canada

Introduction. Experimental Sleep Fragmentation (SF) involves inducing arousals by administering intrusive auditory stimuli periodically throughout the night. It is intended to model the type of frequent and periodic disruption experienced in sleep disorders such as apnea and periodic limb movement. Previous research has shown that SF leads to daytime sleepiness, but evidence of performance impairment has been inconsistent (1). The purpose of this study was to investigate the cumulative effects of SF on daytime functioning. Method. Participants were six healthy adults (3 female, mean age=35.2) with no sleep complaints. Following a screening night, participants spent four consecutive 24-hour periods in the laboratory. The order of conditions was fixed as follows: baseline, SF1, SF2, recovery. On nights two and three, sleep was fragmented using methods similar to Philp et al (1). Stimuli were delivered via an earphone in the right ear for 10s beginning at 65 dB(A). If no EEG arousal resulted, the intensity was increased by 10 dB for each successive presentation to a maximum of 115 dB(A). During the day, participants performed a 35-minute computerized test battery at 2-hr intervals (09:00-19:00). The battery was fixed and included: Profile of Mood States (POMS), Stanford Sleepiness Scale (SSS), Alpha Attenuation Task, reaction time (10 min), an auditory odd-ball (ERP paradigm), serial addition/subtraction, and subjective percep- tion of performance. The battery was practiced prior to the baseline night. Results. On pre- and post-sleep questionnaires, measures of sleepiness and fatigue (7-point scales), PANAS positive mood (2), and a visual analogue mood scale (energetic to sluggish) all revealed significant CONDITION effects (p<.01). For all of the variables, paired t- tests indicated that baseline differed from SF1, SF2, and recovery. SF1 did not differ from SF2. Polynomial analyses indicated a significant quadratic trend for all of these variables (p<.01). Analyses were also carried out on the average of daytime performance sessions for SSS, RT and math. Speed of performance on math or RT did not differ across experimental conditions. There was a significant CONDITION effect (p=.02) and a linear polynomial trend (p=.03) for the number of correct responses on the serial addition/subtraction task. Baseline, SF1, and SF2 were all significantly lower than recovery values. There was no difference between the 1st and 2nd SF conditions. Discussion. Following the 2nd night of SF, there was no change in subjective mood or sleepiness. Thus, there was neither habituation nor an additive effect on successive nights of SF, although it does appear that learning on the math task remains asymptotic during the sleep debt. Subjective measures of sleepiness and mood are somewhat at odds with performance measures on neurobehavioural tasks. We expect that subsequent analyses of electrophysiological measures of arousal and attention (EEG and ERPs) will support the subjective feelings of reduced well-being following this degree of partial sleep deprivation. References. (1) Philip P, et al. Sleep 1994;17:242-247. (2) Wat- son D, et al. J Pers Soc Psyc 1988;54:1063-70. 120 Actas de Fisiología 7, 2001

AGE-RELATED CHANGES IN THE SLEEP QUALITY OF PRESCHOOL AND EARLY SCHOOL AGED CHILDREN

Monique K. LeBourgeois and John R. Harsh Sleep Research Laboratory, University of Southern Mississippi, USA

Introduction: Despite growing awareness of the need for normative developmental data on children’s sleep (Sadeh et al., 2000), such data are still lacking for preschool and early school-age children. The present study describes changes in the sleep of 2- to 8-year-old children along several developmentally meaningful dimensions using a new research instrument, the Children’s Sleep-Wake Scale (CSWS). Methods: Caretakers recruited from the local community were administered the CSWS via telephone. The CSWS is parent report instrument comprised of several subscales, including Going to Bed (7 items, alpha = 0.85), Falling Asleep (6 items, alpha = 0.69), Maintaining Sleep (6 items, alpha = 0.69), Reinitiating Sleep (7 items, alpha = 0.71), and Returning to Wakefulness (5 items, alpha = 0.85). Higher scores reflect better sleep on the respective dimension. Data were collected on 485 children (236 male, 249 female), ages 2 to 8 years (mean = 4.9, SD = 2.0). Response rate was 88%. Results: Mean scores on the CSWS subscales for each age group are shown in the following graphs.

18 Going to bed 17 Falling Asleep 16 Maintaining Sleep

17 F = 3.48, p < 0.01 16 F = 9.27, p < 0.001 15 F = 3.53, p < .01

16 15 14 Mean Scores Mean Scores Mean Mean Scores 15 14 13

14 13 12 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Age Age Age Reinitiating Sleep Falling Asleep

20 14 F = 6.58, p < 0.001 F = 9.34, p < 0.001 13 19

12 18 11 Mean Socres Mean Scores Mean 17 10

16 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Age Age

Discussion: The results of this study suggest a general trend in the sleep-related changes across childhood. Children’s success in going to bed and sleeping through the night increases with age, while their ability to awaken in the morning decreases as they get older.

Sadeh, A., Raviv, A., & Gruber, R. Sleep patterns and disruptions in school-age children. Developmental Psychology,36(3), 291-301. Actas de Fisiología 7, 2001 121

LIGHT EXPOSURE OF CHICK EMBRYO AS A DETERMINANT OF MONOCULAR SLEEP

Gian G. Mascetti a, Daniela Bobboa, Federico Galvania and Giorgio Vallortigarab aDepartment of General Psychology, University of Padua, Via Venezia 8, 35131 Padova, Italy; bDepartment of Psychology and B.R.A.I.N. Centre of Neuroscience, University of Trieste, Trieste, Italy

Birds, like some aquatic mammals, exhibit a unique behavioral and electrophysiological state called “unihemispheric sleep”, in which one cerebral hemisphere is awake and the other is sleeping. Slow-wave sleep in one hemisphere is associated with closure of the controlateral eye, whilst the eye controlateral to the awake hemisphere is open. Closure of both eyes, in contrast, is associated with bihemispheric slow-wave sleep or with REM sleep. During the last three days of incubation chick’s embryo is turned in the egg so that, the right eye lies adjacent to the egg shell and is stimulated by light penetrating the shell and membranes. We measured the number and the duration of episodes of binocular and monocular sleep in chicks hatched from eggs maintained in the dark before hatching and in chicks exposed to light during the last three days before hatching. In the first two day of life light-incubated chicks slept prevalently with their left eye open, whereas dark-incubated chicks prevalently with their right eye open. The difference tended progressively to disappear until, on day five, both light- and dark- incubated chicks preferentially slept with their left eye open. These result suggest the role of enviromental factors in modulatìng complex phenomena as unihemispheric sleep. 122 Actas de Fisiología 7, 2001

EFFECTS OF THE HYPNOTIC DIAZEPAM ON MEMORY IN THE VISUAL AND AUDITORY MODALITY

Annika Smit and Anton Coenen Nijmegen, The Netherlands

Central topic of research is to clarify the relation between level of vigilance and learning and memory processes. In the present study, the classic hypnotic diazepam (vali- um), which is known to have amnesic effects, was used to decrease vigilance. Memory was tested after diazepam intake (10 mg orally). Two main effects of diazepam on memory were investigated: anterograde amnesia (impaired recall of what is learned under drug influence) and retrograde facilitation (enhancement of recall of what is learned just before drug intake). It was questioned whether both memory effects are present in the visual and the auditory modality and in both complex and simple tasks. The experiment was performed in a double-blind, placebo-controlled way. All 20 subjects were exposed to memory tests: an auditory 15 words test (simple auditory task), a short story recall test (complex auditory task), a visual 15 words test (simple visual task) and the Complex Figure of Rey (complex visual task). Anterograde amnesia was found for both modalities, for simple and complex tasks. Retrograde facilitation was found in both modalities, but only for simple tasks, although the effects were in the same direction for the complex tasks. In all, the data demonstrate that anterograde amnesia is not modality-specific and present in both simple and complex tasks. Retro- grade facilitation is not modality-specific either, being present in both the visual and auditory modality for simple tasks. It was demonstrated that the decrease of vigilance level after diazepam leads to 1) An impaired recall of new material, and 2) An improved recall of already learned material Perhaps less resources are available for the consolidation of new material in a diazepam-induced lowered state of vigilance. On the other hand, the process of ongoing consolidation of already learned material faces less interference, as there is less new material consolidated during this decreased level of vigilance. This lack of ‘competi- tion’ in consolidation between new and already learned material, is expressed in a better recall of this already learned material: retrograde facilitation. Actas de Fisiología 7, 2001 123

INCREASED SLEEP SPINDLE ACTIVITY FOLLOWING SIMPLE MOTOR PROCEDURAL LEARNING IN HUMANS

Stuart Fogel, Jeremy Jacob and Carlyle Smith Dept. of Psychology, Trent University, Peterborough, Ont., Canada

A number of recent studies have suggested that Stage 2 sleep is involved in the consolidation of simple motor procedural memory (1, 2). Other studies have found that the number of spindles are positively correlated with procedural learning (3), and intelligence (IQ) (4). Method: Eight female undergraduate students spent three consecutive nights in the sleep lab (acclimatization, baseline and test). The experimental group was exposed to two hours of simple motor procedural tasks, while the controls were not. The experimental group was retested on the tasks one week later. Results: It was found that the number of sleep spindles increased 42% after learning (F = 8.22, p < .03). When spilt according to high and low Performance IQ a planned orthogonal comparison revealed only the high Performance IQ group had a significant increase in the number of spindles (t = 5.58, p < .01). These results were supported by Sigma analysis of Stage 2 sleep in high Performance IQ participants. It was also found that the baseline spindle density and Sigma were related to Perfor- mance IQ but not Verbal or Full Scale IQ. Discussion: The results demonstrate that after exposure to simple motor procedural learning, the number of sleep spindles increases. The results suggest that the function of spindles is to consolidate simple motor procedural memories. The relationship between spindles and IQ indicate that not only are spindles a product of new simple motor procedural learning, but they are also related to aptitude for learning and can serve as a biological marker for both. These results implicate thalamocortical loops in the consolidation of simple motor procedural memory from which spindles originate.

References:

1) Smith, C. & MacNeill, C. (1994). J. of Sleep Res., 3, 206-213. 2) Tweed, S., et al. (1999). Sleep, 22, S241. 3) Briere, M., et al. (2000). Sleep, 23, A219 4) Nader, R. & Smith, C. (2001). Sleep, 24, (suppl.), A160.

Research Supported by Canadian Institutes of Health Research (CIHR) 124 Actas de Fisiología 7, 2001

INTRA-AMYGDALA SCOPOLAMINE INFUSIONS DURING A PARADOXICAL SLEEP WINDOW IMPAIRS CONDITIONED CUE PREFERENCE ACQUISITION

Laura Kenton and Carlyle Smith Dept. of Psychology, Trent University, Peterborough, Ontario, Canada

Introduction: There is now substantial evidence that paradoxical sleep (PS) is involved with memory processing in animals (1,2) and humans (3). In rats, paradoxical sleep windows (PSWs) have been identified for several tasks. PS deprivation during these PSWs results in memory impairments (1,2). Recently, a PSW at 9 -12 hours after acquisition has been identified for the conditioned cue preference (CCP) task (4). In addition, it has been demonstrated that an intact lateral amygdala is required for CCP acquisition (5). In the present study, the acetylcholine antagonist, scopolamine, was used to suppress cholinergic activity in the lateral amygdalae. This was done to examine the involvement of the lateral amygdalae during a PSW for the CCP task. Method: Cannulae were implanted bilaterally into the lateral amygdalae of Sprague Dawley rats (n=15). The rats were then trained on the CCP task for 8 consecutive days. Experimental rats were given bilateral intra-amygdala infusions of scopolamine either 9 hours (to be active during the PSW, n=4) or 12 hours (to be active after the PSW, n=4) after each training session. Control rats were given bilateral intra-amygdala infusions of saline either 9 hours (n=4) or 12 hours (n=3) after each training session. One day after the last training session the rats were tested on the CCP task. The number of light arm entries (correct responses) and dark arm entries (incorrect responses) were recorded. Time spent in the light arm, dark arm, and centre platform were also recorded. Results: An ANOVA revealed that the scopolamine-infused animals made significantly fewer light arm entries than the saline-infused animals [F(1,13) = 6.95, p<0.02]. An additional ANOVA revealed that the scopolamine-infused animals spent significantly less time in the light arm than the saline-infused animals [F(1,13) = 13.10, p<0.003]. Discussion: The scopolamine-infused animals demonstrated significant impairments in CCP acquisition as compared to the control animals. Thus, suppressing cholinergic activity in the lateral amygdalae many hours after training appears to disrupt CCP acquisition. This suggests that the lateral amygdalae may be actively involved in processing memory for the CCP task several hours after training, presumably during a PSW.

References: 1)Smith, C. (1995). Behav. Brain Res., 69, 137-145. 2) Smith, C. (1996). Behav. Brain Res. 78, 49-56. 3) Smith, C. (in press). Sleep Med. Rev. 4) Vallance, K. et al. (1999). Sleep, 22, (suppl.), S243. 5) McDonald, R.J. & White, N. (1993). Behav. Neurosci.,107, 3-22.

Research Supported by Natural Science and Engineering Research Council Actas de Fisiología 7, 2001 125

INTELLIGENCE AND STAGE 2 SLEEP SPINDLES

Rebecca Nader and Carlyle Smith Department of Psychology, Trent University, Peterborough, Ontario, Canada

Introduction: A number of studies have been done to examine the relationship between Stage 2 sleep and learning. One such study has reported that the number of sleep spindles was related to learning efficiency1. Expanding on this idea, we predicted that the number of spindles and the mean amount of sigma power during Stage 2 would be related to the subjects’ intelligence scores on the MAB-II IQ test. Methods: Ten subjects (range: 18 - 29) were used in the study. Subjects were screened for abnormal sleep patterns and excessive drinking/drug use. Subjects’ completed the MAB-II2 IQ test and spent the subsequent two nights (acclimatization and baseline) in the laboratory. The number of sleep spindles (12-16Hz) and the mean Sigma Power (12-14Hz) were assessed for each page of Stage 2 sleep on the baseline night. Epochs with large body movements or major artifacts were excluded. Spindle activity and sigma power were assessed for both the C3 and C4 derivations. Results: The total number of spindles (C3+C4) for the night was highly correlated with Performance IQ (r = .71, p = .022), and with Full Scale IQ (r = .76, p = .010, but not with Verbal IQ (r = .56, p = .094). The mean sigma power of both C3 and C4 for the entire night was highly correlated with Performance IQ (r = .76, p = .011) and Full Scale IQ (r = .77, p = .009). When the night was divided into thirds, the mean sigma power in the last third of the night showed the strongest correlation with both Perfor- mance IQ and Full Scale IQ (r = .87, p = .001, and r = .84, p = .002 respectively). None of the correlations with Verbal IQ were significant. Sigma power was found to be the most highly correlated with two of the sub-tests of the Performance IQ scale: Picture Completion and Object Assembly. Both tasks require perceptual and analytical skills for successful completion. Conclusions: The data suggest that the number of spindles and sigma power from central derivations are both powerful predictors of how well an individual will perform on the Performance portion of an IQ test. This is especially true of perceptual and analytical skills. The reasons for these relationships are not yet clear.

References:

1) Briere, M. Forest, G., Lussier, I. & Godbout, R. (2000). Implicit verbal recall correlates positively with EEG sleep spindle activity. Sleep 23, A219. 2) Jackson, D.N. (1998). Multidimensional Aptitude Battery-II. Sigma Assessment Systems, Inc. Port Huron, MI.

Research Supported by Canadian Institutes of Health Research (CIHR). 126 Actas de Fisiología 7, 2001

SLEEP EDUCATION AND CHANGES IN SLEEP AND SLEEPINESS IN CO- LLEGE STUDENTS

Ling-Ling Tsai and Sheng-Ping Li Department of Psychology, National Chung-Cheng University, Chia-Yi 621, Taiwan, R.O.C.

A two-credit (100 min/week) “Sleep Management” course was offered by the first author in the spring semester in 1998 and 1999 at the National Chung-Cheng Universi- ty, Taiwan. The aim of this course was not just to passively increase sleep knowledge but also to actively enhance sleep awareness and, thereby, practical impacts on the students’ daily sleep and sleepiness might be found. The curricular activity of this course included lectures on basic sleep psychology, physiology, pharmacology, disorders and hygiene, dream, and circadian rhythm, group discussions on sleep old sayings and misunderstandings, and self evaluations on sleep. A web page was designed to provide on-line reviews of the course materials and an after-school discussion board. The students enrolled in this course were required to record 7-day sleep logs in March, April, and May as course assignment. The Epworth Sleepiness Scale (ESS), Morning- ness-Eveningness Questionnaire (MEQ), Pittsburgh Sleep Quality Index (PSQI), and Ullanlinna Narcolepsy Scale (UNS) were administered once in 1998 and three times in 1999. A total of 362 students were enrolled in the 1998 class and of 84 in the 1999 class. The control college student subjects were recruited from the friends of the course-enrolled students in 1998 and from respondents to ads placed on seven popular, university-based Bulletin Board System (BBS) in1999. Results from statistical analyses showed several significant changes in sleep recorded in the sleep log across the three months of the semester. Both the course-enrolled and control students showed progressive decreases in the number of awakenings in night sleep and increases in sleep efficiency on weekdays across the semester. However, only the course-enrolled students showed significant progressively decreased sleep latency on weekdays, increased sleep efficiency and subjective sleep quality on both weekdays and weekend, and decreased ESS scores over the three months. These results suggested that sleep and sleepiness were improved along the progress of this course. Further studies are needed to verify the long-term impact of this kind of sleep course on sleep health. Actas de Fisiología 7, 2001 127

SINGLE AND DUAL TASK PERFORMANCE UNDER CONSTANT ROUTINE CONDITIONS

Alexander P.J. van Eekelen & Gerard A. Kerkhof Department of Psychology, University of Amsterdam, The Netherlands

The purpose of this study was to investigate the endogenous circadian rhythmicity of cognitive performance on different levels of task complexity. In this research, a single task (mental arithmetic) and a dual task (unstable tracking combined with memory search) were utilized within a constant routine protocol. After a night of sleep registra- tion, subjects participated in a 27 hour continuing experiment starting at 9 am (N=17, 8m, aa=22.4 ), or at 9 pm (N=14, 7m, aa=23.6), in order to balance for prior wakefulness. After two adaptation hours, tasks were presented every three hours. All subjects completed practice sessions on the day prior to the experiment. Regarding the dual task, the instruction was to handle a joystick with the dominant hand and to try to divide attention equally over the two subtasks. The mental arithmetic task contained multipli- cations ranked as ‘easy’ (single x single digit) and ‘difficult’ (single x double digit). Both tasks were aproximately 3 minutes in duration. Time of day effects were analysed using repeated measures ANOVAs for measures of speed and accuracy. Results exposed congruent circadian rhythmicity on tracking accuracy , and for response correctness on both tasks, for both groups, with a decrease during night hours and a trough at 8 am. Response latency displayed a similar pattern for the memory search task, but was absent for the mental arithmetic task. Also, response lapses showed this rhythmicity for both tasks. The mental arithmetic task revealed that computation difficulty resulted only in an overall decrease of accuracy as well as speed. Additional trade off analyses on the dual task provided insight in the strategy subjects used to perform concurrently on the two subtasks. Consistent time on task effects were found for each session. Time of day analysis showed no effect, suggesting that the division of attention was not affected by the circadian rhythm. Furthermore, TSD parameters were calculated for the memory task, which showed no compensatory strategy was used during the circadian trough. In conclusion, measures of accuracy appear to be highly dependent on the circadian rhythm, as does response latency, when task complexity is advanced. Strategies used in dual task performance were not influenced by the circadian rhythm, whereas sleep deprivation (<40 hours) has little effect on overall performance. 128 Actas de Fisiología 7, 2001

ADENOSINE ACTIVATES IP3 RECEPTOR-MEDIATED CALCIUM RELEASE AND NUCLEAR TRANSLOCATION OF NF-KB IN BASAL FOREBRAIN CHOLINERGIC NEURONS

Radhika Basheer, Elda Arrigoni, Vijay Ramesh, Hemant Thatte, Robert W Greene and Robert McCarley Harvard Medical School/ VA Medical Center, West Roxbury MA, USA

Previously, we reported that DNA-binding activity of NF-kB in basal forebrain (BF) was increased by sleep deprivation-induced elevation in extracellular adenosine, an effect mediated by the A1 adenosine receptor. However, the signal transduction pathway for NF-kB was not clear, since A1 receptor occupation can produce inhibition of adenylate cyclase or, alternatively, activation of phospholipase C, which, in turn, could activate IP3 receptor-mediated intracellular calcium release. To determine if adenosine mobilized intracellular calcium, acute brain slice preperations in carboxygenated ACSF were loaded with calcium orange. Increases in intracellular fluorescence in individual neurons in response to adenosine treatment were measured using a BioRad multi-photon system in real time. Adenosine significantly increased cytosolic calcium that was independent of calcium in the external medium. Depletion of internal stores of calcium by thapsigargin abolished the adenosine response, further confirming the release of calcium from internal stores. Blocking the IP3 receptor with xestospongin C or 2ABP prevented adenosine-mediated calcium increase whereas the ryanodine receptor blocker DHBP had no effect. Immunohistochemistry showed that all the calcium-responding neurons were cholinergic. Moreover, double labeling with NF-kB and ChAT antibody showed that adenosine treatment induced nuclear translocation of NF-kB only in cholinergic neurons. Together, our results suggest that, adenosine mobilizes IP3-regulated intracellular calcium and consequent nuclear translocation of NF-kB in BF cholinergic neurons. Adenosine-mediated selective activation of the NF-kB may play a role in producing the long term effects of sleep deprivation.

Supported by: SMERF award (RB),NIMH39683 (RWM). Actas de Fisiología 7, 2001 129

DETERMINATION OF HUMAN TUMOR NECROSIS FACTOR ALPHA BY A SENSITIVE ENZYME IMMUNOASSAY

Mehdi Hedayati, Fereidoun Azizi and Razieh Yazdanparast Endocrine Research Center, Tehran, Iran

Tumor necrosis factor alpha is an extremely potent peptide cytokine. TNF-a and its receptors are found in normal brain including areas known to be involved in sleep regulation. We described a simple and sensitive enzyme linked immunosorbent assay for human TNF-a in serum/plasma or cell culture media based on horse radish peroxidase in biotin streptoavidine amplification system with a detection limit of 0.1pg/ml and up to 10 pg/ml.The within run coefficient of variation ranged from 3.7 to 5.9 and between run from 8.0 to 9.9. 130 Actas de Fisiología 7, 2001

MICROINJECTION OF CYCLOHEXYLADENOSINE IN BASAL FOREBRAIN INDUCES NUCLEAR TRANSLOCATION OF NFkB IN CHOLINERGIC NEURONS

Vijay Ramesh, Radhika Basheer, Hemant Thatte and Robert McCarley Harvard Medical School/ VA Medical Center, Brockton, MA, USA

Our lab has found a sleep-deprivation (SD) induced monotonic increase in extracellular adenosine (AD) with each hour of SD; this monotonic increase was selective to the basal forebrain (BF). Moreover, we have shown that either 3h SD or in vitro AD application to BF slices resulted in increased DNA-binding activity of the transcription factor NFkB, mediated via the A1 adenosine receptor. We also have recently reported that 3h SD/ microinjection of AD in vivo induces nuclear translocation of NFkB. To evaluate whether the AD-induced translocation of NFkB is mediated through the A1 receptor, and to determine whether translocation takes place in cholinergic and/or non- cholinergic BF neurons, we unilaterally microinjected the A1 receptor agonist cyclohexyladenosine (CHA; 0.5 ml of 100nM ), targeting the cholinergic zone of the BF (AP -0.30, ML -2.2, DV -8, (n=4)). ACSF-injected rats served as controls. One hour after injection, the rats were perfused and brains collected. 40 mm thick sections were double labeled using antibodies against NFkB and ChAT. Results: We observed that nuclear translocation of NFkB occurred only in ChAT +ve neurons in the CHA-injected side, without any effect on the side contralateral to the injection. Nuclear translocation of NFkB did not occur in ACSF-injected control animals. Our results show that CHA induced nuclear translocation of NFkB occurs only in cholinergic neurons of BF. It is possible that the transcription factor NFkB may promote transcription of proteins important in the long-term effects of sleep deprivation.

Support: NIMH 39683 (RWM), SMERF award (RB). Actas de Fisiología 7, 2001 131

MOLECULAR CORRELATES OF SLEEP AND WAKING IN DROSOPHILA

Chiara Cirelli and Giulio Tononi Department of Psychiatry, University of Wisconsin/Madison, USA

Recently, we (1) and others (2) have shown that Drosophila rest shares many of the critical features of mammalian sleep. Fruit flies exhibit consolidated periods of activity during the light period and consolidated periods of rest during the dark period. The arousal threshold during periods of quiescence is higher than during periods of activity. Drosophila rest is homeostatically regulated, i.e. fruit flies rest more after being rest- deprived. As in mammals, caffeine increases waking in fruit flies, while antihistamines increase rest (1). In both rats (3) and Djungarian hamsters (4) we have shown that sleep and waking can also be distinguished based on molecular criteria. Several genes are specifically induced in the cerebral cortex of rats and hamsters during spontaneous or forced waking but not during sleep, including immediate early genes and transcription factors, mitochondrial genes, growth factors/adhesion molecules, chaperones, neurotransmitter receptors and enzymes. In a previous study (1) we found that a few of the genes upregulated in the cerebral cortex of spontaneously awake or sleep deprived rats and hamsters are also induced by waking in Drosophila. These genes code for the mitochondrial enzyme cytochrome oxidase c subunit I, the chaperone BiP and the enzyme arylalkyamine N-acetyltransferase, functionally homologue to the rat enzyme arylsulfotransferase. The aim of this study was to determine whether other genes known to be upregulated in rats and hamsters by waking and sleep deprivation are induced in rest-deprived flies relative to resting flies. Wild type (Canton-S; 3-5-day old virgin females) flies were maintained on a 12:12 LD schedule and activity patterns were monitored using infrared technology (Trikine- tics). Flies were sacrificed during the dark period, after 11 h of spontaneous rest or 11 h of automated rest-deprivation. Real-time quantitative PCR was performed using di- gested pooled total RNA from the heads of resting and rest-deprived flies (n=50/group). Rest deprivation was found to significantly increase mRNA levels of several Droso- phila homologues to rat “waking” genes, including Stripe A (homologue to the immediate early gene NGFI-A), neuroglican precurson (homologue to the F3 adhesion molecule), and Drosophila glutamate transporter (homologue to the rat glutamate transporter). These findings confirm that significant changes in gene expression occur across behavioral states. Moreover, they support the idea that molecular correlates of sleep and waking are shared across species as different as the fruit fly, the Djungarian hams- ter and the rat. Therefore, molecular correlates can be useful to define sleep in species in which classical criteria, such as electroencephalographic criteria, are difficult to apply.

(Supported by Neurosciences Research Foundation).

(1) Shaw et al., Science 287:1834,2000 (2) Hendricks et al., Neuron 25:129,2000 (3) Cirelli and Tononi, Brain Res 885:303,2000 (4) Cirelli et al., Sleep 24(S):131, 2001 132 Actas de Fisiología 7, 2001

ANATOMICAL ANALYSIS OF INDIRECT EFFERENT PATHWAYS OF THE SUPRACHIASMATIC NUCLEUS TO WAKE-RELATED CELL GROUPS IN RAT

Samuel Deurveilher, Joan Burns and Kazue Semba Department of Anatomy and Neurobiology, Dalhousie University, Halifax, Nova Scotia, Canada

The suprachiasmatic nucleus (SCN) of the hypothalamus is known to house the principal clock for many circadian rhythms in mammals, including the daily sleep/wake cycle. Little is known, however, about the neuronal efferent pathways by which the SCN exerts its control over sleep and wake behavior. We previously showed that the SCN output to the ventrolateral preoptic nucleus, known to contain sleep-active neurons, may be relayed by the medial preoptic area, subparaventricular zone, and dorsomedial hypothalamic nucleus. In this study we examined whether wake-related neurons in the forebrain and brainstem may also receive SCN output via these and other nuclei. Injec- tions of biotinylated dextran amine, alone or with cholera toxin B subunit, were made into selected hypothalamic regions. Wake-related neuronal groups were identified im- munohistochemically. Following injections into the medial preoptic area, subparaventricular zone, dorsomedial hypothalamic nucleus or posterior hypothalamic area, fairly dense anterogradely labeled fibers and terminals were seen close to hypocretin/orexin- containing neurons and histaminergic tuberomammillary neurons in the hypothalamus. Fewer labeled fibers and terminals were seen mixed with serotoninergic neurons in the dorsal raphe nucleus, noradrenergic neurons in the locus coeruleus and cholinergic neurons in the basal forebrain and mesopontine tegmentum. Numbers of retrogradely labeled neurons in the dorsomedial and ventrolateral regions of the SCN varied depending on the injection site. These results suggest that the medial preoptic area, subparaventricular zone, dorsomedial hypothalamic nucleus and posterior hypothalamic area may link the SCN not only to sleep-related neurons in the ventrolateral preoptic nucleus, but also to wake-related neurons in the forebrain and brainstem. The strength of links appears to vary among wake-related neuronal targets. These links may provide an anatomical substrate for the circadian influence on the timing of sleep and waking. Whether SCN efferents directly synapse, within these potential relay nuclei, with neurons projecting to wake-related cell groups remains to be investigated.

Supported by: CIHR (MOP42553) and Fondation Singer-Polignac. Actas de Fisiología 7, 2001 133

CIRCADIAN CONTROL OF THE AUTONOMIC AROUSAL RESPONSE

Michael F Hilton, Rena J Sugarbaker, Andrea J Okruch, John A Trinder, Sally A Sheils and Steven Shea Sleep Disorders Program, Brigham and Women’s Hospital, Boston, USA

Arousal from sleep is associated with an autonomically mediated cardiovascular response. Daily changes in sympathetic tone vary due to posture, activity and sleep. Evidence indicates little to no endogenous circadian contribution (suprachiasmatic nucleus (SCN)) to sympathetic tone. It is unknown if the autonomic nervous system’s (ANS) ability to respond to a stimulus is influenced by the SCN. To examine this question five subjects (1F, 4M) participated in a 13-day protocol conducted in dim light (<8 lux) temporal isolation allowing the SCN to free run at its intrinsic rhythm. During 7-days of the protocol, the subjects’ sleep-wake cycles were desynchronized from their endogenous SCN periods by altering day length to 28-hours (9.33-h scheduled sleep). Subjects were awoken from light sleep to full wakefulness, using an auditory stimulus (incremental tone), on three occasions evenly distributed across each sleep episode yielding arousals across the entire circadian cycle. The onset times of the EEG arousals were manually scored (ASDA criteria). Arousals repro- ducibly elicited monotonically increasing instantaneous heart rate (HR). The cardiac response to arousal was evaluated in terms of [1] magnitude, the size of the HR increase from first cardiac cycle to the 6th cardiac cycle after arousal onset and [2] speed, gradient of the 6 instantaneous HR’s immediately after arousal onset. Results were assigned a circadian phase according to their relationship with core body temperature (CBT) measurements. Heart rate response to auditory arousal in terms of [1] magnitude and [2] speed demonstrated significant endogenous circadian rhythms (p>0.03 and p>0.04 respectively) with peak cardiac response times at 80-circadian degrees (cosinor analysis). 80- Degrees circadian converts to an average clock time of 10:10AM (33-min) for these subjects. The average instantaneous HR increase, in response to arousal from sleep, was 33(3) b.p.m. at circadian phase of 80. The HR increased by 22(5) b.p.m. at circadian phases between 180 and 300-degrees (approx. 5 PM to 1 AM). The data demonstrate a robust, large amplitude endogenous circadian contribution to the ANS response to arousal from sleep. This may suggest that at approximately 10AM the ANS responds the most quickly and with the greatest magnitude to an environmental stimulus. These findings may have implications in cardiovascular disease where it has been shown that most sudden cardiac deaths occur at approximately between 9 and 11AM. 134 Actas de Fisiología 7, 2001

LIGHT EMITTING DIODES CAN PHASE DELAY THE MELATONIN RHYTHM

Helen Wright, Leon Lack, Kelly Partridge School of Psychology, Flinders University, South Australia, Australia

Introduction: Two portable light sources, comprising light emitting diodes (LED) of two different wavelengths, were compared to a standard light box in suppressing and phase shifting nocturnal salivary melatonin. Method: Sixty-six volunteers participated in the 2-day study and were randomly allocated to one of four conditions; white light box, white LED, blue/green LED or no light control group. Light was administered to the experimental groups from midnight to 0200 h on the first night. Half-hourly saliva samples were collected from 1900 h to 0200 h on night one and until 0100 h on night two. The wavelengths compared were a white LED with a narrow peak wavelength at 460 nm and a secondary broader peak at 560 nm and a blue/green LED with a single narrow peak wavelength at 497 nm. The LEDs were attached to spectacle frames with light directed at the pupils from a distance of 12 mm from the corneas of the participants’ eyes. All light sources were equated for illuminance of 2000 lux. Results: Percent melatonin suppression (SE bars) on night one is shown for all groups in Figure 1. A one-way ANOVA showed significant melatonin suppression during light stimulation for all experimental groups in comparison with the control group. The blue/green LEDs produced the greatest suppression (70%) and significantly more than the white LED (50%). Dim light melatonin onset (DLMO) was calculated for nights one and two. Figure 2 shows the mean phase delay (hrs) (SE bars) for each group. A one-way ANOVA indicated that all light stimulation groups had significant phase delay compared to the control group. The blue/green LED had a significantly greater DLMOdelay of 42 minutes than the white LED at 22 minutes and the light box of 23 minutes.

1 80

0.8 60

0.6 40 0.4

20 0.2

Phase delay (hours) 0 0 Mean melatonin % suppression -0.2 -20 Control Light box White LED B/G LED Ctrl Light box White LED B/G LED Figure 1: Mean melatonin Figure 2 Mean DLMO delay (hrs) percentage suppression

Conclusion: These data suggest the portable LED light source is an effective way of delivering light to phase shift the melatonin rhythm, with the blue/green LED being the more effective of the two LEDs. Actas de Fisiología 7, 2001 135

THE BENEFITS OF BRIEF AND ULTRA-BRIEF DAYTIME NAPS

Leon Lack and Amber Tietzel School of Psychology, Flinders University, South Australia

Introduction: Recent studies have shown that brief naps (7minutes and 10 minutes of sleep) improve subjective and objective alertness as much as longer naps (30 minutes) in the afternoon (1,2). Our main question was, Does the recuperative benefit come merely from the initiation of sleep? If so, it would have important applied utility. It also would have methodological implications since the initiation of sleep and a brief period of Stage 1 sleep (30-90 seconds) is usually required in the objective measurement of sleepiness (eg. multiple sleep latency test). In addition, if sleep initiation is as recuperative as longer sleeps, it would challenge the prevalent model of sleep homeostasis (Process S). Therefore, we compared the effects of a brief nap of 10 minutes with ultra-brief naps of 30 and 90 seconds of sleep. Methods: Sixteen (8 male, 8 female) young adult self-reported good sleepers with no history of sleep complaints or daytime napping all participated in four napping conditions (no nap, a 30-second nap, a 90 second nap, and a 10-minute nap) presented in counterbalanced order. On the evening prior to the laboratory sessions subjects sleep was restricted to five hours between 2400 and 0500 hours. The Stanford Sleepiness Scale (subjective sleepiness), Profile of Mood States (fatigue and vigour) and cognitive performance tasks (symbol-digit substitution and letter cancellation) were administered three times during each laboratory session; pre-nap, 5 minutes post-nap and 35 minutes post- nap. Objective sleepiness was assessed by sleep onset latency (SOL) measured both pre nap (the latency to the nap) and one hour after the nap. Results: There were no differences or interactions from pre-nap to post-nap between the no nap control and 30 second and 90 second sleep nap conditions. However, as in a previous study with the same methodology the 10-minute nap again resulted in significant (p<.01) decreases of sleepiness. Sleep onset latency increased from 3 minutes pre- nap to 8 minutes post-nap while there was no change in all other conditions. SSS decreased in comparison with all other conditions. In the 10 minute nap condition, although subjective fatigue decreased and vigour and the cognitive performance tasks showed increases in comparison with the other conditions these interaction effects only approached significance (p<.15). Conclusions: Following a night of sleep restricted to only five hours a 10-minute nap in the early afternoon produced a rapid recovery in alertness sustained for at least one hour following the nap. On the other hand, ultra-brief naps of 30 and 90 seconds sleep had no measureable benefits within the hour following the naps. Therefore, using a single or three 30-second sleep epochs criteria to determine sleep onset in the MSLT should not change subsequent sleepiness measurements. The benefits of the brief 10 min nap must arise from more than the initiation of stage 1 sleep. Whether the benefits of naps only arise from the beginning of stage 2 sleep, the initiation of some delta wave activity, or simply a fixed period of any stage of sleep between 90 seconds and 10 minutes duration needs further investigation.

References: (1) Takahashi M, Fukuda H, Arito H: Brief naps during post-lunch rest: effects on alertness, performance, and autonomic balance. European Journal of Applied Physiology 1998;78:93-98. (2) Tietzel, A.J., Lack, L.C.: The short-term benefits of brief and long naps following nocturnal sleep restriction. Sleep 2001;24:293-300. 136 Actas de Fisiología 7, 2001

NEUROIMMUNE INTERACTIONS IN THE MAMMALIAN CIRCADIAN SYSTEM

Luciano Marpegan, Tristán A. Bekinschtein, Mónica A. Costas and Diego Golombek C y T, Univ. de Quilmes, R.S. Peña 180, Bs. As., Argentina

Mammalian SCN rhythms are controlled by environmental stimuli and endogenous variables that may include circulating immune factors. These factors including proin- flammatory citokynes and bacterial lipopolysaccharide (LPS) are capable to induce sleep in mammals. Immunosupressant drugs produce circadian rhythm alterations in hamsters and many infection diseases produce modifications of some circadian patterns. In this work we have characterized the effects of LPS on circadian locomotor activity rhythms in C57/Bl6 mice, as well as photic entrainment and p50 (a member of the NF-kB complex) in the SCN. The administration of LPS (25 mg/Kg) induced a maximal motor inhibitory effect at CT 15. Under constant dark conditions, LPS produced a low amplitude photic-light phase response curve, with phase delays at CT 15. Light pulses (400 lux, 10 min) at CT 15 induced phase delays of 152±31 min, decreased to 83±10 min by the previous administration of sulfazalasine (an inhibitor of NF-kB activation). After LPS administration we found nuclear localization of p50 in SCN cells, while control tissue exhibited a cytoplasmic localization, suggesting the activation of NF-kB in the clock. Glial SCN cultures (harvested from P4 pups) were transiently transfected with a NF-kB-luciferase construct. LPS or TNF-a administration activated the construct, suggesting a direct action of these immune factors on SCN cells. We found TNF-a and IL- 1 receptors in these cells, which may be active on a feedback loop of clock-controlled cytokine rhythms onto the SCN. These results support the interaction between the immune and the circadian system, which probably regulate each other through fine-tuned mechanisms. Actas de Fisiología 7, 2001 137

RELATIONSHIP BETWEEN PLASMA HOMOCYSTEINE LEVELS AND SLEEP PARAMETERS IN SHIFTWORK BUS DRIVERS

Paulo José Forcina Martins(1), Vânia D’Almeida(1,2), Eduardo Henrique Rosa Santos(1), Marco Túlio de Mello(1) & SergioTufik(1). Departments of (1)Psychobiogy and (2)Pediatrics, UNIFESP-EPM, Brazil

Many work settings require individuals to displace sleep from the normal nighttime hours, leading to disturbed sleep, which may affect performance and safety. Disruption of the biological process that programs daytime wakefulness and nighttime sleep can result in reduced opportunities to sleep and poor sleep quality, affecting the shift work- er’s health. Previous studies have indicated an association between shiftwork and coronary heart disease with a 40% increase in cardiovascular disease risk, probably due to factors like circadian rhythms disturbance. Recent studies with humans and animals have shown that hyperhomocysteinemia is associated with increased risk of cardiovascular disease. Abundant epidemiological evidence shows a parallelism between hyperhomocysteinemia and cardiovascular risk factor that is independent of other conventional risk factors. There are factors like nutritional deficiency of vitamins (folic acid and B12) that participate in homocysteine metabolism, which can lead to metabolic disruption and potentially to hyperhomocysteinemia. The present study investigated the possibility of a relationship between plasma homocysteine levels and sleep parameters in professional bus drivers. Thirty drivers were submitted to polysomnography after an initial night of adaptation to the sleep laboratory. Sleep was recorded for 8 hours. The blood fasting samples for biochemical analysis were collected at 8 a.m. on the day of sleep recording. Homocysteine analysis was determined in plasma by high performance liquid chromatography, with fluorescence detection. Plasma folic acid and vitamin B12 were determined by enzyme immunoassay. Plasma homocysteine levels of bus drivers were higher than those observed in the health population (mean ± SEM) (18.57 ± 1.74mM; normal range 5-15mM); however, no alterations were observed in folic acid (5.08 ± 0.41hg/mL; normal range 1.1-20 hg/mL) and vitamin B12 (367.76 ± 20.86 pg/mL; normal range 211-911 pg/mL) levels. The bus drivers showed sleep alterations including: sleep efficiency was 84.2 ± 2.15 %, arousal per hour were 35.89 ± 3.69 and apnea/hipopnea index was 6.1 ± 2.14. There was weak correlation between homocysteine and sleep onset latency (rs=- 0.58), sleep efficiency (rs= 0.39) and number of apnea/hour (rs= 0.43). These results indicated that neither correlations between sleep parameters and homocysteine levels nor vitamin status can explain the elevated plasma homocysteine levels above those observed in the general population. However, hyperhomocysteinemia may be an explanation for the elevated indices of cardiovascular disease in shift workers, since it is a recognized independent risk factor for these diseases.

Financial support: Fapesp, Capes and AFIP. 138 Actas de Fisiología 7, 2001

IMPACT OF WORK SCHEDULES ON SLEEP DEBT IN HOSPITAL WORKERS

Quera Salva M. A., Cosquer M., Iwatsubo Y., Poissonet C.M. and Veron MA Hôpital Raymond Poincaré, AP/HP, Garches, France

The purpose of the present study was to evaluate the effects on sleep of different shift patterns in healthcare workers. The standard shift work index (SSWI) was used in a large survey on 3210 employ- ees of “Assitance Publique/Hôpitaux de Paris” (AP/HP). From the items of the SSWI, a sleep debt parameter was calculated from the estimate sleep need and the time spent sleeping during the various shifts, i.e. between two nights or two days on duty. The population was divided into 2 groups: group A worked only during the day on different schedules, group B worked on various schedules including always several night shifts. A covariance analysis was realised with sleep debt as covariant and the following independent variables: type of schedule, sex, age, speci- ality, number of children in charge, commuting time to work, job satisfaction score, global health quality and medical consumption. The total population had an estimated sleep need of 8 hours. Twelve hours day or night and night shift workers had an estimated sleep need of 7.9 hours. Among group A (n=1546), personnel with the greater sleep debt were those working on longest extended hours, they had a mean sleep debt of 0.65 hour per day (p=0.014 vs total sample). Personnel working 12 hours in a role had a sleep debt of 0.54 hours/day. Globally all personnel involved in night shifts (n=627) had a mean sleep debt of 1 hour per day versus 15 minutes for personnel working in day shift. Among group B workers, 4 different sub-groups were distinguished: - shift workers (n= 88), people working 12-hour rotating shifts (4), permanent night workers (438), people working 12-hour night shifts (37). Among the 4 sub-groups sleep debt between two nights of work was higher for shift workers while in the night shift, with a mean of two hours more than for the other 3 sub-groups with a mean of 1 hour for each group. Analysis of covariance show that in group A sleep debt was 15 minutes higher for women than for men. It increased by 9 minutes a day for each additional child. It was negatively correlated with work satisfaction score. In group B, sleep debt was 20 minutes higher for women than for men. It increased 17 minutes with each additional child. It was negatively correlated with work satisfaction score. Sleep debt was highest for people working at night than for people working on day schedules. People on shift work were most affected by sleep debt. As for day workers sleep debt was highest for people working longest hours. Sleep debt was highest in women than in men and increased with each additional child at home for day and night workers. On the contrary work satisfaction decreased with sleep debt. Actas de Fisiología 7, 2001 139

DAYLIGHT SAVING TIME AND MOTOR VEHICLE ACCIDENTS IN SOUTHERN BRAZIL

Geraldo NV Rizzo, Marilene F Alam and Euripedes A Santos

Benjamin Frankling, in 1784, observed that in advancing the watches in the summer , the individuals could spend more hours of natural light and become richer, by saving the mon- ey spent with candles. During the I World War this suggestion was applied by Germany , England and United States to save energy. And even after the war ended many countries like Brazil have addopted the so called Summer Time to save eletric energy. The energy saving is less than 5% but around 10% of the population present a lot of problems for a few days after this shift - insomnia, somnolence, irritability, digestive problems and innatention . The innatention due to sleep debt can result in vehicle crashes, job accidents, falls, serious burns, etc. Nowadays it has become increasing clear that insufficient sleep and disrupted circadian rhythms are a major public health problem. Stanley Coren in his book “Sleep thieves” has shown that motor vehicle crashes in Canada have increased by 8 percent following the change to daylight saving time and decreased by 7 percent after the change to standard time. With that in mind, the objective of our study was to examine whether the shift to daylight savings time in Brazil has also short-term effects on the incidence of traffic crashes. We have studied motor vehicle accidents on every national road of Rio Grande do Sul, the southernmost state of Brazil, during 3 days in the spring: the Monday preceding, the Monday immediately after the implantation of Summer Time and the Monday 1 week after it. The data was obtained with the help of the Federal Road Police, during 3 consecutive years: 1998,1999,2000. DATE ACCIDENTS INJURIES DEATHS DATE ACCIDENTS INJURIES DEATHS 05.10.98 36 32 05.10.99 22 02 06.10.98 28 08 11.10.99 23 10 01 12.10.98 46 21 03 12.10.99 28 10 13.10.98 31 05. 01 02.10.00 23 13 01 19.10.98 17 08 02.10.00 25 02 20.10.98 26 05 01 09.10.00 37 13 02 27.09.99 21 05 10.10.00 13 03 01 28.09.99 23 08 02 16.10.00 30 18 01 04.10.99 21 01 17.10.00 20 07 02 Accidents are one of the leading causes of mortality in Brazil and, although we do not know what percentage of total deaths are cause by accidents, a lot of actions have been developed to decrease this number. Regarding the so call Summer Time there is a debate whether the Brazilian Government should stop it or not. We expected to confirm Coren’s hypothesis that a small decrease in the duration of sleep could increase one’s susceptibility to accidents. However the loss of one hour’ sleep associated with the spring shift to daylight savings time did not increase the risk of accidents. This fact would be an argument in favor of those who want to end the Daylight Saving Time. (Trabalhar a estatística, risco relativo, intervalo de confidencia, chi quadrado, P<0.01) Lambe M, Cummings P. The shift to and from daylight savings time and motor vehicle crashes. Accident Analysis and Prevention 32(4):609-611, 2000. Mitler MM, Carskadon MA, Czeisler CA, Dement WC, et al. Catastrophes, sleep, and public policy: consensus report. Sleep 11(1):100-109, 1988. Leger D. The cost of sleep-related accidents: a report for the National Commission on Sleep Disorders Re- search. Sleep 17(1):84-93, 1994. Ferguson AS, Preusser DF, Lund AK, Zador PL, Ulmer RG. Daylight savving time and motor vehicle crashes: the reduction in pedestrian and vehicle occupant fatalities. Am J Oublic Health 85(1):92-95, 1995. Monk TH, Folkard S. Adjusting to the changes to and from daylight saving time. Nature 261: 688-689, 1976. Webb WB, Agnew HW Jr. Are we chronically sleep deprived? Bull Psychonom Soc. 6:47-48, 1975. 140 Actas de Fisiología 7, 2001

EVIDENCE FOR A SUPERSLOW WHOLE-HEAD WAXING-WANING RHYTHM IN MEG SIGNALS DURING LIGHT SLEEP

Norman R. Simona and Fernando H. Lopes da Silvab aDepartment of Physics and Astronomy, University of Nebraska-Lincoln, USA bInstitute of Neurobiology, University of Amsterdam, The Netherlands

Sleep recordings consisting of 10-min sleep periods were obtained from 3 subjects, using a 61-channel, whole-head MEG system. Each 10-min period was homogeneous in that it corresponded to either light sleep or deep slow-wave sleep. We segmented each period into 1200 ½-sec bins and calculated the activity aik for each channel i within each time-bin k. Thus N 1/2 =−1 ()2 aik∑ bb ikn ik N n=1 where bikn is the magnetic field detected in channel i at the nth step (sample) within bin k, bik the average field over time-bin k, and N=125. Thus the activity is a measure of the variation of the magnetic signal during each ½-sec time interval. Relative power spectra were constructed for the activity time series in each channel, i.e., aik (k=1,2,…,1200) and the spectra averaged over all channels at each frequency. Typical examples of channel-averaged power spectra:

1.0 1.0 Slow- wave 0.8 0.8 sleep

Light

0.6 r 0.6 Sleep

0.4

el. powe 0.4 power Rel. R

0.2 0.2

0.0 0.0

0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 f (hz) f (Hz) For light-sleep periods we observed striking peaks in the lowest frequency bins at 0.008 or 0.016 Hz (left plot); but as sleep deepened into the slow-wave phase, this oscillation virtually disappeared (right plot). The high coherence of the light-sleep rhythm can be verified by examining correlations in the activity among detector pairs. Averaging over all channel pairs, we found correlation coefficients of the order 0.55 for light sleep, but falling to 0.15 in the deep slow-wave phase. Thus it seems that the magnetic signals in light sleep (but not in slow-wave sleep) wax and wane collectively at locations all over the head in a superslow rhythm with periodicity 1 – 2 min. This oscillation may be congruent with one found recently in EEG sleep data (P. Achermann, et. al., Neurosci. 81, 1997, 213-222). Its precise origin remains to be elucidated. Actas de Fisiología 7, 2001 141

POSTNATAL REM SLEEP DEPRIVATION AND DEPRESSION: NEW FINDINGS AND HYPOTHESIS

Pingfu Feng Sleep Lab., Department of Psychiatry and Behavioral Sciences, Emory University, Atlanta, USA

It is well documented that postnatal (PN) treatment with clomipramine (CLI) and several other drugs, mostly antidepressant and REM sleep (RS) suppressant, produces a set of behavioral changes exhibited in human depressive disorders. These behavioral changes include decreased sexual activity, decreased aggression, increased locomotion, decreased pleasure-seeking behaviors, extended immobility in forced swim test, increased adult alcohol consumption. These drugs have different aminergic effects. Regardless of their aminergic effects, all tested antidepressant drugs that produce adult depressive behavior have RS deprivation (RSD) effects. These findings suggested the hypothesis that PN RSD by the drugs, rather than their different aminergic effects, mediates the adult depression. Consistent with this hypothesis, Iprindole, an antidepressant drug that does not decrease RS, does not produce adult depressive symptoms after administration to PN rats. RSD by drugs is always confounded by the other, unavoidable effects of the drugs. To circumvent such complications, our lab developed a method for continuous (24 hr/day), long term (weeks) polysomnographic (PSG) recordings and instrumental RSD (IRSD) of PN rats. In our new method, an IRSD and a yoked control (YC) PN rat shared the same cage but were separated by a vertical wall. The cage was attached to the platform of a laboratory shaker. The shaker was turned on when IRSD rats began RS. YC rats were permitted to have RS, while IRSD rats were not. Our study demonstrated that IRSD of PN rats produces adult behavioral changes similar to those elicited by PN treatment with CLI. Compared with YC rats, adult IRSD rats demonstrated diminished sexual behavior, decreased aggressive behavior, increased RS percentage, and decreased body weight. This result further supports the hypothesis that PN RSD produces adult depressive behavior. PN RS plays a critical role in development has been postulated based on the well- accepted behavioral phenomena in the early-age: a very high percentage of RS and during which very frequent phasic muscle twitches occur. Three of our recent studies using our new method demonstrated the similar result, i.e. PN (no elder than 21 days) RSD either by drug or by instrumental method produces large portion (45-69% depending on age) of RS reduction without proportional increase of wake. The RS reduction is largely compensated by NS during PN RSD. This is possibly the answer to the paradox of why adult RSD alleviates depressive symptoms and PN RSD produces depression like behaviors. RS reduction without an increase in wakefulness results in a net loss of stimulation through the reduction of endogenous (self) stimulation generated during RS without the increase of exogenous stimulation by behavioral movement during waking periods. 142 Actas de Fisiología 7, 2001

FUNCTIONS OF SLEEP: SIMPLICITY VERSUS COMPLEXITY

R.V. Rial1, M.C. Nicolau1,A. Gamundi1, M. Akaarir1, and S.R. Pandi-Perumal2 1Laboratoiri de Fisiologia, Departamento de Biologia Fonamental I Ciencies de la Salut, Universitat de les Illes Balears, 07071 Palma de Mallorca, Spain. 2Department of Physiology, UCLA School of Medicine, Los Angeles, CA 90095, USA

An important challenge facing sleep researchers is the fact that sleep, although considered a simple behavior in animals, is rather complex, with many components and multiple functional and anatomical controlling substrates. As additional factors of this complexity, different animals show sleep with different components, with different depths, and with different neural controls. As a behavior, sleep has been broadly defined by 1) motor rest, 2) elevated sensory thresholds, 3) easy reversibility, 4) species specific sleeping posture, 5) resting/sleeping locations, 6) circadian organization, and 7) physiological regulation. The above sum of characteristics however is heterogeneous, because reversibility, rhythmicity, and regulation are factors which pervade other ones, acting as regulators for the amount, intensity, organization, and of the spatial-temporal distribution of sleep. Regulation exists for the amount of motor activity, sensory processing efficiency, sleeping posture, and for search of the sleeping site. The existence of multiple regulatory mechanisms upon every sleep component could be used as an indicator to the cause of sleep, i.e., one could state that animals sleep either to be immobile, to achieve sensory rest, or to maintain a regular contact with the sleeping site as a part of its individual territory. However, there is an ascending inclusiveness of sleep characteristics, from simple to complex organisms: protozoa, (even plants) will regulate their motor rest/activity cycles, while mammals control the whole set. This inclusiveness could be important because it also offers cues on the function of sleep: simple species must sleep only to divide the biological time in accordance with their ability to cope with changes in the environment. Then additional functions can be added as animals become more and more complex. However, one should not blindly assign functions to every anatomic or functional detail found in a living being. Such details could be either mere evolutionary accidents, or unavoidable byproducts of other truly adaptive characteristics, which, in principle, could bear no relation to sleep. The problem could be thus to distinguish between essential characteristics and byproducts. Actas de Fisiología 7, 2001 143

PARADOXICAL SLEEP DEPRIVATION AND THE ACTIVITY OF THE HPA AXIS

Deborah Suchecki, Paula A Tyba and Tatiana M Baía University of Sao Paulo, Brazil [FOCUS GROUP]

Instrumental methods to induce paradoxical sleep (PS) deprivation also result in an immediate activation of the hypothalamic-pituitary-adrenal (HPA) axis, suggesting these methods are stressful. For several years our group has been trying to develop a method of PS deprivation that would contain the least intervening variables as possible, such as social isolation and restriction of movements. Recently, a series of papers showed that animals raised and PS deprived as a socially-stable group in a large water tank containing 14-15 narrow platforms exhibit attenuated stress indexes, compared to animals deprived by the same method, but as a socially-unstable group. This method has been named modified multiple platform method (MMPM) and includes an environmental control group, where animals are placed on a grid inside the water tank. It is known that prolonged enhancement of corticosterone (CORT) levels may hinder further elevations of CORT, due to the influence of negative feedback system. Moreover, we were curious to examine whether animals PS deprived individually (single platform method [SPM]) or as a group (MMPM) would respond differently to stress. To examine whether PS deprivation could alter the dynamics of the HPA axis, we submitted animals PS deprived by the SPM or by the MMPM to a mild stressor (saline injection + novelty) or to a more dramatic stressor (5 min session on the elevated plus maze [EPM]). Appropriate control groups (dry SPM control [SPMC] and the grid control [GR]), in addition to a cage control (CC) group, were used. Animals were either sacrificed immediately after the deprivation period or at the corresponding time-point for control groups (time-point 0) or 5, 20 or 60 min after the stressor. In response to the mild stress, control groups, except GR showed a peak CORT levels at 5 min, which persisted until the 20-min time-point, returning to basal levels by 60 min. PS deprived animals, curiously, exhibited a peak of similar magnitude at 5 min, but precipitously returned to basal levels by 20 min remaining so low by 60 min. In response to the EPM, control groups exhibited elevated CORT plasma levels at 5 min, which were further enhanced by 20 min. Except for the GR animals, CORT levels began to lower but did not return to basal by 60 min. Again PS deprived animals exhibited increased CORT levels in response to stress without showing the abrupt return to basal levels by 20 min, but doing so by 60 min. Interestingly, MMPM animals exhibited a less anxious behavior in the EPM, with more entrances and longer time spent in the open arms. These data indicate that PS deprivation does not hinder the ability of HPA axis to mount an adequate stress response nor impairs the negative feedback system. Howev- er, the behavioral response to the EPM appears to be dependent on whether animals are deprived individually or as a group.

This work was supported by grants from AFIP and FAPESP (CEPID project). 144 Actas de Fisiología 7, 2001

BILATERAL ABLATION OF VOMERONASAL ORGANS DETERMINES THE APPEARANCE OF HIGH AMPLITUDE GAMMA WAVES IN OLFACTORY BULBS, OLFACTORY TUBERCLES AND PIRIFORM CORTEX DURING PARADOXICAL SLEEP OF THE ARMADILLO

Jorge Mario Affanni and Claudio Cervino Neuroscience Institute. (INEUCI). Consejo Nacional de Investigaciones Científicas y Técnicas. Universidad de Buenos Aires. Facultad de Medicina de la Universidad de Morón. Argentina

Little is known about the effect of suppression of vomeronasal input to the brain during Wakefulness and Sleep as well as about the interaction between main and accessory olfactory systems. Male and female armadillos of the species Chaetophractus villosus were submitted to bilateral ablation of the Vomeronasal Organs. The bioelectrical activity of olfactory bulbs, olfactory tubercles and piriform cortices was studied during Wakefulness, Slow Wave Sleep and Paradoxical Sleep both before and after the surgical ablation of the organs. Before the operation, a peculiar bioelectrical rhythm of 8-12 Hz was observed in the olfactory bulbs, olfactory tubercles and piriform cortex coexistent with Adrian´s induced waves and Ottoson´s slow waves. The rhythm was seen only during relaxed Wakefulness. The presence of both peripheral and centrifugal inputs to the olfactory bulbs appears necessary for the rhythm to appear. Neither induced Adrian´s waves nor Ottoson´s slow waves were seen during Slow Wave Sleep and Paradoxical Sleep. After the operation, the peculiar 8-12 Hz rhythm was still observed during Wake- fulness. During Slow Wave Sleep the bioelectrical activity was similar to that of intact or sham operated animals. However, from the beginning of Paradoxical Sleep a striking change persisting throughout this phase was seen: high amplitude 25-35-Hz gamma waves invaded all the above mentioned structures. Those waves are very similar to Adrian waves but they are not originated by impulses from neurons of the olfactory mucosa because they persist after the chemical destruction of the mucosa. Instead, the waves appear as being elicited by the centrifugal input to the olfactory bulbs since they are abolished by complete surgical section of the olfactory peduncles. Those waves were not seen in sham operated animals. These results show a selective effect of vomeronasal ablation on the bioelectrical patterns of Paradoxical Sleep. As the effect of the operation clearly modifies the bioelectrical activity of the main olfactory bulbs, an interaction between both olfactory systems (main and accessory) is strongly suggested. The significance of these findings is discussed. Actas de Fisiología 7, 2001 145

DIFFERENCES IN DREAM STRUCTURE BETWEEN BLIND AND SIGHTED SUBJECTS

Helder Bértolo, Teresa Paiva, Lara Pessoa, Tiago Mestre and Raquel Marques Lab. EEG/Sono – Centro de Estudos Egas Moniz – Faculty of Medicine of Lisbon, Portugal

It is currently assumed that congenital blind have a different dream structure than sighted subjects, namely that they do not dream with images. We evaluated 10 congenital blind (age: 28.2±5.2; 5 males and 5 females) and a control group of 9 sighted subjects (28.2±5.5; 4 males and 5 females), matched for age, sex and education, without medical or psychiatric diseases, and with regular sleep schedules, in order to assess the differences between the dream structure. The dream reports were collected at the volunteers’ home during two consecutive nights. The subjects were awaken four times each night, every 90 minutes, through a programable alarm. After being awakened the volunteer recorded to a voice-activated tape recorder everything he/she was thinking prior to the awakening. Grammatical and content analysis were performed on the dream reports. The first consisted on the following classes: Total Number of Words (TW), Percentage of Verbs (PV), Percentage of Adjectives (PA) and Percentage of Nouns (PN). The content analysis was performed based on the Hall & Van the Castle criteria using two of the categories: Activities and Emotions, and the corresponding classes: Physical (P), Movement (M), Change in Location (L), Verbal (V), Expressive Communication (E), Visual (S), Auditive (A), Thought (T), Fury (FU), Apprehension (AP), Happiness (HA), Sadness (SA) and Confusion (CO). Each recording was analysed independently by two experimenters, unaware of the volunteer’s category. For each analysis the following parameters were calculated: Global Oneiric Activity Index (GAI=(P+M+L+V+E+S+A+T)/TW), Visual Activity Index (VAI=S/TW), Thought Activity Index (TAI=T/TW), Muscular Activity Index (MAI=(P+M)/TW), and Global Oneiric Emotivity Index (GEI=(FU+AP+HA+SA+CO)/ TW). The final value was the mean of the two independent scores. The 2 groups were compared through the Mann-Whitney’s Test. Regarding the grammatical analysis no significant differences were found between the groups. For the content analysis the only significant difference is for the visual class (S) (U=16.5, p<0.05), but this difference disappears when we compute the VAI, i.e., the Visual Activation Index presents no difference between the groups. These results suggest that congenital blind and sighted subjects dream structures are similar. 146 Actas de Fisiología 7, 2001

SLEEP AND DREAM HABITS OF PATIENTS WITH EATING DISORDERS

Brigitte Holzinger (1), Gerhard Klösch (1,2), Silvia Parapatics (3), Petra Schüssler (3), Martina De Zwaan (3) And Peter Gathmann (3) 1 Institute of Consciousness- and Dream Research, Vienna, Austria; 2 Department of Neurology, University of Vienna, Austria; 3 Department of Psychiatry, University of Vienna, Austria

Various epidemiologic studies show that during the last 20 to 30 years the number of patients suffering from eating disorders (ICD-10: F50.0.; F50.2) has increased continuously. Apart from a number of physical symptoms acute weight loss also causes insomnia and decreased amounts of sleep. This study* intended to objectify the sleep and dream habits of 15 inpatients with anorexia nervosa and 15 inpatients with bulimia, utilising sleep and dream diaries and wrist-worn actigraphs. The patients were studied over a period of eight weeks at the Department of Psychiatry, Section of Psychosomatics, University of Vienna. They had to fill out several questionnaires including the Pittsburgh Sleep Quality Questionnaire (PSQI), Eating Disorder Inventory (EDI-II), Personality Inventory (NEO-FFI), Self- rating Anxiety and Depression scale (SAS,SDS) and Quality of Life Questionnaire (QLI) at the beginning and at the end of the treatment period. Results demonstrate that patients have shorter sleep periods (total sleep time and time in bed) at the beginning of therapy. At their weekends at home they also show different sleep habits (extremely short sleep periods) as compared to the days at the hospital. Dream content analysis revealed frequent conflicts with family members and a great amount of scenes containing violence and frightening episodes.

*) The project was supported by the Austrian National Bank Actas de Fisiología 7, 2001 147

HOW DREAMS AND MEMORY MAY BE RELATED

Eugen G Tarnow 3725 Blackstone Avenue, #4E, Riverdale, NY 10463, USA

We present a theory of dreams and long term memory structure that proposes that both entities are closely related. It is based on a variation of Freud’s dream theory: (1) we re-label Freud’s “Unconscious” the “Long Term Memory Structure” (LTMS), (2) we propose that dreams are ever present excitational responses to perturbations of perceptions and thought, during waking life as well as sleep, which only become conscious when the executive function of waking life ceases, and (3) we reinterpret Freud’s “Dream Work” as describing the pre-dream Storage Transformation of perceptions and thought into the LTMS. We make one further conjecture: Memories are stored in the LTMS according to what is already in the LTMS. Our theory solves two problems inherent in Freud’s theory: (1) the strangeness of dreams comes about not because a sophisticated censor is turned on during sleep but because an executive function is turned off and (2) equating the “Unconscious” with the LTMS gives an experimentally verifiable anchor for psycho-dynamic theory. The observables of Freud’s theory remain the same in our theory. Our theory is also consistent with recent experimental findings and suggests a partial basis for personality: the selection process of the Storage Transformation. 148 Actas de Fisiología 7, 2001

ACTIVE SLEEP STRENGTHENS THE PROCESSING OF LOW INTENSITY AUDITORY STIMULATION IN EARLY INFANCY

C. Algarín1, M. Garrido1, R. Uauy1, E.E. Birch2 & P. Peirano1 [1]Sleep and Functional Neurobiology Lab, INTA, University of Chile, Santiago, CHILE [2]Retina Foundation of the Southwest, Dallas, USA

Introduction: To asses the effects of sleep states on transmission through the auditory pathway, brainstem evoked responses (ABER) were obtained in a group of 95 healthy preterm infants evaluated at 50 weeks post-conceptual age. Methods: ABER recordings were assigned to one of two groups by an independent observer according to whether the infants were in active sleep (AS) or quiet sleep (QS). Both sleep states were determined behaviorally while ABERs were evaluated when a sleep state was clearly established. ABER amplitude (uV), latency (ms), and area under the curve (uv/ms) were assessed in all infants at 85 decibels (dB) for waves I, III, and V (25 during AS and 70 during QS), and at 55dB for waves III and V (25 during AS and 70 during QS). At 15dB all except 9 infants were evaluated solely for wave V (23 during AS and 63 during QS). Results: No statistically significant differences in amplitude and latency between AS and QS were observed.The data obtained show that the area under the curve of wave V at 55dB (4.8+0.4 vs 4.0+0.4, p<0.04), 35dB (3.8+0.4 vs 2.9+0.4, p<0.002) and 15dB (2.6+0.5 vs 1.9+0.5, p<0.004) were significantly larger in AS relative to QS. At 85dB the area under the curve was not statistically significant. Discussion: These results indicate that AS modulates the gain in the late synaptic relay of the brainstem auditory pathway (lateral lemniscus) when low intensity stimulation is applied, but does not affect it at high intensity signals. A larger area under the curve may represent an extended neural recruitment. We speculate that AS may help strengthen the processing of low intensity auditory stimulation in early infancy . (Funded by NIH HD 22380 and Fondecyt 1930820 & 1000657 grants). Actas de Fisiología 7, 2001 149

EVALUATION OF SLEEP PROPENSITY AND SLEEP INERTIA USING THE FORCED AWAKENING TEST IN PATIENTS WITH EXCESSIVE DAYTIME SLEEPINESS

Helene Bastuji, Fabien Perrin and Luis Garcia Larrea Hopital Neurologique, 59, bd Pinel, Lyon, France

Although event related potentials (ERPs) are also known to be sensitive to sleepiness, they have been found less powerful than MSLT to detect it in individual subjects. A less investigated aspect of excessive somnolence is ‘ascending sleepiness’ or ‘sleep inertia’, which we explored using a paradigm combining forced awakening from a nap with simultaneous recording of ERPs. Our previous study showed that the forced awakening test might represent an useful adjunct to current strategies for detecting pathological somnolence. We now report the results obtained in 60 patients complaining of excessive daytime sleepiness (EDS) from various origins. During the test, patients learned to perform a typical “oddball” paradigm whereby occasional high-pitched tones (‘targets’) had to be detected amongst a flow of low- pitched sounds. The P300 to target tones was used as an index of controlled processing of task-relevant stimuli, and was recorded (a) during full wakefulness, and (b) after forced awakening following 3 minutes of continuous sleep. ERPs during wakefulness were normal in both controls and patients. While in controls, forced awakening lead to a small (-25%) decrease of P300 amplitude with little or no latency delay, patients with excessive daytime sleepiness showed very significant ERP and behavioural signs of sleep inertia on forced awakening. Furthermore abnormal patterns, specially ERPs containing “sleep negativities”, were significantly more prevalent in patients with narcolepsy and idiopathic hypersomnia, whereas patients with psychiatric somnolence did not exhibit ERPs signs of sleep inertia. Thus, the FA test coupled to ERPs seems to provide a robust and rapid tool to evaluate not only sleep propensity but also acute sleep inertia, another important aspect of EDS. 150 Actas de Fisiología 7, 2001

CONSCIOUSNESS, STIMULUS AWARENESS AND EVENT RELATED POTENTIALS. LESSONS FROM THE FORCED AWAKENING TEST

Helene Bastuji, Fabien Perrin and Luis Garcia Larrea Hopital Neurologique, 59, bd Pinel, Lyon, France

The P3 wave of event-related potentials (ERPs) has been considered as a possible concomitant of the access of sensory information to consciousness and to memory storage. However, the mere presence of a P3 is not a guarantee of stimulus access to consciousness, since a degraded version of this component has also been observed in patients who were apparently unaware of the stimulation (cortical blindness), as well as in sleeping subjects who were not able to recall the stimulus on awakening. We used the forced awakening test to further explore the relationship between ERPs, stimulus awareness and stimulus access to consciousness and recall. Twenty one control subjects and 40 hypersomniac patients learned to perform a typical “oddball” paradigm whereby occasional 70 dB, high-pitched targets had to be silently counted amongst a flow of low-pitched sounds. Tones where presented (a) during full wakefulness, and (b) following 3 minutes of continuous sleep. After the forced awakening test (FA) subjects were questioned about quantitative and qualitative aspects of stimulus recall. In subjects whose the quality of recall was excellent, P3 waves during FA were indistinguishable from those obtained previous to the nap. When P3 was found attenuated, delayed and desynchronised in subjects whose recall was quantitatively degraded (increased counting errors). P3 was concomitant to (or replaced by) high-amplitude sleep negativities in subjects in whom stimulus recall was severely degraded or absent. These results are consistent with the notion that (a) the presence of a P3 is crucial to ensure stimulus encoding in a retrievable format, the quality of encoding being related to the characteristics of the P3 wave, and (b) ‘sleep negativities’ superimposed on P3 degrade very significantly the quality of recall, suggesting that, although instant awareness of the stimulus may still exist in those cases, sleep negativities may act as ‘erasers’ pre- venting accurate memory encoding and retrieval of the stimulus. Actas de Fisiología 7, 2001 151

DIFFERENTIAL PROCESSING OF INCONGRUENCES DURING SLEEP STAGE 2 AND PARADOXICAL SLEEP

Helene Bastuji, Fabien Perrin and Luis Garcia Larrea Hopital Neurologique, 59, bd Pinel, Lyon, France

Recent studies suggest that sleeping persons, in sleep stage 2 & paradoxical sleep, can discriminate deviant tones in a series of monotone tones (Bastuji and García-Larrea, 1999) as well as their own name in a series of other names (Perrin et al., 1999, 2000). However, the extent to which sleeping subjects can process the significance of external stimuli remains open. To further investigate this question, we recorded auditory evoked potentials, from 10 normal adults, in response to semantically related and unrelated pairs of words, during both waking and all night sleep (passive paradigms). Six lists of 72 stimuli were elaborated: all were monosyllabics and presented with a fixed 1,200 ms interval. One third of stimuli were pseudo-words and were intermingled between either semantically related or unrelated pairs of words. During passive waking, a higher N400 potential was elicited in response to unrelated words relative to related words. Pseudo-words also elicited a N400 potential which amplitude was also higher than that observed after unrelated words. During sleep stage 2 —when no K-complexes were simultaneously evoked— a N400-like potential was enhanced to unrelated words and to pseudo-words relative to related words. Conversely to that observed in waking state, the amplitude of this nega- tivity was similar in response to unrelated words and to pseudo-words. During paradoxical sleep, a N400-like potential was elicited in response to unrelated words but not to pseudo-words, which amplitude was intermediated to that of related and unrelated ones. These results suggested that the discrimination of semantic incongruences is still operative during paradoxical sleep and sleep stage 2, but that the processing of pseudo- words is specific to each sleep stage. 152 Actas de Fisiología 7, 2001

A CONTINUOUS 50 HZ ELECTRIC AND MAGNETIC FIELD DURING SLEEP SUPPRESSES SLOW WAVE ENERGY DURING THE FIRST SLEEP CYCLE

Jens Nilsson1, Mats Gillberg2, Torbjörn Åkerstedt1, Bengt Arnetz3 1National Institute for Psychosocial Factors And Health, Stockholm, Sweden, 2Karolinska Institutet, Stockholm, Sweden and 3Department of Social Medicine, University of Uppsala, Sweden

28 healthy subjects, 14 women, mean age 26,6 ys(SE=±1,1), participated in a double blind study regarding effects on sleep of a nocturnal exposition of a continuous vertical 50 Hz electric and magnetic field (EMF). The field was generated from a coil, 2 m in diameter, placed under the floor. The conditions were 0 mT, 1 mT and 2 mT in bal- anced order, one sleep per week, with a habituation sleep one week before the experimental conditions. The Total Sleep Time the night before the experimental conditions was 7,7h(SE=±0,25), monitored with actigraphy and the Karolinska Sleep Diary. The subjects (two per night) arrived at the laboratory at 2000h on the experimental night and had a light standardized meal at 2130h. An i.v catheter was inserted at 2100h and blood samples were obtained eight times(10 ml) between 2130h to 0630h. Lights out was at 2230h. Polysomnography was carried out using: C3/A2, C4/A1, EOG and EMG. Slow Wave Energy(mV2 x minutes of NREM sleep) was calculated per cycle for analysis with a two way repeated measures ANOVA. Sleep was also scored visually according to conventional methods. The main effect of cycle was significant (p<0,01), but there was no significant main effect of condition. The interaction effect of condition and cycle was significant F(2,24)=4,061(p<0,05). SWE was suppressed in the exposure conditions during the first sleep cycle. That relationship was inverted during the second sleep cycle(figure 1). SWS showed a similar pattern as SWE, although not significant. Thus, EMF had a slight effect on the normal sleep architecture. Actas de Fisiología 7, 2001 153

AMBIENT MODULATION OF REM SLEEP OCCURRENCE [SYMPOSYUM]

200 180 d, e, f Fig.1. Time course of the average 160

) relative power of the EEG delta band 140 (0.75-4.0 Hz) for the following 120 REMS intervals (s): c 100 a (30-60); b (61-120); c (121-180); 80 b a d (181-600); e (601-960); f (961- 60

elative delta power (% 1800). r 40 The onset of REMS is indicated by the 20 zero value on the x-axis. 0 -90 -60 -30 0 time before REMS (s)

1) R. Amici et al., J.Sleep Res. (1994) 3, 250-256 2) R. Amici et al., Brain Res. (1998) 781, 252-258 154 Actas de Fisiología 7, 2001

SLEEP IN THE FOLLICULAR AND LUTEAL PHASES OF THE MENSTRUAL CYCLE

Fiona C. Baker, Duncan Mitchell, and Helen S. Driver Wits Sleep Laboratory, Brain Function Research Unit, University of the Witwatersrand, Johannesburg, South Africa

In young women, reproductive hormones vary cyclically over a period of approximately 26 to 35 days, which constitutes the menstrual cycle. Reproductive hormones not only regulate the functioning of reproductive tissues during the menstrual cycle, but through their secondary actions in the central nervous system also may influence other physiological processes, such as sleep. Although sleep has been investigated previously in healthy, young women (See Driver and Baker, 1998 for review), small sample sizes and variable sampling times have prevented consensus of how the menstrual cycle influences sleep. We therefore combined the sleep data from 20 women (age: 21±4 y, mass: 61±6 kg; height: 1.65±0.70 m; menstrual cycle length: 31±5 days) with ovula- tory menstrual cycles who had participated in three different studies, but followed the same protocol and slept in the same laboratory, during the mid-follicular and mid- luteal phases of the menstrual cycle. Although the women tended to report a poorer sleep quality in the luteal phase compared to the follicular phase, objective measures of sleep quality did not differ between menstrual phases; the women took a similar length of time to fall asleep and they spent a similar, small amount of time awake, moving and in light stage 1 sleep, combined. Sleep efficiency was greater than 95% and was the same in both phases of the menstrual cycle. There also was no difference in the latency to slow wave sleep (SWS) or in the amount of SWS in the follicular phase compared to the luteal phase. Sleep homeostasis, as indexed by sleep efficiency and amount of SWS, therefore is maintained despite the hormonal differences in two extremes of the menstrual cycle. In contrast to the lack of effect of menstrual phase on non-rapid eye movement (REM) sleep, REM sleep was significantly affected by menstrual phase. The women had slightly less REM sleep (21±2 %) and a shorter onset latency to REM sleep (65±12 min) in the luteal phase than in the follicular phase (22±2 % and 70±13 min, respectively; paired t- tests, P<0.05). The variation in REM sleep may be related to the body temperature changes or altered circadian processes during the menstrual cycle, or may be caused directly by ovarian hormones. In conclusion, sleep is remarkably stable, apart from a small change in REM sleep, in the mid-follicular and mid-luteal phases of the menstrual cycle in young women with no menstrual-associated complaints.

Reference: Driver, H.S. and Baker, F.C. (1998). Menstrual factors in sleep. Sleep Med. Rev. 2: 213-229 Actas de Fisiología 7, 2001 155

SLEEP AND WAKEFULNESS IN A GENETIC MOUSE MODEL OF HELPLESSNESS

Daniela Popa, Saoussen Bouali, Malika El Yacoubi*, Jean Costentin*, Michel Hamon, Jean-Marie Vaugeois* and Joëlle Adrien INSERM U288, Pitié-Salpêtrière, 75013 Paris, and * CNRS 6036, 76183 Rouen Cedex - France

Sleep architecture undergoes specific alterations in depressed patients, notably difficulties in falling asleep, reduction of the Rapid Eye Movement (REM) sleep latency, and sleep instability These impairments are associated with a general increase in cholinergic and decrease in serotoninergic neurotransmissions (1). In addition, antidepressants might exert their therapeutic action, at least in part, through a facilitation of serotoninergic neurotransmission mediated by desensitization of serotoninergic 5-HT somatodendritic autoreceptors (2). Because there is a need for experimental models, which would allow to investigate the neurobiological mechanisms possibly underlying the psychopathology of depression and its treatments, we have analyzed sleep-wakefulness regulations and the functional status of cholinergic and serotoninergic systems, in a genetic mouse model of helplessness (3). The tail suspension test (of 6 min duration), in which immobility mimics a state of helplessness, has allowed the generation of two lines of mice from the CD1 strain, diverging in their helpless (H: immobility >115 sec), and non-helpless (NH: immobility <35 sec) behaviour. These behavioural traits exhibit great stability, and that of the H line can be partially reversed by chronic treatment with antidepressants. Under pentobarbital (70 mg/kg) anaesthesia, female mice of the H and NH line (8th generation) were implanted with electrodes for polygraphic sleep monitoring (4). After recovery and habituation to the recording conditions (12 h light-dark cycle, lights on at 7:00, food and water ad libitum): 1- Polygraphic recordings of the spontaneous sleep- wakefulness cycles were performed during 96 hours for baseline data, and during 8 hours after IP injection (at 10:00) of the muscarinic agonist arecoline (0.05 and 0.1 mg/kg) for examining the functional status of the cholinergic system. 2- Body temperature was measured (with a rectal probe) after S.C. treatment with 8-OH-DPAT (0.2 to 2 mg/kg), in order to evaluate the hypothermia induced by selective activation of 5-HT1A somatodendritic autoreceptors. Under baseline conditions, both lines of mice exhibited a classical circadian rhythm of sleep and wakefulness. However, enhanced amounts of light slow wave sleep (SWS1) and of REM sleep, as well as a decrease in REM sleep latency, were observed in H as compared to non NH mice. In addition, arecoline induced a precipitation of REM sleep, and 8-OH-DPAT a larger hypothermia, in the H as compared to the NH group. These data indicate that the spontaneous sleep-wakefulness cycles in helpless mice mimic those observed in depressed patients, and that the cholinergic regulation of REM sleep is facilitated as compared to that in non-helpless mice. In addition, the 5-HT1A triggered negative feedback influence on serotoninergic neurotransmission is enhanced in helpless mice. Altogether, these results seem to validate the helpless line of mice as a relevant model for investigation of the neurobiological correlates of depression and of antidepressant treatments.

1- Gillin JC, Sutton L, Ruiz C, Kelsoe J, et al. Arch Gen Psychiatry 1991, 48: 264-270 2- Pineyro G, Blier P. Pharmacol Rev 1999, 51: 533-591 3- Vaugeois J-M, Odievre C, Loisel L, Costentin J. Eur J Pharmacol. 1996, 316, R1-R2 4- Boutrel, et al. J. Neurosci., 1999, 19: 3204-3212 156 Actas de Fisiología 7, 2001

CLINICAL APPLICATION OF SPLIT-NIGHT POLYSOMNOGRAPHY

José P. Arcos and Daniel Lorenzo. Technical Assistance: Lilián Chiappella, Zulma Rodríguez, Nancy Otaño and Héctor Escamendi. Laboratorio de Exploración Funcional Respiratoria, Hospital de Clínicas y Depto. de Medicina, IMPASA, Montevideo, Uruguay

Clinical presentation of sleep apnea syndrome may predict the severity the disorder and suggest the indication of treatment with nasal CPAP. Split-night polysomnography (diagnosis and pressure titration) has been recommended to initiate therapy (ASDA 1997). The purpose of this study was to evaluate the practical use of this method and to assess the predictive value of simple clinical signs. Patients referred for sleep-breathing disorders were scheduled either for full-night or split-night polysomnography if they complained of severe excessive daytime somnolence and if they were markedly obese. The functional outcomes sleep questionnaire (FOSQ) was also applied. Clinical characteristics and polysomnographic data are presented in the following table:

Split-night Full-night Diagnosis Titration N 125000000 5700000000 — Age, yr 48.9 ± 13.3 53.3 ± 10.8 — BMI, kg/m2 31.4 ± 7.3 37.4 ± 7.8 — TST, min 314.8 ± 66.5 111.1 ± 26.1 214.1 ± 43.5 S1 lat, min 23.4 ± 29.4 14.5 ± 15.0 12.0 ± 9.9 S1-2, %TST 78.3 ± 13.7 91.0 ± 11.0 56.5 ± 21.8 S3-4, %TST 7.5 ± 7.4 1.7 ± 4.2 15.8 ± 13.1 REM, %TST 13.0 ± 6.9 8.2 ± 11.4 27.2 ± 13.0 Efficiency Index 0.77 ± 0.23 0.74 ± 0.17 0.81 ± 0.14 AH Index 30.4 ± 25.0 59.9 ± 26.8 — Arousal Index 22.0 ± 25.3 64.5 ± 36.4 — CT<90%,%TST 11.0 ± 19.9 12.2 ± 14.5 — FSOQ 15.0 ± 3.2 11.6 ± 2.9 —

The presence of somnolence and obesity predicted an AHI>40, with 62% sensibility and 92% specificity. FOSQ showed a significant lower score in patients selected for split-night. CPAP titration was accomplished and allowed to reach the effective pressure to suppress apneic events and desaturation in 90% of patients. During the titration period a remarkable increase in slow wave sleep was observed. Follow up of patients who started CPAP treatment confirmed the efficacy of selected pressure in most cases. Actas de Fisiología 7, 2001 157

EFFECTS OF SLEEP DEPRIVATION ON EXTRACELLULAR SEROTONIN IN THE HIPPOCAMPUS AND THE FRONTAL CORTEX OF THE RAT

Bjørn Bjorvatn, Janne Grønli, Frøydis Hamre, Eli Sørensen, Eldbjørg Fiske, Alvhild A Bjørkum, Chiara M Portas and Reidun Ursin Dept. of Public Health and Primary Health Care, Bergen, Norway

Sleep deprivation improves the mood of depressed patients (1), but the exact mechanism behind this effect is unclear. An enhancement of serotonergic neurotransmission has been suggested. In this study, we used in vivo microdialysis to monitor extracellular serotonin in the hippocampus and the frontal cortex of rats during an eight hour sleep deprivation period. These brain regions were selected since both have been implicated in depression. The behavioral state of the animal was continuously monitored by polygraphic recordings during the experiment. Sleep deprivation produced a gradual decline in extracellular serotonin levels, both in the hippocampus and in the frontal cortex. In order to investigate whether the reduction in serotonin was due to other factors than sleep deprivation, i.e. time of day effect, another experiment was performed. Here the animals were allowed to sleep during most of the recording period. This experiment showed the expected changes in extracellular serotonin levels: consistently higher levels during wakefulness compared to during sleep, but no time of day effect. The reduction in extracellular serotonin during sleep deprivation may suggest that serotonin does not play a major role in the mood-elevating effect of sleep deprivation. However, the effect of sleep deprivation on depressed mood may be related to the fact that serotonergic neurotransmission is strongly behavioral state dependent. Both the spontaneous activity of serotonergic neurons and the extracellular serotonin levels in different brain regions are clearly higher in waking than in sleep (2, 3). Thus, maintenance of wakefulness keeps serotonergic neurotransmission on a high level for a prolonged period of time. By eliminating sleep, there may be a net increase in serotonergic neurotransmission during the sleep deprivation period, even though extracellular serotonin decreases.

1. Wirz-Justice A, Van den Hoofdakker RH, 1999. Sleep deprivation in depression: what do we know, where do we go? Biol Psychiatry. 46, 445-453 2. Jacobs BL, Fornal CA, 1991. V. Activity of brain serotonergic neurons in the behaving animal. Pharmacol Rev. 43, 563-578 3. Portas CM, Bjorvatn B, Ursin R, 2000. Serotonin and the sleep/wake cycle: special emphasis on microdialysis studies. Prog Neurobiol. 60, 13-35 158 Actas de Fisiología 7, 2001

NIGHTCAP: A RELIABLE SYSTEM FOR DETERMINING SLEEP ONSET LATENCY

Jose L. Cantero, Mercedes Atienza, Robert Stickgold and John A. Hobson Laboratory of Neurophysiology, Department of Psychiatry, Harvard Medical School, Massachusetts Mental Health Center, 74 Fenwood Road, Boston, United States

Previous studies have demonstrated that the Nightcap home-based sleep-monitoring system can differentiate waking, NREM and REM sleep based on eyelid movements (ELMs) and head movement behavior. The present study aims at determining the reliability of the Nightcap in scoring sleep onset latency (SOL) in normal subjects using a standard multiple sleep latency test (MSLT) protocol. Four naps were recorded in each of ten normal subjects using both polysomnography (PSG) and the Nightcap. The Night- cap scored sleep onset as the first of 4 consecutive 30-sec epochs with less than 5 ELMs. The mean Nightcap and PSG SOLs differed by only 2 sec (7.15 min vs.7.12 min), and the average absolute difference was 45 sec (13.3% of PSG SOL). SOLs differed by less than 1 min in 85% of onsets. Recordings of EEG activity from 90 sec before and after PSG-identified sleep onsets were subjected to spectral analysis. Changes in spectral power in the theta (4-7 Hz) and alpha (8-12 Hz) frequency bands during the transition into light sleep correlated well with eyelid behavior. However, changes in ELM density predicted sleep onset better than did changes in theta and alpha spectral power. Comparisons indicate that the Nightcap predicts SOL in normal sleepers better than other non-PSG sleep monitoring systems. According to these results, the Nightcap may be a valid alternative to the PSG technique in the assessment of excessive diurnal somnolence using standard MSLT protocols. Actas de Fisiología 7, 2001 159

EVALUATION OF TELEMETRIC RECORDING TECHNIQUES TO STUDY SLEEP PARAMETERS IN THE COMMON MARMOSET (Callithrix jacchus)

Sergio Melotto, Alessandro Poffe and Philip Gerrard Sleep Research Dept, GlaxoSmithKline S.p.A., Medicines Research Centre, Verona, Italy

Despite the widespread use of the marmoset as a laboratory species, few reports have been published regarding quantitative assessment of sleep parameters in this animal. The marmoset, being a small diurnal primate, may provide a useful species in which to model human sleep behaviour. This has not been possible until recently due to technical difficulties associated with the recording of electroencephalogram (EEG) in this species. We have used multichannel telemetric probes (TL10M3-F50-EEE, Data Sci- ences Int.) to record EEG, electromyogram (EMG) and electrooculogram (EOG) in unrestrained animals. The aim of the present study was to validate telemetric recording as a method to study sleep parameters in the marmoset. Adult marmosets (n=5) were implanted with telemetric transmitters located into the peritoneum under alphax- olone anaesthesia (SaffanTM) and EEG/EMG/EOG recording electrodes were posi- tioned using standard surgical techniques. Three weeks post surgery, the effects of the hypnotic agents triazolam (TRIAZ, 0.03 and 0.3mg/kg p.o.) and zolpidem (ZOLP, 1 and 3mg/kg p.o.) and the anxiolytic buspirone (BUSP 1 and 3mg/kg p.o.) were tested in a disturbed sleep model. A “white noise” (70dB, normal active colony noise level) was used to reduce the amount and quality of basal sleep over the 5h test period. Hyp- nograms were derived from EEG/EMG/EOG recordings using Fast Fourier Transfor- mation (FFT) and four sleep parameters were analysed; awake, stage 1-2 sleep, stage 3-4 sleep and REM sleep. TRIAZ. and ZOLP. reversed the increase in time awake by increasing stage 3-4 sleep (P<0.05 paired t-test) though failed to reverse the REM deficit. BUSP did not modify sleep parameters at the doses tested, suggesting that the sleep disturbance was not an anxiety response to the white noise. Future studies will aim to validate further this model as a method to investigate the effects of hypnotic agents on sleep parameters in the primate. 160 Actas de Fisiología 7, 2001

EVALUATION OF ACCEPTANCE, IMPACT AND BENEFITS OF A PORTUGUESE SLEEP-SITE: PRELIMINARY RESULTS

Ana P. Nunes*, Teresa Paiva*, Marta Gonçalves* and Armando Teles Araújo** *ISTEL – Inst. Sono, Cronobiologia e Telemedicina, Lisboa & Oporto, Portugal ** Emeline, Lisboa, Portugal

The impact of internet upon healthcare is very important; nevertheless objective evaluations of attitudes and impact are mostly lacking. Sleep sites are increasing in number and complexity, but most are in English, designed for Anglo-Saxon culture. Sites in natural languages are scarce, and rules concerning quality of information content and HTML design either lack or are imprecise. Therefore, the design of this site took into consideration the mentioned aspects and used the following strategy:

Content Quality

Synthetic

Figurative Interested people

Patients Design Site on the Net Med. Doctors www.istel.pt Redesign

Journalist

Content Multidisciplinary Visitors Evaluation assessment

EVALUATION Navigation Self explained

Multidirectional

CONCEPTION IMPLEMENTATION

Conception, done by sleep specialists had 2 main focus: easy and immediate grasp of relevant information; accurate content. Design had figurative/artistic objectives; navigation is friendly, using conventional tools and multidirectional access. Implemented by a site designer, it was online on the International Sleep Day. A questionnaire was then sent to: Internet specialists sleep patients, med. doctors, med. students and jour- nalists (n=15/group). It focused on: content, information, design, usefulness, friendli- ness, relevance, criticism, suggestions. The main results were: 1)The Site was considered useful by the vast majority; the artistic side and clearness of exposure were enhanced. 2)The main criticism was the site weight and downloading difficulties; some referred the need of more detailed information. 3)Some patients considered it difficult to understand; Medical doctors and students considered the site useful and could learn from it. Conclusion: At this preliminary stage the Portuguese site www.istel.pt on sleep, seems to have a positive evaluation and impact, but further improvements are still needed and redesign recommended. Actas de Fisiología 7, 2001 161

PORTUGUESE GENERAL PRACTITIONERS NEEDS AND KNOWLEDGE CONCERNING SLEEP DISORDERS

Teresa Paiva*, Jürgen Zulley**, Madalena Teles Araújo*, Marta Gonçalves*, Thomas Penzel**, Pierre Philip** and Christian Guilleminault** *ISTEL- Inst.Sono, Cronobiologia e Telemedicina, Lisboa and Oporto, Portugal ** ENN Society – Marburg, Germany

Objectives: The development of a multimedia telematic network to support GP’s should be based on the actual needs and knowledge of this group. Thus, the evaluation of the users needs and knowledge is a prerequisite for the start of the development of the different products such as a multimedia tutorial, expert systems and communication network facilities. Material and Methods: The present data, focused in Portuguese GPs, represents part of a multinational questionnaire developed during ENN project, by Sleep specialists. The questionnaire with 20 questions focus on GPs knowledge and needs; It contains general questions (10), a sleep questionnaire on GPs needs (8) and on their knowledge (12). The questionnaire has been translated, the groups of GPs selected and then it was sent out to the respective group of GPs in Lisboa (n= 23) and Oporto (n= 129); To- tal=152. Results: GPs populations have similar characteristics concerning age, gender, professional experiences and patients population size. Most GPs do not work with computers; however in a study of prediction of GPs attitudes, conducted in 50 GPs in Lisbon, it was observed that most of them were willing to use new computer based technologies. In daily practice there is clear impact of insomnia. Most GPs relate insomnia with psychiatric causes. For hypersomnia, the etiological differences found are also ques- tionable: 32.8% make the diagnosis of idiopathic hypersomnia. The vast majority GPs (>97%) feel difficulties in the diagnosis and treatment of sleep complaints. The same applies to the information needed, which concerns several aspects of physiology, diagnosis and treatment of sleep disorders. Females tend to focus on sleep in children and address a multiplicity of problems. Answers concerning basic sleep knowledge were only correct in percentages ranging from 2.6% to 58.3%. The questions with worse scores concerned physiological aspects of sleep and sleep in children. Those with best scores related to treatment of sleep apnoea, long-term hypnotic prescription and drugs inducing insomnia. Females achieved in some items higher scores then males. Conclusions: The answers to the questionnaires raise the suspicions that many treatable cases might go undiagnosed. Treatment and diagnostic difficulties of GP’s induce frequent consultations, excessive and unnecessary medications and useless laboratory tests. Links between GPs and trained specialists are often scarce, with difficult exchange of knowledge. New telematic technologies possess useful tools to change the actual picture. The very low scores achieved indicate the need for teaching sleep medicine. 162 Actas de Fisiología 7, 2001

SENSITIVITY AND SPECIFICITY OF THE MAINTENANCE OF WAKEFULNESS TEST AT DIFFERENT LEVEL OF SLEEPINESS

Montserrat Reséndiz , Rosa M. Campos, Josefina Montes, Violeta Castaño, Victoria Santiago, Guillermo Garcia-Ramos and Matilde Valencia-Flores Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán, PUIS-Facultad de Psicología UNAM. México, D.F.

The Maintenance of Wakefulness Test (MWT) has been proposed for evaluation of daytime sleepiness/wakefulness, the Multiple Sleep Latency Test (MSLT) is considered a reliable test for measurement daytime sleep tendency. The aim of this study was to determine the sensitivity and specificity of the MWT against MSLT at two different sleepiness levels. Subjects were 67 obese patients with obstructive sleep apnea and mean age of 44.4+12.5 years. After one night of polysomnography, two hours after awakening, the patients underwent 20-min per trial version of MWT followed by a standard MSLT. In addition, the Epworth Sleepiness Scale (ESS), the Sleep-Wake Activity Inventory (SWAI) and the Sleep Disorders Questionnaire (SDQ) were applied. The patients were classified according to the mean MSLT. Group-1 < 5 min Group-2 < 10 min. Table shows the sensitivity and specificity parameters for MWT.

Mean MSLT Maintenance Wakefulness Test Sensitivity Specificity

< 5 min 95.2% 44.0%

< 10 min 31.2% 38.2%

Reported sleepiness in ESS and SDQ was more severe in the Group-1.(Group-1 10.7+7.7, Group-2 7.0+4.3).These preliminary data suggest that sensitivity of MWT is highly dependent on the level of sleepiness but not specificty, which is very low, at different levels of sleepiness.

Supported by DGAPA-IN209500 Actas de Fisiología 7, 2001 163

ARE K COMPLEXES AND DELTA WAVES DIFFERENT? A MATHEMATICAL MODELISATION AND ITS APPLICATION

Agostinho Rosa, Christian Guilleminault, Dalva Poyares, Lamia Afifi and Uta Koester Laseeb - ISR - IST, Lisbon, Portugal

In the current literature there is an unclear functional role or distinction between delta waves and K complexes. This work put forward the hypothesis that delta waves and K complexes are morphologically and functionally different. The argument is supported by an extended tonic and phasic EEG generation model with physiological background. This model is able to generate a whole night sleep EEG with preservation of the macrostructure and phasic events. According to this model the delta burst and isolated delta waves are the tonic or permanent response of the model under control of the sleep ultradian NREM-REM sleep regulator by setting the trend and changes of the delta generators synchrony. On the other hand an external stimuli elicits a phasic or as in linear systems analysis terminology, a transient response. The final response to the external stimuli or event-related response is composed by the sum of the two tonic and phasic components simultaneously. This model can explain the whole range of variation in the tonic delta waves and also all grades of variations of the K complexes. This wide range of morphologic differences are controlled by the current tonic states or sleep stage and the significance or intensity of the stimulus. Besides illustration through simulation of the two different activities the model also provides a mean to detect them, through a model based estimation and detection approach. Methodology: Scorers we instructed to classify isolated delta band (0.1 to 4Hz) activity into isolated delta burst or K complexes according to it “non-estacionarity” i.e. sudden change and morphological resemblance to the simulated results. The same data was analyzed by the model based automatic detector, this detector will mark with some probability the K complexes. The selectivity and specificity of the automatic detector in relation to the visual analysis was performed. 164 Actas de Fisiología 7, 2001

ENN PROTOTYPES FOR SLEEP DISORDERS

Madalena Teles Araújo*, Teresa Paiva*, Thomas Penzel**, Caterina Rehm- Berbeni***, Armando Teles Araújo****, Ana Fred*, Christian Guilleminault** *ISTEL – Instituto do sono, Cronobiologia e Telemedicina, Lisboa, Portugal ** ENN Society, Marburg, Germany *** FUTUREtec, Bergisch Gladbach (köln), Germany **** Emeline, Lisboa, Portugal

Introduction: The ENN Project (European Neurological Network) aimed the creation of a multimedia telematic network of units involving different countries and giving support to experts and GP’s, namely in the field of sleep. Objectives: For sleep, the objective was the development of a set of prototypes, namely a database/sleep atlas, an expert system for polysomnography and a tutorial CD Rom, representing modules of an information system, useful for better understanding, diagnosis and treatment of sleep disorders. Prototypes: The database/ Sleep Atlas contains representative electrophysiological signals, which can become a reference for a digital atlas of PSG recordings. Implementa- tion includes the physical structure, software and interface for loading and retrieval of multimedia data, according to existing standards. It consists of 20 fully documented patients: obstructive sleep apnoea; snoring; periodic leg movements; narcolepsy; sleep walk- ing; delayed sleep phase syndrome, etc. It will have case reports for 88 different diagnosis (International Classification of Sleep Disorders), with clinical information and video recordings (medical interviews, sleep disorders, polygraphic signal recordings). The Polysomnography Expert System, implements diagnostic rules defined by experts and provides a final PSG report. It has a user interface with explanation capabilities, access to database and links to the Internet. A “case builder” allows evaluation of real and virtual data of the expert system. This expert system is rule based and requires as input variables the evaluation of a PSG. The Sleep Tutorial encompass basic knowledge within their fields, presented in a case oriented practical way, with a user friendly navigation; Theoretical explanations are provided whenever requested by the user. Applications are developed using Macromedia and the program only requires a standard multimedia PC. Results: A set of prototypes (ENN tool set), process oriented, multilingual (6 languages), with scientific and professional back-up, which are now under some small improvements, in order to become marketable products. Conclusions: All prototypes, combined like pieces of a puzzle, the ENN Informa- tion System, distributed using standard and e-society facilities and process oriented, constitute the ENNIS (European Neurological Network in the Information Society) concept and offer, which in view of its characteristics and results are high potential products. Actas de Fisiología 7, 2001 165

THE ACTIVITY OF ACETYLCHOLINESTERASE (Achase) IN SUBCELLULAR FRACTIONS OF THE CEREBRAL CORTEX FROM RATS DEPRIVED OF PARADOXICAL SLEEP (PS)

Marco A. C. Benedito and Marilene Demasi Departamento de Psicobiologia, Universidade Federal de São Paulo, São Paulo, Brasil.

High tonic cortical activity during PS, similar to waking state, is a characteristic of PS (Neuroscience 40: 637-656, 1991). Cholinergic basal forebrain neurons project topo- graphically to the cerebral cortex and are involved in cortical activation (Progress in Brain Research 98: 61-71, 1993). Cortical acetylcholine (Ach) release is high during waking and PS (Brain Research 671: 329-332, 1995). The level of Ach in the synaptic cleft is controlled by the activity of Achase, the enzyme that hydrolyses Ach. PS deprivation (PSD) for 96 hours in rats induced a decrease in the activity of membrane-bound Achase in the frontal cortex (Brazilian J. of Med. Biol. Res. 30:641-647, 1997); however, there was no change in the cerebral cortex membranes (Brazilian J. of Med. Biol. Res. 34: 103-109, 2001). This work aimed to determine the activity of Achase in subcellular fractions of cerebral cortex after PSD. Adult, male, Wistar rats were deprived of PS by the flower-pot technique (N=6). The control rats (N=6) remained in their home-cages. After 96h of PSD the rats were sacrificed. Achase activity was analyzed photometrically using acetylthiocholine as the substrate (1mM final concentration) in homogenates (Ho: 900xg supernatant), synaptosomes (SP) and synaptic plasma membranes (SPM) from cerebral cortex. The results (nmol of thiocholine formed/mg protein/min, mean±SD) obtained did not detect any statistically significant difference in the activity of Achase after PSD (p>0.05, Student t test).

Ho (900 xg) SP SPM Control 104 ± 9 94 ± 9 153 ±11 PSD 103 ± 9 99 ± 7 159 ± 8

The results obtained in the activity of Achase from subcellular fractions of cerebral cortex are in accordance to the previous reported lack of a change in Achase activity in the whole cerebral cortex after PSD in rats.

Supported by AFIP and CAPES 166 Actas de Fisiología 7, 2001

EFFECTS OF PROLONGED WAKING-AUDITORY STIMULATION ON HUMAN SLEEP

Jose L. Cantero, Mercedes Atienza, Rosa M. Salas, Elena Dominguez-Marin Eliana A. Quintero Laboratory of Neurophysiology, Department of Psychiatry, Harvard Medical School, Massachusetts Mental Health Center, 74 Fenwood Road, Boston, United States

Evidence suggests that changes in sleep architecture are the result of increased brain metabolism during the prior wakefulness. On the other hand, the destabilization caused by the preferential stimulation of specific neuronal groups in detriment of other has been hypothesized to be followed by a mechanism of neural compensation in order to preserve the synaptic superstructure of the brain, these mechanisms of synaptic maintenance being plausible happen during sleep. To test both hypotheses, eight volunteer subjects were presented with auditory stimulation unilaterally during 6 hours prior to sleep in two different days (right and left). Polysomnographic recordings were carried out in four different nights (adaptation, baseline, after right stimulation, and after left stimulation; stimulation sessions were separated one week one each other). Prolonged auditory stimulation was followed by a selective increase in the duration of slow wave sleep (SWS), and a shortening of the latency between SWS periods compared with baseline night. REM sleep parameters were unaltered by the waking-auditory stimulation. At a electrophysiological level, 6 hours of continuous auditory stimulation generated a higher level of synchronization in the range of alpha band and fast sleep spindles over the whole scalp (more accentuated in temporal regions) restricted to SWS, as compared with baseline night. Coherence analysis made complex and hemispheric- dependent functional relationships between different cortical regions clear in SWS and REM sleep. Together, these results demonstrate that the selective and prolonged stimulation of the auditory system markedly alter the SWS regulation and its underlying electrophysiology, supporting the notion of sleep both as a use-dependent mechanism and as an essential period to regulate synaptic unbalances imposed by the prior wakefulness. Actas de Fisiología 7, 2001 167

SINGLE CELL ACTIVITY PATTERNS OF PEDUNCULO PONTINE TEGMENTAL NEURONS PREDICT BEHAVIORAL STATES OF SLEEP AND WAKEFULNESS IN FREELY MOVING RATS Subimal Datta Sleep Research Laboratory, Department of Psychiatry, Boston University School of Medicine, Boston, Massachusetts, USA

Introduction: The pedunculo pontine tegmentum (PPT) contains a prominent group of cholinergic neurons that project widely throughout the brainstem and forebrain (1). Microinjection of L-glutamate into the PPT can increase wakefulness and/or REM sleep depending on the L-glutamate dosage (2, 3). Single cell recordings from the PPT of behaving cats have identified several different classes of cells whose firing rates correlate with both wakefulness and REM sleep (1). Some of these PPT neurons, called REM-on cells, progressively increase their firing rates as the animal moves from wakefulness to slow-wave sleep (SWS) and then to REM sleep. Others, constituting the majority of these cells in the PPT, called Wake-REM-on cells, are tonically active during both wakefulness and REM sleep. The goal of the present study was to record single cell activity patterns of PPT cells in freely moving rats during different stages of the sleep-wake cycle. Methods: Recordings were performed on 12 male Sprague-Dawley rats weighing between 250 and 350 g. With the use of sterile procedures, cortical electroencephalogram (EEG), dorsal neck muscle electromyogram (EMG), electrooculogram (EOG), hippocampal EEG (to record theta wave), and pontine EEG (to record P-wave) recording electrodes were chronically implanted. In addition, a microdrive-microelectrode assembly with 12 flexible wires was stereotaxically implanted for recording single cell activity from the PPT (A:0.70-1.30; L:1.5-2.0; H:2.4-3.2). Following a post-surgical recovery period of 3-7 days, rats were habituated to a sound attenuated recording cage and free moving polygraphic recording conditions for 7 days. All recording sessions were performed between 10:00 and 16:00 h, when rats are normally sleeping. After the adaptation recording sessions, single cell recording sessions began. Unitary action potentials from each cell were recorded continuously for two to three sleep-wake cycles. These neuronal activity data together with polygraphic records are stored on DAT tape for later off-line computerized analysis. Results: A total of 64 cells’ action potentials were recorded in more than two sleep-wake cycles. Histological identifications showed that single cell activity patterns recorded from 48 cells were within NADPH-diaphorase-positive cell compartments of the PPT. The other 16 cells were in the inferior colliculus. All of those PPT cells fired tonically as single spikes. It is interesting to note that only inferior colliculus cells fired as both single spikes and low-frequency bursts (between 70 and 90 Hz within a burst). Of those 48 PPT neurons, 6 cells (12.5%) were more active during REM sleep compared to wakefulness (W) and SWS (REM-on type). Another 11 cells (22.92%) were more active during W than SWS and REM sleep (W-on type). Rest 31 cells (64.58%) were more active during both W and REM sleep compared to SWS (W-REM-on type). The firing rate of these PPT neurons is summarized below (Mean ± S.D. Hz): Cell Type (number, percentage) W SWS REM sleep REM-on (n=6, 12.50%) 0.17 ± 0.12 0.29 ± 0.23 6.70 ± 2.5 W-on (n=11, 22.92%) 1.88 ± 0.29 0.21 ± 0.18 0.12 ± 0.11 W-REM-on (n=31, 64.58%) 12.6 ± 3.2 0.90 ± 0.8 6.7 0 ± 1.8 Conclusions: These results demonstrate that the majority of PPT cells are more active during both W and REM sleep compared to SWS. Population activity analysis showed that 59.45% of the total activity is during W, 4.64% is during SWS, and 35.91% is during REM sleep. Thus, the level of population activity within the PPT area during REM sleep is 60% of the level of activity during W. The results presented here suggest for the first time in the rat that high levels of cellular activity within the PPT may be causally involved in the induction of W and that more moderate levels of activation lead to REM sleep. Low levels of cellular activity within the PPT may facilitate SWS. 1. Datta S. Neuronal activity in the peribrachial area: relationship to behavioral state control. Neurosci. Biobehav. Rev., 1995, 19:67-84. 2. Datta S, Siwek DF. Excitation of the brainstem pedunculopontine tegmentum cholinergic cells induces wakefulness and REM sleep. J. Neurophysiol., 1997, 77:2975-2988. 3. Datta S, Spoley EE, Patterson EH. Microinjection of glutamate into the pedunculopontine tegmentum induces REM sleep and wakefulness in the rat. Am J. Physiol., 2001:R752-R759. Research was supported by NIH research grants MH59839 and NS34004. 168 Actas de Fisiología 7, 2001

THE TIME COURSE OF 1-Hz EEG POWER VALUES DURING THE SLEEP ONSET PERIOD

Luigi De Gennaro, Michele Ferrara and Mario Bertini Department of Psychology - University of Rome “La Sapienza”, Rome, Italy

Microstructural electroencephalographic changes during the wakefulness-sleep transition have been investigated by comparing two definitions of sleep onset: the first occurrence of stage 1 and of stage 2. Power values were calculated across a 1-28 Hz frequency range in a 1-Hz bin resolution in the sleep recordings of 26 normal subjects. Quantitative changes were assessed after averaging individual time series, aligned with respect to the first occurrence of stage 1 or of stage 2. The time course of a single-Hz activity revealed a linear increase of power in the 1- 6 Hz range and a linear decrease in the 9-12 and 16-28 Hz ranges during the stage 1 transition. During the stage 2 transition, EEG power linearly increased in the 1-7 and 14-15 Hz ranges and decreased in the 18-28 Hz range, while the 8-12 Hz range fitted a second-order polynomial curve . The two “switch” points were also compared in their ability to differentiate Hz by Hz wakefulness from sleep: a lower mean power was found after Stage 1 onset in the 9- 11 Hz and 20-28 Hz bins and a higher one in the 1-5 Hz bins, while a higher power was found in the 1-8 Hz and 12-16 Hz bins and a lower one in 18-28 Hz bins after stage 2 onset. The time course of three electroencephalographic frequency ranges [delta/theta/ sigma (1-7 and 12-16 Hz); beta (17-28 Hz); alpha(8-11 Hz)], grouped on the basis of a Principal Component Analysis, fitted a first-order polynomial curve for the first two ranges, and a second-order polynomial curve for the last, with a progressive decrease during wakefulness, a minimum point during stage 1, and a subsequent increase during stage 2. The uniformly increasing electroencephalographic power across the 1-16 Hz frequency range during stage 2 and the shift of functional meaning for the alpha power during Stage 1 point to the start of stage 2 as a more reliable boundary between wakefulness and sleep. Actas de Fisiología 7, 2001 169

ANTERO-POSTERIOR EEG CHANGES DURING THE WAKE-SLEEP TRANSITION

Luigi De Gennaro, Michele Ferrara, Giuseppe Curcio and Riccardo Cristiani Department of Psychology - University of Rome “La Sapienza”, Rome, Italy

Objectives of the study were to investigate the brain topography of the human sleep EEG along the antero-posterior axis during the wakefulness-sleep transition, by means of both a single-Hertz analysis and a grouped-frequency analysis of EEG changes. Methods: EEG power values were calculated across a 1-28 Hz frequency range in a 1- Hz bin resolution during the wakefulness-sleep transition of 7 normal subjects. Topo- graphical changes were assessed from C3-A2, C4-A1, Fpz-A1, Fz-A1, Cz-A1, Pz-A1, Oz-A1 recordings, after averaging individual time series, aligned with respect to the onset of stage 2. Results: The single-Hz analysis showed that before sleep onset, the <7 Hz slow frequencies were more prominent at the more anterior scalp locations; this anterior prominence was counterbalanced by a reciprocal prevalence across the >8 Hz frequencies of EEG activity from the occipital areas; while the >13 Hz fast frequencies were not characterized by significant antero-posterior differences. After sleep onset, more EEG power was found in the range of slow frequencies at the centro-frontal scalp locations and a second peak of EEG activity was also revealed within the range of the sigma frequency, higher at the centro-parietal scalp locations. No consistent topographical changes were observed within the range of faster EEG frequencies. Grouped-frequency analysis confirmed these results, also pointing to different changes in the alpha frequency as a function of the sleep onset point. Conclusions: Results suggest that: a) the alpha rhythm spreads anteriorly as the transition progresses; b) several anterior areas first synchronize EEG activity; c) the functional meaning of the EEG bands during the sleep onset period should be partially revised with regard at least to alpha rhythm; d) sleep onset coincides with the start of stage 2. 170 Actas de Fisiología 7, 2001

UNITARY AND FIELD ACTIVITY IN HIPPOCAMPUS DURING CLASSICAL CONDITIONING IN ALERT CATS

A. Múnera, A. Gruart, M.D. Muñoz, R. Fernández-Mas, J.M. Delgado-García División de Neurociencias. Laboratorio Andaluz de Biología, Universidad Pablo de Olavide, Ctra. de Utrera, Km 1, 41013 Sevilla, Spain

The electrical activity of hippocampus is related to cognitive states involved in memory and learning during alertness, including the acquisition of new motor abilities. The population and unitary activities of pyramidal cells were recorded during the classical conditioning of eyelid responses in alert behaving cats. The experimental preparation included: 1) a coil implanted in the upper lid to record lid movements using the magnetic searching technique; 2) a bipolar stimulating electrode implanted in the fornix; and, 3) either fixed (metal) or mobile (glass) electrodes directed to the CA1 layer. The unconditioned stimulus (US) was an air puff applied to the cornea, whereas the conditioned stimulus (CS) was either a tone or a weak, short air puff. The conditioning paradigms used were either of delayed or trace type. The analysis of the population activity in the pyramidal layer of CA1 showed that, during the paired CS-US presentation: 1) the CS-evoked potential amplitude was larger than the US-evoked potential one; 2) there was a shift from theta towards high frequency (mostly beta) oscillations; 3) there was an enhancement of the field potential response to fornix electric stimulation; and, 4) there were no correlation between population activity parameters and the kinetic properties of eyelid conditioned responses (CR). Scopolamine, a muscarinic antagonist, administered to animals with well-established CRs not only distorted the information processing in the hippocampus, but also impaired specifically the execu- tion of the CR. These findings suggest that cholinergic inputs participate in hippocampal information processing in a way that allows precise CR performance and memory trace formation. Identified CA1 pyramidal cells characteristically responded with brief bursts to unpaired stimuli presentations. However, after paired CS-US presentations, pyramidal cell responses to CS were enhanced, whereas their responses to US remained unchanged. These responses were independent of the conditioning paradigm and of CS sensory modality. There was no correlation between the firing rate of pyramidal cells and the kinetic parameters of conditioned or reflex blink responses. In conclusion, the hippocampus processes CS-US temporal relationship and the firing of its pyramidal cells signal the predictive value of the CS. Moreover, a characteristic beta oscillatory pattern seems to be present during the acquisition process.

Supported by grants from DGICYT PM98-0011, Fundación La Caixa 00/032-00, and Junta de Andalucía CIV122. Actas de Fisiología 7, 2001 171

SPECTRAL ANALYSIS OF EEG BEFORE AND DURING AROUSALS

De Carli Fabrizio1, Nobili Lino2, Beelke Manolo2, Ferrillo Franco2 1 Center for Cerebral Neurophysiology, National Research Council, Genoa, Italy; 2 Sleep Disorder Center, DISMR, University of Genoa, Italy

ASDA arousals has been defined as abrupt shift of EEG frequency to theta, alpha or beta range. They can be induced by endogenous or exogenous stimuli and can appear during different sleep stages. Thus different arousal patterns can be found and it has been noted that burst of slow waves often precede arousals. The aim of this study is to investigate EEG activities preceding and constituting arousals by means of frequency analysis and to seek for a relationship with preceding slow waves. We have analyzed 11 polysomnographic recordings from patients with different sleep-related pathologies (breathing disorders, PLMs, epilepsy, narcolepsy and psy- chophysiologic insomnia). Each tracing has been inspected to select artifact-free arousals clearly corresponding to the ASDA criteria. Three consecutive EEG segments have been associated to each event, corresponding to: background activity (20 sec), pre- arousal period (3.5 sec), arousal (variable length). The EEG signal recorded from 2 bipolar channels (F4-C4 and C4-O2) has been analyzed by means of the discrete wavelet transform (using filters optimized for frequency selectivity) and the logarithms of the delta, theta, alpha, sigma and beta power have been associated to each segment. Chi-square test has been applied to evaluate the rate of delta increases preceding the arousal and ANOVA has been applied for testing between subjects and between stages differences. ANOVA showed significant differences between subjects and between stages. Dur- ing the pre-arousal period the band power was mainly stage dependent while the frequency composition of the arousal was more subject dependent. Pre-arousal delta activity was increased respect to background activity in 75% of the cases (p<0.001). All the fast activities (alpha, sigma and beta) increased during the arousals but the ratio between sigma and alpha plus beta decreased during the same interval. The quantitative analysis confirmed the frequent association between the arousals and short burst of preceding delta waves, characterizing sleep stage 2 and 3. This suggests that these patterns could be of a global phenomenon involved in the reaction to a stimulus and connected to the cyclic alternating pattern (CAP). On the contrary, during stage 1 and REM the arousal was characterized by the direct appearance of alpha/theta activities. Beside the high between-subject variability the arousals showed a considerable steadiness within each subject (perhaps partially connected to the pathologic conditions). 172 Actas de Fisiología 7, 2001

SEROTONIN MODULATION OF A HYPERPOLARIZATION-ACTIVATED CURRENT IN SUPRACHIASMATIC NUCLEUS NEURONS

Jennifer A Flett and Christopher Colwell 1606 Barry Ave #1, Los Angeles, CA, USA

Our laboratory is interested in the mechanisms by which serotonin (5-HT) modulates the mammalian circadian system based in the suprachiasmatic nucleus (SCN). Various lines of evidence indicate that some 5-HT effects on the circadian system are mediated in part by modulation of intrinsic membrane currents in SCN neurons. In the current study, we report evidence that 5-HT modulates a hyperpolarization-activated nonselec- tive cation current (Ih), in SCN neurons. Visualized, whole-cell patch clamp recording techniques were used in acutely-prepared hypothalamic slices from young rats (postnatal day 10-16). Under voltage clamp, hyperpolarization of SCN neurons produced an inwardly rectifying current that exhibited a voltage and time dependence consistent with Ih. The current recorded was insen- sitive to barium but sensitive to the bradycardic agent ZD7288, a known blocker of Ih, which reduced the current amplitudes by an average of 33%. This current was observed in the majority of cells recorded in the SCN. Preliminary data indicates that there may also be a day-night difference in Ih current density, with daytime Ih currents having a higher density than night. Notable for its regulation of rhythmic oscillations within the brain, Ih is a widely distributed current that is modulated by cAMP via direct interaction with the cytoplasmic side of Ih channels. Some 5-HT receptors, including those expressed in the SCN, have been shown to be positively coupled to adenylyl cyclase. The biophysical parameters of Ih were therefore evaluated both before and during extracellular application of 5-HT. 5-HT significantly increased the amplitude of Ih from 10-30% within this dosage range, with maximal effects found between 10 and 25microM 5-HT. This increase in current amplitude persisted in the presence of Ba2+, but was attenuated in the presence of ZD7288. We believe that the effects of 5-HT on Ih act in concert with the modulation of other ionic currents to mediate the effect of 5-HT on the membrane properties of SCN neurons. Actas de Fisiología 7, 2001 173

INTRACELLULAR STUDY OF EXCITABILITY CHANGES IN HYPOGLOSSAL MOTONEURONS DURING NATURALLY-OCCURRING ACTIVE SLEEP

Simon J. Fung, Jack Yamuy, MingChu Xi, S.R. Pandi-Perumal, Francisco R. Morales and Michael H. Chase Department of Physiology and Brain Research Institute, University of California at Los Angeles, School of Medicine, Los Angeles, CA 90095, USA

Previously we reported that hypoglossal motoneurons are postsynaptically inhibited during the active sleep-like (AS-carbachol) state induced by the microinjection of carbachol into the nucleus pontis oralis of anesthetized as well as decerebrate cats. Be- cause a true validation of the responsible synaptic mechanisms can be derived only from direct measurements at a cellular level, we studied the excitability changes of hypoglossal motoneurons using intracellular recording techniques during naturally- occurring AS in chronic, intact cats. Accordingly, adult cats were prepared for chronic intracellular recording from hypoglossal motoneurons during naturally-occurring states of sleep and wakefulness (using 2M potassium citrate- and 3M potassium chloride- filled glass microelectrodes). Hypoglossal motoneurons were identified by antidromic stimulation (spike size = 55-90 mV) via a chronically implanted cuff electrode (0.05 ms, 0.8-4 V). Rheobasic currents were assessed by intracellularly injecting a depolarizing pulse (50 ms, 1 Hz) which was sufficient to elicit a direct spike in over 90% of the trials. Motoneurons were less excitable during quiet sleep compared to wakefulness and were least excitable during AS. At the onset of AS, motoneurons became hyperpo- larized and their spontaneous firing rate decreased; during AS these cells ceased discharging except phasically in conjunction with episodes of intense rapid eye movements and/or myoclonic twitches. An increase in the initial segment-somatodendritic delay of the action potential was observed during AS. There was also a marked increase in rheobase for cells recorded during AS (from 2 to 10 nA) compared to those monitored during wakefulness or quiet sleep (from 0.3 to 2 nA). Spontaneous, large- amplitude (approximately 1 mV) inhibitory postsynaptic potentials (IPSPs) were evident during the transition to AS and throughout AS. When chloride ions were injected into hypoglossal motoneurons, the AS-specific IPSPs were reversed in polarity, which indicates that they are chloride-dependent IPSPs. These data indicate that hypoglossal motoneurons are postsynaptically inhibited during naturally-occurring AS, presumably by glycine receptor-mediated synapses that are activated specifically during this state. The present findings provide key data in the chronic cat during naturally-occurring sleep and wakefulness related to the postsynaptic control of hypoglossal motoneurons, which provides a basis for examining their activity vis-à-vis obstructive sleep apnea.

This research was supported by USPHS grants HL60296, MH43362, NS23426, NS09999 and AG04307. 174 Actas de Fisiología 7, 2001

VISUAL NEURONS EXHIBIT FLUCTUATING THETA RHYTHM PHASE- LOCKING THROUGHOUT THE SLEEP-WAKING CYCLE

Gambini J.P., Velluti R.A. & Pedemonte M. Neurofisiología, Departamento de Fisiología. Facultad de Medicina, Universidad de la República. Montevideo, Uruguay.

The hippocampal theta (Hipp θ) rhythm has been associated with movements, attention, auditory processing, autonomic functions as well as learning and memory. It has also been postulated as a temporal organizer in auditory and visual systems, adding a temporal dimension to the processing during behavior (1, 2, 3). The phase-locking of sensory neurons to the θ rhythm depends at least on the behavior, the θ power, or both. The aim of our work is to search for a physiological condition that may produce this temporal correlation. Adult guinea pigs were chronically implanted with bipolar electrodes to record Hipp θ and for standard behavioral control. A craniotomy over the location of the lateral geniculate nucleus (LGn) was made and left open for micropipette penetration. Neuronal firing pattern and control recordings were stored and processed in a computerized system (Spike2 software). The cross-correlation between Hipp θ field activity and spikes was computed by spike-triggered averaging of 30-50 spikes during spontaneous and evoked firing. This analysis was performed during wakefulness (W), slow wave sleep (SWS), and paradoxical sleep (PS). Visual neurons in LGn exhibited a fluctuating temporal correlation with Hipp θ. The phase-locking appears for periods of time, followed by periods of apparent no correlation, to re-appear later related to known or unknown phenomena. The known reasons we found were: 1) changes in flash stimulation, e.g., on, off, or changes in the frequency of the flashes; 2) shifts in the animal behavioral states, e.g., passing from W to SWS or from SWS to PS. These results support the tenet that θ rhythm may act as a time giver related to attention as a consequence of novel situations, e.g., when external sensory inputs show changes. Moreover, it is well known that sensory input resets and enhances θ rhythm, increasing the probability of phase-locking. From this viewpoint, shifts in behavioral states may also be considered novel conditions.

1. Gambini J.P., Pedemonte M & Velluti RA. (1999) WFSRS, 3rd International Congress, Sleep Res Online, 2(Suppl 1): 99. 2. Pedemonte M., Pérez-Perera L., Peña J.L. & Velluti R.A. (2001) Sleep Res OnLine, 4(2): in press. 3. Velluti R.A., Peña J.L. & Pedemonte M. Biol Signals Recept, (2000) 9: 297-308. Actas de Fisiología 7, 2001 175

EXCITATORY NEUROTRANSMITTERS IN THE INFERIOR COLLICULUS DURING SLEEP AND WAKEFULNESS

Natalia Goldstein-Daruech, Marisa Pedemonte, Alejandra Inderkum, and Ricardo A. Velluti Neurofisiología, Departamento de Fisiología. Facultad de Medicina, Universidad de la República. Montevideo, Uruguay

The contribution of N-methyl-D-aspartate (NMDA) receptors to auditory tone-bursts evoked responses in the guinea pig inferior colliculus, was analyzed. Single-unit activity was recorded before and after iontophoretic ejection of a receptor specific antagonist, 2-amino-5-phosphonovaleric acid (AP5), during sleep and waking. On the other and, the same inferior colliculus auditory neuron was studied during both kynuretic acid (Kyn) iontophoresis and auditory cortical stimulation in anesthetized guinea-pigs and during wakefulness (n=37). Two experimental protocols were used: a) “acute” animals under general anesthesia (n=16) and b) “chronic” animals for recordings in physiological conditions, i.e., during the sleep-waking cycle (n=21). Bipolar cortical stimulating electrode was placed in AI region. Double barreled micropipettes were used in both preparations and filled with, a) 3M Sodium acetate and Sky blue and, b) AP5 or Kyn acid. Unitary data obtained in acute animals and during periods of W, SWS and PS were analyzed with selected epoch windows. Peri-stimulus time histo- grams were performed, studying the number of discharge and firing pattern. The AP5 determined a decrease in most of the units recorded, mainly in the later part of the response. A low proportion of them exhibited a firing increase, in wakefulness and sleep phases. It is postulated that the physiological firing increase reported in SWS, could be due to NMDA action enhancement. Kyn acid, a broad-spectrum antagonist of ionotropic excitatory amino acid receptors, was used in comparison to auditory cortex (AI) electrical stimulation in anesthetized animals during wakefulness. Both Kyn iontophoresis and cortical stimulation evoked a decreased firing rate in inferior colliculus units while a low proportion of them increased their discharge, in anesthetized guinea pigs and during wakefulness. The main findings were: 1. The cortico-ICc efferent effects are at least partially mediated by ionotropic excitatory amino acid receptors. 2. Both, Kyn acid and AP5 blockers provoke ICc firing decreases and also discharge increments through direct and indirect circuital actions. 3. Actions of excitatory amino acid on the ICc cells are present during wakefulness and sleep phases, i.e., independently on anesthetic drug. 4. NMDA-blocker, AP5, may affect peripheral incoming auditory information and/or actions of cortical origin, the auditory efferent system. 5. The physiological firing increase reported (1) during SWS may be due to enhanced NMDA action on ICc neurons through ascending and/or descending, efferent fibers.

1. Morales-Cobas, Ferreira, Velluti, J. Sleep Res. 1995

W, wakefulness; SWS, slow wave sleep; PS, paradoxical sleep 176 Actas de Fisiología 7, 2001

TOPOGRAPHICAL CHANGES OF LATE POSITIVE COMPONENTS OF THE EVENT-RELATED POTENTIAL IN THE SLEEP ONSET PERIOD

Tadao Hori, Nao Kataoka, Hiroshi Nittono, and Mitsuo Hayashi Department of Behavioral Sciences, Faculty of Integrated Arts and Sciences, Hiroshima University, Japan

There are two opposing views about the late positive components (P300 and P400) of the event-related potential in the sleep onset period. One is that P400, which develops after the appearance of vertex sharp waves in the electroencephalogram (EEG), has the same characteristics as P300 in the waking state. The other is that P400 is independent of P300 and may reflect a different brain function than that in wakefulness. The purpose of this study was to evaluate the topographical changes of P300 and P400 in the sleep onset period and clarify the nature of P400. Sixteen healthy students (7 men and 9 women, age = 20-26 years) participated in the study. The stimuli were 1000Hz (p = 0.85, Standard) and 2000 Hz (p = 0.15, Tar- get) tones presented through earphones (duration = 50 ms, ISI = 1450 ms, 60 dB). Eight participants were told to make a finger-lift response to target tones (Attend condition). The other eight participants were told to ignore the tones (Ignore condition). The EEG was recorded from 21 scalp sites referenced to the nose tip and was scored in units of 5 s according to Hori’s criteria (1994). ERPs were obtained separately for two different tones, six stages (wake and five EEG stages), and two conditions. P300 appeared only in the Attend condition. The amplitude of P300 decreased in the EEG flattening period (S2) and became almost zero in the theta wave period (S3). In contrast, P400 developed after the appearance of vertex sharp waves in the EEG (S4) and was observed in both Attend and Ignore conditions. The latter result indicates that voluntary attention to target tones does not exert a marked effect on ERPs after S3. Topographical differences were also observed: P300 was dominant at centro-parietal sites, while P400 was dominant at temporo-occipital sites. In sum, P300 and P400 were sensitive to different experimental variables and had different scalp distributions. These results suggest that P300 reflects a cerebral process related to voluntary attention, while P400 reflect a different, sleep-specific process. Actas de Fisiología 7, 2001 177

AROUSAL EFFECT OF OREXIN A DEPENDS ON THE ACTIVATION OF

HISTAMINERGIC NEUROTRANSMISSION VIA H1 RECEPTORS

Zhi-Li Huang1, Wei-Min Qu1, Wei-Dong Li1, Takatoshi Mochizuki1, Naomi Eguchi1, Takeshi Watanabe2, Yoshihiro Urade1,3 and Osamu Hayaishi1 1Dept. of Molec. Behav. Biol., Osaka Biosci. Instit., Osaka, Japan; 2Dept of Molec. Immunol., Med. Instit. of Bioregulation, Kyushu Univ., Fukuoka, Japan; and 3CREST, Japan Sci. and Tech. Corp., Osaka, Japan

Orexin neurons are exclusively localized in the lateral hypothalamic area and project their fibers to the entire central nervous system including histaminergic tuberomammillary nucleus (TMN). Dysfunction of the orexin system results in the sleep disorder narcolepsy, but the role of orexin in physiological sleep-wake regulation and the mechanisms involved remain to be elucidated. Here we provide several lines of evidence that orexin A induces wakefulness via the TMN and histamine H1 receptor (H1R). Perfusion of orexin A (5 and 25 pmol/min) for 1 hr into the TMN of rats through a microdialysis probe promptly increased wakefulness for 2 hr after starting the perfusion by 2.5- and 4-fold, respectively, concomitant with a reduction in rapid eye movement (REM) and non-REM sleep. Microdialysis studies showed that application of orexin A to the TMN increased histamine release from both the medial preoptic area and the frontal cortex by approximately 2- fold over the baseline for 80 to 160 min in a dose-dependent manner. Furthermore, infusion of orexin A (1.5 pmol/min) for 6 hr into the lateral ventricle of mice produced a significant increase in wakefulness during the 8 hr after starting infusion to the same level as the wakefulness observed during the active period in wild type mice, but not at all in H1R gene knockout mice. These findings strongly indicate that the arousal effect of orexin A depends on the activation of histaminergic neurotransmission mediated by H1R. 178 Actas de Fisiología 7, 2001

EFFECT OF SELF-AWAKENING ON EEG DURING A SHORT NAP

Kosuke Kaida, Hiroshi Nittono, Mitsuo Hayashi and Tadao Hori Department of Behavioral Sciences, Faculty of Integrated Arts and Sciences, Hiroshima University, Japan

Introduction: Self-awakening (SA) is defined as waking up on time when we are going to wake up. The time is decided before sleeping. It has been controvertial whether SA changes the sleep structure of nocturnal sleep. However, it has not been examined as for diurnal nap. We found that self-awakening from a short nap can reduce sleep inertia (Kaida et al. 2001). In the present study, we investigated the effect of self-awakening on the sleep structure in diurnal nap. Methods: Participants were 10 healthy volunteers (4 men and 6 women, 21-23 years old, mean 21.7 years). They took a nap for around 15 min in SA condition or for 20 min in Forced-awakening (FA) condition from lights off at 14:00. Each participant took part in both conditions. The order of conditions was counterbalanced across participants. According to Hori et al. (1994), the hypnagogic EEGs (C3) was classified into 9 stages every 5 s: 1) alpha wave train, 2) alpha wave intermittent (>50%), 3) alpha wave intermittent (< 50 %), 4) EEG flattening, 5) ripples, 6) vertex sharp wave solitary, 7) vertex sharp wave bursts, 8) vertex sharp waves and incomplete spindles, and 9) spindles. Results and Discussion: Participants slept for 13.8(±1.18) min in SA condition and 18.3 (±0.61) min in FA condition after sleep stage 1. Fig. 1 shows the frequency of appearance of EEG stages 7+8+9 for 10 minutes before awakening from the nap. In SA condition, the frequency of these stages were lower than those in FA condition (χ2 = 110.97, df = 9, p < .001).

This result suggests that SA 0.8 Self-awakening suppressed the progress of 0.7 Forced-awakening the EEG stages. It is known that sleep inertia appear 0.6 strongly when one wakes 0.5 up from deep sleep. The 0.4 present results are consis- 0.3 tent with our previous re- 0.2 sults. The effect of SA on sleep inertia would be due Frequency of appearance 0.1 to continual light sleep just -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 before awakening. Fig 1. Time course of EEG stages 7+8+9 during a nap from 10 min to just before awakening. Actas de Fisiología 7, 2001 179

LONG LASTING EXCITATION BY OREXIN ON THE CHOLINERGIC NEURONS IN THE LATERODORSAL TEGMENTAL NUCLEUS

Kazumi Takahashi(1), Yoshimasa Koyama(1), Yukuhiko Kayama(1) and Mitsuaki- Yamamoto(2) 1Department of Physiology, Fukushima Medical University, School of Medicine, 2Laboratory of Neurophysiology and Bioinformatics, Graduate School of Information Sciences, Tohoku Univ., Japan

Orexin is proposed to be involved in the control of arousal state and deficit in orexinergic neural system causes severe attack of nalcorepsy. Orexin neurons are located in the lateral hypothalamus and send diffuse projections throughout the brain including the brainstem nuclei involved in the regulation of sleep-waking cycles. Cholinergic neurons in the laterodorsal tegmental nucleus (LDT), one of the main targets of the orexin neurons, have crucial roles in the regulation of waking and paradoxilcal sleep. The present study was performed to examine effects of orexin on the LDT cholinergic neurons. In urethane-anesthetized rats, single neuronal activity was recorded extracellularly in the LDT through a glass pipette microelectrode. Cholinergic neurons in the LDT can be distinguished from non-cholinergic ones by the width of their action potentials. Or- exin-A (100 M), dissolved in an artificial cerebrospinal fluid and filled in a multi-barrel glass pipette glued to the recording electrode, was applied by pressure to the close vicinity of the recording neurons. Glutamate (1 mM) or desmopressin (a vasopressin analogue, 85 ƒÊM) was applied for comparison. Orexin caused an increase in firing rate in some of the cholinergic neurons. The excitation lasted dose-dependently for 80 to 280 sec, though firing rate during this long lasting excitation did not exceed 150 % of the control level. The orexin induced excitation was quite different in fashion from those of other neurotransmitters; glutamate evoked intensive and prompt excitation which disappeared soon sfter cessation of application, and excitation by the vasopressin analogue lasted longer but disappeared within several seconds. The present results suggest that orexin-A acts as a long lasting transmitter or modulator on the cholinergic neurons in the LDT to participate in the regulation of sleep and wakefulness. 180 Actas de Fisiología 7, 2001

LOCAL GABA-ERGIC MECHANISMS ARE A MAJOR DETERMINANT OF THE LEVEL OF ACTIVITY IN HYPOTHALAMIC OREXIN-CONTAINING NEURONS

Vitaliy Marchenko1,2, Graziella Mann1, Victor Fenik1,2, Richard Ross2,3,4, Adrian Morrison1,2,3 and Leszek Kubin1,2 1Department of Animal Biology, 2Center for Sleep and Respiratory Neurobiology, 3Department of Psychiatry, University of Pennsylvania and 4Veterans Affairs Medical Center, Philadelphia, PA 19104, USA

Hypothalamic orexins (ORX)-containing neurons emerged recently as a major neuronal group mediating signs of arousal in the CNS (reviewed by Willie et al., Annu. Rev. Neurosci. 2001, 24:429-458). Consequently, the effects of ORX at many CNS sites have been investigated, but the mechanisms controlling the level of activity in ORX neurons remain largely unknown. We investigated the effects of lateral hypothalamic injections of various relevant agonists and antagonists on cortical EEG, motor and cardiorespiratory activity in urethane-anesthetized, vagotomized rats, and on the sleep-wake behavior in chronically-instrumented, behaving rats. Under anesthesia, unilateral injections of quinpirole (dopaminergic D2 receptor agonist; 1 mM, 50 nl; n=1), eticlopride (D2 receptor antagonist; 1-5 mM, 50 nl; n=4), or clonidine alpha2- adrenergic agonist; 0.75 mM, 50 nl; n=3) had no effect on the EEG, blood pressure and respiratory motor output. Similarly, neither serotonin (5 mM, 50 nl; n=1) nor SKF- 38393 (D1/D5 agonist; 1 mM, 50 nl; n=1) produced any physiologic changes nor did their microinjections result in c-fos expression in ORX-containing neurons. In contrast, bicuculline, a GABAA receptor antagonist (1-4 mM, 10-50 nl, n=5), produced a long-lasting (at least 40 min) activation of cortical EEG and an increase of both the respiratory rate (from 51+/-6(SD) 1/min to 64+/-14 1/min; p<0.05) and blood pressure (by 36+/-19 mmHg; p<0.02). Forty min following bicuculline, all ORX cells present at the injection site expressed c-fos. In three chronic rats, unilateral bicuculline microinjections (1 mM, 200 ul) tended to reduce the amount of REM sleep, whereas muscimol, a GABAA receptor agonist (1 mM, 200 ul), elicited in two rats a slow-wave cortical activity during wakefulness similar to that in slow-wave sleep. Thus, GABA- ergic mechanisms are important in local control of ORX cell activity, whereas the role of local dopaminergic and serotonergic mechanisms appears to be small.

[Supported by: SCOR HL-60287, Sub. 02, HL-47600, MH-42903 and Dept. of Veterans Affairs.] Actas de Fisiología 7, 2001 181

DOPAMINE EFFECT ON MUSCLE ACTIVITY IS MEDIATED THROUGH LOCUS COERULEUS

Yuan-Yang Lai Lai, Tohru Kodama, John H Peever and Jerome Siegel Neurobiol Res (151A3), VAGLAHS Sepulveda, North Hills CA, USA

The ventral mesopontine junction (VMPJ) occupies the area of caudal portion of the ventral tegmental area (VTA), retrorubral nucleus (RRN) and mesencephalic reticular formation in the midbrain, as well as the rostroventral part of the paralemniscal tegmental field in the pons. Dopaminergic neurons are found in the VTA and RRN. Our previous studies found that electrical stimulation in the VMPJ produces muscle tone suppression, whereas lesion to it generates muscle hyperactivity in the decerebrate cat. An increase in phasic (rhythmic) muscle activity, stepping-like activity and twitches, is the major symptom induced by VMPJ lesion. This VMPJ lesion-induced muscle hyperactivity can be attenuated or blocked by nonNMDA agonist and NMDA antagonist injection into the nucleus magnocellularis, which receives axonal projection from the VMPJ, in the medulla. L-dopa, a precursor of dopamine, given intrathecally facilitates motoneuron activity in the decerebrate cat. In contrast, our study found that dopamine release in the motor nuclei is unchanged during brainstem stimulation-induced atonia. We hypothesize that the VMPJ dopaminergic effect on muscle activity may be mediated through the locus coeruleus (LC) because 1) electrical stimulation in the LC increases spinal motoneuron activity (Lai et al., 1989) and 2) dopamine agonists inhibit LC neuron activity (Seutin et al., 1990). Five young adult cats were decerebrated under halothane anesthesia. Splenium and occipitoscapularis of the neck muscles were dissected for EMG recording. Train (500 ms with 0.2 ms, 100 Hz and 10-40 ƒÝA rectangular cathodal pulses) electrical stimulation was applied into the LC through a stainless steel monopolar electrode (A-M Systems). The stimulation in the LC generated muscle tone facilitation, bilaterally. Then, 0.5 ƒÝl of dopamine agonist was injected into the LC. We found that 4 ƒÝg/ƒÝl of quinpirole (Tocris, St. Louis), a D2 agonist, injected into the LC elicited a reduction of neck muscle activity. The amplitude of muscle tone suppression ranged from 12% to 67% with the average of 35% of the baseline level. In contrast, 8 ƒÝg/ƒÝl of SKF38393 (Tocris), a D1 agonist, injected into the LC did not change muscle activity. Our present study finds that electrical stimulation in the LC elicited an increase of muscle activity. We also find dopamine D2 agonist microinjected into the LC induces a reduction of muscle activity. We suggest that the dopaminergic effect on muscle activity may be mediated through the LC. We hypothesize that activation of VMPJ dopaminergic neuronal activity inhibits LC and disfacilitates motoneuron activity resulting in muscle hypoactivity during sleep.

References: Lai YY, HK Strahlendorf, SJ Fung and CD Barnes (1989) The actions of two monoamines on spinal motoneurons from stimulation of the locus coeruleus in the cat. Brain Res 484:268-272. Seutin V, Seuvee-Moreau J, Giesbers I, Massotte L and Dresse A (1990) Effect of BHT 920 on monoaminergic neurons of the rat brain: an electrophysiological in vivo and in vitro study. Naunyn-Schmiedebergs Arch Pharmacol 242:502-507.

Acknowledgment: This work supported by HL41370. 182 Actas de Fisiología 7, 2001

DATA UNIFORMIZATION IN THE EEG/SLEEP LABORATORY

Antónia Lomba, Isabel Henriques, Madalena Teles-Araújo and Teresa Paiva Lab. EEG/Sleep – Centro de Estudos Egas Moniz,Lisbon Medical Faculty, Portugal

Databases and electronic patient records are essential tools in Sleep Clinics. The de- mands of modern databases are very sophisticated since they should deal with 3 main type of problems: data integration and uniformization; multimedia content; and intelligent environments. Data integration and uniformization are essential whenever long- standing archives should deal with the constant software innovation. This work represents a preliminary attempt in data uniformization from different software packages into a standard solution, in order to prepare future applications of multimedia distributes environments. Since 1988 the Lab. EEG/Sleep of Lisbon Medical Faculty uses computer databases to store patients information, sleep questionnaires and EEG/Sleep neurophysiological data. About 8,700 records were stored in 3 different databases: StatView for Apple, MSDOS-Knosys and Windows-FoxPro, using not-related tables and not-validated string fields. The integrity rules were not respected allowing records duplication and other redundancies. A consistent DB, flexible enough to fit clinical and research needs was developed, applying security and validation restrictions. DBMS and sub-languages used are Ac- cess2000, VBA and SQL. Related tables were created: Patient Identification, Sleep Questionnaire, SF36 Questionnaire (quality of life), Economical Evaluation Question- naire (healthcare costs), Routine and Sleep EEG tests, PSG tests, MSLT tests, Epilepsy Monitoring, CPAPi Control Tests. Data types and fields properties were defined and validated avoiding redundancies. A user-friendly VBA interface was implemented enabling management, retrieval and input/output of data. SQL were used to select, insert/ edit, index, validate filtered dynasets based on defined conditions, guaranty consis- tence and integrity. Reports were standardized according to each kind of EEG/Sleep test. The menu bar was personalized hiding programming tools. Links to Word and Excel documents normally used in the routine of the Laboratory, were created. Data stored in anterior DB’s were imported to this DB after formatted and validated. Redundancies and inconsistencies were eliminated. Data tables are automatically exported to Excel sheets linked to statistic software’s (SPSS and StatView). DB is daily updated with new records in order to test performance and behaviour. In this moment we have around 7,000 patient records, 2,000 sleep questionnaires records, 50 SF36 questionnaires, 50 economical evaluation questionnaires, 6,000 neurophysiological records ready to be used by Doctors, Managers and other researchers of this laboratory. Staff satisfaction is considered high and errors have been sporadic. Future plans include the inclusion of multimedia files and adaptation to intelligent environments. Actas de Fisiología 7, 2001 183

ALPHA-2 ADRENERGIC RECEPTORS ON GABAERGIC BASAL FOREBRAIN NEURONS THAT DISCHARGE MAXIMALLY DURING CORTICAL SLOW WAVE ACTIVITY

Ian D Manns, Y P Hou and Barbara Jones Montreal Neurological Institute, McGill University, Montreal, Canada

The basal forebrain (BF) plays an important role in modulating cortical activity across the sleep-waking cycle. In addition to cholinergic neurons that stimulate cortical activation, it contains GABAergic neurons that could play different roles. In fact, we recently found that in contrast to cholinergic neurons, many GABAergic neurons decrease their firing in association with cortical activation and fire instead in association with cortical slow wave activity in urethane anesthetized rats (1). We proposed that such GABAergic cells could be sleep-promoting neurons. Sleep-active neurons have been recorded in the BF and also in the adjacent preoptic area, where they were inhibited by either stimulation of the locus coeruleus or iontophoretic application of noradrenaline (2-4). In vitro studies have identified non-cholinergic neurons in the BF and GABAergic neurons in the ventrolateral preoptic nucleus that are commonly hyperpo- larized and inhibited by NA (5, 6). The inhibitory effects of NA are mediated by the alpha-2 adrenergic receptor (7). In order to determine whether GABAergic, potentially sleep-promoting, neurons in the BF bear alpha-2 receptors, we first applied dual- immunostaining with fluorescent labels for glutamic acid decarboxylase (GAD, using a green fluorescent label) and the alpha-2a adrenergic receptor (using a red fluorescent label) in the rat brain. The majority of GAD+ neurons (>65%) within the BF magnocellular preoptic nucleus (MCPO) were found to be immunopositive for the alpha-2 receptor. We then examined whether the GABAergic neurons that fire maximally in association with cortical slow wave activity were immunopositive for the alpha-2 adrenergic receptor. Cells were recorded in the MCPO in association with EEG activity from limbic cortical regions and their response to somatosensory stimulation determined in association with the resulting cortical activation in urethane-anesthetized rats. The cells were labeled with neurobiotin using the juxtacellular technique and subsequently triple stained for neurobiotin (with a blue fluorescent label), GAD (with a green fluorescent label) and the alpha-2a adrenergic receptor (with a red fluorescent label). The majority of the GABAergic neurons that fired maximally in association with slow irregular cortical activity and decreased or ceased firing with cortical activation were found to be immunopositive for the alpha-2 adrenergic receptor. The results suggest that noradrenergic input from locus coeruleus neurons impinging upon alpha-2 adrenergic receptors of these neurons may be a mechanism for inhibiting their discharge during waking. Accordingly, these GABAergic BF neurons would be released from inhibition during slow wave sleep when they would discharge maximally and potentially actively dampen cortical activation and promote cortical slow wave activity. (Sup- ported by the Canadian MRC.)

1. Manns, I. D., Alonso, A. & Jones, B. E. (2000) J. Neurosci. 20, 9252-9263. 2. Szymusiak, R. & McGinty, D. (1986) Brain Res. 370, 82-92. 3. Osaka, T. & Matsumura, H. (1994) Neuro- sci. Res. 19, 39-50. 4. Osaka, T. & Matsumura, H. (1995) Neurosci. Res. 21, 323-330. 5. Fort, P., Khateb, A., Serafin, M., Muhlethaler, M. & Jones, B. E. (1998) NeuroReport 9, 1-5. 6. Gallopin, T., Fort, P., Eggermann, E., Caull, B., Luppi, P.-H., Rossier, J., Audinat, E., Muhletha- ler, M. & Serafin, M. (2000) Nature 404, 992-995. 7. Bai, D. & Renaud, L. P. (1998) Neuro- science 83, 905-16 184 Actas de Fisiología 7, 2001

EXPRESSION OF ALPHA-1 AND ALPHA-2 ADRENERGIC RECEPTORS ON THE NORADRENERGIC AND CHOLINERGIC NEURONS OF THE PON- TOMESENCEPHALIC TEGMENTUM

Y.P.Hou, I.D.Manns, and B.E.Jones Montreal Neurological Inst., McGill University, Montreal, QC H3A 2B4, Canada

The noradrenergic neurons of the locus coeruleus (LC) discharge tonically during waking, diminish firing during slow wave sleep (SWS) and cease firing during paradoxical sleep (PS) (1). Groups of pontomesencephalic cholinergic neurons have been reported to discharge at high rates during both waking and PS or only during PS (2). Previous electrophysiological studies performed in vitro reported an alpha-2 adrenergic mediated inhibitory action of noradrenaline (NA) on the cholinergic neurons that could underlie a reciprocal activity profile of the cholinergic and noradrenergic neurons during waking and PS (3). Yet, a depolarizing alpha-1 mediated action of NA has also been reported that would suggest the two cell groups could be active together during waking (4). The present study was undertaken to examine by immunohistochemistry whether cholinergic neurons possess alpha-2 or alpha-1 adrenergic receptors. Using antibodies directed against the receptor proteins, it was first established that the vast majority of tyrosine hydroxylase (TH) immunoreactive neurons of the locus coeruleus nucleus were immunopositive for the alpha-2a adrenergic receptor antibody, while none were immunopositive for the alpha-1a adrenergic receptor antibody. The majority of cholinergic neurons in the laterodorsal tegmental nucleus (LDT) and pedunculopontine tegmental nucleus (PPT) were found to be immunopositive for the alpha-2 adrenergic receptor, but a minority were definitely immunopositive for the alpha-1 adrenergic receptor. The lack of alpha-1 and presence of alpha-2 receptors on LC neurons corroborates earlier data indicating that these neurons have inhibitory alpha-2 autoreceptors. The appearance of alpha-2 receptors on the majority of the cholinergic neurons supports work demonstrating the hyperpolarizing effect mediated by alpha-2 receptors on the cholinergic cells. Such neurons would be expected to be inhibited by NA during waking and thus only active during PS. The presence of alpha-1 receptors on a minority of the cholinergic neurons confirms other electrophysiological evidence indicating that a proportion of the cells are depolarized and excited by noradrenaline. These neurons could thus be active during waking when stimulated by noradrenergic input and also during PS when stimulated by other extrinsic or intrinsic influences. The results from this study thus suggest that there may indeed be two subgroups of cholinergic pontomesencephalic neurons, one which is active during both waking and PS and another which may be active solely during PS.

Supported by the Canadian MRC.

1. Hobson, J. A., McCarley, R. W. & Wyzinski, P. W. (1975) Science 189, 55-58. 2. El Mansari, M., Sakai, M. & Jouvet, M. (1989) Exp. Brain Res. 76, 519-529. 3. Williams, J. A. & Reiner, P. B. (1993) J. Neurosci. 13, 3878-3883. 4. Muhlethaler, M., Khateb, A. & Serafin, M. (1990) in The Diencephalon and Sleep, eds. Mancia, M. & Marini, G. (Raven Press, New York), pp. 31-48. Actas de Fisiología 7, 2001 185

GLYCINERGIC MECHANISMS AT THE HYPOGLOSSAL MOTOR NUCLEUS SUPPRESS GENIOGLOSSUS MUSCLE TONE AND ACTIVITY IN RESPONSE TO HYPERCAPNIA

Janna L Morrison, Xia Liu, Sandeep Sood, Hattie Liu and Richard Horner University of Toronto, Toronto, Ontario M5S 1A8, Canada

Introduction: Sleep is associated with suppression of muscle tone that is most pronounced in rapid-eye-movement (REM) sleep. For pharyngeal muscles, such as the genioglossus (GG) muscle of the tongue, this suppression can lead to airway narrowing and obstructive apneas. Obstructive apneas are most common in REM sleep but the neural mechanisms mediating GG suppression in REM sleep are unknown. It is known, however, that postural lumbar motoneurons receive glycinergic inhibitory inputs in REM sleep (1) but evidence is conflicting for hypoglossal motoneurons innervating GG muscle (2,3). Nevertheless, studies in neonatal tissue slices demonstrate functional glycinergic inhibition of individual hypoglossal motoneurons in-vitro (4). However, whether glycinergic inhibition modulates total hypoglossal motor outflow to GG muscle has not been fully characterized in-vivo, nor has the interaction of this putative pathway with the respiratory control system been determined. Hypotheses: This study tests the hypotheses that increasing glycine at the hypoglossal motor nucleus will (i) produce graded suppression of GG muscle tone and (ii) selectively reduce GG responses to hypercapnia. Methods: Eight urethane-anesthetized adult male rats were studied. The rats were tracheotomized and vagotomized to avoid the potential confounding influences of changes in upper airway resistance and breathing pattern on GG activity. We recorded elec- trocortical activity, GG and diaphragm muscle activities and blood pressure. Microdi- alysis probes (1 mm length, 240 m diameter) were implanted into the hypoglossal motor nucleus for delivery of glycine (6 doses, 0.0001-10 mM) or vehicle alone (artificial cerebrospinal fluid, control). Respiratory muscle responses to 7% inspired CO2 were determined at each dose of glycine. Results: Increasing glycine at the hypoglossal motor nucleus produced graded suppression of GG activity (F6, 36=13.677; p<0.0001) with ~80% decrease at 10 mM glycine. CO2 stimulated GG activity across all doses of glycine (F2,12=11.08, p=0.002) although the absolute magnitude of GG activity in the presence of CO2 was progressively reduced to ~20% of control by increasing glycinergic inhibition. These respiratory effects were selective for GG muscle because neither diaphragm amplitude (F6,36=0.914, p=0.496) nor diaphragm response to CO2 were affected by glycine (F12,72=1.14, p=0.343). Conclusions: Increasing glycine at the hypoglossal nucleus causes progressive decreases in GG muscle tone and activity in response to hypercapnia. Whether this inhibitory pathway plays a role in suppression of GG activity in REM sleep remains to be determined in naturally sleeping animals.

(1) Chase MH, et al. J Neuroscience 9: 743-751, 1989. (2) Kubin L, et al. Brain Res 611: 300-312, 1993. (3) Yamuy J, et al. Neuroscience 94: 11-15, 1999. (4) O’Brien JA, Berger AJ. J Neurophysiol 82: 1638-1641, 1999. 186 Actas de Fisiología 7, 2001

SYNAPTIC RESPONSES AND NEUROTRANSMITTERS INVOLVED IN THE ACTIVATION OF NEURONS IN THE ORAL PONTINE RETICULAR NUCLEUS

Angel Nuñez, Margarita Rodrigo-Angulo, Isabel de Andrés and Fernando Reinoso-Suárez Departamento de Morfología, Facultad de Medicina, Universidad Autónoma de Madrid. Spain

The ventral part of the oral pontine nucleus (vRPO) has been proposed as the REM sleep induction site. The analysis of the synaptic interactions of vRPO neurons with other structures related to REM sleep is an important step to know the mechanism of REM sleep generation. The purpose of the present study was to determine the firing activity of morphologically characterized vRPO neurons as well as their synaptic response properties elicit by electrical stimulation of the penduculopontine tegmental nucleus (PPT) and the contralateral vRPO. Unit recordings were performed in 31 cats anaesthetized with urethane, provide with recording macroelectrodes in the hippocampus and cerebral cortex and stimulating electrodes in the PPT and contralateral vRPO. Three-barrel glass micropipettes were used for unit recordings and simultaneously applying pharmacological agents by microiontophoresis. After completion of the experimental protocol, biocytin was ejected extracellularly. In spontaneous conditions two types of vRPO neurons were identified according to their discharge pattern. Type I neurons (77%) were characterized by a non-rhythmic spontaneous firing at 7.6±1.3 Hz. Type II neurons (23%) displayed single spikes at 12.9±2 Hz and a prominent rhythmic activity at 7-22 Hz. Electrical stimulation of the contralateral vRPO elicited responses in type I and II neurons (57% and 31%, respectively) with a mean latency of 6.5±3 ms and 4.8±1.2 ms for type I and type II neurons, respectively. This effect was blocked by the antagonist of the non-NMDA glutamatergic receptors (CNQX). Electrical stimulation of the ipsilateral PPT evoked orthodromic responses in type I neurons (41%) at 7±0.7 ms of latency. However, PPT electrical stimulation induced an inhibition of the firing rate of all type II neurons during 50-100 ms. Both effects were blocked by the muscarinic antagonist atropine and unaffected by CNQX. Results suggest that vRPO neurons are, at least, partially glutamatergic and also receive an important cholinergic input from PPT. These results indicate that for generation of REM sleep an assembly of neurotransmitters may concur on the vRPO. Cholinergic and glutamatergic vRPO afferents have a main role in activation of vRPO neurons.

Supported by Grant DGICYT PM98-0005 and PB98-0097 Actas de Fisiología 7, 2001 187

AUDITORY CORTEX UNITARY ACTIVITY CORRELATED TO SLEEP- WAKEFULNESS AND THETA RHYTHM IN RESPONSE TO NATURAL SOUNDS

Lucía Pérez-Perera, Claudia Bentancor, Marisa Pedemonte, Ricardo A. Velluti. Neurofisiología, Departamento de Fisiología. Facultad de Medicina, Universidad de la República. Montevideo, Uruguay.

Auditory neurons are functionally influenced by other inputs apart from the specific sound stimuli, e.g., behavioral states and hippocampal theta rhythm (Hipp θ). Some auditory neurons change the characteristics of their discharges according to behavior: wakefulness (W), slow wave (SWS) and paradoxical sleep (PS). It was reported(1) that about half (57.8%) of the AI cortex neurons did not exhibit changes in their evoked firing while asleep in comparison with W. However, some units presented increments (18.3%) and decrements (23.9%) in their discharges on passing from W to SWS. When passing from SWS to PS the percentages were similar. A correlation between neuronal discharges and the Hipp q (phase-locking) was also described (2). Our aim was to analyze the temporal correlation between cortical auditory unit’s firing and Hipp q rhythm during different behavioral states when neurons are stimulated with a possibly significant natural sound, the guinea pig vocalization (whistle). Adult guinea pigs were chronically implanted with electrodes for behavioral control. Two light metal bars attached to the skull permitted a painless stereotaxic position reconstruction during the recording sessions. Hipp θ was recorded through bipolar macroelectrode placed at CA1 hippocampal region. A small craniotomy was made, over the stereotaxic parameters of the primary auditory cortex AI region, for microelectrodes penetration. Glass micropipettes, filled with Sodium acetate and Sky blue, were advanced perpendicularly to the cortical surface searching for sound evoked unitary discharges. Acoustic stimuli consisted of previously taped guinea pig vocalizations (whistle of 750 ms, “cuick”). The recording site was iontophoretically marked with Sky-blue. The “cuick”, acting as a possible more significant stimulus, showed in some cases neurons responding with a low discharge number. In relation to the Hipp θ rhythm, some neurons that responded to the “cuick” also exhibited temporal correlation, phase- locking with Hipp θ rhythm. This temporal correlation was dynamic, being present or not -in the order of seconds- depending on known and unknown factors. The known factors were: attention shifts and changes in the characteristics of the stimuli or the behavioral state. - According to the unitary activity, the processing of auditory information is partially different in W and sleep. No neuron stopped its firing during sleep. - The hippocampal theta rhythm is postulated as a time giver for the information processing, i.e., the auditory cortical unitary discharge. - The unitary cortical activity evoked by a natural sound during W also shows phase- locking with the Hipp θ rhythm.

(1) Peña, Pérez-Perera, Bouvier, Velluti. Brain Research, 816: 463-470, 1999. (2) Pedemonte, Pérez-Perera, Peña, Velluti. Sleep Res. Online, 4: 51-57, 2001 188 Actas de Fisiología 7, 2001

ELECTROENCEPHALOGRAPHIC EFFECTS OF VIBRISSAL AFFERENT BLOCKAGE

Abigail Prchal and Emilio E. Décima Cátedra de Neurociencia Universidad Nacional de Tucumán and Dept. Fisiología y Neurociencia, CONICET, S. M. de Tucumán, Argentina

We have shown previously that bilateral deafferentation of the vibrissae produces depressive effects in behavior. We thought that these results could be related with the sleep deafferentation theory. Thus, the goal of the present work was to study wether these effects of vibrissal pad anaesthesia are accompained with electrocorticographic (ECoG) changes. Adult wistar male rats, were chronically implanted for ECoG and EMG recording under general anaesthesia. A week after surgery reccording were made under normal conditions for at least three days. Vibrissal afferents were then, blocked by subcutane- ous injection of 2 cc of Procaine (0.5%) in vibrissal pad under light ether anesthesia. On a different day reccording were also made after the same light eter anaesthesia but whithout whisker blockage. Our results show that this acute deaferentation produces an important EEG synchronization whithout any behavioral sign of sleep. This effects was never observed after light eter anaesthesia alone. These results support the afferent sleep hypothesis in showing the importance of afferent activity for maintain the awake and alert state. They also indicate the importance of reversible approaches for the assesment of Nervous System impairments.

Actas de Fisiología 7, 2001 189

MORPHOLOGIC ANALYSIS OF JUXTACELLULARLY STAINED NEURONS OF THE ORAL PONTINE RETICULAR NUCLEUS

Margarita L Rodrigo-Angulo, Angel Nuñez, Isabel de Andrés and Fernando Reinoso-Súarez Dept. Morfología, Fac. Medicina, UAM; c/Arzobispo Morcillo 4, Madrid, Spain

The ventral part of the oral pontine nucleus (vRPO) has been proposed as the REM sleep induction site. The purpose of the present study was to determine the morphology of electrophysiologically identified vRPO neurons. Unit recordings were performed in 31 cats anaesthetized with urethane through a barrel glass micropipette filled with a solution of NaCl (0.5 M) plus 2% biocytin; after completion of the experimental protocol (see Nuñez et al., companying abstract), biocytin was ejected extracellularly. After two hours survival time, animals were perfused transcardially with 4% paraformalde- hyde and 0.1% glutaraldehyde in 0.1M phosphate buffer, pH 7.4 (PB) followed by increasing concentrations of saccharose in PB. Brains were frozen sectioned at 60 mm and sections processed immunocytochemically using avidin-biotin-peroxidase reagent (Elite Kit, Vector) in PB, in order to reveal the presence of biocytin stained neurons, which were reconstructed under a microscope provided by camara lucida. Two types of neurons have been described in vRPO according to their firing pattern (see Nuñez et al., accompanying abstract). From morphological analysis, neurons physiologically defined as type I, usually displayed ellipsoidal somata measuring 40 X 20 mm from which emerged 4-6 thick dendritic trunks which arborizate in two or three branches, extending dorsally and ventrally often in obliquae approach in the sagital plane. Nei- ther dendritic spiny nor varicosities could be observed. Axons could be identified arising from the soma or the initial dendritic segment but they could not be clearly followed. No local axonal collateralization was observed. Type II neurons displayed a polygonal somata measuring 36 x 23 mm with dendrites emerging radially from the soma reaching shorter extension than those of type I neurons. Axons could be identified emerging from the soma and followed in a short distance where they slightly arborizate into the field of the dendritic tree. The different pattern of axonal arborisation close to the soma in type II neurons may have a special relevance for synchronization of the activity of neighbouring neurons or activation of feedback inhibitory pathways.

Supported by Grant DGICYT PM98-0005 and PB98-0097 190 Actas de Fisiología 7, 2001

MUSCIMOL INACTIVATION OF THE CENTRAL NUCLEUS OF THE AMYGDALA: EFFECTS ON SLEEP

Larry D. Sanford and Xiangdong Tang Sleep Research Laboratory, Department of Pathology and Anatomy, Eastern Virginia Medical School, Norfolk, USA

It is becoming established that the amygdala has a strong influence on arousal state, particularly rapid eye movement sleep (REM). Electrically activating the central nucleus of the amygdala (CNA) can increase subsequent REM and enhance REM-related phenomena. However, drugs that may be inhibitory to CNA typically have been reported to reduce REM. To test the hypothesis that suppressing activity in CNA could inhibit

REM, we reversibly inactivated CNA using the GABAA agonist, muscimol, and examined the effects on sleep and wakefulness. Rats (90 day old male Sprague-Dawley; n=12) were implanted with electrodes for recording EEG and EMG. Guide cannulae (26 ga.) were implanted with their tips aimed 1.0 mm above CNA (A 6.3, ML ± 4.0, DV 7.0) for microinjecting muscimol (0.001, 0.3 and 1.0 mM/ 0.2 ml saline; 0.1 ml / min). Each animal received bilateral microinjections of muscimol or saline alone in a counterbalanced order followed by six h sleep recordings. Sleep was scored from computerized records in 10 s epochs. Totals, number of episodes and durations were determined for REM, NREM and wakefulness. All dosages of muscimol reduced total REM, number of REM episodes and REM percentage compared to baseline (Table 1). NREM was minimally affected. Measures of NREM were not affected at the low dosage and reductions in total NREM at the medium and high dosages did not reach significance. Total sleep and sleep efficiency were significantly reduced at the medium and dosages, but not at the low dosage.

Table 1. Baseline 0.001 µM 0.3 µM 1.0 µM Total REM 36.7 ± 3.9 14.2 ± 3.7** 12.7 ± 2.8** 16.2 ± 2.3** REM Episodes 17.6 ± 1.4 7.5 ± 1.6** 7.4 ± 1.4** 9.6 ± 1.5* REM Percent 13.6 ± 1.5 5.7 ± 1.4** 5.9 ± 1.5** 6.7 ± 0.8** Total NREM 235.0 ± 7.2 236.7 ± 8.3 207.9 ± 10.1 220.4 ± 4.6 Total Sleep 271.7 ± 6.3 250.9 ± 7.9 220.5 ± 8.9** 236.5 ± 6.0* Sleep Efficiency 75.5 ± 1.8 69.7 ± 2.2 61.3 ± 2.5** 65.7 ± 1.7* * = p < .01; ** = P < .001

The results demonstrate that inactivating CNA with low concentrations of muscimol can produce a selective suppression of REM with minimal effects on NREM and overall sleep. Together, with evidence suggesting that enhancing activity in CNA could promote REM, these findings raise the question of whether spontaneous activity in CNA could facilitate REM initiation and/or maintenance.

This work was supported by NIH grants NS36694, NS35281 and MH61716. Actas de Fisiología 7, 2001 191

GABAERGIC AND ADRENERGIC SUPPRESSION OF REM IN N. SUBCOERULEUS (SubC) AND RETICULARIS PONTIS CAUDALIS (RPC)

Larry D. Sanford1, Xiangdong Tang1, Richard J. Ross2 and Adrian R. Morrison2 Sleep Research Laboratory1, Eastern Virginia Medical School, USA and Laboratory for Research on the Brain in Sleep2, University of Pennsylvania, USA

The SubC and RPC are implicated in the generation of REM. In addition, electrolytic lesions involving SubC and RPC can eliminate the atonia of REM and produce REM without atonia (REM-A) in cats and rats. In order to assess the role of cell bodies in these regions in the regulation of REM and atonia, we microinjected the GABAA agonist, muscimol (MUS), and the alpha-2 adrenoceptor agonist, clonidine (CLON), into SubC and RPC and examined sleep. Rats (90 day old male Sprague-Dawley) were implanted with electrodes for recording EEG and EMG. Guide cannulae (26 ga.) were implanted aimed into SubC or RPC for microinjecting MUS (1.0 µM) and CLON (0.2; 1.0 µM). Each animal received bilateral microinjections of MUS, CLON or saline (0.2 ml/ 0.1 ml / min) alone into each region in a counterbalanced order followed by six h sleep recordings. Sleep architecture after microinjections into SubC (Table 1; MUS, n=5; CLON, n=3) and RPC (Table 2; both drugs, n=2) are presented below. MUS and CLON reduced REM when microinjected into both regions. The effects with 1.0 µM CLON did not reach significance because of the small n and large variability. The low n prevented statistical analyses for RPC; however, the effects followed the trend observed with microinjections of both substances into SubC. Although REM was reduced, individual episodes were characterized by atonia.

Table 1. Saline MUS 1.0 µM CLON 0.2 µM CLON 1.0 µM Total REM 36.87 ± 4.01 22.67 ± 3.77 ** 17.33 ± 2.75 * 17.50 ± 8.93 REM Episodes 17.60 ± 1.63 11.40 ± 2.01 * 9.33 ± 1.33 * 9.33 ± 5.21 REM Percent 13.91± 1.18 9.32 ± 1.27 * 7.43 ± 1.09 * 7.02 ± 3.54 Total NREM 226.27± 7.72 219.93 ± 13.59 217.06 ± 17.87 176.72 ± 45.62

Table 2 Saline MUS 1.0 mM CLON 0.2 mM CLON 1.0 mM Total REM 40.00 ± 9.17 14.34 ± 3.34 23.00 ± 11.17 15.34 ± 1.84 REM Episodes 10.50 ± 0.50 6.50 ± 1.50 11.00 ± 4.00 7.50 ± 0.50 REM Percent 13.881± 2.84 5.99 ± 1.75 8.56 ± 3.89 6.02 ± 0.34 Total NREM 246.58± 7.42 229.42 ± 18.91 240.83 ± 0.50 238.50 ± 14.17 * P< .05; ** P< .01

These data indicate that neurons in SubC and RPC are involved in the regulation of REM sleep; however, microinjections of neither MUS nor CLON eliminated REM atonia. This suggests that fibers of passage in these regions must be damaged to eliminate REM atonia.

This work was supported by NIH grants NS36694 and MH61716. 192 Actas de Fisiología 7, 2001

ROLE OF THE EXTENDED VENTROLATERAL PREOPTIC NUCLEUS IN REM VS. NON-REM SLEEP

Clifford B. Saper and Jun Lu Department of Neurology and Program in Neursocience Harvard Medical School, Beth Israel Deaconess Medical Center, Boston, MA, USA

The ventrolateral preoptic nucleus (VLPO) was originally defined as a small cluster of galanin- and GABA-containing neurons just lateral and dorsal to the optic chiasm, which innervate the components of the ascending arousal system. These neurons are most active during sleep, and they are thought to inhibit the arousal system, thus promoting sleep. Excitotoxic lesions of the VLPO cause profound and long-lasting insomnia, without changes in body temperature, thus suggesting that the VLPO is a critical component of the brain’s circuitry for causing sleep. Loss of neurons in the VLPO following excitotoxic lesions is tightly correlated with the degree of loss of NREM sleep. Although most animals with VLPO lesions also show loss of REM sleep, the degree of REM loss does not correlate well with the loss of neurons in the VLPO cluster. These experiments suggest that there is another site, nearby this cluster, that is important for the control of REM sleep. Some neurons that share neurochemical and connectional properties with the VLPO are located outside the cluster. These GABA- and galanin-containing neurons which extend dorsally and medially from the VLPO cluster also project to components of the arousal system, and they are sleep active. Interestingly, loss of neurons in these regions correlated with the degree of REM, but not NREM sleep impairment. We therefore examined the role of the dorsal and medial extended VLPO (exVL- PO) in sleep. In experiments in which animals were exposed to light during the first part of the normal dark cycle, they experienced 2-3 times as much REM sleep as control animals that slept in the dark. We found that there was increased expression of Fos protein by neurons in the exVLPO in the animals that were light-treated, and that this correlated closely with the amount of REM sleep the animals experienced. More than 90% of the Fos–positive cells in the VLPO cluster and 80% of the cells in the exVLPO during light-treatment contained galanin mRNA. We then examined the projections to the major monoamine groups from the exVL- PO vs. the VLPO cluster. We found that the major source of inputs from the preoptic area to the locus coeruleus (LC), the dorsal raphe nucleus (DR), and the laterodorsal tegmental nucleus (LDT) was the extended VLPO, rather than the VLPO cluster. An- terogradely labeled axons innervated the cell bodies of the LC and DR neurons, but terminated on interneurons intercalated with the cholinergic neurons of the LDT. These projections are consistent with the exVLPO differentially inhibiting the LC and DRN, and perhaps disinhibiting the LDT. The differences in connectivity between the VLPO and the exVLPO may explain their differential roles regulating NREM vs. REM sleep. Actas de Fisiología 7, 2001 193

SLEEP-WAKE CYCLE AND NITRIC OXIDE (NO) IN AGED RATS AND SAMP8 MICE

Nicole SARDA, Pierre CLEMENT, Damien COLAS, Abdallah GHARIB and Raymond CESPUGLIO INSERM Unit 480, Claude Bernard University., F-69373 Lyon - France

Nitric oxide (NO), first known as a toxic free radical, is now accepted as a biological messenger diffusing easily within the tissues. Since the discovery of the brain NO- synthases (NOS), several studies have been carried out in order to determine the function attached to the NO synthesised either by the NOS expressed constitutively within the neurons or by the NOS expressed in the astroglia (iNOS) in response to an immune challenge. In this respect, it is reported that NO donors or nNOS inhibitors respectively facilitate or reduce paradoxical sleep (PS) amount when infused in the nRD (n. raphe dorsalis) [Burlet et al.,1999]. It is also further suggested that an overproduction of iNOS might lower the activity of the central nervous system (CNS) during ageing. As yet, however, regarding ageing the influence of NO remains poorly investigated. The aim of the present study was thus focused on the role of iNOS in the sleep-wake architecture of aged Wistar rat (20-24 months) and SAMP8 mouse (2-8 months) versus respective control, i.e. adult rat, SAM R1 mouse. Polygraphic methods were first used to examine the sleep-wake parameters in basal conditions. The effects of AMT (2-amino5,6-dihydro-6-methyl-4H-1,3 thiazine), a specific inhibitor of the iNOS, was then after tested at different doses (i.p.) on the sleep-wake parameters. Finally, NADPH-diaphorase (NADPH-d) histochemistry was employed for brain nNOS and iNOS labelling. Over a 24h recording period (12h-12h light –dark cycle) and versus control animals, aged Wistar rats exhibited a decrease in diurnal amounts of PS, an increase in PS latency at the light-dark transition and a decrease in the amplitude of PS circadian rhythm. Moreover, power spectral density of d, f and s bands for slow wave sleep (SWS) and f band for PS where decreased in keeping with the circadian rhythm modifications. A moderate NADPH-d activity was noticed in the brain of such animals. Similar changes were observed with two months aged SAMP8 mice versus SAM(1999). R1 control, except for the NADPH-d labelling which remained unchanged. Either in rats or in mice, AMT administered at 10 or 20 mg/Kg i.p. induced a full suppression of PS during the first 4-6h following the administration while any significant effect was induced in control animals. By this time, the power spectra of the electroencephalographic bands was reduced. In conclusion, our approach, in keeping with previous studies, indicates that ageing alters the sleep-wake parameters and mainly those related to PS. NO produced by iNOS is pointed out as determinant for the occurrence of PS in the aged animal.

Burlet S, Leger L, Cespuglio R, Nitric oxide and sleep in the rat: a puzzling relationship. Neu- roscience1999, 92: 627-639. 194 Actas de Fisiología 7, 2001

DIURNAL VARIATION IN OREXIN-A LEVELS IN THE LATERAL HYPOTHALAMUS (LH) OF RAT

Robert E. Strecker*, Lynda J. Dauphin, Michael Wilkemeyer, Vijay Ramesh, Christopher M. Sinton, Robert W. McCarley, Mahesh M.Thakkar Departments of Psychiatry & Neurology, Harvard Med. Sch.& VA Boston Healthcare System, Brockton MA, 02301, USA

The hypocretin/orexin containing neurons are exclusively located in the LH, but they send projections throughout the CNS, including a number of regions that regulate sleep and wakefulness (W). Dysfunction of the orexin peptide-receptor system is linked to cataplexy/narcolepsy, suggesting that orexin may be involved in the regulation of natural changes in behavioral state. To examine the fluctuations in extracellular orexin- A in response to changes in behavioral state and circadian time, the present study used microdialysis (MD) sample collection coupled to biochemical analysis by enzyme linked immuno-sorbent assay (ELISA). MD probes placed in a 1 µM standard solution of orexin-A recovered 0.44 ± 0.16% of the orexin-A in the solution (N = 7), indicating that orexin-A (MW =3,562) can pass through the MD membrane. Probes were placed in the LH of freely moving rats (coordinates: AP -3.1, ML ± 1.5, DV -9.0), and samples were collected using a MD flow rate of 1.0 ul/min. Orexin-A was detectable in 50 to 100 ul of sample. The concentration of orexin-A in the LH microdialysate from 0700h to 1300h was 0.59 ± 0.19 ng/ml (N=7). Relative to this time period LH orexin-A levels in 6 h samples were lower at other times of day (1300h to 1900h = 0.34 ± 0.08 ng/ml; 1900h to 0100h = 0.25 ± 0.05 ng/ml; 0100h to 0700h = 0.29 ± 0.06 ng/ml; P<.05, rmANOVA). During the light period rats predominantly sleep (inactive period), and during the dark period they are typically awake (active period). Hence, we used a 2 to 3 h period of forced W (produced by gentle handling) during the light period, as a first step to disassociate the influence of diurnal rhythm from behavioral state changes & diurnal differences in locomotor behavior. Baseline samples were collected before and after this forced W period. In this forced W experiment, orexin-A levels did not increase in 7 of 8 cases (based on the criteria of > 0.4 ng/ml change; paired t (7) - 1.6, p = 0.17). The forced W experiment will be repeated controlling for the diurnal variation observed (i.e., using a between day design). In conclusion, these data indicate that the time of day has a greater effect on LH extracellular orexin levels than does the sleep-wakefulness state of the animal.

Support: NIMH 39683 & 01798; Dept Vet. Aff. Actas de Fisiología 7, 2001 195

ROLE OF NOREPINEPHRINE IN THE REGULATION OF BASAL FOREBRAIN WAKE-ACTIVE NEURONS

Mahesh M. Thakkar and Robert W. McCarley Psychiatry, Harvard Medical School, VA Medical Center, Brockton, MA 02301, USA [FOCUS GROUP]

There is a plethora of evidence implicating the norepinephrinergic neurons of the locus coeruleus in the regulation of sleep wakefulness. Extracellular unit recordings of neurons in the cat documented that these neurons were maximally active during waking, and virtually ceased discharge during REM sleep. However the exact mechanisms and the anatomical substrates through which norepinephrinergic system mediates it effects on sleep-wakefulness is unclear. We propose that one important action of norepinephrinergic neurons of the locus coeruleus in regulating wakefulness is exciting the wake-active neurons of the cholinergic basal forebrain. Our hypothesis is supported by the fact that the cholinergic magnocellular regions of the basal forebrain are critical in the regulation of wakefulness. Ma- jority of the neurons in the cholinergic basal forebrain show highest discharge activity during active waking. Norepinephrinergic neurons of the locus coeruleus project directly to the cholinergic basal forebrain. In vitro studies indicate norepinephrine excites the basal forebrain cholinergic neurons whereas in vivo microinjections of norepinephrine in the cholinergic basal forebrain increases wakefulness. Furthermore, the cholinergic basal forebrain is suggested to be the key anatomical substrate, mediating the actions of adenosine, a putative sleep factor, and orexin-A, a wakefulness inducing peptide, in control of behavioral states. Thus, we hypothesize that microdialysis perfusion of norepinephrine in the cholinergic basal forebrain will selectively increase the discharge activity of wake-active neurons of the cholinergic basal forebrain. To test this hypothesis, we used the combined unit recording and microdialysis technique developed in our lab. The brief description of the methods is : Using standard aseptic surgical procedures, male adult cats were implanted with standard sleep recording electrodes and bilateral microdrives targeted towards the cholinergic basal forebrain. After at least 15 days of post-operative recovery and habituation, the experiment was begun. Single unit along with electrographic recording were digi- tized and recorded on a computer. Once a unit was encountered, microdialysis probe (CMA-11 membrane length 2 mm, 0.24 mm diameter, CMA/Microdialysis, Acton, MA) was inserted. After at least 12 h of post insertion recovery, ACSF was perfused continuously at 1.5 ul/min. The microdrive was slowly advanced until a single neuron was encountered (signal to noise ratio = 2:1). The discharge activity of the neurons was recorded across one complete sleep-wake cycle with continuous perfusion of ACSF and 30 min with microdialysis perfusion of norepinephrine. Once the experiment was complete, the animals were sacrificed and the brain processed for histology and immunohistochemistry to identify the recording zone in the cholinergic basal forebrain. Preliminary studies suggest that norepinephrine increases the discharge activity of wake-active neurons (n=2) in the cholinergic basal forebrain supporting our hypothesis that norepinephrine excites wake-active neurons of the basal forebrain. This along with other studies done in our lab suggest that the cholinergic basal forebrain may be a key anatomical ‘gate’, integrating and gating the effects of various neurotransmitter /neuromodulators to control of sleep- wakefulness. 196 Actas de Fisiología 7, 2001

FOS IMMUNOREACTIVITY IN HYPOCRETINERGIC NEURONS OF THE CAT DURING ACTIVE WAKEFULNESS, QUIET WAKEFULNESS AND QUIET SLEEP

Pablo Torterolo, Jack Yamuy, Sharon Sampogna, Francisco R Morales and Michael Chase Department of Physiology and the Brain Research Institute, UCLA School of Medicine, Los Angeles CA, USA

Hypothalamic hypocretinergic neurons have been implicated in the control of behavioral states. Recently, we determined that hypocretinergic neurons in the cat expressed c-fos (a marker of neuronal activity) during wakefulness that is accompanied by locomotor-explorative activity, but not during quiet wakefulness in the absence of motor activity 1. In addition, we found a subpopulation of hypocretinergic neurons that expressed c-fos during active sleep induced by carbachol 1. In the present report, in order to differentiate c-fos expression in hypocretinergic neurons produced by general arousal from that produced by motor activity, we compared Fos immunoreactivity in cats sacrificed after periods of active wakefulness induced by constant sound stimulation (but without motor activity) with that in animals sacrificed after a period of locomotor-explorative behavior. C-fos expression during quiet wakefulness and naturally- occurring quiet sleep (QS) was also studied. Adults cats were maintained during 60 to 90 minutes prior to being sacrificed in the following behavioral states: a) quiet wakefulness (QW, n = 3), wherein the cats were awake and restrained; b) active wakefulness with locomotor-explorative activity (AW- L, n = 3), wherein the animals were freely exploring a new environment and exhibited almost continuous locomotor activity; c) active wakefulness with sound stimulation (AW-S, n = 1), in which the animal was restrained, awake and was stimulated with loud clicks; and, d) QS (90 % of QS and 10% of QW, n = 1). Thereafter, the animals were sacrificed and perfused for immunocytochemistry. Double immunostaining for Fos and hypocretin 2 were performed and the distribution of hypocretin+ and hypocretin+ Fos+ neurons were analyzed. The greatest number of hypocretinergic neurons expressed c-fos during AW-L (186 ± 18.5, 79 %, P < 0.0001, in comparison to all other conditions). The animals sacrificed during QW and AW-S showed a similar low number of Fos immunoreactive hypocretinergic cells (3.0 ± 1.9, 1.7% and 2.0 ± 0.6, 1.5%, respectively). No Fos immunoreactivity was present in hypocretinergic neurons during QS. The significant difference in the number of c-fos-expressing hypocretinergic neurons between AW-L and wakefulness without motor activity (AW-S and QW) indicates that these neurons may not be responsible for maintaining wakefulness, per se, but are likely to play a role in locomotor-explorative behaviors. In addition, these data suggest that hypocretinergic neurons do not play a role in QS.

Research supported by USPHS grants NS09999, NS23426, MH43362, AG04307 and HL60269.

(1) Torterolo et al. (2001). Sleep Res Online, in press. Actas de Fisiología 7, 2001 197

THE LACK OF A SENSORY INPUT MODIFIES THE CORTICAL POWER SPECTRA IN SLEEP AND WAKEFULNESS

Giancarlo Vanini, Alberto Rodríguez, Marisa Pedemonte & Ricardo A.Velluti Neurofisiología, Departamento de Fisiología. Facultad de Medicina. Universidad de la República. Montevideo, Uruguay

The auditory cortex exhibits a functional-anatomical reorganization after the partial lesion of the receptor, the cochlea(1). Moreover, the total lack of auditory input -deaf animals- showed changes in the sleep/waking cycle(2),(3) introducing this sensory input as an influence on sleep-waking cycle organization. It was our aim to study the power spectra shifts at three cortical loci that may occur when a guinea pig is totally deprived of auditory input, in sleep and waking. Adult guinea pigs (Cavia porcellus) were chronically implanted, under pentobarbital anesthesia. Bipolar electrodes were placed to record the auditory cortex electro- corticogram, the dorsal hippocampus, the motor cortex and neck muscles electromyogram. After 5 days of recovery the animals were recorded -as their own control- main- tained in a 12:12 light-dark schedule in a sound attenuated chamber with food and water ad libitum. A second surgical procedure, the bilateral destruction of both cochle- ae, was carried out under anesthesia and the region was filled with dental cement. The animal deaf condition was tested. A computerized polygraph (Nautilus Plus, PSG), was used for quantification of the time spent in wakefulness (W), slow wave sleep (SWS) and paradoxical sleep (PS). Then, the Fast Fourier Transform analysis (FFT) was applied to the different cortices electrogram, compared among them during each behavioral condition. The electrocorticograms frequencies analyzed were in the lower range: delta, 1.0-3.5; theta, 4.0-11 cycles/s; in the higher range: 12-15, 15.5-25 and 25.5-32 cycles/s. Out of a total of 75 recorded hours, 30 windows of 5s for each behavioral state were selected for the FFT analysis. The Student’s T test was applied for the comparison between the FFT power in control and the deaf guinea pigs, both in sleep and waking. 1. The results exhibited by one animal were a power significant decrease (p<0.05) in the low frequency range (<15 cycles/s) in the three cortices explored during W, SWS and PS up to the 10th day after the lesion. Thereafter, a power recovery was observed the 13th day. 2. During W the animals presented the same low frequency power change. 3. In the high frequencies range there was not a clear power shift in any behavioral state. 4. The primary auditory cortex exhibited the same power shifts as the motor cortex and hippocampus.

1) Robertson & Irvine. J. Comparative Neurology, 282: 456-471, 1989 2) Pedemonte, Peña, Torterolo & Velluti. Neurosci. Lett., 233: 1-4, 1996 3) Cutrera, Pedemonte, Vanini, Goldstein, Savorini, Cardinalli & Velluti. Arch.Ital.Biol., 138: 285-293, 2000 198 Actas de Fisiología 7, 2001

EVOLUTION OF SLEEP AND SLEEP EEG AFTER HEMISPHERIC STROKE

Jacqueline Vock, Peter Achermann, Matthias Bischof, Milena Milanova, Caroline Müller, Corinne Roth and Claudio Bassetti Neurologische Poliklinik, Zürich, Switzerland

The effects of focal hemispheric stroke on sleep and sleep EEG and their evolution over time are poorly known. We prospectively studied 27 patients (pts) with first hemispheric ischemic stroke and no sleep apnea. Clinical assessment included estimated sleep time/24h (EST) and Epworth Sleepiness Score (ESS) before stroke, as well as EST, ESS and stroke outcome 12 (2-19) months after stroke. Sleep EEG data in the chronic phase (5-24 months after stroke) were compared with those in the acute phase (1-8 days), and with those of 11 hospitalized controls. Changes in EST (>2h, 38 % of pts) and ESS (>3 points, 26%) were frequent but correlated poorly with sleep EEG changes. In the chronic phase no significant differences in sleep EEG between hospitalized controls and patients were found. In patients, high sleep efficiency (SE) and low wakefulness after sleep onset (WASO) in the acute phase were associated with a good long-term outcome. Both SE and WASO improved significantly from the acute to the chronic phase. After hemispheric stroke sleep-wake complaints (insomnia and hypersomnia) are relatively frequent, whereas sleep EEG changes are modest and non-specific. A good sleep continuity in the acute phase heralds a good stroke recovery.

Supported by: Swiss National Science Foundation Grant 32-49853.96 Actas de Fisiología 7, 2001 199

CHANGES IN ELECTROPHYSIOLOGICAL PROPERTIES OF LUMBAR MOTONEURONS DURING BICUCULLINE-INDUCED ATONIA IN THE CAT

Ming-chu Xi, Francisco R. Morales and Michael H. Chase Department of Physiology and the Brain Research Institute, UCLA School of Medicine, Los Angeles, CA 90095, USA

We have recently reported that a behavioral state that resembles naturally-occurring active sleep arises following the microinjection of bicuculline, a GABAA antagonist, into the nucleus pontis oralis (NPO) of the pontine reticular formation (Xi et al. 1999, 2001). However, it remained to be determined if the atonia induced by the injection of bicuculline into the NPO was produced by processes of postsynaptic inhibition or disfacilitation, and if a postsynaptic inhibitory mechanism was responsible, whether it involves the same processes that result in atonia during naturally-occurring active sleep? Therefore, the present study was undertaken to explore the neuronal mechanisms responsible for the inhibition of motoneurons that occurs following the microinjection of bicuculline into the NPO in conjunction with an examination of the basic electrophysiological properties and the state-dependent synaptic activity impinging on spinal cord motoneurons. Six cats were prepared under halothane anesthesia for intracellular recording from lumbar motoneurons. The animals were then anesthetized with a-chloralose (60 mg/ kg, i.v.) after the completion of all surgical procedures. Intracellular recordings were obtained from antidromically identified motoneurons. The electrophysiological properties of motoneurons were measured before and during muscle atonia induced by the microinjection of bicuculline (0.25ml, 10 µM in saline) into the NPO. Following the microinjection of bicuculline into the NPO, lumbar motoneurons exhibited significant changes in their electrophysiological properties which consisted of a decrease in input resistance and membrane time constant, and an increase in rheobase. In addition, discrete, large-amplitude “active-sleep-specific” IPSPs were also observed in motoneurons. Furthermore, stimulation of the medullary nucleus reticularis gigantocellularis (NRGc) evoked a large-amplitude IPSP in motoneurons following, but never prior to, the injection of bicuculline, which reflects the motor inhibitory aspect of “reticular response-reversal” (Chase and Babb 1973). These data indicate that lumbar motoneurons are postsynaptically inhibited following the injection of bicuculline in a manner that is comparable to that which occurs during the atonia of active sleep. Based on the present study, we conclude that the brainstem-spinal cord inhibitory system which is responsible for motor inhibition during active sleep is tonically inhibited by pontine GABAergic mechanisms during wakefulness and quiet sleep, and become activated during active sleep which results in atonia during this state.

This work was supported by USPHS Grants NS 23426, NS 09999, MH 43362, AGO 4307 and HL 60296. 200 Actas de Fisiología 7, 2001

DIURNAL VARIATION OF CSF AND EXTRACELLULAR HYPOCRETIN/ OREXIN LEVELS IN RATS: IN RELATION TO SLEEP/WAKE CYCLE AND RESPONSE TO FOOD DEPRIVATION

Yasushi Yoshida, Nobuhiro Fujiki, Tomoko Nakajima, Hitoshi Matsumura, Hiroshi Yoneda, Emmanuel Mignot, Beth Ripley and Seiji Nishino 1201 Welch Rd. MSLS Rm113, Palo Alto CA, USA

Hypocretins/orexins are newly identified neuropeptides that are produced in neurons located exclusively in the lateral hypothalamic area (1, 2). Hypocretins are reported to play important roles in various hypothalamic functions, including regulation of sleep, neuroendocrine function, feeding and energy homeostasis. It has also been recently discovered that impaired hypocretin neurotransmission is involved in both animal and human cases of the sleep disorder narcolepsy (3, 4, 5). Hypocretin-1 is consistently detectable in cerebrospinal fluid (CSF) of healthy subjects, but is absent in narcoleptics; however, functional roles of CSF hypocretin are largely unknown. In the current study, we therefore developed a survival method that allowed for repeated cisternal CSF taps in rats to examine diurnal variation of CSF hypocretins, as well as to determine whether CSF hypocretin levels reflect the activity of hypocretin neurotransmission in the brain during a food deprivation paradigm. 1) Diurnal variation: We divided rats into group 1 (n = 7) and group 2 (n = 8). Cister- nal CSF collection times for group 1 were at 0, 4, 8, 12, 16, 20 hr circadian time (CT) and for group 2 were at 2, 6, 10, 14, 18, 22 hr CT. 2) Food deprivation: At CT 8, one day before food deprivation, cisternal CSF collection was performed in 18 rats, and taps were repeated at 24 and 72 hr after the commencement of food deprivation. The entire experiment was repeated in the same 18 rats with taps occurring at CT 0. CSF Hypocre- tin-1 was measured with commercially available 125I RIA kits (Phoenix Pharmaceuti- cals, CA). Hypocretin-1 was high during the dark period (14 to 22 hr CT) when animals were active, but decreased by 40 % during the light (rest) period (lowest at CT 8 hr). In a food deprivation session, CSF hypocretin-1 was significantly increased after 72hr but not after 24 hr at CT 8. Levels after 72 hr of food deprivation at CT 8 were comparable to those seen during the baseline dark period. Interestingly, the same manipulation at CT 0, when basal levels are high, did not result in further increase in hypocretin-1. Major findings of this study were 1) hypocretin-1 levels in CSF showed daily diurnal variation, and were highest during the dark period when animals were active, and this variation may reflect the activity of hypocretin neurons. 2) food deprivation increased the CSF hypocretin-1 level at CT 8 (rest phase), but not at CT 0 (active phase). This suggests that the activity of hypocretin neuron may reach its maximum, unaffected by external stimuli, at the active phase in rats. In parallel with these repeated CSF tap experiments, we also measured hypocretin-1 levels in microdialysis perfusates collected in the lateral hypothalamic area with simultaneous sleep recordings. Fluctuations of extracellular hypocretin levels along with changes in sleep parameters will also be presented.

1, Sakurai T et al., Cell 1999, 2, De Lecea L et al., PANS 1998, 3, Chemelli RM er al., Cell 1999, 4, Lin L et al., Cell 1999, 5, Nishino S et al., Lancet 2000 Actas de Fisiología 7, 2001 201

DELTA HOMEOSTASIS AND KINETICS IN YOUNG AND ELDERLY NORMAL SUBJECTS

Irwin Feinberg and Ian G. Campbell VA/UCD Sleep Laboratory, TB 148, UC Davis, CA, USA

In four separate sessions, sleep EEG was recorded on baseline, naps and post-nap sleep in 19 young adult (20-26 yrs) and 19 normal elderly (65-80 yrs). The naps were at 0900,1200,1500 and 1800, with Ss in bed for 2 hours in each nap. EEG was analyzed with power spectral (FFT) and period-amplitude methods with PASS PLUS. Data analysis is still in progress but the following is established thus far for the FFT results. The elderly as well as the young adults maintain delta conservation across naps and post- nap sleep. REM sleep is not conserved; REM occurring in daytime naps does not reduce REM in post-nap sleep. Delta power grows linearly rather than exponentially with increasing wake duration in both young and elderly.The decline in delta power across NREMPs is also linear in both groups. While the slopes of the delta power decline across sleep are similar in young and elderly, the delta growth curve in the elderly is significantly flatter than in the young adults. These findings have implications for homeostatic models of NREM delta and for the nature of brain aging and associated sleep changes. 202 Actas de Fisiología 7, 2001

DOES EARLY IRON DEFICIENCY ANEMIA PROVOKE LASTING EFFECTS ON AUTONOMIC CONTROL DURING SLEEP?

Patricio D. Peirano, Cecilia R. Algarín, Marcelo I. Garrido, Eduardo H. Muñoz and Betsy Lozoff

Introduction: The essential role of iron on myelin process formation could be the clue for explaining the disruption produced by early iron-deficiency anemia (IDA) on the neurofunctional development of the autonomic nervous system (ANS). In the human, vagus myelination is partly postnatal and parasympathetic functional maturation follows that of the sympathetic system. To determine whether ANS changes are long- lasting, we evaluated heart rate and their variabilities as a function of sleep states in former IDA children aged 3-4 years. Methods: All-night standard polysomnographic recordings were done in a group of healthy 3- to 4-year-old children who were treated for IDA (n=24) or were nonanemic (controls, n=26) in infancy. The ECG signal was processed off-line by a signal-to-noise ratio algorithm that detected the peak of the R wave for each heart beat and quantified sequential RR intervals in ms. Another algorithm based on the Fast Fourier Transform provided the spectral amplitude (in ms) in frequency bands: high- , mid- , and low- frequency HRV. These corresponded to variations of RR intervals within periods of 3- 8, 10-25, and 30-100 heart beats, respectively. NREM stages III and IV were grouped as SWS. For the individual child, mean RR interval and each HRV were computed over each REM, NREM stage 2, and SWS and then analyzed according to the successive thirds of the night. Results: The nocturnal organization of RR intervals and their variabilities within sleep states differs between groups. In controls, RR intervals were longer in all states and lengthened during the 2nd third of the night in NREM stages. In former IDA subjects HR values remained similar in all thirds. In all sleep states, controls showed a higher extent of high-frequency HRV, especially during the first two thirds of the night. A similar temporal pattern of differences between groups was present for low-frequency HRV in REM and NREM stages. On the contrary SWS differences appeared by the end of the night. Conclusions: Our results indicate that early IDA is associated with long-lasting alterations in the ANS functioning in both sleep states. This emphasizes the role of iron on the brain integration process that participates in establishing the pattern of temporal organization in the ANS balance. Since related physiological and behavioral processes depend on the functional integrity of the ANS, we suggest that long-lasting IDA effects on the temporal organization of the ANS during sleep may well represent a potential factor for altering developmental outcomes in childhood.

(Support: Grants from NICHD (HD33487) and Fondecyt (CONICYT, Chile 1000657). Actas de Fisiología 7, 2001 203

LASTING EFFECTS OF EARLY IRON DEFICIENCY ANEMIA ON SPONTANEOUS MOTOR ACTIVITY AND ITS TEMPORAL MODULATION

Beatriz C. Valdebenito, Marcelo I. Garrido, Cecilia R. Algarín, Betsy Lozoff and Patricio D. Peirano

Introduction: Our most recent study has demonstrated that iron-deficiency anemia (IDA) in infancy affects the quantitative patterns of motor activity and is characterized by a reduction in spontaneous movements. Alterations in locomotor activity rhythms have also been reported in IDA rodents. These changes are not reverted with iron therapy when IDA occurs in rat pups. Although these persisting modifications have not yet been explained, they may relate to altered CNS transmission, myelination, and/or neuronal metabolic activity since iron is critically involved in all these processes. To determine whether motor activity changes are long-lasting in humans we evaluated movement patterns in former IDA (FIDA) children aged 3-4 years. Subjects and Methods: Actigraphic recordings were performed at home for 24- hours in a group of otherwise healthy FIDA (n=27) and non-FIDA (control, n=33) children. For each child motor activity was evaluated by the total amount of movements on an hourly basis for the whole 24-hour period. Each corresponding value was classified as high (HIL, 4th quartile), medium (MIL, 2nd-3rd quartile), or low (LIL, 1st quartile) intensity level. Comparisons between groups were performed for HIL and LIL. Results: FIDA children demonstrated a reduced amount of movements during the 24-h period (p <.05). The same was also true for both HIL (all between 9 and 22 hrs) and LIL (all between 23 and 8 hrs) values (p <.000 and <0.012 respectively). Differ- ences for HIL among groups reached statistical significance at 11 and 13 hrs (p <.006), and for LIL at 1, 3 and 4 hrs (p <.04, <.05 and <.002, respectively). Discussion: These results demonstrate a persisting effect on spontaneous motor activity in FIDA children. This influence is not merely restricted to its quantitative patterns since the temporal distribution is involved as well. Although several mechanisms may contribute to why early IDA might alter motor activity regulation over time, we suggest that sustained reduction and eventual disruption in motor activity functions to affect developmental outcomes that characterize FIDA during childhood and adolescence.

(Support: Grants from NICHD (HD33487) and Fondecyt (CONICYT, Chile 1000657). 204 Actas de Fisiología 7, 2001

SLEEP BRUXISM AND BEHAVIOURAL DISTURBANCE IN CHILDREN: A POPULATION-BASED STUDY

Maria H. Azevedo, Vanda M. Clemente, Celsa M. Pissarra, Ana C. Gomes, António P. Pereira, Eva M. Silva and António F. Macedo Psicologia Médica, Faculdade de Medicina, Coimbra, Portugal

Abstract: AIM. To investigate the relationship between sleep bruxism and behavioural disturbances in Portuguese children attending primary school. METHODS. A sleep-waking questionnaire was sent to the parents of 1381 children attending primary schools in a parish of Coimbra, Portugal. To assess psychiatric symptomatology, the Portuguese version of Rutter’s Children’s Behaviour Question- naire for completion by teachers was used. A total deviance score is derived from the sum of scores for the individual items (n= 26), an emotional sub-score obtained from the sum of scores of four items (worried, miserable, fearful, tears on arrival at school) and a conduct sub-score obtained from the sum of scores of six items (destructive, fights, disobedient, lies, steals, bullies). The ratings were made by the teachers in the last term of the school year. RESULTS. Of the 988 questionnaires returned (71.5%), complete information concerning sleep bruxism was obtained for 864 children (438 girls and 426 boys; mean age: 7.9 ± 1.31 years; range= 6-11 years). Bruxism during sleep was reported as frequent or always present (Bruxers) in 76 children (8.8%), as occasionally present in 164 children (19.0%), and as never present (Never Bruxers) by 624 children (72.2%). The mean total score on the Rutter scale was 5.06 ± 5.37 (range= 0-33). Comparing Bruxers and Never Bruxers no significant differences were found with respect to behavioural disturbance (Mean total score =4.51±4.94 vs Mean total score =4.96±5.26; NS), emotional disturbance (Mean=1.74±1.07 vs Mean=1.89±1.12; NS) and conduct problems (Mean=3.47±1.42 vs Mean= 3.88±2.19; NS) as assessed by teachers on the Rutter scale. Actas de Fisiología 7, 2001 205

CEREBROSPINAL FLUID LEVELS OF HYPOCRETIN (OREXIN) IN HYPERSOMNOLENT PATIENTS WITHOUT CATAPLEXY

Claudio L. Bassetti, Matthias Gugger, Johannes Mathis, Christian Sturzenegger, Bogdan Radanov, Beth Ripley, Seji Nishino and Emmanuel Mignot Neurologische Polikinik, Zürich, Switzerland

Abstract: Background: Positional cloning has identified hypocretin-receptor-2 gene mutations as the cause of narcolepsy in a canine model (Lin et al., Cell 1999; 98: 365- 376). Hypocretin-1 was undetectable in the CSF of seven of nine narcoleptics with cataplexy (Nishino et al., Lancet 2000; 355:39-40) and a loss of hypocretin neuronal expression was demonstrated post mortem in six narcoleptics (Peyron et al., Nat Med 2000; 6: 991-997). Aims: 1) Test the hypothesis that CSF hypocretin is reduced in narcoleptic patients without cataplexy (so-called monosymptomatic narcolepsy) and in patients with other forms of (mainly neurogenic) hypersomnia including idiopathic hypersomnia. 2) Assess the value of a multimodal diagnostic approach in this same patient group. Patients: Twenty consecutive patients (9 women and 11 men; mean age=38 years, range 16-53) with narcolepsy with cataplexy (n=1), symptomatic narcolepsy following Bickerstaff’s encephalitis (n=1), narcolepsy without cataplexy (n=4), idiopathic hypersomnia (n=4), hypersomnia associated with mood disturbances (n=4), HIV-encephalop- athy (n=1), thalamo-mesencephalic stroke (n=1), periodic hypersomnia (n=1), and post- viral illness (n=1). In two patients, hypersomnia remained of undetermined origin despite extensive work-up. Methods: Clinical examination, standard sleep questionnaire, Ullanlinna Narcolepsy scale (UNS, Hublin et al. Ann Neurol 1994; 35: 709-716), conventional polysomnography, multiple sleep latency test (MSLT), 1-week actigraphy, HLA-typing, clinical and psychometric evaluations, and measurement of CSF hypocretin levels as previously described (Nishino et al., Lancet 2000; 355:39-40). Diagnoses were made according to standard criteria (International Classification of Sleep Disorders). Results: 1) Hypocretin-1 was detectable in 19 of 20 patients. Only in the single patient with idiopathic narcolepsy and clear-cut cataplexy were hypocretin concentrations below the detection limit of the assay (<40 pg/mL). 2) An ESS >14, a UNS >14, and so- called REM symptoms (sleep paralysis, hallucinations), that are commonly considered as suggestive of narcolepsy were seen in non-narcoleptics and even in patients with normal sleep latencies and no SOREMPs on MSLT. Conversely, the presence of very short sleep latencies and SOREMPs, was not necessarily associated with very high ESS, UNS or REM symptoms. Finally, hypersomnia related to mood disorders was found to share clinical features with idiopathic hypersomnia, including sleep drunkeness, shortened sleep latencies on MSLT, and abnormal actigraphic findings. Conclusions: Normal CSF hypocretin concentrations frequently accompany narcolepsy without cataplexy and other (neurogenic) hypersomnolent syndromes. Thus, hypocretin dysfunction does not necessarily represent the “final common pathway” in the pathophysiology of hypersomnolent syndromes that fall on the borderline for a diagnosis of narcolepsy. The observed clinical overlap among these hypersomnolent syndromes implies that current diagnostic categories are not entirely unambiguous (Bassetti and Aldrich, Brain 1997; 120: 1423-35). 206 Actas de Fisiología 7, 2001

SLEEPING DIFFICULTIES, ANXIETY AND DEPRESSION AMONG 50-75 YEARS OLD WOMEN WITH RESTLESS LEGS SYNDROME

Harriet Bengtson, Jan-Erik Broman, Jerker Hetta and Ulfberg Jan Psychiatry, University Hospital, Uppsala, Sweden

Abstract: Introduction: Restless legs syndrome (RLS) affects many people and may also have a significant impact on sleep. The aim of this study was to investigate if depression and anxiety was more common among women with both RLS and sleep difficulties. Method: In 1998 a questionnaire about sleep and sleep complaints was sent to a random sample of 2478 women, ages 50-75 years, living in Uppsala municipality in mid-Sweden. 64% responded to the questionnaire. 18 months later a subgroup of 300 women with sleep-problems and 300 without sleep-problems, in total 611, was approached again with another questionnaire. Responses were obtained from 535 women (88%). The second questionnaire included 4 minimal symptoms questions for the diagnosis of RLS, by The International Restless Legs Syndrome Study Group (IRLSSG), and the Hospital Anxiety and Depression Scale (HAD). Results: RLS prevalence in women with sleep complaints was 40.08 % and in the group without sleep complaints was 25.26%. This study shows that women with RLS had more often than women without RLS anxiety, 39.9% versus 27.7% (p=.0053) and a tendency towards depression, 21.4% versus 14.8%(p=.0584). But it in a multiple logistic regression sleep complaints anxiety and depression as predictors of RLS are analysed, only sleep complaints is significant, OR for sleep problems 1.690, for anxiety 1.298 and for depression 1.140. Conclusion: We found that women with RLS have more anxiety than women without RLS only if they also have concomitant sleep problems. Actas de Fisiología 7, 2001 207

CHARACTERISTICS OF SLEEPLESSNESS IN LATIN AMERICA: RESULTS OF 2000 LASS SLEEP SURVEY

Margarita Blanco Centro Neurológico, Hospital Francés - Buenos Aires, Argentina

Difficulties with sleep and wakefulness are found in up to 38 percent of the world population.1 This survey obtained the incidence of sleep disorder sufferers and measured and evaluated the attitudes, perceptions, beliefs, and behaviors of sleep disorder sufferers in three Latin American countries. The Latin American Sleep Society (LASS), in conjunction with ACNielsen BA- SES, conducted two phases of door-to-door and telephone interviews with a sample of adults living in main urban areas, from three Latin American countries. In Phase One, adults (N = 1776) aged 18 to 70 were interviewed in Argentina (N = 800), Brazil (N = 976). In Phase Two, adults aged 18 to 77 were interviewed in Argentina (N = 300), Brazil (N = 300), and Mexico (N = 300). For the regional analysis, the country results were weighted by population. About 68% of those surveyed in the main urban areas report some type of sleep problem within the last twelve months. Over three-quarters believe that not getting enough sleep impairs work performance and affects how a person functions and feels the next day. Among those with sleep problems, about two-thirds believe that missing sleep may lead to other health problems and may accelerate the aging process. Howev- er, only one-third admit that their own sleep problems are a danger to their health; almost half (44 percent) have accepted their sleep difficulty and live with it. The majority (65 percent) have never asked a physician about their sleep problems because they believe their sleep problems are not serious. In a typical month, about one-fifth of sufferers claim to use an herbal beverage (i.e., chamomile, homemade remedy) to help with sleep difficulties. The most desired characteristic in a sleep medication is next-day energy level. Other important factors are next-day concentration and the number of hours slept after taking a sleep medication. These results demonstrate that sleeplessness affects 68% of Latin Americans living in main urban areas and that Latin Americans believe sleep problems have an impact on mental and physical functioning. Although Latin Americans understand the overall consequences of poor sleep, greater education is needed for patients to recognize their own need for adequate sleep and to seek proper treatment for their sleep problems. Although Latin Americans understand the overall consequences of poor sleep, greater education is needed for patients to recognize their own need for adequate sleep and to seek proper treatment for their sleep problems.

Reference: 1 Gyllenhaal C, Merritt S, Peterson S D, Block K, Gochenour T (June 2000) Efficacy and Safety of Herbal Stimulants and Sedatives in Sleep Disorders. Sleep Medicine Reviews. Vol 4 No 3,229-251.

ACNielsen BASES Survey funded by Wyeth-Ayerst Pharmaceuticals 208 Actas de Fisiología 7, 2001

PREVALENCE THE SNORING AND OBSTRUCTIVE SLEEP APNEA SYNDROME IN PRESCHOOL CHILDREN FROM MEXICO CITY

Caleco Maricela, Zuloaga Soraya, Perez-Padilla Rogelio, Vazquez Juan Carlos and Meza Sonia

Very few studies have been published about the prevalence of obstructive sleep apnea syndrome among children, and to our knwoledge none has been done in Latin Ameri- can countries. This syndrome differs from that in adults with regard to clinical manifestations and is often not recognized. We investigated the prevalence of snoring, obstructive sleep apnea and their related symptoms in children from 6 months to 6 years old in México City. All participants were studied with a standardized questionnaire regarding respiratory symptoms, sleep habits and nocturnal breathing. A simplified physical examination including measuring heigth, weigth, neck circumference, and visualization of the throat. Daytime symptoms were explored only in children with habitual snoring (HS): those who snored most or all nights. A total of 2431 children (52% males) were recruited from 20 nursery schools. The participation rate was 85%. 2159 children did not snore habitually (NHS). HS was identified in 162 children (7%), 64% of them were males (p<0.01). Average age was 3 y.o. Thirteen of these children (0.6%) were reported of having witnessed apneas. Com- paring HS Vs. NHS, we found the following differences (average will be expressed as percentage): Nocturnal breathlessness (22% vs 9% p<0.001), restless sleep (73% vs 55% <0.001), heavy sweating (56% vs 35% p<0.001), poor sleep (51% vs 20% p<0.001), enuresis (29% vs 15% p<0.001), daytime sleepiness (20% vs 7% p<0.001), adenotonsillar hypertrophy (28% vs 17% p<0.001), otitis (28% vs 17% p<0.001), passive smoking (54% vs 38% p<0.001), chronic cough (36% vs 15 p<0.001), wheezing and breathing dificulty (35% vs 16% p<0.001), nocturnal oral breathing (46% vs 3% p<0.001), bad breath (38% vs 20% p<0.001), sore throat (31% vs 18% p<0.001), asth- ma (7% vs 3% p<0.003), atopy (25% vs 16 p<0.005), malnutrition (7% vs 2% p<0.001), We conclude that the prevalence of HS in preschool children from México City is 7% and the lower limited of prevalence for sleep apneas is 0.06%. Children having HS are asociated with restless sleep, noctunal oral breathing, adenotonsillar hypertrophy, passive smoking, respiratory and daytime symptoms. Actas de Fisiología 7, 2001 209

CO-MORBIDITY OF SLEEP COMPLAINTS IN A STRATIFIED SAMPLE OF THE GENERAL POPULATION IN BELGIUM

Raymond Cluydts and Myriam Kerkhofs University of Brussels (VUB) and Vesalius Hospital Charleroi, Belgium

1000 individuals (554 females & 446 males, age range 20 to 60 y.) were interviewed during approx. 1 hour during the month of October by trained interviewers. The selection of the participants was based on stratified quota for the Belgian population. The interview included the Pittsburgh Sleep Quality Inventory, the Epworth Sleepiness Scale, the SF-36 and detailed questions on sleep-wake habits, use of products affecting sleep-wake state and general information on familial, professional and social life as well as medical history. In this paper we want to present some preliminary data on a factor analysis that was used to identify factor loadings that encompasses more than just the sleep factor. 5 factors with sufficient eigenvalue were extracted and labeled as follows: 1. ‘Good sleepers’ (N=752 or 75.2%) These are those individuals who fall asleep easily, wake up rarely at night and have no other sleep complaints. Their perceived sleep quality is very good to excellent. On an occasional poor night they attribute this to family and/or professional problems as well as to environmental noise. We find more men in this category and individuals who are more active in general. They present few health concerns. 2. ‘Troubles keep me from falling asleep’ (N=152 or 15.%) Here we find those persons who complain primarily about sleep onset problems. The also say their sleep is agitated and ‘in chunks’. They also wake up during the night or too early in the morning without being able to fall asleep again. They describe themselves as nervous and tense and attribute most of their sleep problems to the familial and/or psychosocial situation. The majority of them claim to have no special medical concerns. This group very rarely will take a sleep promoting product. 3. “Agitated sleep” (N=17 or 1.7%) In this group we find persons who either wake up very frequently during the night, or complain about sleep hydriasis, or have nightmares, or breathing problems. They associate their agitated sleep with nervousness and depression.. They are users of caffeine, go to bed early and have problems to get out of bed in the morning. They use all kind of sleep promoting medications, especially OTC. In this group we find physical and depressive complaints. More woman are found in this group. 4. ‘Long sleepers who do not get enough sleep’ (N=35 or 3.5%). These individuals go to sleep rather early and stay in bed for a long time. They cannot fall asleep easy and wake up at different times. They present with breathing problems and diffuse pain. Sleep quality is very poor. They describe themselves as depressive. They take all kind of sleep promoting agents prescribed by their treating physician. We find again more females in this group in the age group of 50 to 60. They describe their health status as very unsatisfactory. 5. ‘Snorers’ (N=24 or 2.4%). They fall asleep very quickly and without describing any specific sleep complaint they assess their sleep quality as very poor. They snore, cough, and have difficulties breathing during sleep. However they pretend to have no health problems. They often take sleep promoting agents prescribed by their G.P. We find more males in the age group of 50 to 60 who have a liberal profession. 6. ‘Diffuse sleep problems but I try to cope with them’ (N=20 or 2.0%): They have a long sleep onset time and wake up again during the night. They have diffuse pain and cough regularly during the night. They describe their sleep as agitated and say they are ‘nervous’. They have difficulties getting out of bed in the morning. They are reluctant to take sleep-promoting medication. We find more younger males in this group presenting ‘depressive’ complaints. Conclusion: This factor analytic approach presents an overview of the co-morbidity associated with sleep disturbances. It represents an important factor and rationale why to treat patients suffering from insomnia. Future studies are needed to find out if treatment of insomnia can reverse this morbidity that affects patients suffering from insomnia. Acknowledgment: This epidemiological study was funded by Wyeth-Lederle Belgium 210 Actas de Fisiología 7, 2001

SLEEP ARCHITECTURE AND AROUSAL FROM SLEEP IN CHILDREN WITH OBSTRUCTIVE SLEEP APNEA SYNDROME (OSAS)

Marie-Josèphe Challamel, Mathieu Baconnier, Jacqueline Louis, Patrick Froëhlich Centre Hospitalier Lyon-Sud, Pierr Benite, Rhone, France

Abstract: SUBJECTS and METHODS: 58 children (41 males), aged between 20 months and 8 years, with adenotonsillar hypertrophy without any medical condition, were documented by overnight polysomnography for a possible OSAS. Children were divided in 3 groups: Group 1: 29 children with mild or moderate OSAS, IAOH= 2.8 (2.9) ; Group 2 :13 children with severe OSAS, IAOH= 19.9 (17.19) ; group 3: 16 children with no OSAS, IAOH = 0.13 (0.20). Twenty one of the patients with OSAS were re- evaluated after tonsillectomy. Patterns of arousals (>1 sec.) were studied in 10 children with severe OSAS and compared to those of 10 children without OSAS . RESULTS: Sleep parameters were not significantly different between any of the 3 groups. However the quality of sleep was improved after adenotonsillectomy with a decrease in the percentage of stage 1- 2 and an increase in Stage 3-4 of nonREM sleep ( P<0.01 and P<0.05 respectively ). Preliminary results for the study of arousals suggest that patterns of arousal are different in children with severe OSAS when compared with children with no OSAS: with a significant increase in the number of arousals per hour of sleep (median 18.8 versus 8.3). The differences are more pronounced during REM sleep. CONCLUSION: The interest of this study lies in the fact that it was done in a very homogenous cohort of patients ( small range of age, identical physiopathology). Re- sults indicate that the architecture of sleep is affected by OSAS in young children. Actas de Fisiología 7, 2001 211

DQB1*0602 IN NON-NARCOLEPTIC PATIENTS WITH EDS.

J.H.M. de Groen* and J.W.P.H. Soons+ *Center for Sleep- and Wake Disorders Kempenhaeghe, Heeze, Netherlands +Dep. for Clinical Chemistry, St. Anna Hospital, Geldrop, Netherlands

In 92 consecutive patients complaining of Excessive Daytime Sleepiness (EDS) and referred to the sleep center of Kempenhaege because of suspicion of narcolepsy, and in 86 normal subjects, DQB1*0602 typing was done with the PCR-SSP technique. All 92 patients got a MSLT and two nights full polysomnography. If equivocal, the investigations were repeated. Narcolepsy was diagnosed in 40 of the patients. All 40 narcoleptic patients were positive for DQB1*0602, whereas the frequency of the allel in the control group was only 24%. This is not surprising as DQB1*0602 is highly associated with narcolepsy (1). Surprising, however, was the finding that 41% of the 52 patients without narcolepsy were positive. Although not as high as in the narcoleptic patients, the frequency of the allel in the non-narcoleptic patients appeared to be significantly higher than in the controls. Apparently subjects positive for DQB1*0602 are at greater risk of developping EDS, even if there is no narcolepsy. Next, we compared DQB1*0602 positive and negative non-narcoleptic patients as for a number of clinical and polysomnographical parameters: age, age at start of the disease, disease duration, symptoms of the narcoleptic tetrade, ESS score, complaints of sleep maintenance and of dreaming, hypnogram parameters, AH-index, PLM-index, and MSLT parameters. DQB1*0602 positive non-narcoleptic patients appeared to have a significantly decreased percentage of deep NREM sleep and shorter REM-latency, but less sleep attacqs during daytime. Mignot et al.(2) compared DQB1*0602 positive and negative normal subjects and found that positive subjects had shorter REM-latency, increased sleep-efficiency and decreased % of stage I. Hong et al.(3) did the same in narcoleptic patients and found that DQB1*0602 positivity was correlated with more cataplexy, more disrupted sleep, higher incidence of PLM, shorter REM-latency and more MSLT abnormalities. Thus, both in the studies of Mignot and of Hong and in our study the association of DQB1*0602 positivity with a shorter REM-latency is the common denominator. Conclusion: DQB1*0602 modulates sleep, in particular REM-latency.

1. Mignot et al, Sleep 1997; 20: 1012-20 2. Mignot et al, Sleep 1999; 22: 347-352 3. Hong et al, Sleep Medicine 2000; 1: 33-39 212 Actas de Fisiología 7, 2001

PREVALENCE OF OBSTRUCTIVE SLEEP APNEA-HYPOPNEA AND RELATED CLINICAL FEATURES IN THE ELDERLY. A POPULATION-BASED STUDY IN THE GENERAL POPULATION AGED 71-100

Joaquín Durán1, Santiago Esnaola2, Ramón Rubio1, Germán De La Torre1, Joan Sollés3, A Goicolea3 and Vanessa Zorrilla1 1Sleep Unit. Txagorritxu Hospital; 2Research Unit Department of Health; 3School of Pharmacology. Basque Country University. Vitoria. Spain

We conducted a two-phase population-based study, to estimate the prevalence of obstructive sleep apnea-hipopnea (OSAH) in the general population aged 71-100, and living in the city of Vitoria (Spain). Subjects were selected by stratified one-stage cluster sampling. In the first phase, all subjects were submitted to house hold interviews on sleep symptoms (Basic Nordic sleep Questionnaire), and to a nocturnal portable recording device (Mesam IV). In the second phase, all individuals were invited to attend the sleep laboratory for overnight polysomnography (PSG) . Of 1,369 eligible subjects, the first phase was completed by 802 (423 men, 379 women). All of them, were offered to undergo a PSG which was accepted by 428 (53.4%). Based on the questionnaire, 37% of men and 27% of women were habitual snorers. Breathing pauses during sleep were reported by 20% of men and 6% of women, and 16% of men and 13% of women had excessive daytime sleepiness (EDS). Table shows the prevalence of OSAH, according to the PSG results for several apnea-hypopnea index (AHI) cut-off points:

Men Women AHI % (CI 95%) % (CI 95%) > 5 81.3 (76-86) 79.7 (74-86) > 10 67.5 (62-73) 62.1 (55-69) > 15 56.5 (50-63) 48.9 (42-56) > 20 43.5 (37-50) 37.4 (33-44) > 30 26.4 (21-32) 21.4 (15-27) > 10 + EDS 20.5 (14-27) 15.0 (8-22)

Prevalence of OSAH are very high in the elderly and the AHI increases with age. The association of an AHÍ > 10 plus EDS, as definition of OSAH syndrome was found in 20.5% of men and 15.0% of women.

Funded by: FIS (97/0844), Dept. of Health. Basque Government (1998), SEPAR 414/97), Edu- cation Ministry (2FD97-0766-C03-03), and MAP Medizintechnik für artz und patient (GMBH). Actas de Fisiología 7, 2001 213

A RANDOMISED, DOUBLE BLIND, CROSSOVER, PLACEBO-CONTROLLED TRIAL OF A MANDIBULAR ADVANCEMENT DEVICE FOR THE TREATMENT OF SNORING AND MILD OBSTRUCTIVE SLEEP APNEA- HIPOPNEA SYNDROME

Joaquín Durán1, Santiago Esnaola2, Ramón Rubio1, Germán De La Torre1, Ramón Kutz3, Ana Parra3 , Eduardo Anitua3, Silvia Zubia3 1Sleep Unit. Txagorritxu Hospital; 2Research Unit Department of Health; 3 Oral Rehabilitation, TMJ and Dentistry Centre. Vitoria. Spain

We conducted a randomised, double blind, placebo-controlled, crossover study designed to evaluate the effect of a mandibular advancement device (MAD) -KlearwayTM- for the treatment of habitual snoring and mild OSAH defined by an apnea-hipopnea index (AHI) 5-30. As placebo we used the device in centric occlusion without any advancement, which had no effect on mandibular position. With the MAD the mandible was incrementally advanced until the symptoms were resolved or the maximum comfortable limit was reached during an acclimatisation period of 12-18 weeks. After a washout period of four weeks, patients were randomised to undergo three evaluations, basal, placebo and MAD, with intervening washout periods. Outcome measures were subjective responses measured by questionnaire, PSG variables, including AHI, Arousal index (AI), and objective snoring at home by microphone. Complete treatment success (for patients with an AHI > 10) was defined as a resolution of symptoms and an AHI <10/hr. Partial success was defined as a > 50% reduction of AHI. Results: 36 consecutive patients diagnosed of mild OSAH (AHI 5-30 and habitual snorering) accepted to participate, and 33 completed the study (29M, 4F): age (mean + SEM) 46.5 +9.9, BMI 27.3+ 2.9, AHI 13.8+8, AI (17.5 + 0.9). There were no significant changes in BMI or alcohol consumption during the study. The maximum possible mandibular advancement was 8.9 mm + 0.6 and the maximum obtained with the MAD was 8.0 +0.5. The MAD was well tolerated and the compliance was good (6.5 + 0.2 hr/ d placebo and 6.5 + 0.4 hr/d MAD). Compared with placebo the MAD resulted in a significant reduction in AHI (22.5+ 4.5 vs 11.7+ 3.0), AI (25.7+ 2.4 vs 18.7+ 1.4) and in objective snoring (35% vs 10%). Subjective improvements in snoring and sleep quality were reported by 93% and 72%, respectively. Complete success (for patients with AHI > 10) was achieved in 55% and partial success (of all the sample) in 45%. On the other hand 8 (24%) suffered an impairment of AHI with the MAD. We concluded that MAD can be beneficial for some patients with mild OSAH and snoring.

Funded by: FIS (97/0844), Dept. of Health. Basque Government (1998), SEPAR 414/97), Education Ministry (2FD97-0766-C03-03), and Medipro SA. 214 Actas de Fisiología 7, 2001

THE GENDER PERSPECTIVE OF INSUFFICIENT SLEEP AND PERCEIVED EFFECTS OF SLEEP LOSS IN PATIENTS WITH STABLE CORONARY ARTERY HEART DISEASE

1Edéll-Gustafsson U., 2Swahn E., 3Svanborg E., 1Gustavsson G. & 1Kinnerud M. 1Department of Medicine and Care, Nursing Science, Faculty of Health Sciences, Linköping University, Linköping, 2Department of Medicine and Care, Clinical physiology, University Hospital, Linköping, 3Department of Neuroscience, Clinical Neurophysiology, University Hospital Linköping, Sweden.

Few studies have considered effects of sleep loss in females with stable coronary heart disease (CAD). OBJECTIVES: To compare sleep quality and reported psychophysiological effects of sleep loss in women and men with stable CAD. Further, to evaluate whether individual reactions to sleep loss predict psycho-physiological symptoms. PATIENTS: Forty-seven women (mean age 62.0 + 9.6 yr.) and 88 men (mean age 62.7 + 8.1 yr.) with stable CAD were included. METHOD: Structured interviews, measured insufficient sleep, sleep quality and perceived effects of sleep loss. RESULTS: Sleep duration for women was 6.77 + 1.36 hr and for men 6.92 + 1.32 hr. (p=0.67). The women reported significantly worse sleep quality (p<0.0001), longer sleep latency (p<0.03), more initiating sleep difficulties (p<0.02) and physical effects of sleep loss than males. The males reported earlier final morning awakening time (p=0.002). Nearly 75% of the women and 30% of the men complained of insufficient sleep. Compared to those with sufficient sleep, these patients had significantly shorter sleep duration, earlier final morning awakening time, worse sleep quality, and more difficulty maintaining sleep. There were no sex differences between those with insufficient sleep, with exception for sleep latency, which was significantly longer among women (p<0.001). Perceived cognitive changes, altered mood and physical effects of sleep loss were significantly higher among those with complaints of insufficient sleep than not. For those with sufficient sleep the levels correspond to that of men and women in the general Swedish population. Being a women with stable CAD increased the likelihood for worse sleep quality, 3.4 times (OD, 3.4, 95% C.I. 1.4 to 8.2, p=0.007). CONCLUSIONS: Insufficient sleep was more common among women with stable CAD, than amongst men.

Keywords: Insomnia, sleep quality, coronary artery heart disease, gender differences. Actas de Fisiología 7, 2001 215

SLEEP QUALITY IS ASSOCIATED WITH EXHAUSTION AND POOR HEALTH IN A SCHEDULE WITH EXTENDED WORK SHIFTS (15,5H)

Mirjam Ekstedt, Göran Kecklund, Björn Samuelsson, and Torbjörn Åkerstedt

Objective: Poor sleep is often associated with severe fatigue and impaired health. The aim of this study was to investigate how an extreme work schedule (15,5 h shift) relates to individual differences in sleep quality and weather sleep is associated with severe fatigue (exhaustion), subjective health and biological stress markers. Methods: Eighty-three construction workers (mean age = 40 yrs, range 21-62), drill- ing for a road tunnel system, participated in this longitudinal study. The shift cycle started on Wednesday with a single shift (SS: 14.30h -22.00h) followed by two double shifts (DS: 06.45h -22.00h) and a free weekend. They continued with DS Monday and Tues- day, and finished with a single shift on Wednesday (06.45h 14.30 h). The shift cycle ended with a week off. The participants filled in a sleep/wake diary (with questions about sleep, wake, stress, fatigue, workload, and different health complaints) during a shift period, 8 times throughout one year. During this time they also completed a questionnaire at three times. A subgroup of 54 persons wore an actigraph during one shift cycle, and 24 participants gave saliva samples for determination of cortisol (5 samples/day) and blood samples for determination of lipids and testosterone. Results: Individual differences in sleep quality were related to the level of fatigue and biological markers. Subjects having poor sleep (in general) had higher sleepiness on the DS (4.2±0.5 vs. 3.3±0.3, p<0.05) compared to subjects´ rating their habitual sleep quality as good. The poor sleepers also rated the DS as more stressful and had a lower testosterone level (18.8±0.9 vs. 24.1±.09, p<0.001), higher morning cortisol (in saliva; 8.8±0.4 vs. 2.4±0.8, p<0.01) and a higher LDL/HDL ratio (3.1±0.4 vs1.9±0.2, p<0.05). The groups did not differ in age but the poor sleepers had a higher body mass index (25.2±1.2 vs. 21.8±0.8, p<0.05). In a stepwise multiple regression analysis poor sleep efficiency measured with acthigraphy was related to a high level of sleepiness (b=0.64, p=0.008), high testosteron (b=0.88, p=0.004) and cortisol (in saliva) (b=0.37, p=0.050) (R2=0,78, F=7,54, p=0,008). Stress (b=0,35, r2change=0,11, p=<0.007) and commuting time between shifts (b=0,25, r2 change=0,05, p=<0.003) had a negative influence on sleep quality (R2 total = 0,16, F=6,3, p<0.003). Age, and psychosocial factors didn’t correlate with sleep. However, a significant covariation was found for severe fatigue. Sleep quality (b=-0,49, r2=0,22, p<0.001) and work satisfaction (b=0,25, r2change=0.09, p<0.001) correlated with exhaustion (R2 total=0.35, F=13,1, p<0.001). Exhaustion was predicted by biological stress markers; morning cortisol (in saliva r2 =0.35, p=0.015) and low prolactin (in blood, r2change=0.17; R2total=0.51, F=7.9, N=24, p=0.008). Conclusions: The results show that persons who had more sleep problems showed more sleepiness, fatigue and an impaired biological restoration (indicated by a reduction in testosterone) and a higher stress level (higher cortisol level and HDL/LDL ratio). Sleep and psychosocial stress were associated with exhaustion. The differences in biological markers suggest that persons with poor sleep quality are less tolerant to the extended work hours, and may have a higher risk to develop stress related ill health and exhaustion. 216 Actas de Fisiología 7, 2001

RIGHT TO LEFT SHUNTING THROUGH PATENT FORAMEN OVALE CAN OCCUR DURING OBSTRUCTIVE SLEEP APNEA

Beelke Manolo 1, Angeli Silvia 2 , De Carli Fabrizio3 , Nobili Lino 1, Gandolfo Claudio 2, Del Sette Massimo 2, Ferrillo Franco 1 1 Center for Sleep Medicine, DISMR, University of Genoa; Italy; 2 Centro Ictus, Clinica Neurologica, University of Genoa, Italy; 3 Center for Cerebral Neurophysiology, National Research Council, Genoa, Italy

Patent foramen ovale (PFO) can potentially give rise at ischemic stroke by means of paradoxic embolization. In obstructive sleep apnea syndrome (OSAS) right to left shunting (RLS) can occur through PFO during periods of nocturnal apnea similar to Valsal- va maneuver (VM) due to elevated right-sided pressures. The aim of our study was to evaluate the prevalence of PFO by means of transcranial Doppler (TcD) in patients with obstructive sleep apnea syndrome (OSAS) and investigate the presence of RLS during sleep these patients. 78 consecutive subjects with OSAS (mean age 53 ± 12 years) and 89 normal controls (mean age 48 ± 9 yrs.) underwent TcD with injection of agitated normal saline solution mixed with 1 ml of air (contrast medium, CM). The test was performed with patients at rest and during Valsalva maneuver. 10 patients with OSAS (mean age 53 ± 11 years) and prior diagnosed PFO underwent also TcD with CM during sleep. PFO was present in 21 of 78 patients with OSAS and in 13 of 89 control patients. Seventeen of 21 patients with OSAS showed PFO only during VM with respect to 12 of 13 subjects of the control group. Prevalence of PFO in OSAS was statistically different respect to the control group (p<0.05). During sleep obstructive apnea longer than 17 seconds RLS was always present. Prevalence of PFO in subjects with OSAS is significantly higher than in normal controls, however the difference is small. The shunt is frequently present only during VM. In patients with OSAS and concomitant presence of PFO, RLS can occur during sleep when apneas longer than 17 seconds occur. PFO could increase in patients with OSAS the risk for stroke. Actas de Fisiología 7, 2001 217

REVERSAL OF NREM-REM SLEEP PRESENTATION: AN UNUSUAL FINDING IN FATAL FAMILIAL INSOMNIA (FFI)

Arturo Garay Hospital IEAC Neuropsiquiátrico Dr. A.Korn; Centro de Educación Médica e Investigaciones Clínicas (CEMIC); Buenos Aires, ARGENTINA

Since our observation, in a proven case of FFI (1-2), atypical REM sleep preceeding atypical NoREM sleep remained unexplained and unobservable in others reports. The understanding of the pathophysiology of the sleep disorder, the pathogenic mechanisms involved in neuronal loss, and the penetrance of the disease are still unresolved issues in FFI (3-4). Thus, this peculiar finding could help to reevaluate the impact of the pathogenic prion protein (PPrP) on sleep, and over the course of the disease. Briefly, this study was performed in a 42 years-old,male, that developed all the clinical spectrum of FFI with genetic and post-mortem confirmation (1). Interestingly, in addition to typical prion mutations D178N-129M, he had a 24bp deletion in the R3 to R4 region of the octapeptide coding repeats between codons 51 and 91. Reanalysis of PSG’s performed 4 to 1 months prior to death showing a marked reduction of Total Sleep Time (TST): 8.3±3.8% of TRT, REM sleep: 5.3±2.4% of TRT and NoREM sleep:2.4±1.6% TRT. REM sleep onset (6/6 PSG’s) and REMsleep preceeding NREM (3/6 PSG’s) were observed during these studies. Non significant differences were noted between REM cycles (RCy1:98.1±30.4min; RCy2:96.1±31.6 min). A possible explanation to these findings necessarily include, the loss of spindling activity, with the consequent inability to initiate physiological sleep, thus allowing the appearence of REM sleep that occurs typically in FFI (5-7). A recent report, which demonstrates the influence of met/met polymorphism at codon 129 on the decrease of the spindling activity, support the suggestion that an early expression of pathogenic PrP is related to the appearence of this pattern (8). Moreover, the preservation of REM sleep cycling, and the presence of NREM periods also indicates the persistence of ultradian and homeostatic mechanisms in the later course of the disease. Recent findings demonstrating a role of deletions in the modulation of prion replication and disease presentation might help to understand this phenomenom (9-10). Since deletions of the octapeptide repeat is an uncommon polymorphism in FFI, the unanswered question of whether the reversal of the NREM-REM sleep could depend on a different expression of the pathogenic PrP linked to this polymorphism should be considered.

References: 1) Reder A.T. et al. Neurology 1995; 45:1068-1075; 2) Garay A. et al. Neurology 1994; (Sup- pl.2):218-219; 3) Delisle M.B. et al. Clin. Exp. Pathol 1999; 47(3-4):176-80; 4) Cortelli, P. et al. J. Sleep Res. 1999; 8 (Suppl.1):23-9; 5) Esforza E. et al. Electroenceph. clin. Neurophysiol. 1995; 94: 398-405; 6) Lugaresi E. Neurology 1992; 42(Suppl. 6):28-33; 7) Lugaresi E. et al. Brain Pathol. 1998; 8(3):521-6; 8) Plazzi, G. et al. Neurology 2001; 56(Suppl.3):A9; 9) Flech- sig E. et al. Neuron 2000; 27(2):399-408; 10) Mastrianni J.A. and Roos R.P. Sem. Neurol. 2000; 20(3): 337-352. 218 Actas de Fisiología 7, 2001

SEVERITY OF DISEASE AND CIRCADIAN PATTERNS OF PERIODIC LIMB MOVEMENT ACTIVITY IN RESTLESS LEGS SYNDROME: A PRELIMINARY REPORT ON 23 PATIENTS

Diego Garcia-Borreguero, Yolanda de la Llave, Soledad Barrio, Oscar Larrosa, Juan J. Granizo and Richard P. Allen Fundación Jimenez Diaz, Av.Reyes Católicos 2, Madrid, Spain

Introduction: The presence of periodic limb movements during sleep (PLMS) is considered to be the main objective feature of Restless Legs Syndrome (RLS) (1). In addition, RLS patients also exhibit Periodic Limb Movements during Wakefulness (PLMW)(2). Although both PLMS and PLMW seem to be modulated by a circadian oscillator (showing maximal activity in the early night hours)(3), it is not known to what extent circadian activity reflects the severity of the disease. The main objective of this study was to investigate the relationship between circadian variation of periodic leg movement activity and severity of disease. Methods: 23 untreated patients diagnosed with RLS according to the criteria of the IRLSSG (1) were included in the study. Diagnostic procedures included medical history, physical exam, laboratory analysis, and polysomnography. All patients underwent between 22:30-24:00 and 7:00-9:00 a 45-60 minute Suggested Inmobilization Test (SIT). Patients were asked to lay down in bed and only to move if they experienced symptoms (3). The following parameters were calculated for each patient based on the PSG, SIT and IRLS rating scale: PLMS-index, PLMW-index (2), circadian index (CI) on PSG for both PLMW and PLMS (difference of PLM-index between the first and last two hrs of PSG), CI on SIT (difference in PLMW between night and morning SIT), IRLS (self-rating score). Statisti- cal analysis was performed by means of Spearman´s correlations. Results: Preliminary results on 23 patients (mean age + S.D.: 52y +13) of this ongoing study were: a) A significant correlation was found between IRLS self-rating score and PLMW (p<0.05). b) A significant correlation was found between the indexes of PLMW and PLMS (p<0.01). c) There was a marginally significant association between circadian index on SIT and PSG: CI-SIT with CI-PLMS (p<0.05) and CI-SIT with CI-PLMW (p< 0.1). d) No association was found between either PLMW or PLMS and circadian indexes (on either SIT or PSG). Conclusions: Our data do provide external validation of the IRLS severity scale, suggesting a good correlation between subjective severity and PLMW. A significant association was found between all-night indexes of PLMW and PLMS, supporting a common mechanism. However, no association was found between all-night indexes of PLMS or PLMW and indexes of circadian variation, supporting the view that severity of disease and circadian variation of symptoms might be caused by separate mechanisms, and represent two different dimensions in the assessment of RLS.

References: 1.Walters AS. Toward a better definition of the restless legs syndrome. The International Rest- less Legs Syndrome Study Group. Mov Disord 1995;10:634-642. 2.Nicolas A, Michaud M, Lavigne G, Montplaisir J. The influence of sex, age and sleep/wake state on characteristics of periodic leg movements in restless legs syndrome patients. Clin Neu- rophysiol 1999;110:1168-1174. 3.Hening WA, Walters AS, Wagner M, et al. Circadian rhythm of motor restlessness and sensory symptoms in the idiopathic restless legs syndrome. Sleep 1999;22:901-912. Actas de Fisiología 7, 2001 219

CIRCADIAN VARIATION IN NEUROENDOCRINE RESPONSE TO THE ADMINISTRATION OF L-DOPA IN PATIENTS WITH RESTLESS LEGS SYNDROME: A Pilot Study

Diego Garcia-Borreguero, Oscar Larrosa, Teresa Saiz, Cristina Horcajada, Yolanda de la Llave, Soledad Barrio and Wayne Hening Fundación Jimenez Diaz, Av.Reyes Católicos 2, Madrid, Spain

Introduction: Several lines of evidence suggest that a central dopaminergic dysfunction might play a keyrole in the pathogenesis of Idiopathic Restless Legs Syndrome (I- RLS) (1,2). The severity of RLS seems to be modulated by an independent circadian process, with an increase of both motor and sensory symptoms in the evening and at night (3). Dopaminergic function has ben found to exhibit circadian variations in humans both in CSF and in plasma. The objective of this study was to investigate circadian changes in dopaminergic function in patients with RLS. Methods: Preliminary data of an ongoing study include 7 patients diagnosed with I-RLS according to the criteria of the IRLSSG and 4 healthy volunteers. Diagnostic procedures included medical history, physical exam, laboratory analysis, iron plasma levels and polysomnography. All patients had been diagnosed recently and had never received dopaminergic treatment. Following randomization for the order of treatment (night first vs. morning first), patients underwent the L-DOPA challenge at two different times (2300 and 0700), at least 7 days apart. After an insertion of an antecubital iv line for blood drawing, patients were asked to lie down in bed for 90 min. An oral administration of 200 mg l- dopa/50 mg carbidopa was provided at 2300 one occasion, and at 1100 on the other. Blood was drawn 20 and 5 minutes before administration of the drug, as well as 15, 30, 45, 60, 75, 90, 105 and 120 minutes after, and analyzed for plasma values of GH and prolactine (PRL). Statistical analysis was performed both on the time series of (baseline corrected) hormonal plasma values as well as on the area under the response curve (AUC) by means of non-parametric test (Wilcoxon test). Results: Preliminary results (patient´s age: 56.7 + 13.3) are shown on the attached graph. No statistically significant differences were seen between 2300 and 0700 baseline values of GH and PRL. Following nighttime administration of L-DOPA, an increased nighttime responses of GH at mins 75 (p<.05) and 90 (p< .1) was seen in I-RLS but not in the control group. So far, no group differences were seen on PRL responses. Conclusions: Our preliminary data of an ongoing study suggest an enhanced stimulation of GH and inhibition of PRL following nighttime administration of L-DOPA in RLS patients, supporting an increased sensitivity at night of dopamine receptors in I- RLS patients. We hypothesize that these changes in dopamine receptor sensitivity reflect circadian variations in dopaminergic function, with a nighttime hypofunction at night. An increase of the sample size is warranted. 220 Actas de Fisiología 7, 2001

PERIODIC AROUSALS OR PERIODIC LIMB MOVEMENTS DURING SLEEP?

José Haba, Luc Staner And Jean Paul Macher FORENAP 27, Rouffach, France

A 44 year-old woman presenting with symptoms of poor sleep quality, underwent three consecutive nights of sleep recordings. The first night she had typical PLMs, often associated with arousals, with an index of 20 movements per hour of sleep. During the second night she had rare PLMs but “periodic arousals” with an interburst interval corresponding to those observed the night before (index of 20 arousals per hour of sleep).The third night, PLMs were again present (index of 18 PLM per hour of sleep). This observation suggests that limb movements in PLM Disorder are not primary but raher a phenomenon associated to an underlying arousal disorder. Actas de Fisiología 7, 2001 221

REDUCED BRAIN AND CSF IRON CONCENTRATION IN RLS WITH PLMS: A 3.0 T MRI ESTUDY

José Haba, Luc Staner, Thomas LOENNEKER, Francisek HENNEL and Jean Paul Macher FORENAP 27, Rouffach, France

Four subjects presenting RLS with PLMs and four age and sex matched controls underwent MR imaging with a 3.0 T system. The severity of the symptoms were assesed by the Questionnaire of the IRLSSG and PLMs index. Relaxation rates were lowest in patients in the Caudate Nucleus (R2'=0.057+/-0.01 kHz vs 0.109+/-0.06 kHz) and in intraventricular CSF (R1=3,29+/-0.1 sec vs 4.03+/-0.35 sec).CSF iron estimation was correlated with severity scores (r=-0.8664;p<0.5).No differences were found in the Substantia Nigra, Putamen, Pallidum and Red Nucleus.This preliminary study agree with the hypothesis that an altered management of iron in the brain can produce the dopaminergic dysfunction seen in RLS and PLMs patients. 222 Actas de Fisiología 7, 2001

QUALITY OF SLEEP AMONG NEW IMMIGRANTS IN ISRAEL

Jack Hadjez, Ilan Modai, Alexander Ponizovsky, Yael Ratner and Michael Ritsner Institute for Psychiatric Studies, Sha’ar Menashe Mental Health Center, Mobile Post Hefer 38814, Hadera, Israel

Ninety newcomers and 35 Israeli native controls completed the Pittsburgh Sleep Qual- ity Index (PSQI), a self-rated questionnaire for assessing sleep quality and disturbances, the Beck Depression Inventory (BDI), Perceived Immigrant Stressor Scale, and Multidimensional Scale of Perceived Social Support (MSPSS). T-tests and multiple regression analyses were used to examine the relationships between parameters of the interest. Sleep disturbances were revealed in 44.4% of immigrants, but only in 20% of the controls (p<.01). Compared to controls, immigrants reported greater sleep disturbances (1.1 vs. 0.9 p<.05), less sleep duration (0.9 vs. 1.5 p<.01), and greater daytime dysfunction (1.1 vs. 0.6 p<.01). Immigrants reporting the high PSQI scores (>5) also reported moderate levels of depressive symptomatology according to the BDI rates (>8). The best predictors for sleep disturbances were depressive symptoms, lack of friend support and information-related stressors that accounted for 47%, 17% and 6%, respectively, of the total variance in the PSQI scores. Results of the study suggest that the sleep quality disturbances are common among recent immigrants, and are associated with depressed mood, lack of friend support and information deficits. Actas de Fisiología 7, 2001 223

DAYTIME SLEEPINESS (AS MEASURED BY MWT) AND INITIAL NCPAP COMPLIANCE AS PREDICTORS OF NCPAP TREATMENT ADHERENCE

Raphael Heinzer, Emilia Sforza*, Jean-Paul Janssens, Pascal Lebas, Thierry Rochat Division of Pulmonary Diseases, Geneva University Hospital, Switzerland *Department of Psychiatry, Geneva University Hospital, Switzerland

Aim of the study: To determine the influence of daytime sleepiness as measured by Maintenance of Wakefulness Test (MWT) and Epworth Sleepiness Scale (ESS), apnea- hypopnea index (AHI), age and initial compliance to nasal CPAP (nCPAP) on the probability of pursuing nCPAP treatment. Methods: Retrospective analysis based on a computer database including all patients treated by nCPAP from 1991 to 2001 in Geneva hospital outpatient clinic. Data base also includes regular measurements of compliance through nCPAP counters with a first control 1-4 months after nCPAP initiation. Daytime sleepiness was assessed with MWT and ESS on 47 consecutive patients. Patients were separated in two groups according to their adherence to nCPAP (interrupt or pursue). Probability of pursuing nCPAP according to MWT, ESS, age and CPAP using time at first control was determined with Kaplan Meier curves. The difference in sleepiness, age, initial compliance and AHI between the two groups of patients was assessed by unpaired t-tests. Patients: 530 patients from Geneva area (aged 57,8 +/-12 years, range 27-87; 77% men 23% women) treated by nCPAP were studied. Mean follow-up was 29.2 (+/- 24) months. Results: Considering the 47 patients who had sleepiness assessments, MWT was correlated with apnea hypopnea index (Spearmann: P=0.0192) and with ESS (Spear- mann : P= 0.0053). MWT was higher (p =0.027) and ESS lower (p= 0.05) among patients who interrupted their nCPAP treatment. CPAP usage at first control was also lower (p=0.019) for these patients. Age, AHI, BMI were not different between the two groups. Probability of interrupting nCPAP was significantly higher (p< 0.043) among patients with a MWT over 25 min than for patients with MWT < 25 (respectively 50% vs 13% interruption rate) Considering all patients of the database (530), a lower compliance to nCPAP at the first control (<3 hours of CPAP usage) was associated with a higher rate of treatment interruption (54% vs 19.9% p < 0.001). Neither age with a threshold of 65, 70 or 75 years nor ESS with a threshold of 10/24 could predict the pursuing of CPAP treatment. Conclusion: Predicting which patient will be adherent to nCPAP treatment is a challenge for the clinician. These data suggest that a higher daytime sleepiness as measured by MWT at baseline (with a threshold of 25 min) and a nCPAP usage of more than 3 hours at first control are associated with a lower treatment interruption rate. 224 Actas de Fisiología 7, 2001

SEASONALITY AS MEASURED BY THE GLOBAL SEASONALITY SCORE IN 40-45 YEAR OLD MEN AND WOMEN IN A NORWEGIAN COUNTY

Fred Holsten, Bjørn Bjorvatn and Reidun Ursin Department of Psychiatry, Department of Public Health and Primary Health Care, and Department of Physiology, University of Bergen, Norway

The data collection of this study was conducted as part of HUSK (the Hordaland Health Study ’97-’99) in collaboration with the Norwegian National Health Screening Ser- vice, the University of Bergen and local health services. The study population included all individuals in Hordaland county born 1953-57 (29.400). A total of 8598 men and 9983 women participated, yielding a participation rate of 57 % for men and 70 % for women. Self-administered questionnaires provided information on various health behaviors. Also, baseline measurements including height, weight, blood pressure etc were performed. Subgroups of 3432 men and 8113 women answered the Global Seasonality Score (GSS) questionnaire from the Seasonal Pattern Assessment Questionnaire (SPAQ) which is used for evaluation of possible Seasonal Affective Disorder (SAD). Six questions on seasonal changes in sleep, mood, weight, energy, social activity and appetite were scaled 0 for no seasonality, 1 for light, 2 for moderate, 3 for marked and 4 for strong seasonality. High GSS score (> 11), suggesting possible SAD, were seen in 18.4 % of the men and 22.2 % of the women, while 18.6 of the men and 18.0 % of the women had GSS score between 8 and 10 suggesting sub-SAD. Frequent daytime sleepiness was higher in subjects with high (> 11) GSS score ( 20% in men and 23 % in women) than in those with low (0-7) GSS score (8% in men and 11 % in women). Forty-two per cent of men and 47 % of women with high score (> 11) reported that they did not get enough sleep, vs. 29 % and 31 % of those with low GSS score. Insomnia (as defined by difficulty sleeping once or more times per week) was reported by 16 % of men and 21 % of women with GSS score > 11 vs. 7 % of men and 10 % of women with low score. Suicidal thoughts during the last 12 months were reported by 11 % of men and 12 % of women with GSS score > 11 versus 5 % of both men and women with low GSS score. Actas de Fisiología 7, 2001 225

DISTURBED NOCTURNAL SLEEP IN NARCOLEPSY PATIENTS

Lois E. Krahn MD, Michael Silber MBChB Mayo Sleep Disorders Center, Mayo Clinic, Rochester MN

Introduction: The two most common symptoms of narcolepsy are excessive daytime sleepiness (100% of patients) and cataplexy (64%) (1). The less common symptoms include disturbed major sleep period, sleep paralysis, hypnagogic hallucinations and automatic behaviors. Disturbed nocturnal sleep has received less attention in narcolepsy than other symptoms possibly because it was not included in the traditional “tetrad” of core symptoms (2). Furthermore disrupted nocturnal sleep is a non-specific symptom which can be caused by periodic limb movements and obstructive sleep apnea, conditions which are known to be more common in narcolepsy (3). Although hypnotics are sometimes prescribed to improve nocturnal sleep (4), (in addition to psychostimu- lants for excessive daytime sleepiness, antidepressants for cataplexy and optimal sleep hygiene), few published reports address this practice. Methods: A retrospective study was conducted of polysomnographic (PSG) data and naturalistic follow-up data collected on 55 narcoleptic patients. Patients were not taking any medications that might influence the PSG. Patients with an apnea hypopnea index > 10 were excluded. Patients with and without cataplexy were compared. Results: Cataplexy was present for 28 patients (51%). The mean age of patients with cataplexy was 47 yrs ± 18 vs 30 yrs ± 13 (p <0.001). Total arousals per hour for patients with cataplexy were 22.1 ± 14.7 vs 13.5 ± 7.8 (p <0.001), sleep efficiency 83.6 % ± 11 vs 90.7 ± 6.5 (p < 0.005) and periodic limb movements per hour 23.6 ± 28.89 vs 9.7 ± 13.7 (p <0.05). There were no significant differences in initial REM latency, wake after sleep onset, movement arousals and breathing arousals. All patients were prescribed a psychostimulant. Hypnotics were prescribed equally to patients with or without cataplexy, however, dopaminergic agents were more likely provided to patients with cataplexy. Discussion/Conclusion: Patients with cataplexy in this sample appear to have less satisfactory sleep during PSG compared to those without cataplexy. These patients conceivably have a more severe form of narcolepsy. Understanding the severity and cause of disturbed sleep in narcolepsy aids clinicians in deciding whether and how to treat this under-recognized symptom of narcolepsy.

References 1. Silber MH, Krahn LE, Olson EJ, Pankratz VS. Epidemiology of Narcolepsy in Olmsted County, MN: A Population-based Stud (abstract). Sleep 2001; S 24: A98. 2. Yoss RE, Daly DD. Criteria for the Diagnosis of the Narcoleptic Syndrome. Proc Staff Meet Mayo Clin 1957; 32:320-328. 3. Harsh J, Peszka J, Hartwig G, Mitler M. Night-time sleep and daytime sleepiness in narcolepsy. J Sleep Res 2000; 9:309-316. 4. Yoss RE, Daly DD. On the Treatment of Narcolepsy. Med Clinics of North America 1968; 52 (4): 781-786. 226 Actas de Fisiología 7, 2001

GENETIC ASPECTS OF SLEEPWALKING

Michel Lecendreux1, Claudio Bassetti2, Geert Mayer3, Yves Dauvilliers4, Elisabeth Neidhart5, Mehdi Tafti5 1Service de Psychopathologie de l’Enfant et de l’Adolescent, Hopital Robert Debré, Paris-France; 2Neurology Policlinic, UniversitätsSpital, Zürich-Switzerland; 3Hephata Klinik, Schwalmstadt-Treysa-Germany; 4Service de Neurologie B, Hôpital Gui-de-Chauliac, Montpellier-France; 5Neurophysiology Unit, Department of Psychiatry, University of Geneva-Switzerland

Sleepwalking (SW) is a highly frequent sleep disorder of childhood affecting up to 20% of children. Although it has been suggested that SW generally disappears at adul- thood, recent data in twins indicated that adult cases are found in a proportion of 1 to 3% and most of them have suffered from SW since childhood. Epidemiological sur- veys including familial and twin studies, have shown a strong genetic implication in SW. Prevalence of SW in first degree relatives of an affected subject has been estimated at least ten times greater than that in the general population. HLA DQB antigens are known to be associated with disorders of rapid eye movement (REM) sleep, narcolepsy (DQB1*0602) and REM sleep behavior disorder (DQBw1), both characterized by high parasomnia comorbidity. Therefore, we tested the hypothesis as to whether specific HLA DQB subtypes may also be associated with sleepwalking, the most prominent NREM parasomnia. To date 57 patients and their families, 42 sporadic and 15 familial cases, have been included from 4 European sleep centers. Forty-five ethnically matched subjects recruited from the same sleep centers and without any diagnosed sleep disorder constitute the control group. Among HLA-DQB1 subtypes tested, DQB1*05 positivity (number of subjects) was 28 in SW and 11 in control subjects (?2=6.5, p=0.01). In sporadic cases, DQB1*05 positivity was 23 and 11 in SW versus control subjects (?2=8.4, p=0.0038). In familial cases, DQB1*05 positivity was not significantly associated with SW (Fisher test, p=0.51). Four sporadic cases were homozygous DQB1*05 (allele frequency comparison with controls: 27/84 vs 11/90, ?2=10.1, p=0.0015, OR=3.4, 95%CI=1.6-7.4). This result remained highly significant even after correction for multiple testing (12 alleles, p=0.018). Further analyses in nuclear families should reveal whether or not DQB1*05 is preferentially transmitted to SW patients. Our data confirm a strong implication of genetic factors in SW, in particular HLA class II genes. Since DQB1*05 has also been involved in REM sleep behavior disorder, a common genetic predisposition to SW and REM sleep behavior disorder may explain the coexistence of both disorders in some patients with the so-called parasomnia overlap disorder. More generally, these findings suggest that DQB1*05 may be implicated in disorders of motor control during sleep. Actas de Fisiología 7, 2001 227

THE ETIOLOGICAL EXPLORATION OF NORMALISATION OF BRAIN SHRINKAGE BY A POTENT WAKING MOLECULE

Liu Shiyi and Ye Yunhua Shanghai Yuake Institute of Sleep & Anti-dementia, School Of Chemical & Molecular Engineering, Peking University, China

Alzheimer’s disease(AD) in general refers to the disease of increasing brain shrinkage with a prevalence of 5% after 65 and 30% after 85 and its incidence enhances exponentially after 90 years. Since the aged population grows worldwide in the 21th century,, AD serves as a major public health problem for both developed and developing natio- ns. Early studies mainly focus to the neurotransmitter acetylcholine (Ach), so most drugs (e.g. tacrine, donapezil) for the treatment of AD are cholinergic, especially cho- linesterase inhibitors (ChEi). It seems that no such ideal drug has been found until now (Nash, 2000). One reason is that they are related to “symptomatic” rather than “etiologic” treatment. The pathological hallmarks of AD are most frequently localized in the temporal lobes, hippocampus, amydaloid nuclei,etc., where senile plaques (SPs) and neurofibil- lary tangles (NFTs) are also mostly take place. But more recent studies revealed that the etiologic cause of brain shrinkage is more related to the decrease of glucose turnover (glycolysis) associated with deformation or apoptosis of mitochondria in cells of these regions(Moley & Mueckler, 2000). Thus, the decrease of of glucose transporta- tion – ATP depletion – mitochondria apoptosis is linked to the earliest steps of pro- grammed cell death in brain shrinkage. It is known that beside PGE (Morairty et al.,1995), Qus5-DSIP (Xu and Liu, 1992) and modafinil (Edgar et al, 1995),2 etc., less information has been reported with regard to waking substances. Based on the isolation from the roots of wild Panax ginseng C.A.Meyer associated with structural modification, we recently found that a potent crystalline molecule GGE-3 in microgramme exerts consistent and significant waking effects in animals. It can penetrate the blood-brain barrier (BBB), enhance cerebral blood flow (CBF), induce long-term potentiation (LTP) in aged rats (SD, >17-18 months), etc. only by a single administration of GGE-3 (100 mg/kg for i.v., 150 mg/kg, for i.p. or p.o.). It is also worthy of note that consecutive daily oral administration of GGE-3 for one month (equivalent to 150 mg/kg. p.o., 1m) can normalise the vacuoliza- tion, paucity of glucogen granules and deformation of mitochoddia, etc. in cells of the temporal regions of aged mice (KM, >11-12 months), when compared to aged controls (0.9% NaCl, p.o., 1m).

(1) Liu S.Y., 2000 Congress of Chin. Assoc. of Sci. & Techn. 2000, 1117, Xian. (2) Liu S.Y. .3rd ASRS Congress, 2000, A94, Bangkok. 228 Actas de Fisiología 7, 2001

SLEEP APNEA SYNDROME. CLINICS DIAGNOSIS TREATMENT

López Varela M.V., Anido T., Otaño N., da Rosa A. Pulmonary Function Laboratory. Centro de Asistencia del Sindicato Médico del Uruguay

Introduction: Obstructive Sleep Apnea Syndrome (OSAS) has become a common diagnosis due to a better knowledge of this entity and the availability of less expensive validated simplified studies. The diagnosis is suggested by clinics and confirmed by polysomnography (PSG). The treatment of airway collapse is through continous positive airway pressure (CPAP). We started to study and treat sleep disorders at our Lab- oratory in december 1997. Objective: We present the findings on clinic, diagnosis and treatment of patients studied at our Laboratory up to July 2001. Methods: 233 patients were studied with questionnaire, Epworth Sleepiness Scale (ESS) and simplified unattended home PSG. We titrated with autoCPAP those patients with an Apnea-Hypoapnea Index (AHI) over 10 and then controlled the response to treatment with ESS. We evaluated the number of studies in the last year. Results: 175 patients (75%) diagnosed as OSAS compared with 58 (25%) with a negative study showed: mayority of males (77% vs 57%), higher AHI (36 ± 21,7 vs 4,9 ± 8,7, more respiratory pauses seen by the sleep partner (86% vs 55%), morning headaches (45% vs 22%) sleep movements (47% vs 22%), arterial hypertension (58% vs 34%) and higher ESS (10.6± 6.8 vs 7.4 ± 8.6). In 39 patients with OSAS who started treatment with CPAP with pressure titrated of 9.53 ± 1.33 and efective pressure of 9.55 ± 1.39 (r=0.83, p=0.001) we evaluated somnolence with ESS. There was a significant change in the scale (p=0.0001 t test) value pre treatment = 13.39 ± 4.61, post treatment 1.71 ± 2.40. The increase of studies to diagnose OSAS was of 40% during 2000-01. Conclusions: The OSAS is an increasing cause of consults. Patients are more often men, with hypertension, daytime sleepiness, morning headaches, respiratory pauses and movements during sleep. Treatment on CPAP titrated with auto CPAP is effective as evaluated through the ESS. Actas de Fisiología 7, 2001 229

POLYSOMNOGRAPHIC AND PSYCHOMETRIC MEASURES IN OCD: A FACTOR ANALYSIS

A. Matos-Pires, F. Cavaglia, A. Atalaia, E. Lara and F. Arriaga Department of Psychiatry, Faculty of Medicine of Lisbon and Sleep Unit, British Hospital, Lisbon, Portugal

Objectives: The aim of this study was to identify and interpret eventual factors extracted from a set of measures referring to a group of OCD patients. We think that sleep EEG can be helpful in the nosological positioning of OCD. Methods: OCD cases were outpatients attending a psychiatric clinic of a university hospital. Patients were selected according to DSM IV criteria (APA, 1994)for “obsessive compulsive disorder” (OCD) (n=14). Their age ranged between 20 and 61 years (mean = 33.29; SD= 12.9). Depressive comorbidity was excluded. All patients were medication free. The following evaluations were applied: Eysenck Personality Inventory (EPQ); Oswald and Norris scales, Symptom-Check-List-90-Revised (SCL-90); Hamilton Anxi- ety Rating Scale (HARS); Hamilton Rating Scale for Depression (HRSD); Beck Inven- tory for Depression (BID) and the Yale-Brown Obsessive Compulsive Scale (Y-BOCS). Sleep records were scored according to Rechtschaffen and Kales criteria (1968) by a trained electroencephalographer. Factor analysis and principal components analysis was used to study the interrela- tionships among the variables without reference to a criterion. The rotation method chosen was the Varimax. Results: Our results showed that the Y-BOCS (p=0.938), Y-BOCS Compulsions (p=0.852) and Y-BOCS Obsessions (p=0.908) loaded highly on factor 2. The same happened with initial insomnia (p=0.719), total sleep period (p=-0.847), sleep latency (p=0.893), phase 4 latency (p=0.792) and phase 3 latency (p=0.769). Factor 1 was loaded by the following measures: SCL psychoticism (p=0.96), SCL interpersonal sensitivity (p=0.957), SCL symptomatic index (p=0.943), SCL total score (p=0.909), morning awake (p=0.864), EPQ psychoticism (p=0.845) and SCL paranoia (p=0.768). Conclusions: After examining which variables load highly on each factor, we think that factor 1 seems to address a psychotic domain while factor 2 relates mainly with the obsessive-compulsive area. We would expect that the obsessive-compulsive measures would load on factor 1, which would be the most successful in describing the variation among the variables. Instead, we observe an unexpected relevance of the so-called “psychotic measures”. Is this an indicator that OCD is wrongly placed within the anxiety disorders and that is true affinity is with the “thought disorders”? The severity of OCD disease encompasses a troubled sleep induction, an association already approached by Insel et al (1982). He considered (like we do) the disrupted sleep induction as consequent to the existence of obsessions.

References - RECHTSCHAFFEN A, KALES A. A Manual of Standardized Terminology, Techniques and Scor- ing Systems for Sleep Stages of Human Subjects.. Washington DC: US Public Health Service, Government Printing Office, 1968. - APA. Diagnosis and Statistical Manual of Mental Disorders (4th ed.). Washington, DC: American Psychiatric Association, 1994. - INSEL TR, GILLIN JC, MOORE A, MENDELSON WB, LOWENSTEIN RJ, MURPHY DL. The sleep of patients with obsessive compulsive disorder. Arch Gen Psychiatry 1982; 39: 1372-1378. 230 Actas de Fisiología 7, 2001

VALUE OF AMBULATORY POLYSOMNOGRAPHY FOR DIFFERENTIAL DIAGNOSIS OF PSYCHIATRIC DISORDER IN COMMON SLEEP COMPLAINS

Christine Norra1, Ralf Kaminski1, Stefan Züchner2, Ann Assmus2, Marcel Hungs3, Alexander Gundel4, Rudolf Töpper2 1)Department of Psychiatry and Psychotherapy & 2) Department of Neurology, Medical Sleep Center, Technical University of Aachen, Germany, 3) Stanford Sleep Research Center, Stanford University, Palo Alto, CA, U.S. 4) Institute for German Aerospace and Navigation, Cologne, Germany

Objective: Psychiatric disorder are among the most common causes of sleep disturbances therefore sleep complains need distinction between primary sleep disorder or mental disorder (1). Ambulatory polysomnography is established in sleep medicine with comparable technical quality to sleep laboratory (2,3) and may offer specific advantages in lowering costs and capacity problems of sleep laboratories. Method: 50 non-diagnosed referrals with main symptoms of hypersomnia and insomnia (25 of each group) were consecutively recruited (33males, 12 females, mean age 47.95) for a telematic controlled validation study with Vitaport 2. Polysomnography was located twice in the lab and twice at home. Apart from clinical examination, completion of sleep diaries and psychometric testings 44 subjects agreed to undergo semi-structured psychiatric evaluation. Finally medical, sleep and psychiatric diagnoses (ICSD, ICD-10, DSM-IV) were concluded by all main investigators. Results: More than ½ of participants revealed some form of psychiatric disturbances: Almost 1/3 of hypersomniacs showed comorbidity with mood and somatoform disorder, whereas 7/10 of the insomniacs suffered from psychiatric disorder, predominantly mood, anxiety/adjustment and personality disorder. 2/3 of the latter group were diagnosed insomnia secondary to Axis I psychiatric disorder, another 1/3 showed sleep apnea, restless legs syndrome or narcolepsy. - Altogether 2/3 of patients diagnosed with psychiatric disorder never had experienced any previous psychiatric contact nor treatment. Conclusions: Enhanced psychiatric morbidity in subjects with common chronic sleep complains (4,5) underline the relevance of psychiatric screening or co-investigation to polysomnography in order to initiate adequate therapy (6). Due to easier access and home based advantages ambulatory polysomnography could qualify as substantial and beneficial tool in this field by improving and accelerating differential diagnosis of sleep complains.

References 1. Benca RM, Obermeyer WH, Thisted RA, Gillin JC (1992) Sleep and psychiatric disorders: a meta-analysis. Arch Gen Psychiatry 49:651-668. 2. Edinger et al. (1997) Sleep in the laboratory and sleep at home: comparisons of older insomniacs and normal sleepers. Sleep, 20:1119-1126. 3. Züchner S., Kaminski R., Hungs M., Norra Ch., Gundel A., Töpper R. (1999) Telematic controlled ambulatory polysomno-graphy is comparable to the sleep laboratory. Sleep Res Online 2:592. 4. Buysse DJ, Reynolds CF3rd, Kupfer DJ, Thorpy MJ, Bixler E, Manfredi R. Kales A, Vgontzas A, Stepanski E, Roth T et al. (1994) Clinical diagnoses in 216 insomnia patients using the International Classification of Sleep Disorders (ICSD), DSM-IV and ICD-10 categories: a report from the APA/NIMH DSM-IV Field Trial. Sleep 17,7:630-637. 5. Ohayon MM (1997) Prevalence of DSM-IV diagnostic criteria of insomnia: distinguishing insomnia related to mental disorders from sleep disorders. J Psychiatr Res 31,3:333-346. 6. Chokroverty S (2000) Diagnosis and treatment of sleep disorders caused by comorbid disease. Neurology 54,5 supl1:8-15. Actas de Fisiología 7, 2001 231

MORBID OBESITY AND OBSTRUCTIVE SLEEP APNEA SYNDROME

Rosa Peraita-Adrados, Pilar López-Esteban Unidad de Sueño y Epilepsia-Neurofisiología Clínica. Hospital General Universitario Gregorio Marañón. Madrid

AIM: Retrospective study to determine the incidence of OSAS in a series of morbidly obese patients who presented to the Sleep Unit, prior to surgery for obesity. The selection of patients for surgery was made without taking into account the results of sleep recordings. PATIENTS AND METHODS: We evaluated 61 patients (24 males, 37 females, mean age 41.42 +/- 11.8 years (range 21-70). For each patient the following were made; a medical report, sleep questionnaire, Epworth Sleepiness Scale, anthropometric measures (weight, height, body mass index (BMI)) and blood pressure. The mean BMI was 52.3 +/- 13.6 (range 33-93). A standard video PSG (EEG, EOG, ECG, submental and Tibialis anterior EMGs, oro-nasal airflow, thoraco-abdominal respiratory effort and SaO2) was performed in each patient. 27 patients (44.2%, 15 males and 12 females) underwent subtotal gastrectomy and biliopancreatic shunt (modified Scopinaro technique) to correct their obesity. RESULTS: 26 patients (42.7%, 9 males and 17 females) did not present OSAS. In 35 patients (57.3%, 20 males and 15 females) we found a moderate-to-severe OSAS. RDI=38.7+/- 23.3 (range 10-99) and in 24 cases (75%, 11 males, 13 females) we started CPAP treatment prior to surgical intervention in order to correct apneas and nocturnal hypoventilation associated with obesity. In 6 patients who underwent surgery, we carried out a post-surgical control a year after the operation by means of nocturnal pulse-oxymetric recording. The BMI was 35.1 +/- 1.7 and in all patients we observed a remission of OSAS. CONCLUSIONS: 1.The incidence of OSAS in morbidly obese patients is very high in our series (57.3%) in relation to the normal population. 2. The percentage of OSAS in morbidly obese females represents 46.8 % in our series. 3. In 6 CASES the OSAS went into remission after surgery. KEY WORDS: SURGERY, MORBID OBESITY, OSAS. 232 Actas de Fisiología 7, 2001

SYMPTOMATIC NARCOLEPSY

Claudio Sergio Podesta, Mario Massaro And Maria Elena Mazzola Buenos Aires, Argentina

Symptomatic narcolepsy is a very uncommon disoder that can occur as a consequence of, or during the course of an underlying disease. It has been described associated with brain tumors, multiple sclerosis, degenerative and vascular diseases, etc. We present here three patients (2 males and 1 female) who were refered to us because all of them had hypersomnolence and two had cataplexy attacks too. In one subject the narcoleptic symptoms began after a head trauma and in the other two those appeared during the course of a neurological disease. Polisomnographic and MSLT studies were done in all patients. The SORMPs were: 2/5 (male 20 ys), 4/5 (male 19 ys) and 4/5 (female 14 ys). The HLA typing was negative in two of them and in one patient it could not be performed. Actas de Fisiología 7, 2001 233

HEART RATE VARIABILITY DURING SLEEP-DISORDERED BREATHINGS IN UARS AND MILD OSAS SUBJECTS

D Poyares, A Rosa, U Koester, C Guilleminault, Sergio Tufik* Stanford University Sleep Disorders Center * Psychobiology Department, Federal University of São Paulo

We evaluated autonomic activation and arousal responses during abnormal breathing events in UARS and mild OSAS patients. Autonomic activation was investigated by analysis of Pulse Transit Time (PTT), heart rate frequency, and determination of its low and high frequency components. EEG arousals were scored visually, and were considered as greater than 1.5 sec. Methods: 10 UARS mean age=39.7 ± 8.6, and 10 mild OSAS mean age= 46.5 ± 10.0, with BMI < 35 kg/m2 were studied with polysomnography. Respiration was monitored by nasal cannula with the flow measured on a pressure transducer; mouth thermocouple; inductive respiratory plethysmography; esophageal pressure (Pes), calibrated on supine position; and pulse oximetry. PTT events were defined as a fall in PTT curve greater than15 msec. Period Amplitude Analysis and power spectrum analysis were performed to ECG signal to assess the RR Interval and heart rate variability, respectively, on 148 respiratory events randomly selected in Nrem and Rem sleep. The events were defined according to presence/absence of visual EEG arousals, and of PTT response at the end of respiratory events. There were only 37 events without PTT change. This number represented the limiting factor for selection of events. Statistics: Two-way ANOVA was applied considering the following two major effects: a) RR interval, and b) UARS vs OSAS, PTT vs NO PTT, and EEG vs NO EEG. A similar two-way ANOVA design was also performed to analyze the sympathetic and parasympathetic component of heart rate variability. Wilcoxon ranked pairs test was applied to compare sympathetic and parasympathetic heart rate components, before and after respiratory events, during Nrem and Rem sleep. Results: A total of 1168 respiratory events were identified for all patients. The events were then subdivided according to presence or absence of EEG visually scored arousal (>1.5 sec) and of PTT. Significant increase in heart rate was found after respiratory event termination. Mild OSAS subjects presented persistent shorter RR interval when compared to UARS patients, i. e., OSAS subjects exhibited consistent pattern of increased heart rate. There was no association between RR interval and PTT event. However, when a cortical arousal was visually detected, there was a significant increased heart rate following the respiratory event. The analysis of heart rate variability showed consistent changes after NRem sleep respiratory events. There was a significant increase in sympathetic, and decrease in parasympathetic components in response to abnormal breathing effort. No significant changes could be detected during Rem sleep. The parasympathetic component distinguished UARS from mild OSAS subjects. The formers presented a significant decrease in parasympathetic component after respiratory events. PTT was associated with significant decrease in parasympathetic component of heart rate after breathing events. The presence of PTT event and EEG arousal, for both group of subjects, was not associated with significant change in the sympathetic component of heart rate neither during NRem nor during Rem sleep. Acknowledgement: Dr Poyares is recipient of grants from FAPESP, Sao Paulo-Brazil. 234 Actas de Fisiología 7, 2001

RESISTANCE TO INSULIN ACTION, AS EVALUATED BY THE GLUCOSE CLAMP TECHNIQUE, IN OVERWEIGHT AND OBESE SUBJECTS WITH OBSTRUCTIVE SLEEP APNOEA SYNDROME (OSAS)

MA Savarese, *N Pannacciulli, B Brancasi, *A Minenna, M Palumbo, P Montesanti, *M Bellacicco, A Petruzzellis, D Stomati, *G De Pergola, *R Giorgino, FM Puca I Neurological Clinic, Centre for Sleep Disorders, University of Bari, Bari, Italy; *Section of Internal Medicine, Endocrinology, and Metabolic Diseases - Department of Emergency and Transplants, University of Bari, Bari, Italy

Background. Body fat accumulation has been demonstrated to be associated with several complications, including both respiratory and metabolic abnormalities, such as obstructive sleep apnoea syndrome (OSAS) and insulin resistance, respectively. Furthermore, the impairment in insulin action has been shown to be higher in patients affected by OSAS with respect to age- and BMI-matched controls. Therefore, the aim of the present study was to evaluate the possible relationship between polysomnographic parameters and insulin sensitivity, as measured by the glucose clamp technique, in overweight and obese patients with OSAS. Subjects and methods. Insulin sensitivity (M, mg glucose/kg body weight x min), as evaluated by the euglycaemic hyperinsulinaemic glucose clamp technique; apnoea- hypopnoea, and oxygen desaturation indexes (AI, HI, and ODI, respectively), as calculated by polysomnography; and fasting concentrations of glucose, insulin and lipids (total cholesterol and triglycerides), as measured by commercially available kits, have been determined in 6 overweight and obese nondiabetic subjects (5 men and 1 woman; age: 44.4 ± 10.7 years; BMI: 34.3 ± 9.2 kg/m2) affected by OSAS according to standard polysomnographic criteria. Pearson’s correlation coefficients were used to quantify the univariate associations among variables. Multiple regression analysis was performed to test the joint effect of age and BMI on AHI (dependent variable). Results. AHI was inversely associated with M-value (P<0.05). No significant association has been found between polysomnographic parameters and fasting concentrations of glucose, insulin, and lipids. Moreover, AHI did maintain its independent association with M-value even after multivariate analysis and correction for age and BMI. Conclusion. The findings of the present study suggest that respiratory abnormalities in overweight and obese OSAS patients may cause an impairment of insulin action, irrespective of age and BMI, thus leading to or worsening a condition of insulin resistance, often associated with OSAS. Actas de Fisiología 7, 2001 235

APNEAS IN PRETERM INFANTS: POLYSOMNOGRAPHIC EVALUATION

Cristina Scavone1, Lilian Chiappella2, Alberto Rodríguez2, Daniel Lorenzo3 1Cátedra de Neuropediatría, Facultad de Medicina. 2Escuela Universitaria de Tecnología Médica, Facultad de Medicina. 3Instituto de Investigaciones Biológicas Clemente Estable. Montevideo, Uruguay

The results of the first series of preterm infants studied in our country with polysomnographic (PSG) recording are presented. Seventy four infants (51 males, 23 females) born before 38 weeks of gestational age (26-37 weeks) were studied exhibiting apneas or episodes cataloged as apparent life threatening event (ALTE). At the time of the recording, 21 infants had chronological age of 37 weeks or less and the remaining 53 were between 38 weeks and 5 months old. PSGs were performed during naps, with adequate control of temperature and ambient light. The infants slept in their habitual positions: 24 in supine, 18 in prone, 4 in lateral, 28 changed the sleeping position more than twice during the study. The recordings were classified as abnormal when presented one or more of the following events: a) apneas longer than 15 seconds irrespective of their mechanism. b) apneas shorter than 15 seconds with concomitant changes in cardiac frequency and/or low O2 saturation. c) changes in heart rhythm and/or low O2 saturation without apnea. d) inadequate sleep organization, immaturity of EEG activity or epileptiform activity. RESULTS. Fifty seven infants showed pathologic recordings: n=29 (50.8%) with apneas of different mechanisms, n=18 (31.5%) with apneic events and EEG alterations, n=8 (14%) presented only abnormal EEGs and n=2 (3.5%) showed episodes of brady- cardia and/or low O2 saturation. Among the 21 infants studied at chronological ages <38 weeks, n=19 (90.4%) had pathologic PSG. Conversely, only 50% of the recordings were abnormal in the 53 infants with more than 38 weeks of gestational age. CONCLUSIONS. Preterm infants present a high risk of cardiorespiratory alterations during sleep and the probability of an abnormal PSG increases with lower gestational age and birth weight. The locally obtained results appears according to the corresponding literature.

1) Guilleminault, C et al. Mixed and obstructive sleep apnea and near miss for sudden infant death syndrome: comparison of near miss and normal control onfants by age. Pediatrics, 64 (6) 882-891. 1979. 2) Beckerman, R et al. Respiratory control disorders in infants and children. Edited by: Williams & Wilkins. Baltimore, 1992. 3) Gaultier, C. Cardiorespiratory adaptation during sleep in infants and children. Pediatric Pulmology. 19 :105-117, 1995 4) Curzi-Dascalova, L. Developpement du sommeil et des fonctions sous controle du systeme nerveux autonome chez les nouveau-ne premature et a terme. Arch Pédiatr 2: 255-262, 1995. 5) Kato, I. Sleep, 23 (4) 487-492, 2000. 236 Actas de Fisiología 7, 2001

AN INVESTIGATION OF INSOMNIA AROUSAL: MCMI-III PROFILE CLUS- TERS FOR MEN AND WOMEN DIAGNOSED WITH INSOMNIA

Kathy Sexton-Radek1, Ramon Eremia2, Rene Pichler3, Rob Fleming3, Mandy Grant3 Brianne Lukowski1, Erin Allbright1 and Peter Freebeck Elmhurst College1/Suburban Pulmonary and Sleep Associates/Hinsdale Hospital4, Elmhurst, Illinois; National Lewis University, Glen Ellyn, Illinois2; Illinois Institute of Technology, Chicago, Illinois3

In an effort to extend the hypotheses about increased arousal in individuals with Insom- nia, personality profiles were studied. With a 10% prevalence of chronic insomnia disorders, the significance of diagnosis/identification of the pathologic triggers is essential. A DSM IV field trial for sleep disorders identified insomnia related to mental disorders as the most frequent sleep diagnosis (46%) and that primary insomnia was the next most prevalent sleep diagnosis (22%) among patients referred with insomnia complaint at five sleep disorder centers. While reliable and valid diagnoses are made from assessments composed of clinical interviews, sleep and nap studies and self-report measures, the consideration of prespecified contributing factors such as arousal to the insomnia symptomology is not done. Both emotional/cognitive and physiological arousal levels are sufficient triggers to imitate and sustain insomnia symptomology. Item analysis and cluster analytic procedures were employed to identify groups of profiles by item and subtest score for individuals diagnosed with insomnia at an outpatient sleep clinic. The study cohort consisted of 200 patient seeking medical and psychological services for Insomnia symptoms. Results from the Millon Multiaxial Clin- ical Inventory III was examined for this study and compared to Psychopharmacologi- cal and Psychological professional service use. The item analysis findings indicated some commonly selected items by individuals with chronic insomnia. Profile clusters for males and females were not differentiated; heightened scores on scales that include components of arousability were noted. Replication of the cluster analysis is necessary and planned for a future study. Actas de Fisiología 7, 2001 237

COSTS AND IMPACT OF OBSTRUCTIVE SLEEP APNEA ON GENERAL HEALTH STATUS AND QUALITY OF LIFE

Madalena Teles Araújo*, Teresa Paiva*, Ana Barros**, Nuno Batista** *Hospital de Santa Maria, Lab. EEG e Sono, Lisboa, Portugal ** Faculdade de Medicina de Lisboa, Lisboa, Portugal

Introduction: Obstructive sleep apnea has a high prevalence in general population with detrimental effects on general health status and quality of life of patients, and important economical consequences due to the associated morbililty, comorbility and mortality. Objectives: To evaluate the costs of obstructive sleep apnea and the impact on patients general health status and quality of life, in a sample of 20 patients (1st time consultation) from Hospital de Santa Maria, Sleep Consultation, chosen during a 3 month period and according to their willingness to participate. Material and Methods: 2 questionnaires were applied to the sample: 1) Economic Evaluation, developed by the team, is divided in 4 groups – I. Socio- demographic characterisation (age, gender, Graffard Index); II. Direct costs for health care services (primary care, secondary care, medicines, exams, transports...); III. Direct costs external to health care services (health care supported by the patient and/or relatives, care given by relatives, transports, domestic work...); IV. Indirect costs (productivity, time and earnings lost). Questions refer to the previous year. Questionnaires were introduced in a database, all items quantified, and subject to statistical analysis. The sample results were compared with the means for general population in Portugal (National Health Inquiry and Statis- tics); 2) General Health Status Evaluation, Short Form 36 (SF 36), from the “Medical Outcomes Trust”, which covers physical and mental health, in the Portuguese version; it was scored in accordance to the model. This questionnaire allowed the quantification of intangible costs such as pain, suffering, anxiety, social isolation. Results: Patients suffering from obstructive sleep apnea have a worst general health status (direct impact on intangible costs); they have higher direct and indirect costs in the previous year. As an example, they consume more consultations than the general population (mean for Portugal is 2.7 consultations per inhabitant/year). Relatively to indirect costs, the vast majority has productivity decreased, namely due to excessive sleepiness. Conclusions: Before arriving to a sleep consultation these patients are higher con- sumers of health care resources and have a deteriorated general health status, with impact on their quality of life; Treating them is a priority. As the main objectives of HC Systems are reducing global costs and minimising opportunity costs, while improving the quality of interventions, evaluating the costs of this disease and the effectiveness of treatments is of utmost importance. Improvements in the quality of life related to health could be the most important measure of effectiveness. A future study will evaluate the improvements on health status and quality of life of these patients (effectiveness), after treatment. 238 Actas de Fisiología 7, 2001

REM RELATED CARDIAC ARRHYTHMIA AND CPAP

Alejandro Tobon and Edgar Osuna Clle 116 9-02 Of. Neurologia, Bogotá, Cundinamarca, Colombia

A 37 year old male patient was referred to our sleep clinic for evaluating snoring. The polysomnogram showed during REM sleep a second grade AV-block drifting to a total AV-Block that lasted up to 12 seconds. There was no correlation between the arrhythmia and sleep apnea. A 24 hour holter recording was performed and shows episodes of arrhythmia between 3am and 6am. Two additional sleep studies with CPAP were realized evaluating the possibility of improving arrhythmia correcting the underlying sleep disorder (Obstructive sleep apnea). There was no change in the duration or the number of arrhytmic periods. We concluded that CPAP does not solve REM related cardiac arrhythmia and other therapeutic strategies should be considered in these patients. Actas de Fisiología 7, 2001 239

SLEEP AND SLEEP PROBLEMS IN 40-45 YEAR OLDS IN A NORWEGIAN COUNTY

Reidun Ursin, Bjørn Bjorvatn and Fred Holsten Department of Physiology, Department of Public Health and Primary Health Care, and Department of Psychiatry, University of Bergen, Norway

The data collection of this study was conducted as part of HUSK (the Hordaland Health Study ’97-’99) in collaboration with the Norwegian National Health Screening Ser- vice, the University of Bergen and local health services. The study population included all individuals in Hordaland county born 1953-57 (29.400). A total of 8598 men and 9983 women participated, yielding a participation rate of 57 % for men and 70 % for women. Self-administered questionnaires provided information on various health behaviors including sleep problems, also, baseline measurements including height, weight, blood pressure etc were performed. A subgroup of 8860 persons (3531 men and 5329 women) answered a questionnaire with more detailed information on sleep habits and problems. Response rate varied between questions. Subjective sleep need was estimated to 7.3 h in men and 7.8 h in women, however as many as 35 % (33 % of men and 36 % of women) reported that they did not get enough sleep. Twentyone per cent (19 % of men and 22 % of women) regularly slept 1 h less than their estimated need. Fourteen per cent (12 % of men and 15 % of women) reported frequent sleepiness. Five per cent reported sometimes falling asleep at work. Difficulty sleeping 1-2 nights a month was reported by 19.6 % (19.1 % of the men and 20.1 % of the women), while 13 % (11.9 % of the men and 13.5 % of the women) had insomnia, defined here as difficulty sleeping once or more times a week. Insomnia was related to low income and low level of education. Self reported good health was higher in good sleepers than in insomniacs (92 % vs. 68 %). The sleep of the insomniacs was characterized by difficulty, most days or every day, of going to sleep (52.8 % in insomniacs vs. 1.8 % in good sleepers), repeated awakenings (39.6 vs. 1.6 %), and early morning awakening (31.1 vs. 3.6 %). Sleep medication was used for a shorter or longer period by 22 % of the women insomniacs and 14 % of the men. 240 Actas de Fisiología 7, 2001

SURGICAL WEIGHT REDUCTION IN MORBIDLY OBESE PATIENTS WITH OBSTRUCTIVE SLEEP APNEA

Matilde Valencia-Flores, Alejandra Castaño, Josefina Montes, Miguel Herrera, Montserrat Resendiz, Victoria Santiago, Rosa M. Campos, Jorge González- Barranco, and Guillermo García-Ramos Instituto Nacional de Ciencias Médicas y Nutrición “Dr. Salvador Zubirán” and Universidad Nacional Autónoma de México, México, D.F.

Obesity is known to be a risk factor for obstructive sleep apnea (OSA). In patients with morbid obesity, weight loss may improve sleep apnea. We evaluated the effects of weight loss on nocturnal polysomnography and daytime Multiple Sleep Latency Test (MSLT) in morbidly obese patients. Fifteen morbidly obese patients were studied. There were 6 males and 9 female aged 22-56 years with a mean of 40.4+10.8 years. One night plysomnographic recording was performed in all patients preoperatively and 13 months thereafter.

Before Surgery After Surgery p Body Mass Index kg/m2 54.2+12.1 38.2+8.4 0.0001 Neck circumference (cm) 46.2+5.2 40.4+4.5 0.0001 Snoring (SDQ item 20) 4.4+1.0 2.7+1.2 0.002 Epworth Sleepiness Scale 12.2+7.4 6.0+5.3 0.009 Mean MSLT (min) 7.1+6.7 6.8+6.3 ns Awake

Mean % SaO2 86.1+8.5 92.5+8.2 0.01

Etco2 (mm Hg) 44.1+6.8 36.7+5.5 0.02 Sleep Apnea/Hypopnea Index 44.5+37.2 18.6+25.0 0.004

Mean % SaO2 in NREM 76.4+15.3 88.5+4.5 0.005

Etco2 (mm Hg) in NREM 45.4+7.2 37.2+5.0 0.007

Mean % SaO2 in REM 67.6+18.3 86.2+6.3 0.0007

Etco2 (mmHg) in REM 44.8+9.2 39.9+7.8 ns Total time of oxygen Desaturation <65% (min) 83.8+14.7 1.8+5.2 0.02

A decrease in body weight of 30% in obese OSA patients, resulted in significant improvement in obstructive sleep apnea, with an improvement in sleep pattern with a reduction in light sleep (Sleep stage 1%) but not change on the objective sleep tendency during day, although the patients reported less probability to doze off. Additional weight loss might result in further improvement in both breathing pattern during sleep and daytime sleepiness.

This study received support from UNAM DGAPA-IN209500 Actas de Fisiología 7, 2001 241

RESTLESS LEGS SYNDROME IN PATIENTS WITH DIALYSIS-DEPENDENT RENAL FAILURE

Vecchi Carlos*, Risacher Guillermina*, Barreneche Mirta#, Butta Patricia#, Di Pino Patricia#, Matoso Marinela#, Mele Pablo#, Pierdominici María#, Rudolf Guillermo# * Association Medical Hospital - Sleep Unit; # Fresenius Medical Care Argentina Bahia Blanca City - Buenos Aires - Argentina

Introduction: Sleep disorders in patients with dialysis-dependent treatment were reported. The restless legs syndrome (RLS) was described with high prevalence in these sleep disorders. According with the bibliography, almost 40% of dialysis dependent renal failure patients could show this motor sensitive syndrome (RLS). Population: A questionnaire was performed in order to classify the prevalence of sleep disorders in patients with end-stage renal disease (ESRD) treated in Bahia Blanca hemodialysis Units. Habits as alcohol and tobacco, actual medications, current disease and laboratory reports (ferritin, hemoglobin, creatinine, KTV, ferritin saturation). Exclusion criteria were patients less than one year in hemodialysis treatment or with severe cognitive compromise. Results: 69 patients were evaluated (43 male and 26 female). Median age was 52 (23-78) years old. Forty seven patients (68%) informed about unpleasant leg symptoms as pruritus, swarm or burning. At least 25 patients had minimum criteria syntoms to be enroll as RLS; the majority (90%) with low extremities symptoms. There was no significant difference in laboratory reports between RLS group and a control group (patients without unpleasant leg symptoms). Conclusions: Sleep problems are very common in dialysis-dependent renal failure and the prevalence of RLS is very high in these patients. It’d be necessary to make an earlier diagnosis of this syndrome, to elaborate a proper treatment to reduce substantial morbidity. 242 Actas de Fisiología 7, 2001

STRESS AND INSOMNIA IN ANXIOUS AND DEPRESSED PATIENTS

Véronique Viot-Blanca and Chrystèle Deschauxb aHôpital Lariboisière, Paris, bSleep laboratory, Grenoble - France

It has been postulated that the clinical course of depression could be stressful life–events, then anxious disorder and finally depression, insomnia being present as a symptom at all three stages. Thus we have examined whether there is any clinical feature in insomniac patients that supports such a continuum. 250 GP were asked to complete 4 clinical questionnaires for insomnia, either chronic or not, together with a Hospital Anxiety Depression Scale. Among 976 patients, 770 completed the HAD: 27.8% were purely anxious (A group: A³10, D<10), 26% were anxious and depressed (AD group: A³10, D³10) 5,1% were purely depressed (D group: A<10, D³10) and 26% were non-psychiatric (N group: A<10, D<10). A, AD and N groups were compared using either Student T-test or Mann-Whitney (non- parametric) for quantitative variables and Chi-2 for qualitative comparisons. Anxiety, alone or coexisting with depression was more prevalent among women (N=59%, A=76%, AD=68%-N vs A: p<0.001, N vs AD: p<0.05) and was associated with specific clinical features: more awakenings during sleep (N=56%, A=70%, AD=80% - N vs A: p<0.001, N vs AD: p<0.01), more premature waking (N=32.5%, A=41%, AD=41.5% - N vs A: p<0.05, N vs AD: p = 0.05) more non-restorative sleep (N=33%, A=59.5%, AD=50% - N vs A: p<0.001, N vs AD: p <0.001). There was however no difference between groups in the difficulty to fall asleep. As a consequence of a poor night sleep, daytime sleepiness was more prevalent in the anxious groups (N=11%, A=18%, AD=20.5% - N vs A: p<0.05, N vs AD: p = 0.05) and the patients were also more tense and irritate (N=36%, A=54%, AD=64.5%-N vs A : p<0.001, N vs AD : p<0.001) while there was no difference regarding daytime fatigue, frequency of poor nights, sleep habits or possible medical causes of insomnia. When depression was associated with anxiety (comparison between group A and group AD), the difficulty to fall asleep was more prevalent (A=74%, AD=82.5% - p<0.05), as were sleep awakenings (A=70%, AD=80.5%- p=0.01) and non-restorative sleep (A=50%, AD=59% - p=0.05) but unexpectedly not premature waking. The classical clinical presentation e.g. difficulty falling asleep being related to anxiety and premature waking to depression, was not relevant in the overall population. In the depressive subgroup, chronic insomnia (N=55%, A=63%, AD=70% - N vs A: ns, N vs AD: p<0.001) and frequent poor sleep nights i.e. 4 or more during a week (N=28%, A=36%, AD=57% - N vs A: ns, N vs AD: p< 0.001, A vs AD: p<0.05) were more often present. Daytime consequences were more severe: they felt more often tense (A vs AD p>0.05), more often gloomy (N=23%, A=25%, AD=41% - N vs A: ns, N vs AD: p<0.001, A vs AD p<0.001). Anxiety alone was not related to a longer duration or a higher severity of insomnia. However, when depression was present, insomnia was more severe regarding frequency, duration and daytime consequences. The absence of other clinical difference is supporting the hypothesis of common mechanisms for insomnia in both anxiety and depression, therefore the hypothesis of a continuum between stress, anxiety and depression. Some depres- sions could be related to repetitive chronic stresses associated with an over-reactivity of the HPA axis followed by a progressive. In this context insomnia could be a marker of stress. Major claim: The absence of clinical differences (except severity) between anxious and depressive insomniacs is supporting the hypothesis of common mechanisms for insomnia in both disorders. Actas de Fisiología 7, 2001 243

CPAP TREATMENT FOR SLEEP BREATHING DISORDERS IN ADULT POPULATION: ACCEPTANCE, SUBJECTIVE TOLERANCE AND OUTCOME VARIABLES

GL Zanniello; JL Olaza; S Parrilla; R Salazar; L Bauso Toselli Sleep Disorders Unit & Epilepsy Center of the Dpt of Neurology of Hospital Italiano de Buenos Aires, Argentina

Rationale: Sleep Breathing Disorders (BSD) is a medical condition with high degree of morbidity and social dysfunction that can be functional reversed with nasal nocturnal continuous positive airway pressure (nCPAP). We evaluated tolerability, outcome variables and changes in diurnal symptoms In patients with sleep pane syndrome that were on treatment with nCPAP. Methods: One hundred thirty three patients under treatment with nocturnal nasal CPAP were analysed retro and prospectively. Inclusion criteria were all patients were evaluated in our Unit and CPAP tritation was determined by specific protocols All patients had a previous diagnostic polysomnographic recordings with oximetry Severe cases were evaluated with split night polysomnography. 15.8% patients were excluded because inadequate follow up (addressed at distal cities) and 4 patients were dead by non related causes. The group of patients with effective follow up had a range of age of 16 to 91 yrs (x of age of 59.2 yrs). Patients that were on treatment with CPAP were compared with the group of patients that refused the use of CPAP. The range of follow up was of 2 to 62 months (x of follow up: 18.8 months). An standard questionnaire was attained by patients every twelve months from the beginnig of the treatment. Pre and post treatment variables were evaluated and data were confronted with those patients that continue and those that discontinued the use of CPAP. The variables analysed were gender, polygraphic findings (Apnea Hipopnea index, mean and lower oxygen saturation, arrythmias), antropometric (body mass index) subjective symptoms (Epworth scale, visual analogyc scales for excessive diurnal sleepiness, somnolence, well being perceptions and morning drowsiness. For bedroom partners a 4 point scale for intensity of snoring pre and post treatment were used. A ratio of use of CPAP per hours of sleep and per days of week were performed. Results: Pre and post treatment variables showed an statistical differences in respect with Epworth scale ( mean: 18.9 vs 1.73; p< 0.05), snoring scale (p < 0.01), visual analogyc scales for excessive diurnal sleepiness (EDS) (p< 0.01), somnolence (p< 0.05), well being perceptions (p< 0.01), and morning drowsiness (p< 0.05). The group that continued on treatment with CPAP was the 64.3 % of the total number of patients with adequate follow up. A direct and significant association was found with IAH and CPAP tolerability. Subjective symptoms diminish over 90 % in the group on treatment and refereed a global use of 88 % of hours per night and days per week. Those who refused the use of CPAP 31 % did it exclusively for economic reasons. Conclusions: We observed a high subjective efficacy (>90 %) and acceptance (64.4%) in patients with BSD under treatment with CPAP. A significant association was found with Epworth scale, EDS and severity of sleep apnea syndrome. In the group that refused CPAP a 31% did it for economical reasons. 244 Actas de Fisiología 7, 2001

SURGICAL TREATMENT FOR SNORING AND SLEEP APNEA SYNDROME IN ADULTS

GL Zanniello, C Boccio*, S Ruggeri*, O Velan# Sleep Disorders Unit & Epilepsy Center of the Dpt of Neurology and *Dpt of Otorhinolaryngology and #Dpt of Imaging of the Hospital Italiano de Buenos Aires, Argentina

Objectives: Analyse surgical outcome in adults patients with snoring and special cases of obstructive sleep apnea syndrome (OSAS) and compare surgical results with presurgical findings. Methods: Since April 1993 to September 2000 one hundred thirty patients primary snoring or sleep apnea syndrome were surgical treated at our institution. Patients were presurgical evaluated with an specific clinical history, endoscopic evaluation with Muller manoeuvre, dynamic CT scan (during awake and sleep states) and nocturnal polysomnography with oximetry. For bedroom partners a 4 point scale for intensity of snoring pre and post treatment were used Surgical procedures were laser uvulopalatoplasty (LUPP), electrocautery uvulopalatoplasty (EUPP) and uvulopalatopharingoplasty (UPPP). Results: Thirty nine patients underwent EUPP and UPPP. Snoring decreased >80 % in 34 patients (87.1%) in 23 patients excessive diurnal sleepiness was present and disappeared or diminished in 21 patients (89.4%) after surgery. Two patients with OSAS were included in this group and had a remission of apneas with an apnea-hipopnea index (AHI) < 4 per hour. LUPP was used in ninety one patients. Snoring decreased >80 % in 87 patients (95.6%). From a total of fifty patients with OSAS 26 patients (52 %) had a remission with AHI < 4 per hour. No major complications were found Conclusions: Adequate presurgical evaluation and selection of patients with snoring and OSAS provided a good surgical outcome with EUPP or UPPP and LUPP with a decreased > 80% of snoring in 87.1 % and 98.4% respectively, with a remission of AHÍ < 4 per hour in 53.8 % of patients with OSAS that could not receive nasal CPAP treatment. Actas de Fisiología 7, 2001 245

ANALISYS OF DISTRIBUTION OF INTERICTAL EPILEPTIC DISCHARGES IN SLEEP MICROSTRUCTURE

GL Zanniello and L Bauso Tosselli Sleep Disorders Unit & Epilepsy Center of the Department of Neurology Hospital Italiano de Buenos Aires, Argentina

Rationale: Brain activity characterises by an aleatorious and rhythmic patterns. The organization of sleep structure during No Rem stages in repetitive periods of EEG burst fluctuations (arousal and phase A of cyclic alternating pattern: CAP), and separated by phase B intervals produced by thalamocortical mechanisms synchronize brain activity and favours the expression of interictal epileptic discharges (IED) frequently with a wider electrical field. The Cyclic Alternating Pattern (CAP) is an endogenous rhythm of No Rem sleep stages that occurs with a periodicity of 20 to 60 seconds with a specific EEG activity, muscle tone and autonomic phenomena. Our hypothesis is that a specific distribution of spikes with a tendency of grouping near phase A and that the transition towards and forward this phenomenon facilitates the expression of the IED even rather than during the burst pattern per se. Methods: We analysed EEG sleep recordings in ten patients. We evaluated macrostructure and microstructures variables and correlated them with the presence of IED and their field distribution, neuroimages and epileptic syndromes. We divided each cycle in proportional segments that corresponded to the 25% of the total length of phase B. We did this to be able to compare cycles of different duration. Results: The mean number of IED analysed per patient was 16.6 (range: 7 to 26). All the patients (100%) showed IED in initial or final 25 % of the total length of phase B in a similar way. The frequency distribution of IED in these portions of phase B for each patient were as follow: 92.3%, 81.2%, 84%, 72.2%, 69,2%, 69.5%, 100%, 86.6%, 80%, 75% The mean frequency was 81.0% and was statistically significant (p< 0.005). IED associated with subtype A1 pattern (89% of cases) produced no changes of sleep stage. IED followed by a subtype A2 showed a tendency to produced changes of sleep stage after 4 to 10 sec. Conclusions: IED occurred predominantly during phase B in all the patients. This frequently occurred in the seconds immediately previous or subsequent of phase A pattern (0.1 to 6.5 sec.) in a fashion that corresponded to the 25 % or less of total length of phase B. This finding was independent of the epileptic syndrome. This transitional period, where shifts of cortical-subcortical mechanisms struggles, could be a critical period as it misbalance structures that would promote IED occurrence instants previous or subsequent a normal EEG burst pattern is fully expressed. 246 Actas de Fisiología 7, 2001

THE EFFECTS OF VAGUS NERVE STIMULATION (VNS) ON SLEEP IN DEPRESSION

Roseanne Armitage, Mustafa Husain, Robert Hoffmann, John Rush 5323 Harry Hines Boulevard, Dallas, TX, USA

Introduction: Numerous studies have demonstrated that major depressive disorders (MDD) are characterized by a variety of sleep and quantitative EEG abnormalities (1). Moreover, it appears that some of the sleep disturbances are not normalized by antidepressant therapies (2). A recent study has demonstrated the efficacy of vagal nerve stimulation (VNS) in treatment-resistant depression (3). Since the vagus nerve has extensive projections via the nucleus tractus solitarius to many brain regions, including the reticular formation, parabrachial nucleus and locus coeruleus, it is not unreasonable to expect VNS stimulation to impact on sleep (4). The purpose of this preliminary study was to investigate the effects of VNS on sleep and biological rhythms in EEG in four treatment-resistant depressed patients. Method: Sleep studies were performed at baseline and after 10 weeks of VNS delivered by a NeuroCybernetic Prosthesis (NCP™). Changes in sleep macroarchitecture, based on standard stage scoring (i.e. total sleep time, % sleep stages, REM characteristics, etc.) and microarchitecture, based on quantitative EEG analyses (i.e. temporal coherence, periodicity and amplitude of EEG rhythms) were evaluated (1). Results: Pre-treatment, baseline sleep studies revealed severe sleep abnormalities, as shown in Table 1 below. Note that baseline sleep was more disturbed than that usually reported in non treatment-resistant patients. Sleep microarchitecture was also particularly abnormal at baseline, with severely dampened ultradian EEG rhythms. After VNS self-reported sleep quality improved after in all patients. Sleep macroarchitectural changes included less non-restorative Stage 1 sleep and intermittent wakefulness, accompanied by increased Stage 2 sleep. The most dramatic effect, however, was on the strength or amplitude of EEG rhythms. Large increases were found for the amplitude of beta, theta and delta rhythms, to near normal levels. See Table 1. Conclusions: Unlike most antidepressant medications, VNS treatment improved both subjective and objective sleep characteristics and reversed the dampening of EEG rhythms. These effects may be of clinical benefit, since persistent sleep disturbance increases the risk of relapse and recurrence. This study also suggests that sleep abnormalities in treatment-resistant depression may differ from those who are not resistant to treatment. A follow-up study is underway to confirm our findings. Table 1 Sleep EEG Measure Pre-VNS (Baseline) Post-VNS Total Sleep Period 422.6 (78.9) 455.9 (83.6) Sleep Latency 15.2 (4.4) 31.4 (17.2) REM Latency 142.4 (22.7) 157.3 (41.7) % Stage 1 33.4 (16.7) 22.3 (12.3) % Stage 2 46.1 (10.4) 58.2 (9.7) % SW 0.6 (0.9) 0.5 (0.7) % REM 12.8 (0.9) 14.3 (2.3) % Awake 7.0 (8.9) 4.7 (4.4) Av Rhythm Strength (Beta & Delta) 1442.3 (198.1) 2441.7 (328.1)

References: (1) Armitage R et al. 1999 Psychological Medicine, 29: 1435-1448. (2) Armitage R 2000 Canadian Journal of Psychiatry, 45:31-37. (3) Rush AJ et al. 2000 Biological Psychiatry, 47: 276-286. (4) George MS et al. 2000 Biological Psychiatry, 47: 287-295. Research supported by Cyberonics, Inc. Actas de Fisiología 7, 2001 247

SLEEP MICROARCHITECTURE IN FEMALES AT RISK FOR DEPRESSION

Kusumakar Vivrek, Kutcher Stanley, LeBlanc John, Armitage Roseanne and Rachel Morehouse 5323 Harry Hines Boulevard, Dallas TX, USA

Introduction: Major depressive disorders (MDD) are more prevalent in women, run in families and are associated with a variety of sleep abnormalities, particularly those derived from sleep microarchitecture (1,2). Low temporal coherence, reflecting a breakdown in the synchronization of sleep EEG rhythms, has been shown to characterize children and adults with MDD and to persist into clinical remission. Thus, low coherence may be a vulnerability marker for MDD (2). As such, it should also be evident in those at high-risk for MDD. Method: Sleep studies were conducted on 41 high-risk adolescent (12-15) girls, who had a maternal history of depression. An age-matched control group of girls (N=40) with no maternal depression were also included. Neither group had a history of depression. All girls spent two consecutive nights in the sleep laboratory. Cross-spectral time series analysis evaluated temporal coherence of sleep EEG rhythms (3). MANO- VA and multiple regression analyses contrasted between-group differences. Follow-up interviews were conducted 2 years after initial sleep study. Results: Temporal coherence was significantly lower in the high-risk group (p<.05). Regression analyses correctly classified 71% of the high-risk group, and 95 % of controls. In addition, 22/41 (54 %) of the high-risk group had coherence values that were 2 sds below controls. Upon follow-up, 11/41 (27%) high-risk girls showed depressive symptoms, 6 who met DSM-IV criteria for MDD. Most importantly, all 11 girls had extremely low coherence values. Only1/40 (2.5%) control girls showed evidence of depression at follow-up and was identified as a false positive by the regression analysis. Conclusion: Low temporal coherence of sleep EEG rhythms is evident in girls at high-risk for depression, even before symptoms are revealed. Coupled with previous findings in children and adults with MDD, low coherence appears to be a vulnerability marker of the illness.

References: Armitage R. 1995 Biological Psychiatry, 37: 72-84. Armitage R, Hoffmann R. 1997 Current Review of Mood and Anxiety Disorders, 1: 139-151. Armitage R et al. 1999. Psychological Medicine, 29: 1435-1448.

Research supported by: The Canadian Psychiatric Research Foundation ; The Nova Scotia Department of Health Designated Mental Health Research Fund; The Stanley Foundation 248 Actas de Fisiología 7, 2001

CORRELATION BETWEEN SLEEP ABNORMALITIES AND COGNITIVE DYSFUNCTIONS IN DRUG-NAIVE PATIENTS WITH SCHIZOPHRENIA

Roger Godbout, Emmanuel Stip and Julie Poulin Centre de Recherche Fernand-Seguin, 7331 Hochelaga, Montreal, Canada

Objective. Many studies have analyzed sleep in schizophrenia in order to better understand the pathophysiology of this disease but only one reported on the relationship between cognitive performance and sleep.1 This study, however, only analyzed REM sleep variables and reported only on explicit memory, whereas patients were recently withdrawn from chronic neuroleptic treatments. For the present report, we have performed sleep recordings and a thorough neuropsychological evaluation of attention and memory in drug-naive patients with schizophrenia, a group of patients very rarely described in the literature. Method. Eight schizophrenia patients never treated with neuroleptics (2W, 6M; 31.0 ± 19.9 years old) were recorded for two consecutive nights in a sleep laboratory. Sleep stages and phasic events (sleep spindles and rapid eye movements (REMs)) of night 2 were scored according to standard criteria. In the morning following night 2, attention and memory were evaluated using a computer-assisted battery. Correlation between sleep variables and cognitive performances was calculated using Spearman’s Rho. Results. As expected, patients with schizophrenia showed impaired performance in attention and explicit memory and normal performance in implicit memory. Reaction times to the sustained attention task correlated with sleep onset latency (.75, p<.05). Reaction times to the selective attention task correlated with total sleep time (-.79, p<.04), wake after sleep onset (.76, p<.05), sleep efficiency (-.76, p<.05), stage 2 percentage (.82, p<.02), SWS percentages (-.93, p<.003) and spindles density (-.90, p<.05). Memory span correlated with REM sleep latency (-.84, p<.008) and implicit memory performance correlated REM sleep percentage (.73, p<.04). Discussion. The present results show that the relationship between attention/memory and sleep organization in neuroleptic-naive patients with schizophrenia is comparable to what is found in healthy subjects.2,3 Given the fact that patients with schizophrenia display impairments in attention and memory, these results suggest that neural networks involved in cognitive processing during sleep could be impaired in this disease.

References. 1. Taylor SF, Goldman RS, Tandon R, Shipley JE. Neuropsychological function and REM sleep in schizophrenic patients. Biol. Psychiatry 1992; 32: 529-538. 2. Riopel L et al. Performance on a selective attention task correlates with sleep organization in normal young subjects. Sleep Research 1997; 26: 211. 3. Smith C. Sleep states and memory processes. Behavioral Brain Research 1995; 69: 137-145.

Sponsored by the CRSNG and the Fonds de la recherche en Santé du Québec Actas de Fisiología 7, 2001 249

DREAM CONTENT OF SCHIZOPHRENIC, NON-SCHIZOPHRENIC, AND COMMUNITY CONTROL ADOLESCENTS

Jack Hadjez M.D*, Daniel Stein M.D** and Vadim S. Rotenberg M.D., Ph. D.*** *Shaar Menashe Mental Health Center, Hadera, **Sheba Medical Center Tel- Hashomer, ***Abarbanel Mental Health Center Affiliated with theSackler Faculty of Medicine, Tel Aviv University, Israel

To date, only few studies have systematically evaluated the manifest dream content of schizophrenic patients. According to some authors dream may be consider a manifes- tation of real emotional problems and the attempt to solve them in an open or meta- phoric way. We hypothesized that the manifest dream content of schizophrenic patients would reflect the psychotic organization in which patients conceive and experience their world. We compared the manifest dream content of 20 schizophrenic adolescent inpatients, with 21 adolescent inpatients with other mental disorders, matched for age and gender, and 31 matched community controls. Inpatients’ medications were stable for at least 4 weeks. All participants received standardized Formal Dream Content Rating Scale (FD- CRS). The Positive and Negative Symptoms Scale (PANSS) was administered to the two patient groups. Controls reported more involvement (p<.05) and emotional expression (p<.01) when compared with schizophrenic patients. They also showed more implausibility (p<.05) and had a greater duration of dream report (p<.03) when compared with non-schizophrenic inpatients group. Among the schizophrenic group only, elevated scores on the negative subscale of the PANSS correlated with lower scores of involvement (p<.05), recall (p<.05), and emotional expression (p<.01). No association was found with positive and general psychopathology subscales of the PANSS and any of the FDCRS dimensions. These results suggest that psychopathology per se, is associated with impoverish- ment of dream content, and that negative schizophrenic symptomatology is positively correlated with changes in specific dimensions of dream content, possibly through a common defect. 250 Actas de Fisiología 7, 2001

FLUOXETINE, SLEEP AND SEX DIFFERENCES IN DEPRESSION

Andrzej Honory, Madhukar Trivedi, Robert Hoffmann, John Rush and Roseanne Armitage The University of Texas Southwestern Medical Center at Dallas, TX 75390

INTRODUCTION: Fluoxetine, a serotonin reuptake inhibiting antidepressant, prolongs REM latency, suppresses REM sleep and increases light Stage 1 sleep(1-2). It has been suggested, however, that the sleep effects are more dramatic among women with depression(3). The present study investigated the effects of 10 weeks of treatment with fluoxetine on sleep EEG in 28 women and 20 men with major depressive disorders (MDD). METHOD: All outpatients met DSM-III-R criteria for nonpsychotic major depressive disorder, single or recurrent type, with moderate to severe symptoms (HRS-D ¡Ý 17). Each subject spent two consecutive nights in the laboratory at baseline, while symptomatic and unmedicated and after 1, 5 and 10 weeks of treatment with 20mg of fluoxetine. Standard visual stage scoring was conducted on each sleep record. EEG was also quantified suing period amplitude analysis. Repeated-measures MANOVA was conducted on sleep parameters from those subjects who completed all four acute- phase sleep evaluations. RESULTS: Of the 58 patients at baseline, 36 completed 10 weeks of treatment with a significant decrease in HRS-D (21.2 vs 5.2; p<.02). Fluoxetine did significantly suppressed REM sleep, but no sex difference was evident. For NREM sleep, however, % Stage 2 was decreased in women and increased in men. Both men and women showed increased Stage 1 sleep, but the treatment effects was larger in women (p<.05). Moreover, the number of arousals and % awake were significantly increased in women but unchanged in men. Treatment by sex interactions were also evident for incidence and amplitude of beta and delta activity (p<.05). Sex main effects were obtained for alpha, theta and sigma measures (p<.02). Women with MDD showed treatment-related increases in beta incidence and amplitude in all sleep stages. Men, on the other hand, showed decreased beta amplitude in Stage 1 but increased beta in Stage 2 sleep. CONCLUSIONS: The effects of fluoxetine on sleep EEG in depression are strongly influenced by gender. Analyses of sex differences in treatment response are currently underway.

Research supported in part by the Sarah M. And Charles E. Seay Center for Brain and Applied Research in Psychiatric Illness and by Eli Lilly and Company.

REFERENCES: 1. Trivedi MH, Rush AJ, Armitage R, et al (1999): Effects of fluoxetine on the polysomnogram in outpatients with major depression. Neuropsychopharmacology 20:447-459. 2. Armitage R, Trivedi M, Rush AJ (1995): Fluoxetine and oculomotor activity during sleep in depressed patients. Neuropsychopharmacology 12:159-165. 3. Armitage R (2000): The effects of antidepressants on sleep in patients with depression. Cana- dian Journal of Psychiatry 45:803-809.

Actas de Fisiología 7, 2001 251

EFFECT OF AGE IN SEXUAL BEHAVIOR OF SLEEP DEPRIVED RATS AFTER COCAINE ADMINISTRATION

Monica Levy Andersen; Magda Bignotto, Ricardo Borges Machado; Sergio Tufik Department of Psychobiology – Universidade Federal de São Paulo – UNIFESP, São Paulo, SP, BRAZIL

Introduction: It has been suggested that paradoxical sleep deprivation (PSD) could sensitize dopaminergic systems, probably by inducing supersensitivity of the post synaptic receptor (TUFIK et al., 1978). The mesolimbic dopamine (DA) system, composed of DAergic neurons in the ventral tegmental area and their projections to the nucleus accumbens and other forebrain structures, has been implicated in the reinforcing and locomotor-activating properties of cocaine (HOGER et al., 1999). In 2000, ANDERSEN et al., showed that acute cocaine administration elicited sexual behavior in young PSD rats. It is known that there is a decline in sexual arousal and copulatory activity in male rats with advancing age, as well as significant changes in various sleep parameters. Thus, the objective of the present study was to examine the behaviors of penile erection and ejaculation in young and old-age rats submitted to PSD after cocaine administration. Methods: Young (3 months) and middle-age (22 months) male rats were PSD for 96 hours by the modified multiple platform methods or stayed in their cages (control group-CTRL). At the end of this period, the animals received an i.p. injection of cocaine (15 mg/kg) and the erections and ejaculations were assessed for 60 minutes. Results: The cocaine-PSD study revealed a significant difference between the 3months- PSD and the three other groups with regard to sexual behavior. The young PSD rats had the greatest number of erection and ejaculation. Among the old-age rats, the PSD group only had erection and sexual behavior was absent in the 22 months control group.

Table 1:

9 Conclusions: Sexual function common- 8 7 ly decreases with age, but the interac- 6 tion of sleep deprivation and the action 5 * ERECTION EJACULAT ION 4 of cocaine in the brain by enhanced DA 3 * transmission can facilitate and increase 2 1 # the sexual behavior in old-age rats. # * # 0 3m-PSP 22m-PSP 3m-CTRL 22m-CTRL GR OUP S References: ANDERSEN, M.L.; PALMA, B.D.; RUE- DA, A.D.; TUFIK, S. – The effects of acute * # - different from 3m-PSD animals; p<0,0005 cocaine administration in paradoxical sleep- by ANOVA test deprived rats. Addiction Biology, 5: 417- 420, 2000. HOGER, B.A.; IYASERE, C.A.; BERNOW, M.T.; ESSER, C.J; NESTLER, E.J.; TAYLOR, J.R. – En- hancement of locomotor activity and conditioned reward to cocaine by brain-derived neurotrophic factor. The Journal of Neuroscience, 19(10):4110-4122, 1999. TUFIK, S.; LINDSEY C.J; CARLINI, E.A. – Does REM sleep deprivation induce a supersensitivity of dopaminergic receptors in the rat brain? Pharmacology, 16: 98-105, 1978.

Support: This work was supported by Associação Fundo de Incentivo à Psicofarmacologia (AFIP). 252 Actas de Fisiología 7, 2001

HIGH-AFFINITY [3H]-OUABAIN BINDING TO THE Na+/K+-ATPase FROM DISCRETE BRAIN REGIONS OF REM SLEEP-DEPRIVED RATS

Marco A. C. Benedito, Umberto Andrade and José Gilberto B. Carvalho Departamento de Psicobiologia, Universidade Federal de São Paulo, São Paulo, Brasil

REM sleep deprivation increases brain excitability (Physiol. Behav. 7: 103-106, 1971). Na+/K+-ATPase has a critical role in restoring transmembrane gradients of Na+ and K+ after neuronal excitation (Amer. J. Physiol. 279: C541-C566, 2000). Ouabain, a Na+/ K+-ATPase inhibitor, binds to the phosphorylated catalytic subunit of Na+/K+-ATPase with a 1:1 stoichiometry, providing and estimation of the number of Na+/K+-ATPase molecules. Three Na+/K+-ATPase isoforms are expressed throughout the brain regions in a great complexity. These enzyme isoforms differ in their affinity to ouabain, a Na+/ K+-ATPase inhibitor. The aim of this work was to determine the high-affinity binding of [3H]-ouabain to Na+/K+-ATPase from discrete brain regions of rats deprived of REM sleep (REMSD). Adult, male, Wistar rats were deprived of REM sleep for 96h by the flower-pot technique. A home-cage and a large platform group (stress control group) were also used. [3H]-Ouabain binding assays were carried out in a crude membrane preparation (30,000 g pellet) from striatum (40 nM), hippocampus (68 nM) and brainstem (60 nM). One-way ANOVA detected a statistically significant difference in [3H]- ouabain binding (pmol/mg protein, mean ± SD) in all brain regions assayed (striatum:

F2,15=5.7, p=0.01; hippocampus: F2,15=6.3, p=0.005; brainstem: F2,13=15.3, p=0.0004). Post-hoc statistical analyses showed that both REMSD and stress control group differed significantly from home-cage group (*p<0.01, two-tailed).

Striatum Hippocampus Brainstem Home-cage 2.7±0.9 (5) 6.8±1.3 (6) 15.3±0.4 (4) Stress control 4.7± 0.9* (6) 9.8±2.0*(6) 19.3±1.8* (5) REMSD 4.5±1.3* (7) 9.2±1.1* (6) 20.7±1.7*(7) (N) number of rats

The results suggest an increase in Na+/K+-ATPase molecules in the brain areas assayed. The increase may be due to stress or may indicate that stress control group was also deprived of REM sleep.

Supported by AFIP and CAPES Actas de Fisiología 7, 2001 253

EFFECT OF MIRTAZAPINA IN DEPRESSIVE PATIENTS WITH INSOMNIA

Francesca Canellas and José Maria de Pedro Psychiatry Department, Hospital Son Dureta, Palma de Mallorca, Spain

Introduction: Depressive disorders are severe and prevalent illnesses, which can be considered an important public health problem. Sleep disorders are integral component of depressive disorders and approximately 95% of depressive patients suffer insomnia. Depressive patients with insomnia have a diminution of cognitive and motor performance with higher impairment in life quality than the rest of depressive patients. Objective: Evaluate the efficacy and tolerance of Mirtazapine in the treatment of depressive disorders associated with insomnia in Primary Health care setting. Study Desing: This was a 6-week observational, prospective, non controlled, national, multicentric study. Three visits (basal, day 15 and day 45) were performed for each patient. Montgomery-Asberg (MADRS) scale for depression, Spiegel Sleep Ques- tionnaire and an Adverse Events Questionnaire were performed in each visit. All patients with at least a follow up visit were used for efficacy analysis and all patients who took at least one dose of medication for tolerability analysis. All patients received a single nightly dose of 15 mg of Mirtazapine the first four days and 30 mg since the 5th. Practitioners were allowed to adjust the dose in each control visit whiting a range of 15-45 mg/day. Results: A total of 538 General Practitioners have studied 2064 ambulatory patients with depression and insomnia (DSM IV criteria) that could be included in the study. In this preliminary report we present the data of the first 131 patients having finished the protocol, 33 men (25,19%) and 98 women (74,81%). MADRS Scale: Total punctuation of scale was 30,55 (+/-7,76) in the basal visit, 20,61 (+/- 8,94) at the 15 days visit and 9,06 (+/- 7,70) at the final one. Decrease in punctuation was significant (p< 0,0001) since the first control. Spiegel: Patients also improved the sleep quality and duration, being significant the results since the first control. No patient was withdrawn from the study due to lack of efficacy. Nine patients (6,82%) showed adverse reactions being the most usual one the diurnal somnolence (3 subjects, one of them had to be withdrawn). The remaining complaints were: bradypsychia (1), suicide attempt (1), constipation (1), hypotension (1), rash (1) and finally a patient had a non-specified crisis. Conclusion: Mirtazapine has been shown to be an efficient and well tolerated treatment for depressive patients with insomnia.

Bibliography: Thase, M.E. Antidepressant treatment of the depressed patient with insomnia. J Clin Psychiatry 60(suppl 17):28-31, 1999) 254 Actas de Fisiología 7, 2001

DISTINCTIVE EFFECTS OF MODAFINIL AND d-AMPHETAMINE ON THE HUMAN EEG POWER SPECTRUM DURING 60 UNINTERRUPTED WAKING HOURS

Florian Chapotot1, Ross Pigeau2 and Alain Buguet1 1Unité de Physiologie de la Vigilance, Centre de Recherches du Service de Santé des Armées, France. 2Human Factors of Command Systems Section, Defence and Civil Institute of Environmental Medicine, Canada

Modafinil, a new wake-promoting agent, reduces the need for long recovery sleep [1] and decreases the rebound in EEG slow wave activity following sleep deprivation [2]. These diachronic effects suggest an action of modafinil on a homeostatic sleep-wake regulatory process. The aim of this study was to determine whether modafinil, in comparison to d-amphetamine and placebo, would affect the homeostatic process that also influences EEG activity during waking [3]. Thirty-three healthy subjects were investigated during 60 hours of prolonged wakefulness in a double-blind placebo-controlled mixed-design study. A 4-min eye-closed maintenance-of-wakefulness test was administrated hourly allowing the measurement of sleep latency and waking EEG activity. The effects of 300-mg of modafinil and 20- mg of d-amphetamine administered at three different times were evaluated. Both stimulants increased sleep latency during 10-12 hours following ingestion, independently of the time of administration. At the level of the waking EEG spectrum, d-amphetamine strongly attenuated the natural circadian rhythms and suppressed the sleep deprivation-related increases in low frequency activity (0.5-7 Hz). In contrast, the evening administration of modafinil had little effect on the EEG circadian rhythms and allowed for the maintenance of slow alpha (8.5-11.5 Hz) frequency activity, which exhibited a homeostatic decrease proportional to time awake under placebo. These findings demonstrate that the wake-promoting action of modafinil and d-amphetamine involves separate neurophysiological regulatory systems: d-amphetamine may inhibit the expression of sleep processes, probably through a direct cortical activation masking circadian rhythms in EEG bands; modafinil, through a synchronic effect, may disrupt the homeostatic process controlling the expression of waking EEG activity.

References 1. Buguet A, Montmayeur A, Pigeau R, Naitoh P. J Sleep Res 4: 229-241, 1995. 2. Lin JS, Gervasoni D, Hou Y, Vanni-Mercier G, Rambert F, Frydman A, Jouvet M. J Sleep Res 9: 89-96, 2000. 3. Cajochen C, Knoblauch V, Krauchi K, Renz C, Wirtz-Justice A. Neuroreport 12(10): 2277- 2281, 2001. Actas de Fisiología 7, 2001 255

SCOPOLAMINE DECREASES RHO BAND (20-30 Hz) EEG POWER DURING REM SLEEP: IMPLICATIONS FOR MEMORY PROCESSES

Oscar Díaz-Ruiz1, Isabelle Beaulieu1 and Roger Godbout1,2 1 Centre de Recherche, Hôpital Sacré-Coeur, 2Départment de Psychiatrie, Université de Montréal, Québec, Canada

Administration of the muscarinic receptor antagonist scopolamine acutely decreases REM sleep and impairs memory.1 Forty-eight hours after treatment, however, no effects on REM sleep are found2 while memory impairments are present.3 We used qEEG analysis of REM sleep to evaluate further this apparent dissociation. Special attention was paid to Rho EEG activity, a cortical EEG frequency selectively associated with REM sleep in the rat.4 Methods. Long-Evans rats aged 3-4 months were implanted for sleep recordings and EEG electrodes were placed over frontal (A= Bregma+3.5 mm; lat. +1.0mm), parietal (A= Bregma-2.0 mm; lat. + 3.5mm), and centro-medial (midway between Breg- ma and Lambda; lat. +1.0) cortices, referred to cerebellum. After full recovery, rats received either scopolamine (0.1 mg/kg, s.c.; n=9) or vehicle (0.5 cc NaCl 0.9%, s.c.; n=14) at 8 AM, 2 h after lights on; sleep recordings started immediately, for 4 h. Five of the scopolamine-treated rats were recorded again, 48 h later. Fifteen 4-sec epochs of artifact-free EEG were selected from REM sleep and submitted to Fast Fourier Trans- form. Absolute power amplitude was calculated and six frequency bands were created: Delta (0.75-3.75 Hz), Theta1 (4-6.75 Hz), Theta2 (7-9.75 Hz), Sigma (10-13.75 Hz), Beta (14-19.75 Hz), Rho (20-30 Hz). Results were compared with t-test for independent samples. Results. Compared to saline-treated rats, acute scopolamine decreased the percentage of REM sleep (16.3±1.3 vs 7.4±1.1; p<.05) and increased the percentage of wake after sleep onset (12.7±1.4 vs 55.8±7.0; p<.05); both measures returned to saline values 48 h later. qEEG analysis showed that acute scopolamine increased all Frontal frequencies except Rho, which was decreased, further decreasing after 48h. Acute scopolamine increased Delta and decreased Rho at the Pari- etal site while it increased Beta activity and decreased Rho at the Centro-medial site. Discussion. The reported long-term effects of scopolamine on memory are only par- alleled by REM sleep EEG Rho activity. This suggests that common neurophysiological networks may be involved.

1. Smith C. Behav Brain Res. 1995; 69:137-145; 2. Beaulieu et al., Sleep 1998; 21: 25; 3. Quirion R. et al J Neurosci. 1995; 15:1455-1462; 4. Campbell et al. J Neurophysiol. 1993; 69:1368-1371. Supported by FRSQ and NSERC Canada. 256 Actas de Fisiología 7, 2001

EFFECT OF MODAFINIL (PROVIGIL) ON THE SINGLE-UNIT ACTIVITY OF SEROTONINERGIC DORSAL RAPHE NEURONS IN BEHAVING CATS

Monica MC Gonzalez, Barry L. Jacobs, Casimiro Fornal Laboratory for Neuromodulation & Behavior, VAMC Philadelphia, Pennsylvania, United States

Claim of Abstract: Modafinil has non-significant effect on serotoninergic neuronal activity. Abstract: Modafinil {2-[(diphenylmethyl)sulfinyl]acetamide} is a novel wakefulness-promoting agent, used in the treatment of narcolepsy and hypersomnia, which has a pharmacological profile distinct from that of amphetamine and other central nervous system stimulants. Although its mechanism of action is not known, recent evidence suggests that the drug may act, at least in part, by enhancing the release of serotonin in the brain (Ferraro et al., 2000). To further characterize the effects of modafinil on the central serotonergic system, the present study monitored the discharge rate of serotonin-containing neurons in the dorsal raphe nucleus of cats, following systemic administration of this drug. Under pentobarbital anesthesia, adult cats were implanted with insulated nichrome microwires for recording neuronal activity and with electrodes (EEG, EOG, EMG) for monitoring behavioral state, as described previously (Fornal et al., 1999). Administration of modafinil (5 or 10 mg/kg, p.o.) induced a state of continuous wakefulness, lasting 10-18 h, without apparent behavioral excitation or hyperactivity. Cats typically adopted a quiet, recumbent posture with eyes open, during the period of enhanced waking. Serotonergic neuronal activity was not significantly altered from baseline levels at any time after modafinil administration. A slight increase in firing rate (up to 30%) was noted for some cells, however, most cells displayed less than ± 10% change in neuronal activity. Overall,these results indicate that the pronounced waking effect observed with modafinil in cats is not accompanied by marked changes in the activity of brain serotonergic neurons.

Supported by: MH 23433 & IBRO Fellowship to M.M.C.G. Actas de Fisiología 7, 2001 257

SLEEP-WAKEFULNESS EEG ANALYSIS AND CENTRAL NERVOUS SYSTEM EFFECTS OF TOTAL AQUEOUS EXTRACTS OF VERVAIN HASTATA (VERBENACEAE) IN RATS

Moses A. Akanmu, Kazuki Honda, Shojiro Inoué Department of Biocybernetics, Institute of Biomaterials and Bioengineering, Tokyo Medical and Dental University, Tokyo, Japan

The sleep-wakefulness EEG analysis and central nervous system effects of the total aqueous extracts (TAE) of aerial portion of vervain hastata (Verbenaceae) was studied in male rats in order to establish the scientific basis in herbal medicines. The plant has been empirically used in the herbal preparations as sedative, relaxant, nerve tonic, analgesic and thymoleptic among other uses, and no data is available. The investigation was conducted using the EEG analysis, barbituric-hypnosis test, acetic acid-induced writhing test and, open-field behavioral study procedures with doses ranging from 125 to 2000 mg/kg per oral (p.o.) or intraperitoneally (i.p.). In the barbituric-hypnosis test, administration of TAE in three doses of 500, 750, and 1000 mg/kg p.o. prior to an injection of pentobarbital (30 mg/kg i.p.) was carried out and at the same time investigated the involvement of GABA -benzodiazepine receptors system by administering a specific antagonist flumazenil A(3 mg/kg i.p.) 15 min before the barbiturate-hypnosis test in another group of rats. The positive control group of rats was treated with diazepam (3 mg/kg i.p.). The EEG sleep-wakefulness study was analyzed in individually housed rats, maintained under 12:12 light/dark cycle after the implantation of electrodes in the cortex and neck muscle for EEG/EMG recordings. Continuous 24 h EEG recording was done during baseline day (vehicle only) and experimental periods before and during the administration of TAE. The total aqueous extract was administered via drinking water at concentration of 30 and 50 mg/ml. The anti-nociceptive test was carried out using the acetic acid-induced writhing method. The results showed that TAE significantly (p<0.01) potentiated the pentobarbital induced-hypnosis by reducing sleep latency and increased sleeping time in a dose- dependent manner. Flumazenil reversed these effects in all animals. Comparison of the EEG data between baseline and extract administration periods showed that TAE augmented total sleep time, time spent in rapid eye movement (REM) sleep and time spent in nonREM sleep at the expense of wakefulness, which were consistent with the barbituric-hypnosis test results. The anti-nociceptive results showed that the extracts possess significant (p<0.05) dose-dependent activity by increasing the pain threshold. In conclusion, the results clearly show the scientific validity for the use of this plant as a sedative, mild analgesic and possibly as a nerve tonic material. 258 Actas de Fisiología 7, 2001

A DOUBLE-BLIND, PLACEBO-CONTROLLED STUDY OF THE EFFICACY OF TRAZODONE IN THE ALCOHOL POST-WITHDRAWAL SYNDROME: POLYSOMNOGRAPHIC AND CLINICAL EVALUATIONS

Olivier C Le Bon, Guy Hoffmann, James R Murphy, Nicolas Kormoss, Monique Kentos, Philippe Dupont, Karin Lion, Isidore Pelc and Paul Verbanck Pl. Van Gehuchten 4, Brussels, Belgium

Alcohol detoxification is accompanied by sustained difficulties in sleep initiation and maintenance. These difficulties are thought to be an important cause of relapse to alcohol use. However, the choice of a hypnotic drug is made difficult by cross-tolerance between benzodiazepines and alcohol. We evaluated here the capacity of trazodone, a second generation antidepressant with anxiolytic and sedative properties, to increase the sleep efficiency in alcohol dependent patients after detoxification. Eighteen patients completed the trazodone (n=8) or the placebo (n=8) treatment arms. Polysomnographies were performed at baseline, after the first drug intake, and after four weeks of treatment. The main outcome variable was sleep efficiency. Sec- ondary outcome measures included changes in other sleep parameters, Hamilton depression rating and clinical global impression scales. The sleep efficiency was increased in the trazodone group when it was computed after sleep onset, both immediately after drug intake and after four weeks of treatment. No benefit was observed in the placebo group. Sleep improvement under trazodone also included the number of awakenings, intermittent wake sleep time and non-rapid eye movement sleep. Hamilton and clinical global scales were better for the trazodone group. Trazodone is thus an interesting option in the treatment of alcohol post-withdrawal insomnia. Considering the prevalence of alcohol dependence, the reduced rates of remission and the relapses linked to insomnia, even if a fraction of alcohol-dependent patients with insomnia could be helped, a large number of individuals and their environment could be positively affected Actas de Fisiología 7, 2001 259

INTRASTRIATAL ADMINISTRATION OF AMPHETAMINE INCREASE THE C-FOS EXPRESSION IN THE PONTOMESENCEPHALIC TEGMENTUM NUCLEI AS A POSSIBLE MECHANISM OF INDUCED WAKEFULNESS

Juan Mena and Magda Giordano Centro de Neurobiologia, UNAM, Juriquilla, Queretaro, Mexico

It has been shown that rapid eye movement sleep (REMs) deprivation induces increases in dopamine content in the striatum (Farooqui et al, 1996), and c-Fos expression in pontomesencephalic nuclei (Maloney et al, 1999). Anatomical studies have determined that the GABAergic output neurons of the substantia nigra pars reticulata from the basal ganglia reach the cholinergic neurons of the pedunculopontine tegmental nucleus (PPt) (Grofova and Zhou, 1998). In turn, the PPt projects back to the substantia nigra pars compacta, the subthalamic nucleus and the pallidum (Parent and Hazrati, 1995). It is possible that the basal ganglia nuclei through their interaction with the PPt may participate in the modulation of the sleep-wake cycle. To test the hypothesis that direct dopaminergic stimulation of the striatum resulted in c-Fos expression in the PPt region, unilateral injections of d-amphetamine sulfate (15ug/1ul) were delivered into the left caudate-putamen of male Sprague-Dawley animals under deep anesthesia. In order to allow c-Fos expression, animals were perfused one hour after injection and PPt brain sections were processed by immunohistochemical techniques. The results show that there is an increase in c-Fos expression within the cholinergic population of the PPt after the amphetamine injection. To further support the notion of striatal participation in PPt activation, c-Fos expression was also evaluated after striatal kainic acid lesions; these experiments are in progress. 260 Actas de Fisiología 7, 2001

NEUROTRANSMITTERS AND NEUROMODULATORS INVOLVED IN THE REGULATION OF THE WAKING STATE

Jaime M. Monti and Héctor Jantos Department of Pharmacology and Therapeutics. Clinics Hospital. Montevideo, Uruguay [SYMPOSIUM]

It is generally agreed that normal waking (W) and the functions associated with it depend upon the integrity of the entire brain, but that subcortical structures situated primarily in the brainstem are critical for the EEG manifestations of W. In this respect, neurons within the brainstem reticular formation (oral pontine and midbrain reticular formation) activate the cerebral cortex via the nonspecific thalamocortical projection system, and via the subthalamus, posterior hypothalamus and basal forebrain (nucleus basalis of Meynert, nuclei of the diagonal band, and septum). Neuroanatomical, neurochemical and neuropharmacological studies, particularly those employing selective norepinephrine (a1), dopamine (D1 and D2), serotonin (5-

HT1-2-3) or histamine (H1) receptor agonists have shown that they tend to promote W and to suppress slow wave sleep (SWS) and REM sleep (REMS). In addition, different glutamate receptors and peptides (substance P, corticotropin-releasing factor, thyrotro- pin-releasing factor, vasoactive intestinal peptide and neurotensin) have wake promoting effects. On the other hand, acetylcholine (M1 and M2 receptors) plays an integral role in EEG arousal and REMS generation. Moreover, experimental evidence tends to indicate that adenosine (A1 and A2A receptors) and GABA (GABAA receptor) are involved in the induction of sleep and the decrease of W. Nitric oxide (NO) is an unconventional neurotransmitter synthesized on demand by the enzyme NO synthase (NOS). NO has been proposed to serve multiple functions in the central nervous system including the modulation of the sleep-wake cycle and circadian rhythms, which may in part be related to its influence on neurotransmitter release. A number of in vivo and in vitro studies have shown that treatment with NO donors increases the basal release of GABA, adenosine and acetylcholine, and reduces the release of norepinephrine, dopamine, serotonin, and histamine in the mammal brain. A decrease in the availability of NO induces the opposite effects. The anatomical structures involved in the control of W, NREMS and REMS are reciprocally interconected, and the alternation of behavioral states could depend upon the predominant release of GABA and adenosine, acetylcholine or catecholamines, serotonin and histamine at critical sites. NO could have a significant role in this respect. Whether a reduction in the availability of NO could explain, at least in part, the increased incidence of W in primary and secondary insomnia is an open question. - J.M. Monti, H. Jantos and D. Monti. Increase of waking and reduction of NREM and

REM sleep after nitric oxide synthase inhibition: prevention with GABAA or adenosine

A1 receptor agonists. Behav. Brain Res. (in press, 2001). - J.M. Monti. Involvement of histamine in the control of the waking state. Life Sci. 53: 1331-1338 (1993). Actas de Fisiología 7, 2001 261

THE EFFECTS OF 8-BROMO-cGMP ON SLEEP IN RATS

Ana C. Ribeiro and Levente Kapás Department of Biological Sciences, Fordham University, New York, USA

Introduction: Brain levels of cGMP show pronounced circadian variation (1) and cGMP has been shown to play a critical role in circadian entrainment in response to light, glutamate and nitric oxide (NO) in the SCN (2). NO is a neurotransmitter implicated in sleep regulation (3). Since NO is a potent activator of soluble guanylyl cyclase and cGMP mediates many physiological effects of NO, experiments were undertaken to investigate whether cGMP itself modulates sleep. Methods: Male rats were implanted with EEG and EMG electrode and a cannula targeted at either the lateral cerebral ventricle (icv) or the medial preoptic region (anterior diencephalic microinjection). Sleep was recorded for 23 h starting at dark onset from the rats on a baseline day after a saline injection (5 ml icv and 1 ml diencephalic). On the test day, the rats received 8-Br-cGMP icv [6.4 ng (n = 7), 160 ng (n = 7), 4 mg (n = 11), 100 mg (n = 8), 500 mg (n = 11) or 2.5 mg (n = 5)]; or 400 ng (n = 8) intra- diencephalically. Sleep was recorded for 2-11 days. The 4 mg dose of 8-Br-cGMP icv was also tested at light onset. Results: Cerebral injection of 8-Br-cGMP suppressed NREMS during the dark phase and transiently increased REMS throughout the recording period. NREMS amounts were decreased by 28.2 ± 8.9 and 48.5 ± 13.5 min in the first and second dark periods, respectively, after the 4 mg dose. There was a strong decreasing trend in EEG power across all frequency ranges and vigilance states. Similar tendencies in sleep amounts and intensity were observed after icv injection of lower and higher doses at dark onset. Light onset injection of 8-Br-cGMP did not elicit any changes in sleep parameters. Diencephalic microinjection of 400 ng 8-Br-cGMP suppressed NREMS by 47.2 ± 18.3 and 52.9 ± 22.7 min on the test day and on the recovery day, respectively. REMS amounts and EEG delta activity during NREMS were slightly increased after diencephalic 8-Br-cGMP. Conclusion: Cerebral injection of cGMP elicited changes in sleep amounts. The second messenger cGMP may be involved in mediating the effects of other sleep factors, including NO.

References: 1. Choma, P.P. et al. Pharmacology, 19:307-314, 1979 2. Amir, S. Brain Res. 586:336-9, 1992 3. Ribeiro, A.C. et al. Am. J. Physiol. 278:R1048-1056, 2000

Acknowledgments: This work was supported in part by NIH (NS-30514), and by the Altman Travel Fund. 262 Actas de Fisiología 7, 2001

RANDOMIZED, DOUBLE-BLIND CLINICAL TRIAL, CONTROLLED WITH PLACEBO, THE TOXICOLOGY OF CHRONIC MELATONIN TREATMENT

Maria L Seabra, Magda Bignotto, Luciano R. Pinto Junior and Sergio Tufik São Paulo, Brazil

Melatonin (N-acetyl-5-metoxytryptamine) is a secretory product of the the pineal gland. The mechanisms by which melatonin exerts its hypnotic effects are a matter of specu- lation. Some authors believe that this effect results from increased indolamine levels at the beginning of the sleep period, suggesting that endogenous melatonin participates in the regulation of the sleep-wake cycle, leading to a cascade of events that may activate somnogenic structures or that even melatonin metabolites possess a hypnotic effect. However, the existence of a causal relationship between sleep and melatonin is still unclear. Therefore, since melatonin is a natural substance, several studies have proposed its use by humans. Some studies have been carried out to evaluate melatonin’s possible toxicity. In the present study, we also examined the possible toxicological aspects of melatonin as well as its effects on sleep in normal volunteers. The safe therapeutic use of melatonin depends on its pharmacological effects and clinical and toxicological results. Therefore, the present study was conducted to evaluated the melatonin effects on biochemical parameters, as well as on sleep-related behavioral aspects, sleep architecture and adverse effects, in healthy volunteers, who ingested 10 mg of melatonin/day for a period of 28 days. Analysis of the PSG showed a statistically significant reduction of stage 1 of sleep in the group melatonin. No other differences between placebo and melatonin groups were obtained. In the present study we did not observe, according to the parameters analyzed, any toxicological effect that might compromise the use of melatonin at a dose of 10 mg for the period of time utilized in this study.

This work was supported by grants from Associação Fundo de Incentivo à Psicofarmacologia AFIP. Actas de Fisiología 7, 2001 263

DISTRIBUTION OF HYPOCRETIN (OREXIN) AND MCH CONTAINING CELLS IN THE TURTLE

Monica M. Eiland, Thomas C. Thannical and Jerome Siegel UCLA, Santa Monica, CA, USA

Human narcolepsy has recently been attributed to a loss of hypocretin (Hcrt) containing cells. We examined the phylogeny of the Hcrt system by mapping cell distribution in the brain of the turtle Pseudemys. We immunostained series of adjacent brain sections from three turtles with polyclonal antibodies for Hcrt A, Hcrt B, and melanocyte concentrating hormone (MCH). Adjacent sections were also used for a no-antibody control and a nissl stain. MCH, known to be present in reptiles, was used as a positive control. In Pseudemys, we found Hcrt positive cells in the nucleus periventricularis hypothalami (nPH) and the median eminence (ME). ME contained more Hcrt positive cells than nPH. In ME, Hcrt cells were clustered in a radial fashion along the edge of the tissue, with large, bulbous processes that protruded into CSF-containing space. All Hcrt positive cells were associated with processes extending into neighboring areas of the hypothalamus. MCH appeared exclusively in nPH. In mammals, MCH-positive cells are intermixed with Hcrt cells in the lateral hypothalamus. Therefore, in Pseude- mys, the Hcrt cells are not as extensively or exclusively colocalized with the MCH cells as in mammals. The anatomical data suggest that hypocretin cells in reptiles may closely monitor or release substances into the cerebrospinal fluid.

Supported by NRSA MH11817-02 and the VA. 264 Actas de Fisiología 7, 2001

BODY TEMPERATURES AND OXYGEN CONSUMPTION DIFFER BETWEEN ACTIVE SLEEP EPISODES OF COMPLETE AND INCOMPLETE SLEEP CYCLES IN NEONATES

Bach V., Telliez F., Chardon K., Chiorri C.*, Leke A.**, Libert J.P. ETP-APC, Fac. of Medicine, Univ. of Picardie Jules Verne, 3 rue des Louvels, 80036 Amiens France; * Inst. of Psychology, Florence, Italy; ** Pediatric Unit, Amiens, France

In non-thermoneutral environments, sleep and thermoregulatory processes interact. Thus, in animals, disruption of the sleep cycle by wakefulness (W) maintains body homeothermia whereas occurrence of fast wave sleep implies its suspension. These transitions have never been studied in neonates, whose thermoregulatory processes operate fully during both active (AS, first sleep stage after W) and quiet sleep (QS). The analysis of the awakening process is of great importance since a relative inability to arouse has been found in infants at risk for Sudden Infant Death Syndrome. Cool exposure reduces QS at benefits of W and AS. The present study was undertaken to compare the AS episodes that precede a QS episode (AS 1: complete sleep cycles) with those which are followed by wakefulness (AS 0: complete sleep cycles). With regard to the sleep-thermoregulation interaction, it is held that body temperatures govern the sleep cycle, i.e. that their levels and/or response patterns differ according to the AS outcome. Protocol: 37 preterm neonates (37 ± 2 weeks post-conceptionnal age) were explored in an incubator during 2 consecutive morning naps (thermoneutrality and cool condition, 1.5°C below the thermoneutral air temperature). Sleep, oxygen consumption (VÝ o2), esophageal temperature (Tes) and mean skin temperature were continuously recorded. Results: Cool exposure significantly shortened QS total duration and increased the frequency of incomplete sleep cycles (+0.010 ± 0.035 min–1) at the expense of complete sleep cycles (–0.004 ± 0.015 min–1). The duration of the AS 0 and AS 1 episodes was not modified by the thermal condition. o and body temperatures differed according to the outcome of VÝ 2 the AS episode. Whatever the thermal condition, in AS 1 episodes, o was high at the VÝ 2 beginning and decreased throughout the episode, as it continued during the succeeding QS episode. In contrast, AS 0 was characterized by lower values of body temperatures and o at the episode onset. T increased throughout the AS 0 episode at VÝ 2 es thermoneutrality but not in the cool condition. Conclusion: Low levels of T and o at es VÝ 2 the onset of the AS episode enhance sleep cycle interruption by W, during which Tes and o are the highest while the thermal response to cool exposure is the most efficient. In VÝ 2 contrast, QS is the least efficient stage as far as thermoregulation is concerned, but is privileged when energy conservation is needed. During cool exposure, the thermoregulatory function of the neonate overcomes the needs of energy conservation, even though the thermal stress is of low magnitude. This is achieved by preferential outcome of AS episode towards wakefulness. Outcome of active sleep episodes and corresponding body temperatures and oxygen consumption in neonates sleeping in thermoneutral or cool environment. Outcome of active sleep episodes in neonates sleeping in thermoneutral or cool environment: measurements of body temperatures and oxygen consumption. Interruption of neonates’ sleep cycle induced to cool exposure. Active sleep (AS) episodes were separated into AS 1 and AS 0 depending on whether the sleep cycle was complete or not (AS followed by a quiet sleep (QS) episode or by wakefulness, respectively, with or without transitional intermediate sleep). Actas de Fisiología 7, 2001 265

EFFECTS OF ENVIRONMENTAL TEMPERATURE AND SLEEP ON VENTILATORY RESPONSE TO HYPEROXIA IN PREMATURE NEONATES

Chardon K1., Telliez F1., Bouferrache B1., Leke A2., Bach V1., Gaultier C3. 1 ETP-APC, Fac. of Medicine, Univ. of Picardie Jules Verne, 3 rue des Louvels 80036 Amiens France; 2Pediatric Unit, Amiens University Hospital France; 3Department of Physiology, Debré Hospital 75019 Paris France

In neonates, previous studies have shown that the respiratory chemoreceptor sensitivity is modified by air temperature or by sleep stages. However the combined influence of both sleep and thermoregulation on the chemical control of breathing through chemoreceptors has never been studied. To assess whether there is an interaction between sleep, body temperature regulation and chemoreceptor functioning, 11 healthy premature neonates (post conceptional age 35.9 ± 2.1 weeks; 2,000 ± 1,000 g) underwent sleep hyperoxic challenge tests in neutral, cool and warm thermal conditions. It is held that this interaction could trigger chemoreceptor dysfunction which could be implicated in various neonatal respiratory disorders.

The neonates were successively exposed to thermoneutral air temperature (TN), warm (TN + 2°C) and cool (TN - 2°C) environments during 3 morning naps. Abdominal, cheek skin temperatures and rectal temperature were continuously recorded. 71 hyperoxic challenge tests (100% O2) of 30 sec duration each were done during quiet sleep (QS) and active sleep (AS) with a fitting face mask connected to a pneumotachograph. Before the test, 5 min was allowed for sleep stabilization. Respiratory flows were determined and tidal volume, inspired and expired times, respiratory frequency and ventilation were calculated breath-by-breath. Data sampled during the hyperoxic challenge were compared with baseline data (30 sec before the test) during which the neonate breathed room air. The respiratory time was defined as the time elapsed between the test onset and the first significant decrease in ventilation which is considered as an adequate response to the hyperoxic challenge. In cool environment, the drop in ventilation was higher during AS (-36.5 ±10%) than in QS (-18.6 ±11%; p=0.03) whereas this was not significant at thermoneutrality or during warm exposure. Thus, sleep and thermoregulatory processes interact with the chemoreceptor function that disagrees with the finding of Rigatto et al.* (1982). This difference could be explained by the fact that these authors have done the hyperoxic tests in thermoneutral environment, only. Since large increase in the gain of these structures can induce breathing variability, the control of the thermal environment is of utmost importance in sleeping neonates exposed to events during which chemoreceptor control mechanisms operate.

* Rigatto H. et al. (1982) Early Human Development. 7(1):1-10 266 Actas de Fisiología 7, 2001

SPECTRAL ANALYSIS OF HEART RATE VARIABILITY DURING SLEEP IN CHILDREN WITH PARTIAL EPILEPSY

Lilia Curzi-Dascalova, Alexis Arzimanoglou, Marie Bourgeois, Christine Beaud, Maurizio Elia, Sebastiano A. Musumeci and Raffaele Ferri

Objective. Alterations of the autonomic control of cardiac activity in epileptic patients have been reported by several studies in the past an both ictal and interictal modifications of heart rate regulation have been described. Also, alterations of the autonomic control of cardiac activity can play an important role in the sudden unexplained death in patients with epilepsy (SUDEP). However, the eventual presence of specific changes in HRV during sleep, not correlated with seizures, has not been assessed in children with epilepsy; for this reason, we evaluated the features of the cardiac autonomic function during sleep without ictal epileptiform EEG activity in a group of children with partial epilepsy. Methods. Eleven patients (5 males and 6 females; mean age 11.5 years, S.D. 3.65 years) affected by partial epilepsy were admitted to this study; 11 normal subjects (5 males and 6 females; mean age 12.9 years, S.D. 2.72 years) served as a control group. All subjects slept in the laboratory for two consecutive nights. The data were analyzed during the second night. Sleep was polygraphically recorded (including one ECG channel) and signals were digitally stored. A series of 5-minute ECG epochs were chosen from each sleep stage, during periods without evident ictal epileptiform activity in the EEG. ECG signals were analyzed for automatic detection of R waves and, subsequently, a series of time- and frequency-domain measures were calculated. Results. Epileptic subjects tended to show an overall lower HRV in both time- and frequency-domain parameters, mostly during REM sleep and, to a lesser extent, during sleep stage 2. Among the different bands, this decrease was most evident for HF absolute power. For this reason, the LF/HF ratio was always higher in epileptic patients than in normal controls and the difference was statistically significant during SWS and REM sleep. Conclusions. Our results seem to indicate that during sleep, a particular condition of basal modification in autonomic asset occurs (mostly during REM sleep) in partial epilepsy patients; this might represent an important factor contributing to the complex mechanism of SUDEP which takes place most often during sleep and supports the need of studying HRV specifically during this state in subjects with seizures. Actas de Fisiología 7, 2001 267

ALTERED NITRIC OXIDE ACTIVITY IN ARTERIOLES OF EPISODIC HYPOXIA RATS

Eugene C Fletcher, Ziad Tahawi, Natalia Orolinova and Irving Joshua University of Louisville; 530 S. Jackson ST., Louisville, KY, USA

Recurrent EH is a feature of sleep apnea that may result in systemic hypertension. Chronic EH (8 hours days for 35 days) in rats causes eleva-tion of diurnal resting, mean arterial blood pressure (MAP). Using in-vivo video microscopy, we examined arteri- olar reactivity in cremaster muscles of male rats after chronic EH. Arterioles of EH (n=6) and controls (n=6) were exposed to varying doses of norepinephrine, acetylcholine (ACH) and endothelin-1 (END). Separately, EH (n=5) and controls (n=5) were given one dose of NO synthase inhibitor, L-NAME. eNOS mRNA levels from the kidneys were examined. EH rats show a 16 mmHg increase in MAP over 35 days; controls none. Responses to NE and END were similar. ACH vaso-dilatation in EH rats was attenuated compared to controls. The degree of vaso-constriction in response to block-ade of the NO system by L-NAME was significantly less (83% of baseline diameter) in arterioles of EH rats compared to that of controls (61% baseline diameter), implying lower basal resting NO release in the EH rats. Kidney mRNA eNOS levels were not different. These data suggest that chronic elevation of BP associated with EH involves increased peripheral resistance from decreased basal release or production of NO. 268 Actas de Fisiología 7, 2001

GESTATIONAL AND POSTNATAL EFFECTS OF PRONE SLEEPING

Rosemary SC Horne, Dorota Ferens, Pratiti Bandopadhayay, Anne-Maree Watts, Jessica Vitkovic, Brendan Lacey, Sarah Andrew, Susan Cranage, T Michael Adamson Department of Paediatrics and Ritchie Centre for Baby Health Research, Monash University, Melbourne, AUSTRALIA

Arousal from sleep is believed to be an important survival mechanism that may be impaired in victims of Sudden Infant Death Syndrome (SIDS). Prone sleeping has been demonstrated to be the major risk factor for SIDS in world-wide studies. It has previously been reported that arousability is impaired in prone sleeping healthy term infants. Similar studies have not been carried out in preterm infants, a group known to be at increased risk for SIDS. The aim of this study was to longitudinally compare arousal responses of term and preterm infants sleeping both prone and supine. The control group for this study was 24 healthy term infants born at 38-42 wk (mean 40±0.4 wk) gestation with normal birth weights 2765-4190g (mean 3540±78g). 14 preterm infants born at 30-35 wks (mean 32±0.4 wks) gestation with birth weights 920- 2337g (mean 1638±102g) were also studied. Preterm infants had received <2 days of mechanical ventilation, had normal cranial ultrasounds and their medical history was uneventful. All infants were studied using daytime polysomnography on 3 occasions: 2- 3 weeks post-term, 2-3 months post-term, and 5-6 months post-term. Multiple measurements of arousal threshold (cm H2O) in response to air-jet stimulation applied alternately to the nares were made in both active sleep (AS) and quiet sleep (QS) when infants slept both prone and supine. Arousal thresholds were compared between sleep states and sleeping positions with 2-way repeated measures ANOVA. Differences between term and preterm infants and between ages studied were tested with 2-way ANOVA. Data are presented as mean±sem and were considered statistically significant at the p<0.05 level. In term infants arousal threshold was elevated in QS compared with AS when infants slept both prone and supine at all three ages. In the preterm infant group, arousal threshold in QS was significantly elevated compared to AS at 2-3 wks and 2-3 mo in both positions, there was no difference at 5-6 mo. In term infants arousal thresholds were significantly higher in both AS and QS when infants slept prone at 2-3 wk and 2-3 mo, but not at 5-6 mo. In the preterm group, the prone position only impaired arousal at 2-3 mo, and this was in both AS and QS. Postnatal age did not alter AS arousal thresholds in either group of infants in the prone position. When infants slept supine AS thresholds were lower at 2-3 mo than 2-3 wks in both groups. Postnatal age had no effect on arousal thresholds in QS in either position in preterm infants. However, arousal threshold increased significantly with postnatal age in QS, in both positions in term infants. When arousal responses were compared between term and preterm infants at matched ages preterm infants were significantly easier to arouse at 5-6 mo in both sleeping positions. We have demonstrated that the prone sleeping position significantly impairs arousal in both AS and QS. The finding that in preterm infants this impairment was only at 2-3 mo, the age when the risk of SIDS is highest is of significance. Importantly at 5-6 mo sleeping position did not affect arousability in either group of infants. This study provides an insight into why the risk for SIDS is increased when infants sleep prone.

This project was supported by SIDSaustralia and SIDassist. Actas de Fisiología 7, 2001 269

A DEFINITION OF INFANT AROUSAL FROM SLEEP

Rosemary SC Horne, Susan M Cranage, T Michael Adamson Department of Paediatrics and Ritchie Centre for Baby Health Research, Monash University, Melbourne, AUSTRALIA

Currently there is no universally accepted definition of arousal in young infants. This has made comparison of studies investigating arousal responses difficult to compare. The aim of this study was, therefore, to develop a definition of arousal, both spontaneous and stimulus-induced in the human infant which would provide consistency across sleep states, postnatal ages and could be used for both term and preterm infants. 24 healthy term infants and 24 healthy preterm infants where studied using daytime polysomnography. Recordings of EEG, EOG, submental EMG, ECG, instantaneous HR, and thoracic and abdominal breathing movements were made. Term infants were studied on 3 occasions: (a) 2-3 weeks after birth; (b) 2-3 months post-term and (c) 5-6 months post-term. Preterm infants had an additional study at 36-38 wks gestational age. A pulsatile air-jet (frequency 3Hz) delivered to the nostrils of the infant was used as an external arousal stimulus in both AS and QS. The probability of spontaneous arousal was calculated as the frequency of arousal which coincided with calibration of each stimulus. In determining the criteria for an arousal response we used changes in the physiological variables previously reported to be important in the overall arousal response. A change in ventilation pattern of more than 2 breaths, a body movement away from our stimulus, an associated increase in sub-mental EMG and an increase in heart rate of greater than 10% above baseline. All these changes needed to occur within 7 s of the stimulus onset. The 10s immediately preceding the stimulus presentation provided the baseline level used to assess the change in each variable. Frequencies of responses in cardiovascular and behavioral variables for both stimulus-induced and spontaneous arousals were compared between sleep states, infant groups and across ages, using Chi Square analysis, a p value of <0.05 was considered significant. A change in HR was consistently the least frequently occurring response across sleep states and ages in both groups of infants. Consequently, arousal was defined as a change in 3 or more of the 4 variables. Using this definition, changes in the four arousal criteria at the time of arousal were not different between AS and QS in the term infants at any of the ages studied. In the preterm infants only HR changes were significantly less frequent in QS compared to AS at 2-3 weeks post-term (p<0.001). Frequencies of responses in each of the four variables were also not different between the two groups of infants at matched post-conceptual ages. Additionally, there was no difference in the frequency of changes with increasing postnatal age in term infants and in preterm infants the only difference was in HR in QS. In this study we have developed and tested a definition of arousal that can be used in young infants and which is consistent across sleep states, postnatal age and gestational age. The definition is simple to use and does no require sophisticated analysis of EEG.

This project was supported by SIDSaustralia, Sudden Infant Death Research Foundation (South Australia) and SIDassist 270 Actas de Fisiología 7, 2001

ACUPUNCTURE TO THE SACRAL SEGMENT PROMOTES SLEEP STATES AND RELAXES THE URINARRY BLADDER

Yoshiyuki Tanaka1,2, Yoshimasa Koyama1, Yukuhiko Kayama1, Akihiro Kawauchi2 and Tsuneharu Miki2 1Department of Physiology, Fukushima Medical University, School of Medicine, 2Department of Urology, Kyoto Prefectural University of Medicine, Japan

It is known clinically that acupunctural stimulation to the sacral segment suppresses hyperactivity of urinary bladder, resulting in prevention of enuresis. To elucidate the mechanisms of this suppression, we examined effects of the acupuncture on the mobil- ity of urinary bladder in urethane anesthetized rats simultaneously with those on electroencephalogram (EEG). Rats were anesthetized with urethane (1.0 g/kg i.p.). Polyethylene catheter was inserted to the urinary bladder to record the pressure and to infuse saline. Stainless steel bolts for recording EEG were screwed to the skull overlying the frontal and parietal cortices. An acupunctural needle was set at the periosteum of the 3rd segment of sacrum, and the needle was rotated manually at a speed of about 3 turns/sec for 1 minute as acupunctural stimulation. The stimulaiton was applied mainly when the bladder con- tracted spontaneously after infusion of saline (0.4 – 0.8 ml). During experiment under urethane anesthesia, high amplitude slow-wave EEG and slow-wave EEG with slightly faster and lower amplitude appeared alternatively; each condition lasted several seconds or more. Spontaneous bladder contraction was observed mainly in the latter condition. The acupuncture stimulation completely suppressed the contraction for 2 to 23 minutes in 41 of 106 cases. In 28 of 41 successive cases, EEG pattern shifted to high amplitude slow waves. The EEG change was observed only when the stumulation suppressed the bladder contraction. The same EEG change could be observed by the acupunctural stimulation applied when the bladder was empty with no contraction. The results suggest that the acupunctural stimulation to the 3rd sacral segment affects sleep promoting systems as well as micturition systems. Actas de Fisiología 7, 2001 271

THE FINGERS HEAT UP AS WE FALL ASLEEP

Leon Lack and Michael Gradisar School of Psychology, Flinders University, South Australia

Introduction: Originally it was believed that the evening decline in core body temperature was the best predictor of sleep initiation (1). However, more recently the increase of distal skin temperature (hands and feet) has been shown to be more related to sleep latency(2). Further exploration of the physiology associated with sleep initiation would seem to be warranted, especially for its possible relevance to a better understanding of sleep difficulties such as sleep onset insomnia. Most studies have examined distal skin temperatures with relatively long time intervals of measurement only over the typical bed retiring period of time. Therefore, the present study investigated the rapid changes of finger temperature from lights out to sleep onset at various circadian phases. Method: Fourteen healthy, good sleeping subjects (11 males, 3 females, mean age = 28.14 yrs) participated in a modified 48-hour wakeful bedrest constant routine while PSG was monitored continuously and rectal temperature (RT) and skin temperature (ST) of the palmar distal surface of the right index finger was recorded at 30-second intervals. Across this 48-hour period multiple sleep latency tests (MSLTs) were administered every half hour. The sleep latency was the interval between lights out and the first of three consecutive 30- second epochs of sleep after which time the subject was aroused from sleep. Results: Baseline finger ST was recorded half-hourly before the start of each MSLT. There was a significant circadian rhythm of baseline ST with average peak of 33.5 deg C at 0300 hrs and nadir of 31.5 deg C at 1600 hrs. The core RT circadian rhythm nadir and peak respectively followed these ST phases by 2-3 hours. During the MSLT trials ST showed consistent changes. During physical adjustment to supine position at the beginning of each trial there was a drop of ST of about 0.6-0.8 deg C over a period of one minute. This was followed by a steady rise of ST at a rate of about 0.6 deg C per minute until the trial was terminated at the third consecutive epoch of sleep or, in the case of longer sleep latencies, until ST leveled out at an asymptotic value. The asymptotic values had a circadian rhythm in phase with the baseline ST rhythm but 1-4 deg C higher depending on the initial starting baseline values and because ST never exceeded about 35 deg C. Conclusions: The circadian rhythm of finger temperature is almost the reciprocal of core temperature, probably serving the function of helping to drive the circadian rhythm of core temperature through heat dissipation from vasodilation or conservation from vasoconstriction. The initial drop of ST with settling down may result from increased sympathetic vasoconstriction associated with the brief muscular effort required to adjust to the sleeping position. The subsequent, more pronounced increase of ST would be a result of decreased sympathetic tone preceding sleep onset and be a part of what has been described as the “sleepening” process. It would be of interest to compare the magnitude of this normal response with that in sleep onset insomniacs who may instead show a sympathetic arousal as indicated by a further drop of ST when presented with an opportunity to fall asleep. References: (1) Aschoff, J. J. (1970). Circadian control of body temperature. In: Hardy, J. D., Gagge, A. P. & Stolwijk, J. A. J. (Eds) Physiological and Behavioral Temperature Regulation, p 905-919. Charles C. Thomas, Springfield: Illinois. (2) Kräuchi, K., Cajochen, C., Werth, E. & Wirz-Justice, A. (2000). Functional link between distal vasodilation and sleep-onset latency? American Journal of Physiology, 278, R741-R748. 272 Actas de Fisiología 7, 2001

MULTIPLE SITES OF ENDOGENOUS EXCITATION OF THE RESPIRATORY SYSTEM IN REM SLEEP

Lovering A.T., Dunin-Barkowski, W., Vidruk, E.H.*, and Orem, J. Department of Physiology, Texas Tech University SOM, Lubbock, TX, USA *Department of Preventive Medicine, University of Wisconsin, Madison, WI, USA

Background. Physiological regulation in REM sleep is characteristically non-homeostatic. Control occurs with internal feedforward systems rather than reflexively through feedback. The logic, origin and sites of action of this feedforward control are generally not known. In the respiratory system, feedforward control is evident as an endogenous excitatory drive, which can account for the rapid and irregular breathing and the lowered threshold to CO2 of REM sleep. In the present study, hoping to determine more about this endogenous drive, we dissociated the neurons of the respiratory system by mechanically ventilating cats to apnea, in which state respiratory neurons and muscles are excitable, even active, but are not reciprocally bound into a rhythmic network, and compared the activity of medullary respiratory neurons to the activity of the diaphragm. Simultaneous excitation of neurons and the diaphragm in REM sleep would suggest that the endogenous drive has a single source and/or that the effect is hierarchical (i.e., from premotor to motor neurons). Independent excitation would indicate that there are multiple sources and sites of action of the endogenous drive. Methods. Adult, intact and unanesthetized cats were implanted with electrodes for recording the EEG, pontogeniculo-occipital waves, and diaphragmatic electromyo- grams. A fistula was created to allow intubation of the trachea. During recording sessions, the animal breathed and was ventilated through the tracheal fistula, and a microelectrode was passed into the medullary ventral respiratory group. Action potentials of single respiratory neurons were recorded along with sleep parameters, diaphragmatic activity and ventilation. The animal was ventilated to apnea and CO2 was injected into the air delivered by the ventilator to maintain a level that was approximately 85% of the end tidal level in NREM sleep. Results and Conclusion. All possible patterns of excitation of the diaphragm and of a respiratory neuron were observed in REM sleep: i. simultaneous excitation of the neuron and diaphragm, as occurs during spontaneous breathing, ii. excitation of only the neuron, and iii. excitation of only the diaphragm. We conclude that there are multiple sources of endogenous drive to the respiratory system in REM sleep and/or that gating distributes the drive independently to the oscillator, and pre-motor and motor neurons.

Support: NIH HL21257 & HL62589, GAANN P200A80102 Actas de Fisiología 7, 2001 273

HYPOCAPNIA INHIBITS REM SLEEP IN CATS

A.T. Lovering, Dinakar, P., Vidruk, E.H.*, and Orem, J.M. Department of Physiology, Texas Tech University SOM, Lubbock, TX, USA *Department of Preventive Medicine, University of Wisconsin, Madison, WI, USA

At altitude, ventilation is increased, and sleep is disrupted. The increased ventilation in response to hypoxia at altitude causes hypocapnia. To determine whether hypocapnia could contribute to altitude-induced sleep disturbances, we studied the effect of different levels of CO2 and O2 on sleep. Adult cats were prepared for recordings of sleep and respiratory (airflow, tidal CO2, intratracheal pressure and diaphragmatic EMG) parameters. Three protocols in five intact, unanesthetized cats were used. Recording periods were three hours long. In the first protocol (hypocapnic normoxia) the animals were intubated through a tracheal fistula, connected to a ventilator and ventilated to apnea while CO2 was held constant at either 65%, 75%, 85% or 95% of eupneic levels. In the second protocol (hypocapnic hypoxia) animals were intubated and allowed to breathe hypoxic gas mixtures (10-15% O2 in N2) and CO2 was allowed to fall. In the third protocol (isocapnic hypoxia) animals were intubated and allowed to breathe hypoxic gas mixtures and CO2 was maintained at the eupneic level. Control animals breathed room-air through a tracheal tube. The results obtained from observations of more than 300 REM periods revealed that, independent of the fraction of inspired oxygen, REM sleep was least at lower levels of CO2. In hypocapnic normoxia and hypocapnic hypoxia REM sleep episodes and time decreased progressively as CO2 decreased. In isocapnic hypoxia REM sleep episodes and time did not differ significantly from control animals. NREM sleep was not significantly affected by hypocapnia in any protocol. These results indicate that REM sleep disturbances at high altitude may be caused by hypocapnia.

Support: NIH HL21257 & HL62589, GAANN P200A80102 274 Actas de Fisiología 7, 2001

SLEEP INFLUENCES ON HUMAN CUTANEOUS BLOOD FLOW (CBF) STUDIED BY LASER DOPPLER FLOWMETRY (LDF)

Mauro MM Manconi, Valeria Tugnoli, Jay Capone and Enrico Granieri S. Anna Hospital, C. so Giovecca 203, Dpt. of Neurology, Ferrara, Italy

NREM sleep induces a gradual reduction of sympathetic activity that reaches the lowest level during slow waves sleep, whereas REM sleep is characterized by a wide sympathetic variations. CBF is regulated by sympathetic adrenergic innervation that mediate vasocostriction responses. There are some differencies between cranial and limb skin sympathetic innervation. Pletismographic studies evidence during sleep stages difference in CBF related to sympatetic activity in segmental sites. OBJECTIVE: to study CBF variations during the different sleep stages in cranial and segmental sites of human healthy subjects. In particular to evaluate following parameters: 1) mean CBF, 2) vasomotion 3) spontaneous phasic variations, 4) vasodilata- tory response to local thermic stimulation, 5) correspondence between CBF vasoconstriction and sympatethic skin response (SSR) 6) differences between cranial and segmental disctrics of the above mentioned parameters. METHODS: polisomnographic study of 10 subjects (age 24-36) for one adaptation night and for a second recording night, monitoring following parameters: 1) electroen- cephalography (C3, CZ, O2), 2) electroculography (ROC - LOC), 3) miloioideo elec- romiography, 4) forehead and fingertip CBF by LDF 5) forehead and hand SSR 6) skin temperature 7) EKG 8) continous digital blood pressure (Portapress), 9) oronasal flow and toraco-abdominal movements. RESULTS: After following asleep appeare spontaneus phasic vasoconstriction re- sponces (VR) in limb sites not correlated with breathing, body or limbs movements, blood pressure variations. VR frequency decreases gradually during NREM sleep from stage 1-2 to slow waves stages. The higtest VR frequency, in all subjects, is recorded during REM sleep. VR were not observed in cranial sites. Actas de Fisiología 7, 2001 275

THE EFFECT OF SLEEP STATE ON VENTILATORY RESPONSES TO MILD HYPOXIA IN 2-3 MONTH OLD INFANTS

Peter Parslow, Rosemary SC Horne, Richard Harding1, Susan Cranage and T Michael Adamson Department of Paediatrics, Ritchie Centre for Baby Health Research and 1Department of Physiology, Monash University, Melbourne, Australia

An impaired ability to arouse and/or augment breathing in response to a respiratory challenge may be a cause of Sudden Infant Death Syndrome. Previously, ventilatory responses to hypoxia have been investigated in quiet sleep (QS) with little attention given to arousal. Our aim was to compare ventilatory responses to hypoxia in active sleep (AS) and QS in tests inducing arousal and compare them with tests that failed to arouse. 9 term infants, aged 2-3 months, were studied using daytime polysomnography during which ventilatory and arousal responses to hypoxia were characterised. Control data were obtained over one minute prior to 15% O2 inhalation. Tests were terminated at either arousal, after 5 minutes, or if SpO2 dropped to 85%. Breath-by-breath measurements of respiratory rate, tidal volume (VT/kg) and minute ventilation (V’i/kg) were obtained. Summaries of respiratory and blood gas variables were obtained for each 30 second epoch of 15% O2 administration and expressed as a percentage change relative to control levels. We performed 21 tests in AS (8 infants) and 28 in QS (9 infants). All AS tests were terminated by arousal and provided ventilatory data suitable for analysis. Tests that induced periodic breathing or were terminated due to SpO2 reaching 85% were excluded from further analysis. 21 of the QS tests (in 7 infants) were suitable for breath-by- breath analysis. In 8 of these tests (6 infants) the infants did not arouse, while the remaining 13 tests (5 infants) induced arousal. Four infants both aroused and failed to arouse within repetitive QS tests. Repetitive tests were averaged within infants in each arousal or non-arousal condition.

In non-arousal tests during QS infants demonstrated an increase in V’i within 90s of hypoxia onset that decreased towards control levels by 5 minutes. In tests inducing arousal during QS, the initial hyperpnoea was greater than in QS non-arousal and AS tests, however, it too was unsustained. Tests during AS demonstrated the lowest rate of initial hyperpnoea with V’i below control levels at arousal.

As arousal latency was significantly longer in QS than AS (p<0.001), V’i was compared between AS and QS with respect to the point of arousal. While a marked increase in V’i prior to arousal was found in QS, there tended to be a slight decrease in V’i in AS tests prior to arousal. This study demonstrates that infants are able to mount a greater ventilatory response to mild hypoxia in QS than AS, and that arousal may serve as a more potent protective mechanism in AS than in QS.

**Supported by Sudden Infant Death Research Foundation (South Australia) & SIDassist 276 Actas de Fisiología 7, 2001

EFFECTS OF SLEEP STATE AND AGE ON AROUSAL RESPONSES TO MILD HYPOXIA IN INFANTS

Peter Parslow, Rosemary SC Horne, Richard Harding1, Susan Cranage and T Michael Adamson Department of Paediatrics and Ritchie Centre for Baby Health Research, 1Department of Physiology, Monash University, Melbourne, Australia

A major hypothesis to explain Sudden Infant Death Syndrome (SIDS) is that infants fail to arouse when confronted with a life-threatening event. While arousal responses of infants to hypoxia have been investigated in quiet sleep (QS), there is limited knowledge of their responses in active sleep (AS) and the effect of postnatal age on arousability. Therefore, the aim of this study was to compare the effects of sleep state longitudinally in infants up to 6 months of age. 10 healthy term infants born at 38-41 wks gestation (birthweight 3516±178g) were studied at 2-4 wks, 2-3 mo and 5-6 mo post term. Apgar scores at 1 and 5 minutes were 8-9 and 9-10, respectively. Infants underwent daytime polysomnography during which airflow was monitored using a purpose-built pneumotachograph. One minute control periods were obtained prior to each test, following which 15% O was administered. Tests were terminated at either arousal, SpO reaching 85% or at 52 minutes. Replicate tests were obtained. 2 Tests terminated due to SpO reaching 85% were excluded from analyses. The probability of arousal in each sleep2 state was determined using chi-square analysis. Mean arousal latency was then calculated for AS and QS. Paired t-tests were used to investigate sleep state effects at each study age while one-way repeated measures ANO- VA with Bonferroni post-hoc analysis was used to investigate age effects in each sleep state. Significance was taken at p < 0.05. A total of 157 tests were successfully conducted, 94 in QS and 63 in AS. Ten of these tests were excluded from analyses due to SpO reaching 85%. Replicate tests showed no evidence of habituation to hypoxia. All tests2 in AS induced arousal, whereas in QS 88% of tests resulted in arousal at 2-3 wks, 58% at 2-3 mo and 61% at 5-6 mo. Significant differences between sleep states (p<0.01) in probability of arousal were present at 2-3 mo and 5-6 mo. Arousal latency was significantly longer in QS than AS at each age, however no age-related effects were found. Our study has demonstrated that the probability of arousal to mild hypoxia is significantly lower in QS than AS at 2-3 months and 5-6 months post-term age. Furthermore, in infants less than 6 months, arousal latency is significantly longer in QS than in AS and is not affected by postnatal age.

Supported by the Sudden Infant Death Research Foundation (South Australia) Actas de Fisiología 7, 2001 277

MICROINJECTION OF OREXIN A AND B INTO TRIGEMINAL MOTOR NUCLEUS INCREASES MASSETER MUSCLE TONE

John H. Peever, Y.Y. Lai and Jerome M. Siegel Department of Psychiatry and Biobehavioral Sciences, School of Medicine, University of California, Los Angeles & VAGLAHS Sepulveda, North Hills, CA 91343, USA

Orexins (hypocretin) are excitatory neuropeptides that are involved in a wide-range of physiological processes, including sleep-wake homeostasis, and regulation of motor control. Deficiency in orexin neurons is implicated in narcolepsy, which is characterized by poor sleep homeostasis and intrusions of REM sleep-like muscle atonia during wakefulness (cataplexy) (1). Recent evidence suggests that orexin-synthesizing neurons of the lateral hypothalamus are relatively active during wakefulness and inactive during sleep (2). Orexins neurons send axonal projections throughout the brain and spinal cord, including both trigeminal and hypoglossal motor nuclei (3), which contain motoneurons that express orexin receptors (4). Based on orexin projections to motor nuclei and the presence of orexin receptors on motoneurons, it is hypothesized that orexins are directly involved in the regulation of motoneuronal excitability across the sleep-wake cycle. To assess the role of orexins on muscle tone regulation, we tested the hypothesis that microinjection of orexin A and B into the trigeminal motor nucleus would excite masseter muscles, which are innervated by trigeminal motoneurons. In 6 decerebrate, unanaesthetized cats, we unilaterally microinjected 0.5 ml of 0.1 mM orexin A or B into the trigeminal motor nucleus while monitoring EMG activity of left and right masseter muscles. Injection of orexin A into the trigeminal motor nucleus increased (Wilcoxon’s matched-pairs sign-ranked test: df = 5; T = 2; p = 0.028) ipsilateral masseter muscle tone (integrated EMG activity) by 154.3 ± 209.3 % (range: 4.1 482.8 %) compared with baseline control. The latency and duration of the response were: 7.0 ± 6.6 s (range: 1 19 s) and 17.3 ± 23.1 min (range: 1.5 52.0 min), respectively. Injection of orexin B into the trigeminal motor nucleus increased (df = 5; T = 2; p = 0.028) ipsilateral masseter muscle tone by 105.7 ± 153.0 % (range: 4.1 402.4 %) compared with baseline control. The latency and duration of the response were: 5.3 ± 3.1 s (range: 2 10 s) and 2.5 ± 3.5 min (range: 0.58 9.7 min), respectively. There was no significant difference between the effect of microinjection of orexin A or B on masseter muscle tone increase (df = 5; T = 2; p = 0.5), latency to response (df = 5; T = 2; p = 0.916), or response duration (df = 5; T = 2; p = 0.173). Unilateral injection of 0.5 ml of artificial cerebral spinal fluid into the trigeminal nucleus (n = 5) had no effect on the EMG activity of left or right masseter muscles. We conclude that orexin A and B postsynapically excite trigeminal motoneurons, and are directly involved in the regulation of muscle tone. Loss of orexin function may therefore disturb normal motor processes, and thereby contribute to the motor system pathology of narcolepsy/cataplexy.

References: (1)Thannickal TC, et al. Reduced number of hypocretin neurons in human nacrolpesy. Neuron, 27:469-474, 2000. (2)Estabrooke IV, et al. Fos expression in orexin neurons varies with behavioral state. J. Neurosci., 21(5):1656-1662, 2001. (3)Marcus JN, et al. Differential expression of orexin receptors 1 and 2 in the rat brain. J. Com. Neurol., 435:6-25, 2001. (4)Fung SJ, et al. Hypocretin (orexin) input to trigeminal and hypoglossal motoneurons in the cat: a double-labelling immunohistochemical study. Brain Res., 903:257-262, 2001. 278 Actas de Fisiología 7, 2001

AUTONOMIC MARKERS OF MICROAROUSAL DURING NORMAL HUMAN SLEEP

L. STANER1, G. VIARDOT1, F. CORNETTE1, A. MUZET2, R. LUTHRINGER1, J. HABA1, J.C. ROEGEL1, J.P. MACHER1 1. FORENAP – Institute for Research in Neurosciences, Neuropharmacology and Psychiatry, Centre Hospitalier, Rouffach, France ; 2. Laboratoires des Régulations Physiologiques et des Rythmes Biologique chez l’Homme, CNRS, Strasbourg, France

Studies in patients with obstructive sleep apnea or with periodic leg movements have suggested that autonomic responses such as increases in heart rate and blood pressure may provide a sensitive marker of brief arousal during sleep. In healthy subjects, Sfor- za et al. (2000) showed a relationship between heart rate (HR) variation and cortical (i.e., microarousal –MA– or transient activation phases) or subcortical (i.e., bursts of K-complexes or of delta waves) indices of arousal. In the present study we further characterise the autonomic markers of arousal by analysing the relationship between changes in HR and blood pressure (BP) and microarousal occurrence in healthy subjects. After a habituation night, 12 males healthy volunteers (aged 32.5 +/- 6.8 years) underwent one experimental session during which sleep EEG, ECG, and beat-to-beat blood pressure (Portapres) were simultaneously recorded. Recordings were visually scored at 30-sec intervals according to Rechtschaffen and Kales and MA were identified using ASDA scoring rules. Autonomic arousals were defined as either periods of at least six successive decreases in R-R values (HR arousals) or as six successive increases in mean BP (BP arousals). Results showed a higher variability of HR and BP during a MA than during a same time frame preceding the MA (in 82.8 % of MA an increase in R-R intervals variance was observed and in 80.9 % an increase in mean BP variance). During the entire sleep period, modest correlations were found between the number of autonomic arousal and MA occurrences (r=0.33 for HR arousal and r=0.23 for BP arousal); however, during REM sleep, a strongest correlation was observed between MA and BP arousal (r=0.67) but not between MA and HR arousal (r=0.16) whereas the converse was true for SWS (r=0.76 between HR arousal and MA; r=-0.04 for BP arousal and MA ). These results suggest that autonomic and cortical arousals are differently regulated in REM and SWS. Actas de Fisiología 7, 2001 279

THERMOREGULATORY ADAPTATIVE MECHANISMS IN NEONATES’ SLEEP STAGE

Telliez F., Chardon K., Bach V. , Krim G.*, Libert J.P. ETP-APC, Fac. of Medicine, Univ. of Picardie-Jules Verne, 3 rue des Louvels, * Pediatric Unit, Amiens, France.

Introduction Contrary to human adults and animals, body temperature regulation is fully efficient in Active (AS) and Quiet (QS) sleep in neonates. Thus, brief cool exposures increase the oxygen consumption (VO2) through non shivering thermogenesis by stimulation of brown adipose tissue. During a prolonged cool condition, there are thermoregulatory adaptative processes leading to a protection of deep body temperature. In adults, this is explained by an increase of the sensitivity of the thermoregulatory system due to a change in the gain or a shift in the hypothalamic set temperature. In neonates, the changes of these adaptative mechanisms in the different sleep stages have never been studied and it would be unwise to extrapolate the results found in brief cool exposure to prolonged condition. To assess whether adaptative thermoregulatory processes are operative in AS as well as in QS, 6 neonates (33 ± 4 weeks gestation,19 ± 14 days old, 2.3 ± 0.2 kg) body mass were exposed during a prolonged cool exposure.

Material and methods Sleep, esophageal (Tes), skin (Tsk) temperatures and VO2 were recorded during the baseline (B: 33.3 ± 0.6°C), the first and last sessions of the cool acclimation period which lasted 75 hrs (air temperature: 2°C under the thermoneutrality). Protocol

1st last first baseline experiment session session B C1 C4

T incubator (°C) thermoneutrality Cool acclimation period: TN- 2°C (75 h) TN - 2°C days

Results The first cool exposure did not affect Tes whereas Tsk sharply decreased.

VO2 increased by 25% throughout the acclimation period. This increase was significant in both AS (+1.1 ± 0.5 ml.min-1.kg-1; p=0.005) and QS (+1.6 ± 0.4 ml.min-1.kg-1; p=0.005) and was not significantly different between the sleep stages. The sensitivity of thermoregulatory system, described by the ratio VO2/Tes and VO2/Tsk, was significantly increased by cool acclimation in both AS and QS. Conclusion The results pointed out that the adaptative thermoregulatory processes occurring during cool exposure are operative in AS and QS. These processes can be mediated through changes at the level of the central controller (changes in the gain and/ or in the hypothalamic set point temperature) and/or by an increase in the sensitivity of the thermoregulatory effector (brown adipose tissue).

Keywords sleep, neonate, thermoregulation, cool acclimation 280 Actas de Fisiología 7, 2001

EFFECTS OF PROLONGED SLEEP LOSS ON GENE EXPRESSION

Giulio Tononi and Chiara Cirelli Department of Psychiatry, University of Wisconsin/Madison, USA

Knowledge of the molecular correlates of sleep and wakefulness is essential if we are to understand the restorative processes occurring during sleep and the cellular consequences of sleep deprivation. To this end, we have recently performed a systematic screening of gene expression across behavioral states. We have compared mRNA levels of ~10,000 genes in the rat cerebral cortex after 3-8 hours of spontaneous sleep, sleep deprivation, or spontaneous waking (1-3). Only <0.1% of these genes was differentially expressed across behavioral states. A few of them, unknown, had higher expression in sleep relative to waking/sleep deprivation. By contrast, most genes were upregulated in waking/sleep deprivation relative to sleep. They included immediate early genes, mitochondrial genes, growth factors, and chaperones. In this study we want to extend this analysis to characterize the molecular consequences of long-term sleep deprivation. Thus, we have performed a systematic screening of cortical gene expression in rats sleep deprived for 4-10 days using the disk-over-water apparatus. Nine pairs of long-term sleep deprived (TSD) rats and yoked control (TSC) rats were sacrificed when body temperature was at baseline level (n=4), 1°C below baseline (n=3), or 5°C below baseline (n=2). Body temperature in TSC rats was always at baseline level. Non-REM sleep and REM sleep were reduced by 80% and 96% in TSD rats, and by 28% and 37% in TSC rats, respectively. Total RNA was extracted from the cerebral cortex and used to screen ~10,000 genes as before (1-3). Almost all genes previously found (1-3) to be upregulated by waking and short-term sleep deprivation did not change their expression or were down-regulated in TSD rats. In fact, almost all genes whose expression was affected by prolonged sleep loss showed lower mRNA levels in TSD rats relative to TSC rats and to normally sleeping, spontaneously awake, or short-term sleep deprived rats. Only 2 genes, coding for the enzymes arylsulfotransferase (AST) and serum/glucocorticoid-regulated serine/threonine kinase, were upregulated in TSD rats. The induction of AST appeared to be related to the duration and/or intensity of sleep loss. AST, in rodents, is responsible for the catabolism of catecholamines and its strong induction in TSD rats suggests that sleep may be important to interrupt the activity of catecholaminergic systems active during waking. These results also suggest that sustained sleep loss may be associated with decreased levels of many cortical transcripts. (Supported by Neurosciences Research Foundation). (1) Cirelli and Tononi, Molec Brain Res 56:293,1998; (2) ibid, Brain Res 885:303,2000; (3) ibid, J Neurosci 20:9187,2000. Actas de Fisiología 7, 2001 281

CEREBRAL CIRCULATION IN SLEEP: RESPONSE TO HYPOXIA

Adrian M Walker, Daniel A Grant, and Jennene Wild Ritchie Centre for Baby Health Research, Monash Institute of Reproduction and Development, Monash University, Melbourne, Australia

Rapid-eye-movement (REM) sleep is remarkable among sleep-wake states for its elevated cerebral blood flow (CBF). Recently it has been proposed that sensitivity of cerebral vessels to low PO2 is a fundamental property that determines the level of CBF (Lenzi et al 1999). If this were to be true, oxygen levels should be a powerful determinant of CBF in all behavioural states. In two specific tests of the hypothesis we: (a) contrasted the response of the cerebral circulation during REM and non-REM sleep to transient, episodic arterial oxygen desaturations designed to mimic sleep apnea (HY- POXIA IN SLEEP); and (b) determined the changes of CBF associated with REM and non-REM occurring against a background of continuous hypoxia (SLEEP IN HYPOXIA). Lambs (n = 8) were instrumented to record beat-beat cerebral blood flow (CBF) using a 2 mm diameter TRANSONICS™ transit time ultrasonic flow probe implanted around the superior sagittal sinus, and implanted with catheters to record cerebral perfusion pressure (CPP) and electrodes to define sleep-wake states. Arterial oxygen saturation (SpO2) was recorded with a NELLCOR™ pulse oximeter. CBF was contrasted between REM and non-REM sleep occurring naturally during normoxia (FiO2

0.21) and during hypoxia induced by reducing FiO2 to 0.10 either (a) transiently (60 sec) within individual sleep epochs (HYPOXIA IN SLEEP); or (b) continuously (1 hr) across repeated sleep epochs (SLEEP IN HYPOXIA). Under baseline (normoxia) conditions, CBF (ml/min) was significantly greater in REM than in non-REM (19±1 vs. 15±1, P<0.01, mean±SE, n = 4, t-test). During continuous hypoxia, significant increases of CBF from control values occurred in REM (24±7 percent, P<0.05) and non-REM (14±5 percent, P<0.05), and the greater CBF (ml/min) of REM (23±2) compared with non-REM (17±1) was preserved (P<0.02). Similarly, there was preservation of the significantly greater CBF of REM compared with non-REM under conditions of transiently (60 sec) imposed hypoxia (n=6). As the major cerebral blood flow differences of sleep (REM > non-REM) are preserved in hypoxia, regardless of its duration, sensitivity of cerebral vessels to PO2 appears to be a fundamental determinant of cerebral blood flow in sleep.

Lenzi P-L, Zoccoli G, Walker A & Franzini C. Cerebral blood Flow Regulation in REM Sleep: A Model for Flow-metabolism Coupling. Archives Italiennes de Biologie, 1999 282 Actas de Fisiología 7, 2001

BLOOD-BRAIN BARRIER PERMEABILITY TO GLUCOSE IN DIFFERENT BRAIN FUNCTIONAL CONDITIONS

Giovanna Zoccoli, Tijana Bojic, Tullia Cianci, Carlo Franzini, Pierluigi Lenzi, Gioia Longhi, Silvia Predieri and Alessandro Silvani Department of Human and General Physiology, University of Bologna, Italy

Blood-Brain Barrier (BBB) permeability modulation may subserve homeostasis of the brain’s internal environment in front of local metabolic and hemodynamic changes. To clarify the issue we measured regional cerebral Blood Flow (BF) and glucose Perme- ability-Surface area product (PS) on adult Sprague-Dawley rats in different cerebral functional conditions, namely Quiet Wakefulness (QW, n=3) or Active Sleep (AS, n=2) after one week’s recovery from surgery in unrestrained animals, and 1% Halothane, 70% Nitrous oxide anesthesia (HN, n=5). Briefly, 3H-D-Glucose (30 mCi) and 14C- iodoantipyrine (10 mCi) were bolus injected in the inferior vena cava, while blood was withdrawn from an arterial catheter for 12 s at a constant rate (1). The animal was then decapitated with an air-pressure operated blade, controlled from outside the box. Blood and brain tissue sample radioactivity measures (b scintillation spectrometer) allowed BF and PS computation in the medulla, pons, cerebellum, midbrain, diencephalon, hippocampus and cerebral hemispheres. Mean PS value was significantly lower in the HN group than in either QW or AS (9.59±3.64 vs. 15.69±2.49 and 9.59±3.64 vs. 14.49±3.22 respectively, ml/min.100g, means±SD, p<0.001, ANOVA with Dunnet’s T3 correction). In each functional condition (HN, QW, SA) regional PS was highly and positively correlated with regional BF (Pearson, p<0.005). Since the fraction of perfused brain capillaries (hence capillary surface area S) is constant (2), our results can be attributed to changes of BBB glucose transporter kinetics reflecting cerebral energy demand and, within groups, by a positive correlation between mean regional BF and uniformity of individual capillary flow rates: a higher BF increase in the fraction of slowly perfused cerebral capillaries equals in fact functionally to an increase in S (3).

1) La Manna, J. C. e Harik, S. I. Regional studies of blood-brain barrier transport of glucose and leucine in awake and anesthetized rats. J. Cereb. Blood Flow Metab., 1986, 6: 717-723. 2) Zoccoli, G., Lucchi, M. L., Andreoli, E., Bach, V., Cianci, T., Lenzi, P. and Franzini, C. Brain capillary perfusion during sleep. J. Cereb. Blood Flow Metab., 1996, 16: 1312-1318. 3) Kuschinsky, W. e Paulson, O. B. Capillary circulation in the brain. Cerebrovasc. Brain Metab. Rev. 1992, 4: 261-286. Actas de Fisiología 7, 2001 283

SLEEP DISTURBANCES AMONG NURSING HOME RESIDENTS

Bjørn Bjorvatn and Arne Fetveit Dept. of Public Health and Primary Health Care, Bergen, Norway

This study assesses the prevalence and characteristics of sleep disturbances among an entire nursing home population (n = 29), consisting mainly of demented patients. Sleep was evaluated for 14 consecutive days using actigraphy, nursing staff observations and patient sleep diaries. No alterations were made in every-day routines or medications during the observation period. Actigraphy showed a mean sleep onset latency of one hour and a mean wake after sleep onset of more than two hours, while there was no findings of early morning awakening. On average, more than 13 hours were spent in bed, and mean sleep efficiency was 75%. Nursing staff reported sleep onset latency of more than 30 minutes in 158 of the 203 analyzed days, while early morning awakening was reported in only 12 of 203 days. Actigraphy and nursing staff observations gave similar results. The patients themselves rated sleep to be of better quality, but the reliability of the self-reported ratings can be questioned because of the high prevalence of cognitive impairment among the patients. In conclusion, sleep disturbances were common among the residents in this nursing home. Sleep onset latency was prolonged, and the patients experienced frequent wake bouts after sleep onset. The diminished ability of sustained sleep may have been influenced by the prolonged time in bed. 284 Actas de Fisiología 7, 2001

PURCHASING POWER AND PREVALENCE OF SLEEP COMPLAINT ABOUT A SURVEY OF 205 347 PEOPLE FROM 1988 TO 1998

D.Cugy1, J.Paty1, J. Balan3, J. Vinclair3, S. Cugy3, J-L.Lenain3, J-P Giordanella4 (1) Clinique du Sommeil – CHU Pellegrin – 33076 Bordeaux cédex, (2) C.E.S. Bordeaux – ZAC Ravezies – 33000 Bordeaux, (3) CNAM-TS

Introduction: Primary healthcare checkup are regularly performed by French welfare centers. We report analysis of Sleep disorders complaint registered from 1988 to 1998 in Bordeaux-Cauderan and Cenon CPAM welfare centers. The prevalence of sleep disorders is estimated from a total of 205 347 checkup. The population is segmented by age (18-25:19 332, 25-35:46 694, 35-45:51 072, 45-55:51 072, 55-65 :32 658, 65- :7 705), gender (male: 101 801, 1998-9605; female: 103 546) and population catergory (General :143 438, Précaires :22 550, Prioritaires :31654). Results: data shows a relation ship between sex and 30,0 105,0 age (Tab 1, 2). Surprisingly 25,0 103,0 we found a relation between 20,0 101,0 Prev 18-25 Prev 25-35 Purchasing Power and Prev- 15,0 99,0 Prev 35-45 Salaires nets alence of Sleep Complaint 10,0 97,0 (Fig 1). There is a significant 5,0 95,0 correlation (R2 = 0,718, p < 0,0 93,0 0,0079). These Data is in re- 1990 1991 1992 1993 1994 1995 1996 1997 1998 Prev 18-25 9,8 9,7 11,1 14,6 15,0 15,8 16,5 20,2 18,6 lationship with M.Ohayon Prev 25-35 11,7 13,9 14,2 16,1 17,6 16,8 17,5 21,6 20,3 Prev 35-45 17,6 17,9 18,4 21,4 20,9 19,9 23,2 25,9 23,9 Findings relatively to low in- Salaires nets 100,0 100,4 100,8 99,8 99,1 98,9 97,4 98,3 100,2 come and sleep complaint (Fig 1) Prevalence of Sleep Complaint and Purchasing Power

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 18-25 ans 9,4% 8,2% 5,6% 6,6% 7,9% 9,0% 12,2% 11,2% 12,5% 14,9% 15,6% 10,3% 25-35 ans 8,2% 10,6% 10,0% 11,7% 11,6% 12,9% 13,2% 13,4% 15,2% 18,6% 17,9% 12,9% 35-45 ans 14,2% 14,6% 14,8% 14,5% 15,5% 17,5% 17,4% 16,0% 20,7% 23,3% 21,9% 17,0% 45-55 ans 11,0% 17,4% 16,6% 18,0% 17,6% 20,8% 19,9% 18,9% 22,9% 24,3% 24,6% 19,8% 55-65 ans 4,1% 18,9% 18,0% 19,6% 17,9% 20,9% 18,5% 19,2% 20,4% 23,4% 20,4% 19,6% 65+ ans - - - - 16,9% 21,2% 18,5% 23,2% 25,3% 24,5% 22,1% 21,7% Global 10,1% 14,4% 13,9% 14,8% 14,9% 17,5% 17,1% 16,6% 19,3% 22,3% 21,0% (Tab 1) Prevalence in Male Population

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 18-25 ans 14,1% 12,1% 12,8% 11,8% 13,2% 18,2% 16,9% 18,9% 19,0% 22,9% 20,8% 16,7% 25-35 ans 13,8% 15,0% 13,3% 16,0% 16,7% 19,3% 21,5% 19,6% 19,4% 23,4% 22,3% 18,3% 35-45 ans 19,9% 21,6% 20,6% 21,6% 21,7% 26,2% 24,7% 23,9% 25,6% 27,8% 26,2% 23,7% 45-55 ans 22,8% 28,8% 28,8% 29,4% 27,8% 34,4% 32,1% 29,2% 29,4% 36,8% 32,4% 31,0% 55-65 ans 27,8% 34,5% 34,4% 33,1% 33,7% 36,5% 35,1% 34,1% 35,4% 36,8% 35,8% 34,9% 65+ ans - - - - 35,0% 35,1% 36,7% 40,6% 39,5% 40,3% 37,0% 37,3% Global 18,0% 22,1% 21,2% 22,9% 22,8% 27,6% 27,4% 25,5% 26,0% 30,8% 28,6% (Tab 2) Prevalence in Female Population

Reference: Ohayon M. Epidemiological study on insomnie in the genral population. Sleep 1996 Apr;19(3 Suppl):S7-15 Actas de Fisiología 7, 2001 285

COGNITIVE BEHAVIOURAL FATIGUE, MOOD CHANGES AND SOMATIC DISTRESS AS RESPONSES TO INSUFFICIENT SLEEP IN MIDDLE-AGED FEMALES

Edéll-Gustafsson U. Department of Medicine and Care, Nursing Science,Faculty of Health Sciences, Linköping University. Linköping, Sweden

Objective: To examine degrees of cognitive behavioural fatigue, mood changes and somatic responses to sleep loss in middle-aged females with and without insufficient sleep. Furthermore, to explore possible links between effects of sleep loss and specific sleep disturbances. Subjects: SS included 156 females, mean age 43.9 (SD 10.4 years). Settings: Three casualty departments. Method: Questionnaire study. The questionnaire measured sleep quality, general health, strains and symptoms related to working conditions, and effects of sleep loss. Statistics: Descriptive statistics, two-way analysis of variance (independent group) and logistic regression analysis were used. Results: About two fifths of the females had perceived insufficient sleep during the last six months. These females perceived significantly worse sleep quality (p<0.0001) and a higher degree of strains according to working conditions (p<0.0001) than the others. Palpitation and dysphoria as effects of sleep loss was predicted independently by worse sleep quality (OD 4.0, OD 95% C.I. 1.5-10-6, p<0.005 and OD 2.7, OD 95% C.I. 1.3- 5.3, p<0.007, respectively). Cognitive behavioural fatigue was predicted by troubled sleep (OD 2.5, OD 95% C.I. 1.3-5.0, p<0.01). This was increased 10 times by palpitation as effects of sleep loss, which per se associated with increased daytime tiredness. The responses to insufficient sleep were most prominent among the females suffering from gastrointestinal problems and pain several days/week or even every day during the last six months. In conclusion, responses to reduced sleep quality indicate to constitute a form of stress with sympathetic activation, increased susceptibility to infections, moderate cognitive impairment, mood changes and somatic distress in middle-aged females.

Keywords: Insomnia, sleep quality, stress, cognitive impairment, mood, females. 286 Actas de Fisiología 7, 2001

DAYTIME SLEEPINESS: A PREVALENCE STUDY AMONG 9946 WORKERS

Jean-Daniel Guieu, Xavier Lenne, Therese Lebrun, Jean-Claude Sailly and Isabelle Poirot Neurophysiologie, Hopital Salengro, CHRU, Lille, France

Investigation about sleep disturbances and daytime sleepiness prevalence in a general population(n=13500;16-64 year old)in northern France.Instrument survey : questionnaire including a- 48 items about sociodemographic, lifestyle,health and sleep related variables; b- the Epworth sleepiness scale(ESS); c- the Nottigham health profile(NHP). In the subpopulation of workers(n=9946), 21,5% had an ESS score >or= 11(Male 19,4%; Female 24,6%; p<0.001).The sleep functionning of NHP was degraded for workers with excessive sleepiness (p<0.001).Prevalence of daytime sleepiness is higher than expected in this population and associated with quality of life limitation Actas de Fisiología 7, 2001 287

POLISSOMNOGRAFIC EVALUATION IN INTERSTATES BUSDRIVERS

Eduardo Henrique Rosa Santos, Marco Tulio Mello and Sergio Tufik Sao Paulo, Brazil

In shiftworks and in long distance driving, the human mistake can be the more accident determinant. The attention lose, the inadequate observations and the cognitive mis- takes corresponding about 40% of cases. OBJECTIVE: Our objective was make a polissomnografic examination (PSG) for evaluation of sleep patterns and sleep-related complaints of profession bus drivers. Metohods and Results: these study was divided in two phases, after and before one day or night of work. In the first phase (n=30) the evaluation was : 1º day Adaptation of bus drivers at PSG, 2º day overnigth polysomnography, 3º day The Multiple sleep latency test (MSLT). In the second phase (n=18) after one day or night of work, the evaluation was: the PSG examination, and two hours after the bus drivers awake, he started the MSLT. Results: Sleep latency 15 x 4.2 min.*; REM sleep latency 86.6 x 77.7 min.*; Total sleep time 382.5 x 380.2 min.**; Sleep efficiency 8436% x 78 %; Time awaken 51.9 x 79.0 min.; Excessive somnolence 41% x 50%* and Period limb movement 30% x 29**; the cases of Obstructive sleep apnea (OSA) were divides into three groups basead on apnea-hypopnea index (AHI): OSA1-AHI 5-15 (20% x 21% of cases ), OSA2-AHI 15-30 (3.3% x 10% of cases), OSA3-AHI >30 (3.3% x 5.2 of cases); respectively on the first x second phase. (*p>.05; **p-n.s). Conclusions: The data indicated that the bus drivers present sleep latency and REM sleep latency higher at the 1ºphase than 2º phase. Time awaken and excessive sleepiness smaller at the 1º phase than 2º phase. The excessive somnolence were high in two phases of present study. These results indicated that the bus drives can be sleep deprived, and the shift works can perturb theirs sleep.

Finantial support: AFIP Associação Fundo de Incentivo à Psicofarmacologia; CEPID FAPESP. 288 Actas de Fisiología 7, 2001

COGNITIVE PERFORMANCE IN MILITARY OPERATIONS: EFFECT OF A ZOLPIDEM-SLOW RELEASE CAFFEINE COMBINATION

Maurice Beaumont, Denise Batejat, Christophe Pierard, Olivier Coste, Claire Turner, Barbara Stone and Didier Lagarde IMASSA, Dept of Physiology, Bretigny Sur Orge Cedex, France

Most of military operations often demand 24-hour round-the-clock capability so that they need a high level of performances overnight and a good quality of sleep during short rest periods before working. We assessed the usefulness of 300-mg slow release caffeine (SRC)(1), alone and combined with 10-mg zolpidem (ZOL) versus placebo (PBO) on performance overnight and sleep in the preceding rest period. 8 healthy male volunteers completed a 42h work/rest schedule including a 6h period of sleep (21:00 to 03:00) followed by a 6h period of work (03:00 to 09:00), a second 6h period of sleep (09:00 to 15:00), a 18h period of work (15:00 to 09:00) and finally a third 6h period of sleep (09:00 to 15:00). Each subject repeated this schedule four times, in a randomized order, to assess the efficiency of the following drug combinations: PBO-PBO, PBO- SRC, ZOL-PBO, ZOL-SRC. Subjects were given ZOL or PBO at the beginning of the second sleep period (9:00), and SRC or PBO at 0:00, meaning midway through the 18h period of work. Cognitive performance was assessed using a 90-min battery of psychometric tasks (1) at regular intervals and sleep was objectively measured from EEG recordings. Compared to PBO, ZOL increased stage 4 of sleep by 20 min (p < 0.0001) and SWS by 15 min (p < 0.01), without altering REM sleep duration. Performance decreased from 13h to 18h in PBO-PBO and ZOL-PBO conditions (p < 0.0001) but was globally maintained all over the 18h work period in PBO-SRC and ZOL-SRC conditions. To conclude, combined or not with ZOL, SRC maintains performance over a whole night of cognitive work. It is also important to point out that ZOL improves sleep during daytime rest periods before working without inducing any impairment of performance after wake up.

1. Lagarde D, Batéjat D, Sicard B, Trocherie S, Chassard D, Enslen M, Chauffard F. Slow release caffeine: a new response to the effects of a limited sleep deprivation. Sleep 2000; 23(5):651- 661. Actas de Fisiología 7, 2001 289

DAYTIME ACTIVITY AND SLEEPINESS AFTER SLEEP DEPRIVATION: A CROSSOVER STUDY

F. Cavaglia, D. Pires-Barreira, A. Matos-Pires and F. Arriaga Department of Psychiatry, Faculty of Medicine of Lisbon and Sleep Unit, British Hospital, Lisbon, Portugal

The aim of this study was to identify the consequences on daytime activity and sleepiness of sleep deprivation (first half of the night) on healthy subjects. Methods: A sample of 12 healthy volunteers was selected. All of them were screened for past and present psychiatric, neurologic and medical disturbances. A crossover design was utilised in which each healthy subject provided data related to the wake period after being exposed to two different experimental conditions: a “normal” sleep period (23:00 to 07:00) and a deprived sleep period (03:00 to 07:00). Counterbalanc- ing was one of the control procedures of spurious variables like fatigue and being familiarised with the chosen measures. To avoid a carryover effect between the first register and the second one an interval of seven days was imposed. Registers were taken only during weekdays. Activity was measured objectively by means of an actigraph (Ambulatory Monitoring, Ardsley, USA). The actigraphs had to be worn on the non-dominant wrist during a 24-hour period. Epochs of one minute were defined. The number of zero crossings per minute was stored. The measure of activity extracted in this study was the mean activity during the wake period (07:00 to 23:00). The following subjective evaluations were applied: Epworth Sleepiness Scale (ESS), Oswald and Norris visual analogue scales. The Wilcoxon test was employed to compare the two related samples. Results: Sleep quality (Oswald scale) was worse after a deprived sleep period (Z=- 1.97; p=0.048). Two Norris Scales discriminated the two conditions, favouring the non-deprived sleep period (Norris 12; Z=-2.60; p=0.009) - (Norris 17; Z=-1.98; p=0.048). Daytime sleepiness was significantly higher after a deprived sleep period (ESS total score; Z=-2.25; p=0.024). The mean activity along the wake period was the most discriminant measure among all measures taken (Z=-3.07; p=0.002). Conclusions: The finding that sleep quality worsens with sleep deprivation of the first half of the sleep period is curious in the sense that does not support the hypothesis that this kind of sleep deprivation could have an “euphoric” effect on healthy subjects. The ESS is a major measure (even if subjective) in what regards daytime sleepiness. The ESS was also able to detect a difference between the two experimental conditions. However, it was “surpassed” by an objective measure of activity (mean activity). This finding is not only interesting, but preludes that an objective measure of activity could demonstrate a concurrent validity with an established measure of sleepiness like ESS. We conclude that activity measures cannot be excluded from the present controversy about gold standard measures of daytime sleepiness. 290 Actas de Fisiología 7, 2001

INCREASED EXPRESSION OF PREPROHYPOCRETIN MRNA IN RAT HYPOTHALAMUS AFTER SLEEP DEPRIVATION AND SLEEP REBOUND

Vânia D’Almeida1,2, Sergio Tufik1 , Débora C. Hipólide1, Roger Raymond3, K. Barlow3, Mario Pedrazzoli1 and José N. Nobrega3. 1Department of Psychobiology and 2Pediatrics, Universidade Federal de São Paulo, São Paulo, Brazil, and 3Neuroimaging Research Section, Centre for Addic- tion and Mental Health, Toronto, Canada

Hypocretins (orexins), hypothalamic neuropeptides originally associated with feeding behavior, have been causally linked to canine narcolepsy, and have been recently found to be substantially depleted in post-mortem hypothalamus of human narcoleptic patients. Other evidence suggests a wider role for these peptides in the regulation of the sleep-wake cycle. These findings prompted us to examine the effects of sleep deprivation on the expression of preprohypocretin, the precursor of both hypocretin A and B, at the mRNA level. Rats were deprived of sleep using the classical platform method. After 96 hr of sleep deprivation one group was sacrificed; a second group was returned to the home cage and allowed to sleep for 24 hr before being sacrificed (the sleep rebound group); controls remained in the home cages throughout. All brains were rapidly removed, frozen over dry ice and stored at -80 °C. Twenty micron coronal sections were prepared at 0.3 mm intervals covering the entire extent of the hypothalamus. In situ hybridization for preprohypocretin was performed using a 48 base oligonucleotide complementary to bases 177-215, tagged at the 3' end with 35S-dATP. Autoradiographic analyses were performed with the MCID AIS/C system and took into account not only the optic density of signals along the anterior-posterior axis, but also the proportion of area taken by the signals within a fixed size sampling window. Preprohypocretin ISH signals were entirely confined to the hypothalamus but formed a continuum across different nuclei. The mean density of ISH signals (mCi/g) averaged over the entire hypothalamus was increased by 12% in the Sleep Deprived group (6.58±0.43) and by 18% in the Rebound group (6.97±0.44), compared to Controls (5.89±0.29; p <0.05). A second series of measures were taken at the level where the highest number of grains was found in each brain. Mean optical density (mCi/g) at that level was 12% and 18% higher in the Rebound group (11.82± 0.4) than in the Control (10.51±0.62) and Sleep Deprived groups (10.05±0.68; p<0.05), respectively. A measure combining optical density and area revealed a 30% increase in the sleep deprived group (1.81±0.39) and an 88% increase in the Rebound group (1.39±0.21) relative to controls (2.61±0.34; p< 0.001). These observations suggest that sleep deprivation increases preprohypocretin mRNA levels and that this effect becomes significantly more pronounced following sleep rebound. It will be of interest to determine which of the two forms of hypocretin is primarily affected and whether these changes are reflected in levels of expression of receptors, particularly the hypocretin 2 receptor that has been found to be affected in canine narcolepsy. Changes in the hypothalamic hypocretin system may relate to metabolic and energy regulatory changes that are characteristic of sleep deprivation. Supported in part by funds from AFIP (Brazil) and CAMH (Canada) Actas de Fisiología 7, 2001 291

EFFECTS OF TOTAL AND PARADOXICAL SLEEP SELECTIVE DEPRIVATIONS IN RATS

Darchia, ND, Gvilia, ID, Oniani, TN Dept. of Neurobiology of Sleep-Wakefulness Cycle, I. Beritashvili Institute of Physiology, Tbilisi, Georgia

Introduction. The effects of total sleep deprivation (TSD) as well as paradoxical sleep deprivation (PSD) on sleep-wakefulness cycle (SWC) in rats are controversial. A special interest draw results of chronic TSD or PSD obtained by means of the Disk-Over- Water method which suggest that in all the deprivation types a huge rebound of the PS and modest increase of slow wave sleep (SWS) does occur. Because TSD per se, moreover “non-stressful” method for its elicitation would be used, results in a conflict between a need for sleep and impossibility of its satisfaction, it must be accompanied by stress processes. The question is rising: do different data result from different method or length of deprivation procedure? Methods. Chronic adult male rats (n=8) were recorded during 24h. TSD as well as PSD, lasted 8-12h. Subjects were monitored continuously and whenever there appeared the signs of SWS in case of TSD, and PS in case of PSD, animals were awakened by threshold stimulation for behavioral arousal. Recovery periods lasted 12-16h. Behav- iors such as grooming, rising, ambulation and sniffing were recorded digitally. The animals’ body weight, volume of consumed food and the number of bolus were monitored daily. The data were processed statistically by the Student’s t-test. Results. Significant increase in the frequency of SWS onset during TSD and PS - during PSD occurred. Intensification of stimulation parameters was required. The longer was a deprivation procedure, the harder was its accomplishment. After 6-8h TSD, the need for sleep was so pressing that the animal managed to “smuggle” the SWS EEG fragments. In post TSD period the SWS volume as well as delta power increased significantly. There was no compensatory increase of the PS - neither quantitatively nor qualitatively. Only insignificant increase of PS amount was occurred after 12h TSD. As regards to post PSD period, the selective PS rebound was observed. Animals’ behavioral parameters were not changed considerably. Conclusion. The idea that PS rebound compensates for specific loss of both SWS and PS was not supported in present study. The longer was the TSD, the longer was the time of the “smuggled” sleep. Consequently, the long lasting TSD enables us to speak about specific compensatory process caused by sleep total loss. Therefore, it could be suggested that homeostatic compensation of lost sleep occurs if sleep restriction is performed during short time by least stressful method when the stress effects and the volume of “smuggled” sleep do not reach significant values and precise evaluation of TSD effects could be achieved. 292 Actas de Fisiología 7, 2001

THE CYCLIC ALTERNATING PATTERN (CAP) DECREASES AS A CONSEQUENCE OF TOTAL SLEEP DEPRIVATION

Luigi De Gennaro, Michele Ferrara, Valentina Spadini, Giuseppe Curcio, Riccardo Cristiani and Mario Bertini Department of Psychology - University of Rome “La Sapienza”, Rome, Italy

Introduction. In the last years, there has been a growing interest in phasic EEG events of NREM sleep. The Cyclic Alternating Pattern (CAP) corresponds to a scoring system of these events within the framework of two functional states, that is phase A, associated to a higher arousal, and phase B, associated to a lower arousal, oscillating reciprocally. Significant CAP changes have been found in sleep disorders, after benzodiazepine assumption, in ageing, and as a consequence of circadian phase delay and of acoustic noise administration during the night sleep (for a review, [1]). Objective of the present study was to assess the sensitivity of the CAP parameters to the increased need of recuperation that follows one night of sleep deprivation. Methods. Nine normal male subjects were recorded for three nights [adaptation, baseline (BSL) and recovery (REC)]. Baseline and recovery nights were separated by 40-h sleep deprivation. CAP parameters were measured in BSL and REC nights following standard rules [2]. Results and conclusions. Recovery after sleep deprivation was characterized by a decrease of stage 1, stage 2 and sleep latency; an increase of SWS, and of the sleep efficiency index, while REM and NREM time were not different. CAP parameters were significantly affected by sleep deprivation, and the main results were: 1) CAP rate (time spent in CAP phases/NREM time) significantly decreased from 38.85% (SEM=2.65) to 31.47% (SEM=2.84); 2) within phase A subtypes there was a significant decrease of A3 subtypes [from 38.11 (SEM=8.51) to 19.57 (SEM=3.90)]. Fur- thermore, the number of arousals scored according to ASDA rules [3] strongly correlated with A3 subtypes during BSL (r=0.79; F=11.30, p=0.01) and during REC nights (r=0.95; F=62.87, p<0.0001). The present results suggest that recuperative processes after sleep deprivation are also associated with a higher sleep continuity as defined by the reduction of CAP rate.

1. Terzano, MG, Parrino, L. Sleep Med Rev 2000; 4: 101-123 2. Terzano MG, Mancia, D, Salati, MR, Costani, G, Decembrino, A, Parrino, L. Sleep, 1985; 8: 137-145 3. ASDA (American Sleep Disorders Association and Sleep Research Society). Sleep 1992; 15:173-184. Actas de Fisiología 7, 2001 293

CHANGES IN THE AUDITORY EVOKED POTENTIALS UPON AWAKENING FOLLOWING AN INCREASE OF SLOW-WAVE SLEEP AMOUNT

Michele Ferrara, Luigi De Gennaro, Massimiliano Barattucci, Giuseppe Curcio, Riccardo Cristiani and Mario Bertini Dipartimento di Psicologia, Università “La Sapienza”, Roma, Italy

Sleep inertia is a period of transitory hypovigilance, confusion, disorientation of behavior and impaired cognitive and behavioral performance that immediately follows awakening. From a physiological point of view, studies on cerebral blood flow and on EEG power spectra during spontaneous sleep-wake transitions showed that the awakening process is characterized by a slow shift from the sleep EEG substrate to that of wakefulness (for a review see ref. 1). The aim of the present study is to assess Auditory Evoked Potentials (AEPs), as a psychophysiological index of brain functioning, before sleep and upon three awakenings during an undisturbed baseline night and to compare them to AEPs during a night characterized by a recuperative increase of SWS amount as a consequence of two consecutive nights of selective SWS deprivation. Ten male subjects slept in the laboratory for 6 consecutive nights. The first 2 nights were undisturbed. The 3rd night was considered as baseline (BLA). During the 4th and 5th nights, selective SWS deprivation was obtained by means of acoustic stimulation. The 6th night was a recovery (REC). The data reported here were collected during the BLA and the REC night. Ss were awakened three times, after 2 h, 5 h and 7.5 h (final awakening) of sleep, respectively. All the awakenings were carried out from stage 2. The AEP recordings were carried out in bed, while subjects were performing a simple auditory reaction time task. The amplitude of the N1-P2 complex decreased during the sleep-wake transition as compared to presleep wakefulness levels. Similarly, N1 latency increased upon the two nocturnal awakenings, while P2 latency was not affected. Moreover, the N1-P2 amplitude increased during REC at the frontal midline derivation as compared to BLA, while it significantly decreased at Pz and Oz. It is concluded that the N1-P2 amplitude and, to a lesser extent, the N1 latency, are sensitive in showing a state of brain deactivation during the sleep-wake transition. The decrease of N1-P2 amplitude at the parieto-occipital locations during REC is coherent with the hypothesis of a functional link between SWS amount and cortical hypoarousal upon awakening. The unexpected increase of the same variable at Fz is consistent with the results of recent fMRI (2) and LORETA (3) studies, and can be interpreted as the effect of a compensatory effort.

1. Ferrara M, De Gennaro L. (2000) Aviat Space Environ Med 71: 843-848. 2. Drummond S., et al. (2000) Nature 403: 655-657. 3. Szelenberger W, Piotrowski T. (2000) J Sleep Res 9 (Suppl. 1):186. 294 Actas de Fisiología 7, 2001

EEG POWER DURING SLOW-WAVE SLEEP DEPRIVATION AND THE FOLLOWING RECOVERY SLEEP

Michele Ferrara, Luigi De Gennaro, Chiara Corvasce, Giuseppe Curcio, Riccardo Cristiani and Mario Bertini Dipartimento di Psicologia, Università “La Sapienza”, Roma, Italy

We previously showed that selective slow-wave sleep (SWS) deprivation almost completely suppress visually scored delta sleep, without any concomitant sleep fragmentation or sleep curtailment, and is followed by a compensatory SWS rebound during the recovery night (1). The aim of the present study is to assess the homeostatic changes of EEG power in response to the selective SWS deprivation. Ten normal males slept for 6 consecutive nights in the lab. After an adaptation and two baseline nights (BSL; BSLA), selective SWS deprivation was accomplished for two consecutive nights (DEP) by means of an acoustic stimulation technique (1). A recovery (REC) night then followed. The EEG was digitalized with a sampling rate of 128 Hz. After an off-line artifact rejection, the signals from 5 derivations along the antero-posterior axis (Fpz, Fz, Cz, Pz, Oz) were analyzed by a Fast Fourier Transform algorhythm using a 4 sec resolution. Spectra from consecutive 4-sec epochs were then averaged in 1-min periods. Power values were calculated, only for NREM sleep, with a 0.25 Hz resolution; adjacent bins were then averaged to allow a 1 Hz bin resolution. Absolute power values were finally log-transformed. For each derivation, single-Hz power values were submitted to repeated measures ANOVAs comparing DEP and REC to BSLA. During the DEP nights, a significant decrease of the 1-11 Hz and of the 16 Hz bins at Fpz was found, as compared to BSLA; this decrease was restricted to the 1-2 Hz bins at Fz and to the 1-3 Hz bins at Oz. At Cz, the decrease comprised the 1-5 and 8-11 Hz bins, while at Pz it included the 1-5 and 10-11 Hz bins. As regards the REC night, the compensatory increase of EEG powers was limited to the 1-2 Hz bins at Fpz, encom- passed the 1-9 and 23-25 Hz bins at Fz, included the 1-7 Hz range at Cz and the 1-2 and 6-7 Hz bins at Pz. No significant increase of EEG powers was found at Oz. In conclusion, our results show that the selective SWS deprivation was effective in reducing EEG power in the low-frequency range. During the recovery sleep, the increase of EEG power showed an anterio-posterior gradient, being prominent over the fronto-central derivations and absent at Oz. It is confirmed that the time course of EEG low frequencies is not only determined by the length of previous wakefulness, not affected in the present study, but also by the amount and quality of SWS during previous sleep episodes. Interestingly, frontal areas show the stronger need for slow wave activity as indicated both by the minimum decrease of these frequencies at Fz during the deprivation nights and the maximum increase during the recovery.

1. Ferrara M, De Gennaro L, Bertini M. Sleep Res Online 2: 15-19, 1999. Actas de Fisiología 7, 2001 295

SLEEP DEPRIVATION IN DIFFERENT RODENTS

Irma Gvilia, Nato Darchia and Tengiz Oniani Dept. of Neurobiology of Sleep-Wakefulness Cycle, I. Beritashvili Institute of Physiology, Tbilisi, Georgia

Effects of total sleep deprivation (TSD) on the structure of sleep-wakefulness cycle (SWC) were studied in good and poor sleeper rodents. Short lasting TSDs (8, 10 hours) were performed in chronic adult male rats (n=8) and guinea pigs (n=8). Prior to the experiments the animals were placed in the experimental cage (1m3), where optimal conditions for the occurrence of a regular SWC were maintained. Food and water was available ad lib. Implantation of stimulating and recording electrodes (EEG, EMG, EOG) was performed under deep nembutal anesthesia. Baselines of 24h SWC were recorded for 5-6 days. For the purpose of sleep restriction the rats were awakened by the stimulation of hypothalamus, while guinea pigs - by gentle handling. Throughout the procedure and after it’s termination the electrophysiological variables and behavioral reactions were recorded. At standard time of the SWC the frequency of the onset, total duration and ratio of different phases in the SWC were determined. In addition, the amplitude-frequency analysis of the delta and theta waves were made. The changes occuring in the structure of the SWC after the TSD were processed statistically. The TSD in the rats and guinea pigs resulted in significant changes of the animals’ SWC structure. In a course of the deprivation procedure there occurred sharp increase in the frequency of sleep onset. Besides, there started to manifest itself the dissociation of the wakefulness EEG and behavioral indices. In parallel to the accumulation of need for sleep the pressure of sleep increased step by step, there appeared the „smuggled“ sleep and eventually, after 10-12 h of TSD, the maintenance of uninterrupted wakefulness became nearly impossible. In the post-deprivation period the development of compensatory processes was evident in both species. In comparison to control conditions, the volume of slow-wave sleep (SWS) increased significantly, especially deep SWS. The EEG delta power increased as well. In parallel to the slowing of the EEG, behavioral sleep deepened also. The tone of the neck muscles decreased and paradoxical sleep (PS) occurred on the background of an already developed atonia. In the post-deprivation SWC of the rats, there was no significant compensatory increase of PS, neither quantitatively nor qualitatively, while in guinea pigs total PS duration was increased. At the same time, the intensity of the last was no considerable changed and the SWS/PS ratio in the total sleep was not changed against the control. Analysis of the results has shown that the TSD in the rats and guinea pigs induced similar effects: it caused in the body of the animals formation and gradual accumulation of internal need for sleep. The acute deficit of SWS was compensated in the post deprivation period by its sharp rebound. 296 Actas de Fisiología 7, 2001

EFFECTS OF REM-SLEEP DEPRIVATION ON THE THYROSIN HYDROXYLASE (TH) ACTIVITY OF THE RAT DORSAL RAPHE NUCLEUS: AN ENZYMHISTOCHEMICAL STUDY

Seyed Behnamedin Jameie, Gila Behzadi and Hossein Noyan Ashraf Neuroscience Division, Department of Anatomy, Faculty of Medicine, Iran University of Medical Sciences, Tehran,Iran

Although more than half a century has passed from the discovery of the REM sleep, the exact neuroanatomical and neurophysioloigical bases of that is still unknown. Dorsal Raphe Nucleus (DRN) as the main serotonergic nuclei in mammalian brain is a common site of pathology in neurological disorders and is also considered to play a key role in various physiological and behavioral functions, such as learning, memory, sleep- wake cycle and movement. Although, the DRN contains a high percentage of 5-HT containing cell bodies, neuronal perikarya containing other neurotransmitters are present in this nucleus which thought to be responsible in sleep-wake cycle. Sleep -deprivation (SD) is one of the most useful methods used in sleep laboratories in order to study a multitude of aspects of this universal phenomenon. Adult Male Sprague- Dawley rats (200-230g) were used in this study. By using Disk Over Water technique 3 days of REM-SD were done. After REM-SD the animals were deeply anesthetized by lethal dose of sodium pentobarbital. Following transcardially perfusion - fixation by aldehyde solution, brain stems were removed, the 50 µ coronal sections were cut by using a vibrotome. The sections were incubated in 1) rabbit TH anti serum 1:500 for 16 h at room tempreture 2) a biotinylated goat anti- rabbit IgG 1/50 for 2h. Finally the sections were immersed in ABC kit, then processed in DAB and H2O2.Light microscopic examination were done and camera lucida drawing were obtained from selecte sections. TH positive neurons and fibers, definitely dopaminergic, were seen in all parts of the DRN of control and trial groups. The results of the present research showed more TH- positive neurons and fibers in DRN of REM-SD groups than controls. These findings support the involvment of aminergic mechanisms in REM sleep regulation and suggests that pontomedullary region is the site for integration of REM sleep. Actas de Fisiología 7, 2001 297

COMPARISON BETWEEN TWO METHODS OF PARADOXICAL SLEEP DEPRIVATION IN RATS BY ELECTROPHYSIOLOGICAL SLEEP SCORING

Ricardo B. Machado, Débora C. Hipólide and Sergio Tufik São Paulo, Brazil

Introduction: Using the “flower-pot” or single platform (1) (SP) and the modified multiple platform(2) (MMP) techniques of paradoxical sleep deprivation (PSD) in rats, we performed a continuous monitoring of the sleep patterns during the PSD and during recovery period. Methodology: Three months age-matched male Wistar rats were chronically implanted with electrodes for EEG and EMG recordings (3). After three weeks from surgery and three days of adaptation to recording system environment, sleep was recorded continuously during the basal day, 4 days of PSD and 4 days of recovery. For SP technique, one group of twelve animals were placed onto narrow platform NP and the other group (n = 12) in the wide platform WP, as a control method. For MMP technique, five socially-stable rats (4) were placed in an tank with water, but only one animal was recorded. One group of animals was placed onto narrow platform NP. As control methods, a second group in wide platform WP and a third group in the a grid over water (4) G. The total number of recorded animals for each group was ten. Elec- trophysiological signals were acquired with a digital polygraph and the sleep stages were scored manually in wake W, slow wave sleep SWS and paradoxical sleep PS. Data was obtained in minutes and converted to percentage of recording time (24h). Two-way ANOVA for repeated measures, at 5% significance level, and post hoc Tukey HSD tests, was performed. Results: The SP and MMP techniques eliminated all PS during deprivation period, but both methods also reduced the SWS. The proposed controls of both techniques decreased the total sleep time, mainly PS time. At recovery period, we observed PS rebound in the first 24h for all groups in both techniques, except in the Grid control group. When we compared both techniques, no differences were found during the deprivation period. However, in recovery period, the animals in the MMP showed intense rebound of PS as compared with SP technique. Conclusions: We conclude that both techniques are efficient to produce paradoxical sleep deprivation, although no specific. The grid control method utilized appears to be appropriate, in respect to sleep parameters observed at the present study.

References: (1) Cohen, H.B. & Dement, W.C., Science, 150,1318 (1965). (2) Nunes G.P. Jr., Tufik S. & Nóbrega J.N., Brain Res. Bul. 34, 453 (1994). (3) Bergman, B.M., et al., Sleep 12,5(1989). (4) Suchecki, D. & Tufik, S., Physiol. Behav.68, 309 (2000).

Support: This work was supported by AFIP. Machado, RB is fellowship from CAPES, of Brazil government. Hipólide, DC and Tufik, S are fellowship from CNPq, of Brazil government. 298 Actas de Fisiología 7, 2001

THE PHENOMENON OF PARADOXICAL SLEEP SELF-DEPRIVATION

Lia Maisuradze, Tengiz Oniani, Nani Lortkipanidze, Manana Mgaloblishvili-Nemsadze, Marina Eliozishvili and Lela Oniani Department of Neurobiology of Sleep-Wakefulness Cycle, I. Beritashvili Institute of Physiology, Georgian Academy of Sciences, Tbilisi, Georgia

The phenomenon of paradoxical sleep (PS) self-deprivation was detected and described in the cats. Self-deprivation is shown to be acquired just as a classical conditioned reflex during enforced paradoxical sleep deprivation (PSD) both by the water tank technique and by the “classical” PSD procedure through the animal’s awakening in response to sensory stimulation or direct electric stimulation of the activating brain structures. In this situation the transition of the brain from one physiological state into another is a conditioned signal, and sensory stimulation or brain stimulation, resulting in the arousing reaction, serves as an unconditioned stimulus. But in the case of water tank PSD the loss of muscle tonus – due to inhibition of spinal reflexes – as one of the earlier PS component becomes the decisive factor, whereas during “classical” PSD the development of the hippocampal theta rhythm and rapid eye movements – even in the lack of muscle tonus – acquire the signal significance. This indicates that transition of slow wave sleep (SWS) into PS is experienced subjectively and hence it can become a conditioned signal for arousing. For comprehensive analysis of the dissociative aspects of PSD the two described findings are of special importance: (1) during selective PSD by the “classical” method the accumulated unsatisfied specific need for PS puts pressure upon other sleep phases, and some PS components (ponto-geniculo-occipital spikes, hippocampal theta rhythm, rapid eye movements, etc.) can occur before the onset of desynchronization in the electroneocorticogram – during SWS; (2) this process can sometimes develop into genuine PS or in some cases it ends in PS self-deprivation, and a misleading impression is then created of an isolated occurrence of some PS components within SWS. The same is true with respect to the isolated occurrence of PS components during wakefulness as a result of selective PSD. The pressure of excessive accumulation for PS need can be also revealed between deprivation trials within episodes of wakefulness before the restoration of SWS. In that case PS can be triggered during wakefulness without restoration of SWS. PS can form completely or be interrupted by self-deprivation, and in the latter case there is again a false impression of an isolated occurrence of certain PS components during wakefulness. Thus careful analysis and assessment of the phenomenon of PS self-deprivation are of great importance for the elucidation of the neurophysiological mechanisms and functional significance of PS, in general, and for the correct understanding of the dissociative processes of selective PSD, in particular. Actas de Fisiología 7, 2001 299

SLEEP DEPRIVATION INCREASES ENDOTHELIN-1 PLASMA LEVELS

Beatriz D. Palma, Alexandre Gabriel Jr., Magda Bignotto and Sergio Tufik São Paulo, Brazil

Endothelin (ET-1) is a 21 amino acid peptide that posses vasoactive properties, synthesized and released by a variety of cells, but mainly by the vascular endothelium. Both physical and chemical stimuli, such as hypoxia, lead to ET-1 release. ET-1 is one of the most potent endogenous vasoconstrictors identified and augmented concentration in plasma appears to be closely related with pathogenesis of arterial hypertension as well as with obstructive sleep apnea (OSA). OSA patients exhibit repetitive episodes of apnea and hypopnea that result in hypoxia and consecutive arousals. These patients are chronically sleep deprived, which may aggravate the hypertensive features, since literature data show that sleep deprivation (SD) results in hypertension both in humans and in animals. Based on the reported relationship between ET-1, hypertension and SD, the purpose of the present work was to examine ET-1 plasma levels in paradoxical sleep (PS) deprived animals. Twenty Wistar male rats were distributed in 2 groups (n = 10/group) and were PS deprived by the modified multiple platform method. One group was deprived for 24 h and the other, for 96 h. Control animals (n = 10) were maintained in their home-cage, in the same room where the animals were PS deprived. Immediately after the end of PS deprivation, the animals were ether-anesthetized and submitted to blood sampling by cardiac puncture. Blood was collected in vials containing sodium citrate and centrifuged at 2300 rpm for 10 min. Plasma was maintained frozen until ET-1 assay by a commercial radio-immune-assay (RIA) kit (Pharmacia, Amersham). Analysis of ET-1 plasma levels (one-way ANOVA, followed by the Duncan Multi- ple Range Test) revealed that 96 h of PS deprivation induced a significant increase (p = 0.05) on ET-1 release. Our results showed that SD altered ET-1 plasma levels, suggesting that such an increase may participate in the genesis of arterial hypertension observed after SD. Further studies are being performed in an attempt to elucidate the underlying mechanisms involved in SD-induced increase of ET-1 secretion.

Financial support: Associação Fundo de Incentivo à Psicofarmacologia (AFIP) 300 Actas de Fisiología 7, 2001

HYPOCRETIN LEVELS IN MICE BRAIN AFTER PARADOXICAL SLEEP DEPRIVATION

Mario Pedrazzoli, Ling Lin, Vania D’Almeida, Debora Hipolide and Emmanuel Mignot Sao Paulo, Brazil

Recently was reported independently for two groups of researchers in the USA the discovery of a neuropeptide named hypocretin (HPCT) or orexin localized only in a small portion of the hypothalamus (de Lecea et al., 1998; Sakurai et al., 1998). In 1999 was reported that the canine narcolepsy was caused by a mutation in the hypocretin receptor 2 gene (HPCTr2) (Lin et al., 1999) and Chemelli et al. (1999) reported that knockout mice for the pre-prohypocretin and also for the HCPTr2 developed narcolepsy. Now is well established that HPCT has a major role in the sleep regulation and that a mutation within one of the genes of the system or the absence of the peptide in the brain can cause narcolepsy, a sleep disorder related with REM abnormalities. The objective of this study was examine the effects of longer sleep deprivation (96 hours) on the levels of hypocretin A and B in the mice brain. Mice were deprived of sleep using the classical platform method. After 96 hr of sleep deprivation one group was sacrificed, a second group was returned to the home cage and allowed to sleep for 24 hr before being sacrificed (the sleep rebound group). The control group remained in the home cages. All brains were rapidly removed, frozen over dry ice and stored at -80 °C. For the peptide extraction a 2ml 0.1M / 5% ETOH solution was added to the brains, boiled for ten minutes and then kept on ice. The brains were homogenized and centrifuged at 5500 RPM for 25 minutes. The supernatant was transferred to a plate of 96 wells and stored at -80°C. We use the EIA method to assess the hypocretin concentration in the mice brain. The results can be seen in the table:

hypocretin A (ng/g) hypocretin B (ng/g) ctr 8.2±2.1 10.4±1.6 sd 16.6± 2.1 9.61±2.0 reb 19.7±8.9 11.3±2.8

The data show that 96h of paradoxical sleep deprivation did not alter the hypocretins levels in mice brain. Also the hypocretins levels were not altered after twenty-four hours of sleep recovery. The hypocretins exhibits circadian fluctuations and food deprivation do not up regulate CSF hypocretin 1 when the hypocretin level are at the peak (CT 0). Maybe in our experiment we are looking for hypocretins at high levels of expression (we have sacrificed the animals at CT 2) and because that we can not see any difference. Longer sleep deprivation periods and sacrifice the animals in different times of the day are necessary to elucidate this question. Actas de Fisiología 7, 2001 301

EFFECT OF TOTAL SLEEP DEPRIVATION ON EMOTIONAL MEMORY

Portas CM*§, Sunderji I*, Brekke C§, Maguire E*, Strange B*, Frith CD* *Department of Cognitive Neurology, University College London, U.K. § Department of Physiology, University of Bergen, Norway.

Emotionally salient items have privileged access to memory storage (1). Thus, we used an oddball paradigm to investigate the effect of total sleep deprivation (SD) on free recall of emotionally salient and neutral control words. A previous study in normal volunteers showed that emotional words embedded in a semantically homogeneous list of items are remembered better than neutral words (1). To highlight the specificity of this effect we included a ‘perceptual’ oddball in the list of words (words written with different fonts). We expected the perceptual oddball recall not to be affected (or to be affected to a lesser extent than the emotional one) by SD. Finally, we used a battery of cognitive tasks to assess short-term, working, and episodic memory. 21 subjects (9 sleep deprived and 12 controls) aged 19-24 were used in this study. All subjects were tested in two occasions (approx. 6 pm and 6 am of the following day) using: oddball paradigm (19 lists of 14 items each), digit span (forward and reverse), word-stem completion (1 list 50 words), Warrington Recognition Memory Tests for pictures and words (50 items in each), Corsi test (for spatial memory), short story recall, abstract picture recall. Between the first and second session of tests controls went home to sleep while the other subjects were kept awake throughout the night. Results were analysed with paired t-tests (one-tailed, following the assumption that SD can only impair performance). Only results from the Corsi task were analysed using a 2-way ANOVA due to the multiple conditions tested. Controls did not show any significant change in any task between the two sessions. As predicted, the emotional word was recalled better than a neutral control word (p<0.05) both in controls (1st and 2nd session) and in the experimental subjects (1st session only). However, after SD recall of emotionally salient items was significantly decreased (p<0.05) falling to the recall level of the control word (p=0.142 emotional versus control after SD). SD had no effect on recall of the perceptual oddball (p=0.304). The mild protocol of total SD used in this study did not affect performance on a wide range of memory tests. However, it was able to highlight the sensitivity of emotional words recall to SD. The effect of SD on world recognition and abstract pictures recall should be further investigated.

1) Strange et al., Neuroimage, 12(4):425, 2000. 302 Actas de Fisiología 7, 2001

LONG TERM TOTAL SLEEP DEPRIVATION RESULTS IN OXIDATIVE STRESS

Lalini Ramanathan, Seema Gulyani, Robert Nienhuis and Jerome Siegel UCLA, North Hills, California, United States

Prolonged sleep deprivation in animals results in increased eating, altered thermoregulation and eventually death (1). It has been proposed that reactive oxygen species (ROS), which accumulate during waking may be responsible for some of these effects. A variety of antioxidative enzymes such as superoxide dismutase (SOD) and glutathione peroxidase (GPx) help to regulate the level of ROS. Reduced glutathione (GSH) is also a potent scavenger of ROS. Glutathione reductase (GR), though not an antioxidative enzyme, is involved in the GPx/GSH pathway. In this study we investigated the changes in the activities of SOD, GPx and GR as well as total glutathione levels in the hippocampus of sleep deprived rats (DRAT), yoked controls (CRAT) and home cage controls (HC). Male Sprague Dawley rats (350-450 g) were subjected to total sleep deprivation by the disk-over-water method. The rats were placed in constant light and given food and water ad libitum. After surgery, the rats were allowed to recover for at least 5 days before being placed on the disk-over-water apparatus for 4 days of adaptation. The experiment was initiated with 3 days of baseline recording where the animals were subjected to a timed stimulus (disk rotation once every hour for 6s). This was followed by 5-13 days of total sleep deprivation. Six sets of animals were sacrificed in groups of three (HC, CRAT, DRAT). The animals were killed by decapitation and brain regions were dissected and stored at 800C prior to performing biochemical assays. The hippocampus was homogenized and the mitochondrial fraction was used for the analysis of Mn-SOD activity while the cytosolic fraction was used for measuring the activities of Cu/Zn-SOD, GPx and GR as well as total glutathione levels. The paired student’s t-test was used to determine statistical significance. There was a small (10%) but significant reduction in the activity of Cu/Zn-SOD in DRATs compared to CRATs (t=5.9, df=6, p=0.002). Similarly, there was a small (13%) but significant increase in the activity of GPx in DRATs compared to CRATs (t=2.6, df=5, p=0.05) Also, the activity of GPx in DRATs was significantly higher than in HC (t=3.4, df=5, p=0.02) The activity of Mn-SOD was not significantly different between DRATs and CRATs. However, both DRATs and CRATs had significantly lower Mn-SOD activity compared to HC animals (73% of HC; t=2.6, df=5, p=0.05 and 65% of HC; t=6.9, df=5, p=0.002). No significant change in the activity of GR or in the levels of total glutathione was observed between DRATs and CRATs, This study shows that prolonged sleep deprivation results in decreased Cu/Zn-SOD activity and increased GPx activity. We have previously reported decreased Cu/Zn-SOD activity in the cortex of sleep deprived animals (2). Based on our findings, we propose that long term sleep deprivation may lead to oxidative stress, as reflected by changes in the activities of antioxidative enzymes.

References: 1. Rechtschaffen A, Gilliland MA, Bergmann BM, Winter JB. Physiological correlates of prolonged sleep deprivation in rats. Science 1983;221:182-184. 2. Ramanathan L, Gulyani S, Nienhuis R, Siegel JM. Long term total sleep deprivation results in altered Cu/Zn- SOD activity. Sleep 2001;24:A250. Actas de Fisiología 7, 2001 303

PSYCHOPHYSIOLOGICAL MECHANISMS OF THE ANTIDEPRESSANT RESPONSE TO SLEEP DEPRIVATION (SD): A THEORETICAL INTEGRATION

Vadim S. Rotenberg Abarbanel Mental Health Center, 15 Keren Kayemet str., Bat-Yam, Israel

Abstract: In an attempt to answer the question whether the benefitial effect of SD in depression is related to the increased wakefulness or to the SD by itsellf, search activity concept is used. REM sleep is functionally deficient in depression, thus it contributes to the state of helplessness instead of restoring mood and search activity. REM sleep deprivation, either selective or not, is beneficial by breaking a vicious circle: depression in wakefulness-giving up in dream-depression in wakefulness. In addition, the ability to confront a challenge of SD and maintain wakefulness, has a positive outcome on depression, especially when wakefulness is accompanied by active behavior. 304 Actas de Fisiología 7, 2001

ASSESSING SLEEP SPINDLES WITH GABOR TRANSFORM

Günther J.L. Gerhardt*, Emerson L. de Santa-Helena**, Suzana V. Schönwald**, Márcia L.F. Chaves** *Instituto de Física da Universidade Federal do Rio Grande do Sul **Laboratório do Sono, Serviço de Neurologia do Hospital de Clínicas de Porto Alegre, Brasil

Since the Fast Fourier Transform lacks space-time resolution to identify single elements from the human surface EEG which are essential for visual inspection of sleep microstructure, other techniques, such as the the Wavelet Transform, have been used recently for this purpose. We tested the adequacy of a Gabor Transform-based program developed for the identification of second-stage sleep spindles. A correct algorithm should yield a plot of spindle ocurrences very similar to the sigma spectral power variation along the night, because the computation of the Gabor Transform is also based upon the spectral content of the time series. The C3-A2 channel of a sleep study taken from the normal volunteers database of the HCPA sleep lab (pertaining to a 23-old male) was used. The EEG machine was an 18-channel analogic Nihon-Kohden (Neurofax) with posterior digital conversion by Stellate software Rhythm10.0, with 12-bit resolution, and 128Hz acquisition frequency. Low-pass filter was 60Hz, high-pass filter was 0.5Hz and notch filter was 60Hz. A digital off-line FIR was used on the signal. The Gabor Transform was calculated for 30-sec windows restricted to the sigma band (11-15Hz). A Windowed-Fourier Transform was also computed using the same parameters. The plot obtained from the Gabor computation showed very good correlation (r=0.807816) with the plot obtained from the WFT analysis, for a whole night sleep study of a human normal subject. The Wavelet technique, meant basically to assess data microstructure, could be used for identification of sleep macrostructure, at least with respect to second-stage sleep spindles. A limitation of this method, to be studied, is the fact that not all sigma activity present in the sleep EEG corresponds to sleep spindles. Actas de Fisiología 7, 2001 305

DOES EARLY IRON DEFICENCY ANEMIA INTERFERE WITH CHILDHOOD REM SLEEP ORGANIZATION PATTERNS?

Cecilia Algarín1, Patricio Peirano1, Marcelo Garrido1 and Betsy Lozoff2 [1]Sleep Laboratory, INTA, University of Chile, Santiago, Chile [2]Center for Human Growth and Development, U of Michigan, USA

Introduction: The establishment of the sleep/wake cycle and the internal architecture of sleep are fundamental steps in the neurofunctional maturation of the brain. Lasting abnormalities in sleep/wake and activity rhythms have been reported in IDA rodents, but are yet to be explained. It is known that iron is required for the functioning of several neurotransmission systems, myelination, and neuronal metabolic activity; thus, different processes may be related to their alteration. Since one of the main sleep organization changes during early infancy is the reduction and temporal distribution of REM sleep, we evaluated whether early IDA would provoke long-lasting effects on nocturnal REM sleep organization. Methods: All-night polysomnographic recordings were done in a group of healthy 3- to 4-year-old children who were treated for IDA (n=24) or were nonanemic (controls, n=26) in infancy. NREM stages III and IV were grouped as SWS. The duration of each SWS and REM sleep episode was analyzed according to each successive third of the night. Results: Between groups, REM and SWS episodes were differently distributed throughout the night. The pattern of REM sleep episodes was different between groups: in controls, the duration of REM episodes increased with advancing thirds, while in former IDA children the duration was similar in all thirds. Furthermore, former IDA children demonstrated a short latency (p<0.02) and a consistent tendency for a long duration of the first REM episode. Discussion: Altered nocturnal temporal organisation of sleep patterns in former healthy IDA children suggests that iron is essential for the normal progression of sleep patterns. Several characteristics of REM sleep organisation in former IDA children indicate that neurodevelopmental processes do not follow age-related expected modifications. It is conceivable that the disruption of REM sleep temporal patterns represents an underlying mechanism that interferes with optimal behavior during both sleep and wakefulness. In fact, REM sleep patterns are reminiscent of those often observed in young depressed subjects and could be related to the depressive symptomatology observed in former IDA subjects during adolescence.

Support: Grants from NICHD (HD33487) and Fondecyt (CONICYT, Chile 1000657). 306 Actas de Fisiología 7, 2001

RESTLESS LEGS SYNDROME: CLINICAL EXPERIENCE WITH LONG-TERM TREATMENT

Sandra C Clavadetscher1, Matthias Gugger1, Claudio L Bassetti1,2 Sleep Medicine Center, Neurology and Pneumology, University Hospital, CH-3010 Bern, Switzerland1; Department of Neurology, University Hospital, CH-8091 Zurich, Switzerland1,2

Introduction There are almost no data on both long-term treatment efficacy (³ 6 months) and predictors of good treatment response in patients with restless legs syndrome (RLS) outside study protocol conditions. The value of a newly described RLS score (RLSS; [2]) for assessing severity and changes of severity of RLS symptoms is unknown. Methods Over three years 70 RLS patients (pts; 36 men, 34 women; mean age of 59 years, range 29-79) were prospectively studied and seen at least twice in our sleep clinic. Diagnosis of RLS was made according to international criteria [1]. Clinical and polysomnographic data as well as severity of RLS symptoms, as estimated by RLSS, were assessed at study begin. Periodic limb movements in sleep (PLMS) were scored according to conventional criteria. Patients were treated on an individual base. After a follow-up time of 1-106 months (mean=16; 51 pts ³ 6 months, 64 pts ³ 3 months) RLS symptoms were re-assessed by both overall clinical impression (much better, better, unchanged, or worse as compared to study begin) and RLSS. Clinical characteristics and treatment response were compared between naive pts (N, never treated for RLS before study begin) and pts with previous treatment for RLS (T). Predictors of treatment response were searched for comparing pts with good treatment response (G, much better or better on follow-up) and those with bad treatment response (B). Results At follow-up 76% (30/40) N-pts and 77% (23/30) T-pts had a good treatment response. No significant differences were found between the two groups in age, gender, etiology and duration of RLS, familiarity, presenting sleep complaint, RLSS, and percentage of pts with PLMS. PLMS were more common in B- than G-pts (100% vs 58% of pts, p=0.008). In all other variables considered G-pts and B-pts did not differ significantly. In all 70 RLS pts the mean RLSS was 26 (range 12-38) at baseline 19 (range 1-36) at follow-up. There was a significant correlation between improvement of overall clinical impression (better or much better on final follow-up) and reduction of RLSS (p<0.01). Two patients had previously unknown „sleep attacks” while taking pergolide (2mg/die) and pramipexole (0.5mg/die) respectively. These episodes disappeared when medication was tapered off (pergolide) or reduced (pramipexole 0.25mg/ die). Conclusion 1) A good long-term treatment response can be obtained and maintained in a clinical setting in about 80% of RLS. 2) Patients without PLMS have a better long-term treatment response. 3) The RLSS represents a useful tool for assessing changes in severity of RLS symptoms in individual patients over time.

References [1]Mov Disord 1995;10:634-642. [2]Neurology 2001;56(Supl3):A4 Actas de Fisiología 7, 2001 307

WORK ACCIDENTS RELATED TO SLEEPINESS

Marta Claudio*, Margarita Blanco° y Maricel Estragó° *Medicina Laboral - ° Centro Neurológico Hospital Francés - Buenos Aires, Argentina

There are many papers about the sleepiness related traffic accidents. We have study the sleepiness as reason of accidents in the work area, including the itinerary from home, hours of work and hours of sleep and rest. We considere as work accidents simple sharp injuries, with or without suture, traumatisms with contusion, tumbles, fractures, sprain milds and severes, craneal trauma (with o withouth unconciousness). We study 649 patients: Morning Shift: 342; Afternoon Shift: 162; Night Shift: 135. The occurrence of accidents was analyzed having in mind: a) hypoglucemia by prolonged fasting, insuficient nutrition, sleepines and poor sleeping. b) poor rest, daily routine, excesive and obsesive working, multiples activities, fast rythm, personal, familial and labour problems. Variability in accidents horary was observed in thre three shifts and related with different reasons from our point of view. We concluded that this is a great problem to solve, because the consecuences affect as the workers as the owners and more information is necessary about the cost of this accidents.

This research was supported by Fundación Alfredo Thomson para el Desarrollo de las Neuro- ciencias. 308 Actas de Fisiología 7, 2001

INCIDENCE OF EATING HABITS AND SLEEP SCHEDULES IN PATIENTS WITH INSOMNIA AND SNORING

Margarita Blanco, Javier Domínguez y Maricel Estragó Centro Neurológico - Hospital Francés La Rioja 951 (1221) -Buenos Aires - Argentina

The majority of the patients seen in our Sleep Laboratory are suffering from insomnia and snoring. Excessive daytime sleepiness (EDS), cognitive disturbances and mood changes are the most common symptoms. In order to evaluate the eating habits and sleep schedules for those patients they filled out a self-reported questionnaire previously the performance of another diagnosis studies. Based on the self-reported questionnaire results, we asked the patients to reduce the amount of night meals and to extend the time before going to bed. To establish if the eating habits (meals time and amount) and the time between the last meal and the sleep onset it has incidence on the patients reported sleep quality. Simple questionnaire with a graphic was made in order to asses: a) bed time and awaking time, b) meals time, c) amount of every meal (B-L-D) Over a total of 1600 patients we selected 923 (349 women and 574 men [ 563 patients with snoring – with or without apneas- and 360 patients with maintenance insomnia]). Age ranges: 22 to 86 years. All of them live in Buenos Aires City and its area round. The amount of food was classified as: No meal, Light meal, Normal meal and Heavy meal. The time between the last meal and the sleep onset was recorded in hours and minutes. The selected patients completed the questionnaire under our supervision. We found that 82 % of them take Normal and Heavy dinner, 68 % take Light or No breakfast. All of them lasted less than 3 hours to get the sleep onset. 38 % of the patients lasted 90 min and 48 % 120 min. With the prescription of reducing the night meal and extending the time previous to going bed, 57% of the patients showed less EDS, better quality of sleep and improvement of the cognitive disorders and mood changes. We think that those results confirm the importance of the “digestion rol” during the sleep on the origin of these sleep disorders : insomnia and snoring. This is part of a larger research involving the motives that causes these alterations.

This research was supported by Fundación Aldredo Thomson para el Desarrollo de las Neuro- ciencias Actas de Fisiología 7, 2001 309

EFFECT OF GASTROINTESTINAL DISEASES ON INSOMNIA AND SNORING

Margarita Blanco, Javier Domínguez y Maricel Estragó Centro Neurológico -Hospital Francés (Buenos Aires - Argentina)

The influence of gastrointestinal activity during the sleep have had examinated in recent international publications. The International Classifications of Sleep Disorders and several papers confirmed that gastroesophageal reflux disease (GERD) and duode- nal ulcer (UD) may disturb sleep. We studied 100 patients, over a total of 578 (41Females,59Males), age ranges 22- 75 years. They were were recruited between 1996-2000. We found a hight frecuency of excesive daytime sleepiness (EDS), psychosocial problems, attention-deficit and cognitive disturbances, specially in the snoring patients. BMI (Body Mass Index) was an important sign to consider in this group of patients. PSG was realized in 71 snoring patients and 4 insomniacs. 99 subjets underwent gastrointestinal roentgenologic study with double contrast . 11 cases had endoscopic digestive study. The endoscopic evaluation of upper airways was made in 66 snoring patients. The PSG showed Apneas and Hypopneas in 47 snoring patients. The Apnea- Hipopnea Index (AH/I) was normal (behind 10) in 28 cases; mild in 25 patients, in 22 moderated and in 10 of all was severe. The endoscopic study of upper airways examination determinated acid laryngitis in 46 cases. The gastrointestinal roentgenologic evaluation showed hiatal hernia in 87 and GERD in 84 patients. The endoscopic digestive study confirmed the esofagytis and GERD in all of cases. We found a better quality of sleep in the patients who reported GERD symptoms with the prescription of reducing the nigth meals and pro- motions of health food. The most of the patients (85/100) reported an improvement of their main symptoms (e.g. insomnia or snoring) and increased the welfare when the specific therapy was administrated (omeprazol, lanzoprazol). In this study we found a significant association between gastrointestinal disorders and Insomnia and Snoring. Our findings suggest that education and promotion of healthy eating habits and the rigth digestive assess should be advocated in addition of the traditional treatment of this common sleep disorders.

This research was supported by Fundación Alfredo Thomson para el Desarrollo de las Neuro- ciencias 310 Actas de Fisiología 7, 2001

SLEEP DISORDERS IN PATIENS WITH DOWN SYNDROME

M.D. de la Calzada, A. Queralt, S. Giménez, I. Casas, S. Peñarroya, R. Alcoberro1 and J.M. Corretger1 Unidad del Sueño. S. de Neurofisiología Clínica. Hospital Universitario; Vall d´Hebron. 1 Fundación Catalana Síndrome de Down. Barcelona

Introduction: Sleep disorders have important impact on the quality of waking life for children and adolescents. Patients with Down syndrome are a risk group for disturbed breathing during sleep due to their structural characteristics affecting upper airway size. Children with attention deficit disorders show sometime behavior disorders that can be related to sleep pathology. The aim of the study is to know the existence of sleep disorders on patients diagnosed as having a Down syndrome and their repercussion over wake time. Material and Method: Three pages screening questionnaire was administered to parents of Down syndrome patients and for comparison siblings of patients Down syndrome. The questionnaire included 14 items that focused on medical and structural characteristics, 12 items related to sleep behavior and 7 items related to daytime behavior. Both groups results were compared and showed significant statistical differences in open mouth during sleep, noisy breathing, nocturnal snore, legs movements, stand up on bed and observed apneas.

Down S. Group Control Group Yes No Yes No p Observed sleep apneas 29 86 1 46 < .000 Nocturnal snoring 46 74 10 39 < .05 Sleep noisy breathing 53 88 10 30 < .005 Open mouth during sleep 82 38 12 38 < .000 Sleep legs movements 79 40 17 31 < .000

Key words: Down syndrome, sleep disorders, behavior disorders. Actas de Fisiología 7, 2001 311

CHANGES IN SLEEP QUALITY AMONG HIV POSITIVE INDIVIDUALS AFTER SIGNIFICANT CAFFEINE REDUCTION: ANALYSIS OF PITTSBURGH SLEEP QUALITY INDEX SUBSCORES

H. Michael Dreher School of Nursing, University of Pennsylvania, Philadelphia, Pennsylvania , USA

Introduction: Sleep pattern disturbances in persons with HIV have been reported to be as high as 79%, far exceeding the proportion found in healthy populations. High levels of caffeine consumption may have an exacerbating effect on already prevalent HIV- related sleep pattern disturbances. Persons with HIV may have altered hepatic caffeine metabolism, possibly due to high rates of hepatitis co-infection and widespread antiretroviral hepatotoxicity associated with the many drugs taken daily in the management of the disease. Purpose: The purpose of this study was to test whether there were any differences in sleep between a group of persons with HIV (60% AIDS and 40% HIV+ only) who reduced their caffeine intake from baseline by 90% or greater for 30 days (n = 44) versus a group of persons with HIV who continued their usual caffeine consumption (n = 44). Sample: An international sample of 221 HIV+ subjects from the United States, Canada, and Brazil were recruited from print sources and the Internet and initially requested entry in the study. Of these, 88 subjects successfully completed either protocol. Each subject reported significant sleeping difficulties (mean pre-PSQI score = 11.06), taking antiretroviral medication, and consuming caffeine daily (mean mg/caffeine/day = 476). Mean CD4+ T cells were 423 and mean HIV Viral Load was 16,414 copies/ml, although 60% reported “less than 400” or “undetectable viral loads.” Method: An experimental design with convenience sampling and stratified random assignment was used. Subjects were administered pre- and post-test Pittsburgh Sleep Quality Index (PSQI) and MOS-HIV Health Survey instruments, with MOS-HIV summary scores used as a health status covariate. Results: Paired-samples T-tests found significant differences between pre- and post-PSQI subscores on measures of subjective sleep quality (p = .000), sleep latency (p = .000), sleep duration (p = .000), sleep disturbances (p = .000), use of sleeping medication (p = .001), and daytime dysfunction (p = .004) for experimental group subjects, and on measure of subjective sleep quality (p = .002) for control group subjects. Overall, there was a significant difference (F = 14.032, p < .001) between the two groups on ANCOVA analysis for sleep, identifying a 35% improvement in sleep among experimental group subjects. Conclusions: The PSQI subscores which changed the most indicated that experimental subjects had improvements in subjective sleep quality, decreased sleep latency, and decreased occurrence of sleep disturbances. Significant reductions of caffeine may improve (though certainly not eliminate) sleep quality in persons infected with HIV. 312 Actas de Fisiología 7, 2001

STRESS INFLUENCE ON SLEEP AND ABSENCE-EPILEPSY IN GAERS RATS

Hélène Faradji, Colette Rousset and Raymond Cespuglio INSERM U480, Claude Bernard University, Lyon, France

GAERS rats (Genetic Absence Epilepsy Rats from Strasbourg) are defined as a genetic absence- epilepsy model, well known to exhibit spontaneous seizures characterised by generalised spike- wave discharges (SWD) (1). With this animal model, we previously showed that the SWD man- ifestation occurs mainly at the expense of paradoxical sleep (PS), while slow wave sleep (SWS) and wakefulness (W) are lightly modified (2). We also demonstrated that nitric oxide contained in brain, exerts an anti absence-epilepsy role and that the substances used in clinic ( valproic acid –VPA-) to treat this pathology might act through their ability to release the endogenous NO (2). Otherwise, it is also known that stress influences the duration and quality of sleep. In this respect, it has been reported that rats submitted to a short intense immobilisation stress exhibit immediately after, a sleep rebound characterised by an increase in SWS and PS duration (3). Finally, in epileptic patients, stress is also commonly believed to precipitate seizures. According to the above data, the present study tried to better define the nature of the correlation existing between PS and SWD in GAERS rats. For this purpose, an immobilisation stress (IS) was imposed to GAERS rats treated or not treated with VPA. Experiments were conducted on one-year old male GAERS and control rats. For polygraphic recordings, the rats were implanted under chloral anaesthesia, with EEG and EMG electrodes. After surgery, each animal was placed in an individual cage and kept in a sound attenuated room maintained at 22-24°C under a 12h/12h dark-light schedule. The 1-hour IS was performed by enclosing each naïve rat in a plastic tube at the beginning of the dark period. One group of rats was submitted to the IS while a second one received before the IS, a VPA injection (i.p. 200mg/ kg). Polygraphic recordings were performed continuously over the 48 hours of the session including control and experimental periods. The results obtained indicate that as long as control rats were restrained, SWS was strongly decreased and PS suppressed while in GAERS rats, the sleep loss was immediately replaced by seizures increase (+119%). This effect was not observed in the animal group treated with VPA (- 38%). After the 1h IS, control rats exhibited a sleep rebound characterised by an increase in PS (+150%) and SWS (+132%), while in GAERS rats a decrease in PS (-75%) replaced by an increase in epileptic paroxysms (+110%) was observed. After pre-treatment with VPA, the IS did not induce changes in SWS of control rats while PS was significantly increased ( 6thh: +63%; 18thh: +34%). In GAERS rats, it is to be noticed that the magnitude of the changes regarding PS (–42%) and epileptic seizures (+16%) is less important after VPA followed by the IS than after the IS alone. The results obtained on GAERS rats, indicate that stress exerts opposite effects on PS (decrease) and absence seizures (increase). The reduced magnitude of the effect on epileptic seizures observed after VPA treatment might be due to the ability of this compound to trigger an endogenous release of NO. Moreover, according to the tight but opposite relationships existing between PS and epileptic seizures, it appears likely that pontin structures generating PS, also containing a NOergic component (n. pedonculopontin and laterodorsal tegmenti), might be concerned in this respect. Finally, NO previously described as an antiepileptic free radical, might also possess anti-stress properties. References: (1)- Marescaux C, Vergnes M, Depaulis A. Genetic absence epilepsy in rats from Strasbourg - A review. J. Neural Transm. 35 (1992) 37—69. (2)- Faradji H., Rousset C., Debilly G., Vergnes M. and Cespuglio R. Sleep and epilepsy: a key role for nitric oxide? Epilepsia, 41 (2000):794-801. (3)- Marinesco S., Bonnet C. and Cespuglio R. Influence of stress duration on the sleep rebound induced by immobilization in the rat: a possible role for corticosterone. Neuroscience 92 (1999) 921-933. Actas de Fisiología 7, 2001 313

COLLATERALS OF DOPAMINERGIC NIGROSTRIATAL PROJECTIONS INNERVATE THE THALAMUS AND DEGENERATE IN ANIMAL MODELS OF PARKINSON’S DISEASE

Amanda A Freeman and David Rye

The mesostriatal, mesocortical, and mesolimbic dopaminergic (DA) systems influence movement, cognition, emotion, and positive reinforcement. We identified a fourth major pathway originating from mesencephalic DA neurons: a mesothalamic system. Histo- chemical visualization of the dopamine transporter (DAT) was localized to thalamic regions known to modulate state, as well as motor and limbic related nuclei as opposed to sensory nuclei in rats, non-human primates, and humans. Immunoelectronmicrosco- py identified a majority of axon terminals in the monkey reticular thalamus as DAT- immunoreactive (IR) (28 DAT-IR vs. 35 non-IR). Two DAT-IR profiles were juxtapo- sed to cell bodies, 5 to large dendrites, 16 to small and moderate-sized dendrites, 3 to other axons, and 2 were indeterminate. Eight made distinct synaptic contacts with symmetric morphology. Anatomical tracing established axon collaterals of the nigrostriatal pathway as the origin of this innervation. In contrast to the bilateral thalamic projections from the pedunculopontine nucleus (PPN) and the locus coeruleus (LC), the dopaminergic projections from the substantia nigra pars compacta (SNc), as well as parallel projections from non-dopaminergic SNc cells, were exclusively ipsilateral. The SNc projections were also more specific with 3 non-overlapping thalamic injections backfilling only 27% of SNc cells compared to nearly 100% or more of PPN cells and 65% of LC cells. These findings implicate the thalamus as a novel site for disease specific alteration in DA neurotransmission; such as nigral degeneration attending Parkinson’s disease. This was confirmed in hemiparkinsonian rats and monkeys where reduction of thalamic DA innervation occurred coincident with signs of active axonal degeneration. Individual mesencephalic DA neurons therefore have the potential to modulate normal and pathological behavior not only through traditional nigrostriatal pathways, but also via axon collaterals innervating the thalamus. These pathways likely modulate thalamocortical arousal state given the diffuse thalamocortical dysrythmia, impairments of wakefulness and REM-sleep, and reduced cell firing in motor thalamic nuclei in PD.

Supported by USPHS grants NS-36697, NS-40221, and MH-64312 314 Actas de Fisiología 7, 2001

DECREASED TONIC AND PHASIC ACTIVITY DURING REM SLEEP IN UNTREATED PARKINSON´S DISEASE

D. Garcia-Borreguero, A.B.Caminero, Y. de la Llave, S. Barrio, A.B. Caminero2, J.A. Pareja3 Sleep Disorders Unit, Dept of Neurology, Fundación Jiménez Diaz, Madrid, Spain

Impairment of movement with akinesia, rigidity and tremor are cardinal features of Parkin- son’s disease. REM Sleep Behavior Disorder (RBD), a disorder characterized by excessive muscle twitching and lack of atonia with injurious behavior during REM Sleep, is frequently associated to, and may precede the onset of Parkinsonism. When associated, both disorders may share motor dyscontrol during REM sleep. OBJECTIVE: The objective of this study was: a) to assess whether motor dyscontrol during sleep precedes the onset of daytime symptoms in Parkinson´s Disease (PD), and b) to investigate the effect of l-dopa treatment on motor activity during REM sleep in PD patients. METHODS: 14 consecutive, recently diagnosed and treatment-naive PD patients (5 females, 9 males) suffering idiopathic Parkinson´s disease underwent all-night polysomnographic (PSG) studies. Sleep studies were repeated following a mean 184 treatment period with a low dosage of l-dopa (mean dosage: 360 mg). Age (mean: 75.3 years) and gender-matched healthy subjects were used as controls. A sleep study was repeated on 4 patients following withdrawal of treatment for 1 week. PSG- studies were recorded and scored according to standard methods. The tonic and phasic components of REM sleep were scored separately, according to the criteria used by Lapierre and Montplaisir (1992): 1. Tonic motor activity: Each 20 second-epoch was scored as “tonic” or “atonic” depending on whether tonic chin EMG activity was present for >50% or <50% of the epoch. 2. Phasic motor activity was quantified as: a. EMG twitches: The % of 2 seconds epochs containing phasic EMG twitches. The latter were defined as any burst of EMG activity lasting 0.1 to 5 seconds with an amplitude exceeding 4 times the background EMG activity. b. REM density: The number of Rapid Eye Movements (REMs) per minute of REM- sleep. Data were analyzed by means of non-parametric statistics (Wilcoxon and Mann-Whitney tests). RESULTS: No differences were observed between the groups or conditions regarding sleep architecture, PLM(periodic leg movements)-index or apnea-hypopnea-index. Compared to controls, untreated PD patients showed a reduction in phasic twitching activity (2.1 + 2.5 vs 5.4 + 4.6: p<.05). Following treatment with l-dopa, a statistically significant increase in both tonic motor activity (495%: p>0.05) and phasic twitching (280%: p<0.01) could be observed. No differences were observed between treated PD patients and controls. Furthermore, withdrawal of l-dopa treatment led to a reduction of phasic chin EMG activity to pre-treatment values (n.s.). CONCLUSIONS: Phasic and motor activity during REM sleep was reduced in untreated Parkinson’s disease as compared to controls. Treatment with l-dopa not only improved daytime symptoms but increased both phasic and tonic motor activity during REM sleep. The l-dopa induced increase in phasic and tonic motor activity during REM-sleep approached the degree of activity observed in controls. Short term discontinuation of l-dopa treatment in 4 patients resulted in a return to baseline (pre-treatment) values of phasic but not of tonic activity. Taken together, the results suggest that the reduction of phasic activity in de novo PD patients improves as a result of treatment. In contrast, the increase over time in tonic activity takes place as a result of the progression of illness, rather than as a result of treatment. REFERENCES: Lapierre O, Montplaisir J. Neurology 1992; 42: 1371-1374 Vachatimanont P, Pareja JA, Mahowald MW, Schenck CH. Sleep Res 1994: 23: 337 Actas de Fisiología 7, 2001 315

EEG SPECTRAL ANALYSIS BEFORE AND AFTER SLEEP IN HIGH-FUNCTIONING AUTISM

Élyse Limoges,1,2 Julie Poulin,1,3 Laurent Mottron,1,3 Roger Godbout1,3 1Centre de Recherche Fernand-Séguin, Hôp. Louis-H.-Lafontaine & Hôp. Rivière- des-Prairies; 2Dép. psychologie, 3Fac. Médecine, Univ. de Montréal

Studies report evidence of frontal and temporal lobe dysfunction in autism (1). We verified whether these atypicalities could be put to evidence during waking EEG and if results would be sensitive to the effect of nocturnal sleep. Method: Seven participants with autism (Group A, 7M; age: 22.1 ± 14.5) with normal IQ were compared to seven healthy controls (Group C, 7M; age: 25.7 ± 14.0). Waking EEG was recorded from four cerebral regions: Frontal (FP , FP , F , F ), Cen- tral (C , C ), Temporal (T , T ), and Occipital (O , O ) before and after1 the2 night7 8 using a monopolar3 4 montage refered3 4 to linked ears. Subjects1 2 were recorded for five minutes with eyes closed (EC) and 5 minutes with eyes opened (EO) on each moment. Spectral analysis was performed on 10 to 15 four-second epochs on absolute power (mV/Hz, 0.75Hz to 19.75Hz). Four frequency bands were created: Delta (0.75-3.75 Hz), Theta (4-7.75 Hz), Alpha (8-12.75 Hz), and Beta (13-19.75 Hz). Groups were compared with t-tests. Results: Before sleep, in the EVENING WAKING EEG DELTA POWER (0.75-3.75 Hz) EC condition, Group A showed 140 * higher Delta power than con- * Autism 120 trols for the Central and Tem- Controls * * poral regions. In the EO condi- 100 * tion, Group A showed higher * Delta and Theta power in the 80 * * Frontal region. After sleep, EEG 60 power was no more different be- POWER(µV)ABSOLUTE 40 tween groups in any condition. Fp1 Fp2 F7 F8 C3 C4 T3 T4 Discussion: Participants EYES OPEN EYES CLOSED with autism showed high spec- CONDITIONS AND ELECTRODE DERIVATIONS tral power values for slow-wave activity in Frontal, Central, and Temporal regions, a difference that disappeared after a night of sleep. These results suggest that: 1) evening fatigue levels are increased in autism compared to controls, following a day of sustained wakefulness; 2) sleep restorative functions are optimal in autism.

(1) Schultz, R., et al. (2000) Neurofunctional Models of Autistic Disorder and Asperger Syn- drome. In A. Klin, F.R. Volkmar & S.S Sparrow (Eds.) Asperger Syndrome, New York, Guil- ford Press, pp.172-209.

Supported by the “Fondation de l’hôpital Louis-H. Lafontaine” and the Natural Sciences and Engineering Research Council of Canada. 316 Actas de Fisiología 7, 2001

A DOUBLE BLIND PLACEBO CONTROLLED TRIAL OF MODAFINIL IN PARKINSON´S DISEASE PATIENTS WITH EXCESSIVE DAYTIME SLEEPINESS

Birgit Högl, Michael Saletu, Elisabeth Brandauer, Susanne Glatzl, Klaus Seppi, Gregor Wenning, Werner Poewe

The goal of this study was to assess the therapeutic efficacy of modafinil in the treatment of increased daytime sleepiness in Parkinson´s disease (PD). Methods: 12 patients with idiopathic Parkinson´s disease (9 m, 3 f; 65.0 ± 7.6 years, disease duration 6.8 ± 4.1 years) and increased daytime sleepiness (Epworth sleepiness score ESS 10 or more) completed this double blind, placebo controlled, randomized, crossover study. Patients with daytime sleepiness due to otherwise treatable causes were excluded. In two two-week treatment blocks patients received placebo or 200 mg modafinil (100 mg during the first treatment week) as a single morning dose in a randomized crossover order. Antiparkinsonian treatment was kept unchanged for the duration of the study. At baseline and at the end of each treatment block sleepiness was evaluated using subjective (ESS) and objective measures (maintenance of wakefulness test MWT, a variation of the MSLT). Results: Epworth sleepiness scores were significantly improved with modafinil (mean score improvement 3.42 ± 3.90) compared to placebo (0.83 ± 1.99; p = 0.011, paired t- tests). Sleep latency in the MWT was marginally improved; before / after placebo 10.7 (2.2-40) / 12.4 (25-40) minutes and before / after modafinil 10.5 (0.8-40) / 16.8 (4.2- 40) minutes (p < 0.1). Conclusion: The results of this study suggest, that modafinil is an effective treatment for daytime sleepiness in PD patients. Actas de Fisiología 7, 2001 317

SCREENING OF SLEEP AND CIRCADIAN RHYTHMS IN MAJOR DEPRESSION: SLEEP AND WAKE COMPLAINTS (Part 1)

Laredo Judith, Falissard Bruno, Quera Salva Maria Antonia and de Bodinat Christian I.R.I.S., Courbevoie Cedex, France

The questionnaire was aimed at identifying different profiles of depressed patients before treatment on the basis of sleep complaints, circadian typology and daily course of depressive symptoms. The questionnaire included nine items. Sleep and wake complaints were evaluated by the first five items: 1. difficulties falling asleep; 2. repeated night-time awakenings; 3. early morning awakenings; 4 difficulties being wide awake after awakening; and 5. sleepiness during the day. The questionnaire was administered to 574 patients with a major depressive episode (DSM-IV criteria), single or recurrent (HDRS score  22). 92% completed correctly the questionnaire. Patient mean age was 44.5  12.5 years, there were 413 women and 161 men. The mean HDRS score was 26.5  4.0. 58% experienced difficulties falling asleep, 60% experienced repeated night-time awakenings, 52% experienced early morning awakenings, 36% complained of difficulties being wide awake after awakening, and 38% experienced daytime sleepiness. A principal component analysis revealed two clear cut dimensions (see figure), one related to insomnia (the first three items) and the other related to daytime sleepiness (the two last items). These two dimensions did not correlate (r =0,019). These results study are in agreement with those of other studies using different methodologies. More patients experienced symptoms of insomnia than daytime sleepiness. 318 Actas de Fisiología 7, 2001

HYPERSOMNIA IN SUPRATENTORIAL BRAIN TUMORS: A SYMPTOM FREQUENTLY NEGLECTED RESULTS OF A 24 HOURS POLYGRAPHIC RECORDING

Cecilia Orellana, Cecilia Fernández Instituto de Neurología, Hospital de Clínicas, Montevideo, Uruguay

Several clinical reports have denoted the existence of hypersomnia in thalamic and brain stem tumors, according with the presence of sleep centers in those anatomical locations. Concurrently, few communications report the presence of the same symptom in supratentorial brain tumors. INTRODUCTION: A 38 years old female, with no pathologic previous clinical history , and without previous sleep disorders , began suddenly with hypersomnia .She slept almost all days during three months, waking up only to eat or void 2 or 3 times a day. Besides the hypersomnia, general physical examination and clinical neurological explorations were totally normal. She took no medications and there was no history of drug abuse. A psiquiatric origin of her hypersomnolence was first suspected, and her doctor treated her with antidepressants during 2 months, without satisfactory clinical response. She was then sent to Hospital to perform further studies, as the extended sleep continued. MATERIAL AND METHODS: A polygraphic recording of 24 hours was performed at her admittance to Hospital, A Computerized Tomography performed inmediately after hospitalization, showed a pathological image on the superior parasagittal surface of left parietal lobe corresponding to a solid tumor of 7 cm width, which had important anatomical relationship with the venous sinus and the falx and which was homoge- neously contrast-enhanced with Gadolinum. RESULTS: The polygraphic all-day recording prior to the surgical exeresis of the tumor, during which she slept 70% of the time, showed increased amount of slow wave sleep, specially stages I and II. Nocturnal sleep was slightly disrupted, because of “ first night effect” and because she had to be awakened in order to be treated with intravenous corticoids . Pathological exams of the tumor showed that it was a meningioma with no further malignant cells. 48 hours after surgical exeresis, hypersomnia had completely receded, and she regained the normal amount of sleep and circadian rithmicity she had before. Evolution was excellent further on. DISCUSSION: Hypersomnia is frequently misdiagnosed and neglected as clinical neurological symptom which may indicate the presence of organic brain disease, specially in young otherwise healthy subjects. In this case, as the venous sinus was involved, pathological edema and flow disturbances could have affected infratentorial sleep structures, thus producing the hypersomnia. Actas de Fisiología 7, 2001 319

NIGHT SLEEP OF PATIENTS WITH AMYOTROPHIC LATERAL SCLEROSIS

Karel Sonka, Eszter Horvath, Jan Fiksa and Jana Sussova Dept of Neurology, First Medical Fac, Charles Univ, Katerinska 30, Prague 2, Czech Republic

Amyotrophic lateral sclerosis (ALS) is progressive disease characterised by the reduction of central and peripheral motor neurons. The respiration insufficiency is its major and fatal complication occurring within a few months after the first appearance of the symptoms. The respiratory insufficiency is usually more prominent during sleep. The aims of the study were to evaluate the quality of the sleep in subjects suffering from ALS, to evaluate the degree of sleep respiratory disturbance in patients having no signs of respiratory insufficiency by day and finally to select the patients for ventilation support by positive intermittent ventilation. In total 9 subjects were examined: 6 men, 3 women; average age 60.1 (SD=6.9) years, BMI 25.8 (4.3). Norris´s score was 71.4 (11.2) and the average duration of the disease was 1.8 (0.5) years. 5 of them were treated by riluzol. EEG, EOG, EMG of mental and tibialis anterior muscles, ECG, respiratory flow in front of the nose and the mouth, the chest and the abdomen movements, oxygen haemoglobin saturation, the body position and video were registered one night in a separate room by the on-line polysomnography system (Schwarzer Brain Lab). The records were analysed according standard rules (Rechtschaffen and Kales). The sleep efficiency was 65.5 (8.9)% and the sleep latency 26,1 (24,6) min. The average proportions of sleep stages and wakefullness were following: 3+4 NREM sleep 11.2 (6.2) %, REM sleep 12.0 (4.4) %, wake after sleep onset 25.5 (8.7) %. The average AHI was 9.6 (7.8), but the AHI in REM sleep was 16.4 (21.5). Records of 8 subjects showed a completely abnormal sleep structure without normal NREM/REM cycles, with the reduction of 3+4 NREM sleep and the reduction of REM sleep. In 4 subjects important number of apnoeas and hypopneas were found during REM sleep. Nobody was recommended to ventilation support during the night. Periodic leg movements in sleep (PLMS) were found in 4 patients. Respiratory problems and namely PLMS were partly reasons of sleep abnormality. Summary: The sleep of ALS subjects was disturbed. Respiratory disturbances and PLMS did not explain this sleep abnormality completely.

Supported by IGA MZ CR NF5999/3 320 Actas de Fisiología 7, 2001

SLEEP DISTURBANCES AND DISEASE ACTIVITY IN PATIENTS WITH SYSTEMIC LUPUS ERYTHEMATOSUS

Matilde Valencia-Flores, Oscar Pulido, Victoria Santiago, Alejandra Castaño, Montserrat Resendiz, Josefina Montes, Rosa M. Campos, Guillermo García-Ramos, Jorge Alcocer and Donato Alarcón-Segovia Instituto Nacional de Ciencias Médicas y de la Nutrición “Dr. Salvador Zubirán” and Universidad Nacional Autónoma de México, México, D.F.

Introduction. This study describes the polysomnographic sleep changes associated with Lupus disease activity and their relationship with depression and level of fatigue. Methods. Twenty four Systemic Lupus Erythematosus (SLE) female patients, mean age of 36.7+11.3 years with a mean body mass index (BMI) of 25.5+3.9 kg/m2 agreed to participate in the study. All fulfilled at least 4 of the American College of Rheuma- tology revised criteria for SLE. Disease onset occurred 11.1+8.7 years prior to the study. Disease activity was scored according to the SLE Disease Activity Index (SLE- DAI), which was adapted for and validated in a Mexican population (MEX-SLEDAI). Eleven patients were receiving corticosteroids plus immunosuppressants (mean dosage of prednisone 5.3 mg/day, azathioprine 50 mg/day), 3 were receiving antimalarials (chloroquine 275 mg/day), 3 were receiving steroids plus immunosuppressants plus antimalarial drugs (prednisone 10 mg/day, azathioprine 100 mg/day, and chloroquine 240 mg/day). Patients were classified into two groups according to the disease activity score: Active-SLE-Group > 2, Non-Active-SLE-Group < 2. The patients underwent all-night polysomnography on 2 consecutive nights. They also completed the Beck Depression Inventory and the Fatigue Severity Scale (FSS). Results. The polysomnographic data showed that Active-SLE-Group had more light sleep (Stage 1%=13.2+3.7) than Non-Active-SLE-Group (Stage 1%=9.2+3.6), t=2.6, p<0.02; increased number of sleep stage transitions (Active-SLE- Group=168.4+27.8 versus Non-Active-SLE-Group=137.7+33.44, t=2.5, p<0.03); and less delta sleep % (Active-SLE-Group=10.9+4 versus Non-Active-SLE- Group=15.8+6.1, t=2.4, p<0.03) and they also reported a higher level of fatigue (FSS=4.7+1.7) than Non-Active-SLE-Group (FSS=3.0+2.0), t=2.4, p<0.03; but about the same level of depression (Active-SLE-Group=17.5+9.1 versus Non-Active-SLE- Group=12.1+8.3, t=1.6, p<0.13). MEX-SLEDAI score correlated with fatigue level (Spearman rho=0.41, p<0.04), sleep Stage 1% (rho=0.46, p<0.03) and number of sleep stage transitions (rho=0.54, p<.007). Conclusion. These preliminary results suggest that Lupus disease activity has an impact on sleep quality and on the level of fatigue but it is not related with the depression level.

This work was supported by CONACYT34937-H. Actas de Fisiología 7, 2001 321

FREQUENCY OF SLEEP DISORDERS IN PATIENTS WITH EPILEPSY

Daniel Zutter, Adrian M Siegel, Thomas J Loher, Fillipo Donati, Christian Sturzenegger and Claudio Bassetti Postfach 19, Inselspital, 3010 Bern, Switzerland

METHODS AND PATIENTS: Fifty (n=50) consecutive patients (27 men, 23 women, aged 19-69 years, mean 47 years) admitted in a specialized outpatient clinic, completed a questionnaire with 108 standardized questions concerning sleep-wake habits and symptoms. Body mass index exceeded 25 in (n=15) and 30 in (n=5). Epilepsy was partial in 14 and generalized in the remaining 36 cases. Patients were treated with a mean of 1.8 antiepileptic medicaments (range 0-4). The epilepsy was considered as refractory to therapy in 12 patients. At least rare nocturnal seizures appeared also in 12 cases. RESULTS: Sleep-wake complains included sleep-onset and sleep-maintenance insomnia (n=30), restless-legs symptoms (n=26), parasomnia (n=20), hypersomnia defined as a epworth sleepiness score >10 (n=10) and fatigue (n=29). Habitual snoring (always or almost always) was present (n=27), and sleep-apnoe considered probable by questionnaire (SA-SDQ>32 for women and SDQ>36 for men) (n=10). Dreams were recalled at least sometimes (n=33) or every night (n=10). CONCLUSION: This preliminary results of a ongoing project suggest a high prevalence of sleep-wake complaints in a unselected population of patients with epilepsy. Their relevance of this findings for seizure activity and control remains to be assessed. 322 Actas de Fisiología 7, 2001

VALIDATION OF A STRUCTURED TELEPHONE INTERVIEW FOR RESTLESS LEGS ON A MULTICENTER POPULATION

Wayne A. Hening1, Richard P. Allen2, Tinna Washburn2, Debbie Heckler2, Stacey Tanner2, Arthur S. Walters1,3, Christopher J. Earley2. 1University of Medicine and Dentistry of New Jersey/RW Johnson Medical School, New Brunswick, NJ; 2Department of Neurology and RLS Center, Johns Hopkins Bayview Medical Center, Baltimore, Maryland; NJ Neuroscience Institute at JFK Medical Center, Edison, NJ, USA

Background – The “gold standard” for diagnosing RLS remains the expert clinical interview. If reliable and valid epidemiological studies are going to be conducted, there must be instruments which are capable of efficiently and accurately diagnosing RLS without sustained clinical interaction. A telephone diagnostic interview developed at the Johns Hopkins Bayview RLS Center is one such instrument. It is based on the diagnostic features of RLS as established by consensus of the International RLS Study Group. It also excludes other disorders such as muscle cramps which figure in the differential diagnosis of RLS. Methods – We have now tested the instrument on 75 individuals who were recruited from the clinics of the Johns Hopkins Bayview Sleep Center and Neurology departments (N=59) and the Sleep Center of the NJ Neuroscience Institute (ASW; N=16). 38 of the subjects were patients with known diagnosis of RLS and 37 were individuals screened for absence of RLS. 3 physicians (RPA, CJE, WAH) interviewed the subjects without knowledge of their name or diagnosis. They read the telephone diagnostic interview to the subjects and made diagnoses based on the subjects’ answers. Results. It took between 2 and 10 minutes to complete each interview. There was complete and correct agreement on the diagnoses of 72 subjects. Using the standard of the prior diagnosis, there was an overall sensitivity of 99% and specificity of 96% with use of the interview. Pairwise agreement between the interviewers yielded kappa values of .92, .95, and .97. Conclusion: The Hopkins telephone diagnostic interview is a reliable, highly sensitive, and specific instrument for diagnosing RLS by experienced interviewers in a brief, anonymous telephone encounter and should provide a suitable instrument for confirming RLS diagnosis while screening large populations for RLS. Further studies are now needed to confirm the usefulness of this instrument on non-clinic-based populations. Actas de Fisiología 7, 2001 323

EFFECT OF MIRTAZAPINE IN DEPRESIVE PATIENTS WITH INSOMNIA

Francesca Cañellas, MD; José Mª de Pedro, MD

Introduction. Depressive disorders are severe and prevalent illnesses, which can be considered an important public health problem. Sleep disorders are integral component of depressive disorders and approximately 95% of depressive patients suffer insomnia. Depressive patients with insomnia have a diminution of cognitive and motor performance with higher impairment in life quality than the rest of depressive patients.

Objective. Evaluate the efficacy and tolerance of Mirtazapine in the treatment of depressive disorders associated with insomnia in Primary Health care setting.

Study Desing. This was a 6-week observational, prospective, non controlled, national, multicentric study. Three visits (basal, day 15 and day 45) were performed for each patient. Montgomery-Asberg (MADRS) scale for depression, Spiegel Sleep Questionnaire and an Adverse Events Questionnaire were performed in each visit. All patients with at least a follow up visit were used for efficacy analysis and all patients who took at least one dose of medication for tolerability analysis. All patients received a single nightly dose of 15 mg of Mirtazapine the first four days and 30 mg since the 5th. Practitioners were allowed to adjust the dose in each control visit whiting a range of 15-45 mg/day.

Results. A total of 538 General Practitioners have studied 2064 ambulatory patients with depression and insomnia (DSM IV criteria) that could be included in the study. In this preliminary report we present the data of the first 131 patients having finished the protocol, 33 men (25,19%) and 98 women (74,81%). MADRS Scale: Total punctuation of scale was 30,55 (+/-7,76) in the basal visit, 20,61 (+/- 8,94) at the 15 days visit and 9,06 (+/- 7,70) at the final one. Decrease in punctuation was significant (p< 0,0001) since the first control. Spiegel: Patients also improved the sleep quality and duration, being significant the results since the first control. No patient was withdrawn from the study due to lack of efficacy. Nine patients (6,82%) showed adverse reactions being the most usual one the diurnal somnolence (3 subjects, one of them had to be withdrawn). The remaining complaints were: bradypsychia (1), suicide attempt (1), constipation (1), hypotension (1), rash (1) and finally a patient had a non-specified crisis.

Conclusion. Mirtazapine has been shown to be an efficient and well tolerated treatment for depressive patients with insomnia.

Bibliography: (Thase, M.E. Antidepressant treatment of the depressed patient with insomnia. J Clin Psychiatry 60(suppl 17):28-31, 1999) 324 Actas de Fisiología 7, 2001

SPATIO-TEMPORAL DISTRIBUTION OF 24-HR. SLEEP/WAKE STATES INPAIRED SIDS-RISK AND CONTROL INFANTS

Anne Christake Cornwell1 and Arlyne Kim2 1Pediatrics Department, Albert Einstein College of Medicine, Peter Feigenbaum, University of Medicine and Dentistry, NJ; 2Mt. Sinai School of Medicine.

Sudden Infant Death Syndrome (SIDS) - the death of an infant that remains unexplained after an autopsy and death scene investigation - is a principal cause of death among apparently normal infants less than one years old. SIDS peaks at 2-4 mos. and occurs most often at night when infants are asleep, suggesting that sleep may be part of the pathophysiological mechanism of SIDS. In an earlier study we found a significant increase in REM sleep during the early morning hours that coincides with the peak incidence of SIDS in a group of infants at risk for SIDS (R), but not in controls (C). This study examines the duration and frequency of sleep/wake episodes in individually matched R and C infants at three developmental stages from 1- 5 mos. old to determine the stability and maturation of sleep and waking patterns with age. Sleep states were identified in 24-48 hr. continuous electrophysiological recordings of infants at high risk for SIDS and carefully matched normal controls that had no history of SIDS in the family. In each R infant an acute, unexplained life-threatening episode of respiratory arrest, cyanosis and limpness requiring immediate parental or medical intervention to restore normal breathing was documented. A standard protocol was followed: the infant was unrestrained and recorded non-invasively while in the supine position. Each polysomnographic recording began at about the same time of day to ensure comparability, and included: EEG, EOG, EMG, EKG, respiratory activity and behavioral measures as observed by a trained technician who was constantly present. Analog sleep data were manually scored “blind,” based on established criteria for infants. Sleep state results by 30sec.epochs for each 24-hr. recording were scored for REM (active), NREM (quiet) and T (transitional) sleep and entered into a computer. Then the sleep/wake data were subdivided into hourly segments. However, since not all sleep or waking states began or ended exactly at a 60min.time interval, we made the necessary minor adjustments to the data so that each sleep or wake state began and ended at a designated hourly clock time interval, i.e., 3am, 8pm, etc. The duration (in min.) and number of sleep/wake episodes for each hour of the day and night, plotted on a graph, present the amount of time spent in each sleep/wake state during the 24 hour period of recording, since polygraphic monitoring must consider the timing of the longest sleep and waking periods and the distribution of the sleep states. The results based on an initial analysis of two paired infants indicate that the temporal distribution of sleep/wake states undergo substantial changes during the first 6 months of life. Episodes of sustained, uninterrupted sleep develop gradually and are characterized by considerable variability in both R and C infants. Shorter REM and NREM cycle periods predominate at younger ages as compared to older C infants who display an increased trend toward physiological stability. Specifically in C, the amount of REM sleep and the number of episodes begin to decrease at 3 mos. old, while periods of wakefulness increase in duration and occur during daytime hours. In R, this trend is reversed and an increase in REM duration occurs with age characterized by a more fragmented state pattern, mostly during the early morning hours.The shorter cycle time found in C infants lends support to the notion that CNS development is related to cycle length. Development is also marked by the emergence of a mature sleep/waking rhythm. A plausible basis for these developmental changes can be attributed to the time course of myelinization and to the maturation of forebrain structures. Actas de Fisiología 7, 2001 325

TABLE OF CONTENTS

SYMPOSIA ...... 7

FOCUS GROUPS...... 87

POSTERS ...... 115

BEHAVIOR ...... 117

BIOCHEMISTRY ...... 128

BIOLOGICAL RHYTHMS ...... 131

BOOKS, REVIEWS ...... 141

DREAMING ...... 144

EFFECTS OF EXTERNAL STIMULI ...... 148

ENDOCRINOLOGY ...... 154

INSTRUMENTATION AND METHODOLOGY ...... 155

NEUROPHYSIOLOGY ...... 165

ONTOGENY ...... 201

PATHOLOGIES AND DISORDERS ...... 204

PERSONALITY AND PSYCHOPATHOLOGY ...... 246

PHARMACOLOGY ...... 251

PHYLOGENY ...... 263

PHYSIOLOGY ...... 264

SLEEP CYCLE IN THE HUMAN ADULT ...... 283

SLEEP DEPRIVATION ...... 288

SLEEP IN OTHER MEDICAL DISORDERS ...... 305

LAST-MINUTE ABSTRACTS ...... 322 326 Actas de Fisiología 7, 2001

AUTHORS INDEX

A Azevedo, MH, 204 Azizi, F. 129 Abel, T. 24 Azzaroni, A. 18 Acevedo, J. 13 Adamson, TM 268, 269, 275, 276 Adrien, J. 92, 100, 112, 155 B Affanni, JM. 144 Afifi, L. 163 Baconnier, M 210 Akaarir, M. 142 Bach, V. 264, 265, 279 Akanmu, MA. 257 Baía, TM. 143 Åkerstedt, T. 152, 215 Balan, J. 284 Alam, MD. 49 Bandopadhayay, P. 268 Alam, MF. 139 Barattucci, M. 293 Alam, N. 16 Barbagli, B. 67 Alarcón-Segovia, D. 320 Barbanoj, MJ. 11 Alcoberro, R. 310 Barlow, K. 290 Alcocer, J. 320 Barreneche, M. 241 Alexandre, Ch. 100 Barrio, S. 218, 219, 314 Algarín, C. 148, 305 Barros, A. 237 Algarín, CR 202, 203 Basheer, R. 128, 130 Allbright, E. 236 Bassetti, C. 72, 198, 205, 226, 306, 321 Allen, RP. 218, 322 Bastuji, H. 149, 150, 151 Amici, R. 14, 86, 153 Batejat, D. 288 Andrade, U. 252 Batista, N. 237 Andrew, S. 268 Bauso Toselli, L. 243, 245 Angeli, S. 216 Baxter, KD. 119 Anido T. 228 Beaud, C. 266 Anitua, E. 213 Beaulieu, I. 255 Arand, DL. 20 Beaumont, M. 288 Arcos, JP. 156 Beelke, M. 171, 216 Ariagno, RL. 61 Behnamedin Jameie, S. 296 Armitage R. 246, 247, 250 Behzadi, G. 296 Arnetz, B. 152 Bekinschtein, TA. 136 Arriaga, F. 229, 289 Bellacicco, M. 234 Arrigoni, E. 128 Benedito, MAC. 165, 252 Arzimanoglou, A. 266 Bengtson, H 206 Ashraf, HM. 296 Bentancor, C. 187 Assmus, A. 230 Bertini, M. 168, 292, 293, 294 Atalaia, A. 229 Bértolo, H. 145 Atienza, M. 117, 118, 158, 166 Bignotto, M. 251, 262, 299 Actas de Fisiología 7, 2001 327

Birch E.E. 148 Catalán, R. 41 Bjørkum, AA. 157 Cavaglia, F. 229, 289 Bjorvatn, B. 157, 224, 239, 283 Cerri, M. 14, 86, 153 Blanco, M 207, 307, 308, 309 Cervino, C 144 Bobboa, D. 121 Cespuglio, R. 70, 193, 312 Boccio, C. 244 Challamel, MJ 210 Boissard, R. 67 Chamberlin, N. 60 Bojic, T. 282 Chardon, K. 264, 265, 279 Bonnet, MH. 10 Chase, MH 57, 173, 196, 199 Borges Machado, R. 251 Chaves, MLF. 304 Bouali, S. 155 Chiappella, L. 156, 235 Bouferrache, B. 265 Chiorri, C. 264 Bourgeois, M. 266 Cianci, T. 282 Boutrel, B. 92, 100 Cirelli, Ch. 71, 131, 280 Brancasi, B. 234 Claudio, M. 307 Brandauer, E. 316 Clavadetscher1, SC. 306 Brandenburg, U. 12 Clement, P. 193 Brekke, C. 301 Clemente, VM 204 Broman, JE 206 Cluydts, R 209 Brusco, LI. 34 Coenen, A 113, 122 Buguet, A. 254 Colas, D. 193 Burns, J. 132 Colwell, C. 172 Butta, P. 241 Cornette, F. 278 Buzsaki, G. 108 Cornwell, AC 324 Buzsaki, G. 107 Corretger, J.M. 310 Corvasce, C. 294 Cosquer M. 138 C Costas, MA. 136 Coste, O. 288 Calasso, M. 18 Costentin, J. 155 Calbet, M. 101 Cote, KA. 119 Caleco, M C 208 Cranage, S. 268, 269, 275, 276 Calvo, JM. 83 Cristiani , R. 169, 292, 293, 294 Caminero, A.B. 314 Cugy, D. 284 Campbell, IG 201 Cugy, S. 284 Campos, RM. 162, 240, 320 Curcio, G. 169, 292, 293, 294 Cantero, JL. 117, 118, 158, 166 Curzi-Dascalova, L. 61, 266 Cañellas, F. 253, 323 Capone, J. 274 Cardinali, DP. 34 Carvalho, JGB. 252 Casas, I. 310 Castaño, A. 240, 320 Castaño, V. 162 328 Actas de Fisiología 7, 2001 D E

D’Almeida, V. 114, 137, 290, 300 Earley, CJ 322 da Rosa A. 228 Edéll-Gustafsson U. 214, 285 Darchia, N. 295 Eguchi, N. 177 Darchia, N. 21, 291 Eiland, MM. 263 Datta, S. 25, 95, 167 Eisensehr, I. 78 Dauphin, LJ. 194 Ekstedt, M. 215 Dauvilliers, Y. 226 Elia, M. 266 Davies, RO. 109 Eliozishvili, M. 298 de Andrés, I. 186, 189 Eremia, R. 236 de Bodinat, C. 317 Escamendi, H. 156 De Carli, F. 171, 216 Esnaola, S 212, 213 de Castro Moreno, CR. 39 Estragó, M. 307, 308, 309 De Gennaro, L. 168, 169, 292, 293, 294 de Groen, JHM 211 de la Calzada, M.D. 310 F de la Llave, Y. 218, 219, 314 De La Torre, G 212, 213 Fabre, V. 30, 100 de Lecea, L. 30, 101 Fagioli, I, 62 de Mello, MT. 137 Falissard, B. 317 de Pedro, JM. 253, 323 Faradji, H. 312 De Pergola, G. 234 Faul, JL. 106 de Santa-Helena, EL. 304 Feinberg, I. 201 De Zwaan, M. 146 Feng, P 141 Décima, EE. 188 Fenik, V. 109, 180 Del Sette, M. 216 Ferens, D. 268 Delgado-García, J.M. 170 Fernández, C. 318 Demasi , M. 165 Fernández, E. 41 Deschauxb, C. 94, 242 Fernández-Mas, R. 170 Deurveilher, S. 132 Ferrara, M. 168, 169, 292, 293, 294 Di Pino, P. 241 Ferri, R. 266 Díaz-Ruiz, O. 255 Ferrillo, F. 171, 216 Dinakar, P. 273 Fetveit, A. 283 Domeniconi, R. 14, 86, 153 Ficca, G. 62 Domínguez , J. 308, 309 Fiksa , J. 319 Dominguez-Marin, E. 117, 166 Fiske, E. 157 Donati, F. 321 Fleming, R. 236 Dorffner, G 11 Fletcher, EC. 267 Dreher, H. M. 311 Flett, JA. 172 Dunin-Barkowski, W. 272 Florian Chapotot, F. 254 Dupont, P. 258 Fogel, S 123 Durán, J 212, 213 Forcina Martins, PJ. 137 Fornal, C. 256 Actas de Fisiología 7, 2001 329

Fort, P. 67 Godbout, R. 248, 255, 315 Frank, MG. 42 Goicolea, A 212 Franken, P. 102 Goldstein-Daruech, N. 175 Franzini, C. 282 Golombek, D 136 Fred, A. 164 Gomes, AC 204 Freebeck, P. 236 Gonçalves, M. 160, 161 Freeman , AA. 313 Gong, H. 16, 49, 54 Frith, CD. 301 Gonzalez, MMC. 256 Froëhlich, P 210 González-Barranco, J. 240 Fujiki, N. 200 Gradisar, M. 271 Fuller, A. 15 Granieri, E. 274 Fung, SJ. 57, 173 Granizo, JJ. 218 Furio, AM. 34 Grant, DA. 281 Grant, M. 236 Greene, RW. 128 G Grenier, F. 110 Grønli, J. 157 Gabriel Jr., A. 299 Gruart, A. 170 Galvania, F. 121 Gruber, G. 11 Gambini, J.P. 174 Gugger, M. 205, 306 Gamundi, A. 142 Guieu, JD. 286 Gandolfo, C. 216 Guilleminault, C. 161, 163, 164, 233 Garay, A. 217 Gulyani, S. 302 Garcia Larrea, L 149, 150, 151 Gundel, A. 230 Garcia-Borreguero, D. 218, 219, 314 Gustavsson, G. 214 Garcia-Ramos, G. 162, 240, 320 Gvilia, I. 21, 291, 295 Garrido, M. 148, 305 Garrido, MI 202, 203 Gathmann, P. 146 H Gaultier, C. 265 Gerhardt, G JL. 304 Haba, J. 220, 221, Gerrard, P. 159 Hadjez , J. 222, 249 Gershanik, OS. 81 Hamon, M. 92, 100, 155 Gervasoni, D. 67 Hamre, F. 157 Gharib, A. 193 Harding, R. 275, 276 Giganti, F 62 Harsh, JR. 120 Gildehaus, FJ. 78 Hayaishi, O. 104, 177 Gillberg, M. 152 Hayashi, M. 176, 178 Giménez, S. 310 Heckler, D 322 Giordanella, J.P. 284 Hedayati, M. 129 Giordano, M. 259 Heinzer, R. 223 Giorgino, R. 234 Hening, W. 219, 322 Giuditta, A. 43 Hennel, F. 221 Glatz, S. 316 Henriques, I. 182 330 Actas de Fisiología 7, 2001 Herrera, M. 240 J Herrmann, W. M.11 Hetta, J 206 Jacob, J. 123 Hilton, MF. 133 Jacobs, BL. 256 Himanen, S.L. 11 Janssens, JP. 223 Hipolide, D. 114, 290, 297, 300 Jantos, H. 65 Hobson, JA. 84, 158 Jens Nilsson, J. 152 Hoffmann, G. 258 Jessen, C. 15 Hoffmann, R. 246, 250 John, J. 31 Högl, B. 64, 316 Joncas, S. 75 Holsten, F. 224, 239 Jones, B. 183 Holzinger, B. 146 Jones, BE. 46, 52, 66, 107 Honda, K. 257 Jones, CA. 14, 86, 153 Honory, A. 250 Jörg-Hermann, P. 12 Horcajada, C. 219 Joshua, I. 267 Hori, T. 176, 178 Horne, RSC. 268, 269, 275, 276 Horner, R. 185 Horvath, E. 319 K Hou, Y P. 183, 184 Kaida, K. 178 Kaminski, R. 230 J Kapás, L. 261 Kataoka, N. 176 Kawauchi, A. 270 Jones, B.E. 184 Kayama, Y. 179, 270 Kecklund, G. 215 Kemp, B. 10, 11 H Kenton, L. 124 Kentos, M. 258 Huang, ZL. 177 Kerkhof, GA. 127 Hungs, M. 230 Kerkhofs, M 209 Husain, M. 246 Kharraz, B. 78 Kim, A 324 Kinnerud, M. 214 I Klösch, G. 11, 146 Kodama, T. 181 Inderkum, A. 175 Koester, U. 163, 233 Inoué, S. 257 Kormoss, N. 258 Issa, NP. 42 Koyama, Y. 107, 179, 270 Iwatsubo Y. 138 Krahn, LE. 225 Krim, G. 279 Kubin, L. 107, 109, 180 Kumar, VM. 95, 96, 98 Kunz, D. 11 Actas de Fisiología 7, 2001 331

Kusumakar, V. 247 Louis, J 210 Kutcher, S. 247 Lovering, A.T. 272, 273 Kutz, R. 213 Lozoff B 202, 203, 305 Lu, J 53, 192 Lukowski, B. 236 L Luppi, M. 14, 86, 154 Luppi, PH. 67 Lacey, B. 268 Luthringer, R. 278 Lack L. 134, 135, 271 Lagarde, D. 288 Lai, YY. 181, 277 M Lara E. 229 Laredo, J. 317 Macedo, AF 204 Larrosa, O. 218, 219 Machado, RB. 297 Le Bon, OC. 258 Macher, J.P. 220, 221, 278 Lebas, P. 223 Maguire, E. 301 LeBlanc, J. 247 Mahowald, MW. 73 LeBourgeois, MK. 120 Maidment, N. 31 Lebrun, T. 286 Maisuradze, L. 298 Lecendreux, M. 226 Maloney, SK. 15 Leke, A. 264, 265 Mallick, HN. 97 Lenain, J.L. 284 Manconi, MMM. 274 Lenne, X. 286 Mandile, P. 43 Lenzi, P. 282 Mann, G. 180 Leonhardt, E. 74 Manns, ID. 183, 184 Levy Andersen, M. 251 Marchenko, V. 109, 180 Li, SP. 126 Marpegan, L. 136 Li, WD. 177 Marques, R 145 Libert, J.P. 264, 279 Mascetti, GG. 121 Limoges, E. 315 Massaro, M. 232 Lin, L. 300 Mathis, M 205 Linke, R. 78 Matoso, M. 241 Lion, K. 258 Matos-Pires, A. 229, 289 Liu, H. 185 Matsumura, H. 200 Liu, X. 185 Mayer, G. 226 Loenneker, T. 221 Mayer, G. 74 Loher, TJ. 321 Mazzola, ME. 232 Lomba, A. 182 McCarley, R. 47, 103, 128, 130, 194, 195 Longhi, G. 282 McGinty, D. 16, 49, 54 Lopes da Silvab, FH. 140 Mednick, S. 118 López Varela M.V. 228 Mele, P. 241 López-Esteban, P. 231 Melotto, S. 159 Lorenzo, D. 156, 235 Mello, MT. 287 Lortkipanidze, N. 298 Mena, J. 259 332 Actas de Fisiología 7, 2001

Menna-Barreto, L. 38 Nobili, L. 171, 216 Mestre, T. 145 Nobrega, JN. 114, 290 Meza, S 208 Norra, C. 230 Mgaloblishvili-Nemsadze, M. 298 Nunes, AP. 160 Mignot, E 200, 205, 300 Miki, T. 270 Milner, CE. 119 Ñ Minenna, A. 234 Mirmiran, M. 61 Nuñez, A. 186, 189 Mitchell, D. 15 Mochizuki, T. 177 Modai, I. 222 Montes, J. 162, 240, 320 O Montesanti, P. 234 Monti, D 22 Ocampo-Garcés, A. 13, 85 Monti, JM 22, 35, 65 Okruch, AJ. 133 Montplaisir, J. 75 Olaza, JL. 243 Morales, FR. 57, 173, 196, 199 Oliveira, AC. 114 Morehouse, R. 247 Oniani, L. 298 Morrison, A. 180 Oniani, T. 21, 295, 298 Morrison, AR. 55, 79, 191 Oniani, TN. 291 Morrison, JL. 185 Opp, MR. 93 Mottron, L. 315 Orellana, C. 318 Müller, C. 197 Orem, J. 272, 273 Múnera, A. 170 Orolinova, N. 267 Muñoz, EH 202 Osip, SL. 119 Muñoz, M.D. 170 Osuna, E. 238 Murphy, JR. 258 Otaño N. 156, 228 Musumeci, SA. 266 Muzet, A. 278 P

N Paiva, T. 145, 160, 161, 164, 182, 237 Palma, BD. 299 Nader, R. 125 Palumbo, M. 234 Nakajima, T. 200 Pandi-Perumal S.R 142, 173 Nakayama, K. 118 Pannacciulli, N. 234 Neidhart, E. 226 Parapatics, S. 146 Netchiporouk, L. 70 Pareja, J.A. 314 Nicolau, M.C. 142 Parmeggiani, PL. 14, 18, 86, 154 Nienhuis, R. 302 Parra, A. 213 Nishino, S 82, 200, 205 Parrilla, S. 243 Nittono, H. 176, 178 Parslow, P. 275, 276 Noachtar, S. 78 Partridge, K. 134 Actas de Fisiología 7, 2001 333 Paty, J. 284 Q Pedemonte, M. 40, 174, 175, 187, 197 Pedrazzoli, M. 290, 300 Qu, WM. 177 Peever, JH. 181, 277 Quera Salva , M. A. 138, 317 Peirano, J. 13 Queralt, A. 310 Peirano, P. 148, 305 Quintero, EA. 166 Peirano, P.D. 202, 203 Pelc, I. 258 Penzel, T. 11, 12, 74, 161, 164 Peñarroya, S. 310 R Peraita-Adrados, R. 231 Pereira, AP 204 Radanov, B 205 Perez, E. 14, 86, 154 Ramanathan, L. 302 Perez-Padilla, R 208 Ramesh, V. 98, 128, 130, 194 Pérez-Perera, L. 187 Ramírez-Salado, I. 83 Perrin F. 149, 150, 151 Ratner, Y. 222 Pessoa, L. 145 Ray, LB. 119 Petruzzellis, A. 234 Raymond, R. 290 Philip, P. 161 Rehm-Berbeni, C. 164 Pichler, R. 236 Reinoso-Suárez, F. 186, 189 Pierard, C. 288 Reséndiz , M. 162, 240, 320 Pierdominici, M. 241 Rial, R.V. 90, 142 Pigeau, R. 254 Ribeiro AC. 261 Pinto Junior , LR. 262 Riley, BT. 26 Piquer, JM 13 Ripley, B 200, 205 Pires-Barreira, D. 289 Risacher, G. 241 Piscopo, S. 43 Ritsner, M. 222 Pissarra, CM 204 Rizzo, GNV. 139 Podesta, CS. 232 Rochat, T. 223 Poe, GR. 26 Rodrigo-Angulo, M. 186, 189 Poewe, W. 316 Rodríguez, A. 197, 235 Poffe, A. 159 Rodríguez, Z. 156 Poirot, I. 286 Roegel, J.C. 278 Poissonet C.M. 138 Rojas, MJ. 57 Ponizovsky, A. 222 Rosa Santos, EH. 137, 287 Popa, D. 92, 155 Rosa, A. 163, 233 Portas CM. 157, 301 Röschke, J. 11 Poulin, J. 248, 315 Ross, R. 180 Poyares, D. 163, 233 Ross, RJ. 191 Prchal A. 188 Rostig, S. 12 Predieri, S. 282 Rotenberg , VS. 249, 303 Puca, FM 234 Roth, C. 197 Pulido, O. 320 Rousset, C. 312 Rubio, R 212, 213 Rudolf, G. 241 334 Actas de Fisiología 7, 2001

Ruggeri, S. 244 Simón-Arceo, K. 83 Rush, J. 246, 250 Sinton, CM. 194 Rye, D. 56, 313 Smit, A. 122 Smith, C 123, 124, 125 Smith, CT. 27, 44 S Sollés, J 212 Sonka, K. 319 Sailly, JC. 286 Sood, S. 185 Saiz, T. 219 Soons, JWPH 211 Salas, RM. 117, 166 Sørensen, E. 157 Salazar, R. 243 Spadini, V. 292 Saletu, M. 316 Staner, L. 220, 221, 278 Salzarulo, P. 62 Stein , D. 249 Sampogna, S. 196 Steriade, M. 110 Samuelsson, B. 215 Stickgold, R. 118, 158 Sanford, LD. 190, 191 Stip, E. 248 Santiago, V. 162, 240, 320 Stomati, D. 234 Santos, EA. 139 Stone, B. 288 Saper, CB. 53, 60, 192 Stoohs, RA. 106 Sarda, N. 193 Strange, B. 301 Savarese, MA. 234 Strecker, RE. 32, 194 Scavone, C. 235 Stryker, MP. 42 Schlögl, A. 11 Sturzenegger, C. 205, 321 Schönwald, SV. 304 Suchecki, D. 143 Schulz, H. 23 Sugarbaker, RJ. 133 Schüssler, P. 146 Sunderji, I. 301 Schwarz, J. 78 Suntsova, N. 16 , 54 Seabra, ML. 262 Sussova, J. 319 Semba, K. 48, 132 Svanborg, E., 214 Seppi, K. 316 Swahn, E., 214 Sexton-Radek, K. 236 Szymusiak, R. 16, 49, 54 Sforza, E. 223 Shea, S. 133 Sheils, SA. 133 T Shiyi, L. 227 Shram, N. 70 Tafti, M. 72, 102, 226 Siegel, AM. 321 Tahawi, Z. 267 Siegel, J. 181, 263, 302 Takahashi, K. 179 Siegel, JM. 31, 277 Tanaka, Y. 270 Silber, M. 225 Tang, X. 190, 191 Silva, EM 204 Tanner, S 322 Silvani, A. 282 Tarnow, EG. 147 Silvestri, R. 80 Tatsch, K. 78 Simona, NR. 140 Teles Araújo, A. 160, 164 Actas de Fisiología 7, 2001 335 Teles Araújo, M. 161, 164, 182, 237 V Telliez , F. 264, 265, 279 Thakkar, MM. 32, 103, 194, 195 Vadilonga, D. 72 Thannical TC. 263 Valdebenito, BC, 203 Thatte, H. 128, 130 Valencia-Flores, M. 162, 240, 320 Thompson, CM. 26 Vallortigarab, G. 121 Tietzel, A. 135 van Eekelen, APJ 127 Timofeev, I. 107, 110 Vanini, G. 197 Tobon, A. 238 Van Someren, E. 18 Tononi, G. 131, 280 Vaugeois, JM. 155 Töpper, R. 230 Vazquez, JC 208 Torterolo, P. 57, 196 Vecchi, C. 241 Trenker, E. 11 Velan, O. 244 Trenkwalder, C. 78 Velluti, R.A. 174, 175, 187, 197 Trinder, JA. 133 Venturi, S. 14, 86, 154 Trivedi, M. 250 Verbanck, P. 258 Tsai, LL. 126 Veron MA 138 Tufik, S. 114, 137, 233, 251, 262, Vertes, RP. 107 287, 290, 297, 299 Vescia, S. 43 Tugnoli, V. 274 Viardot, G. 278 Turner, C. 288 Vidruk, E.H. 272, 273 Tyba, PA. 143 Vinclair, J. 284 Tysor, MJ. 26 Viot-Blanc, V. 94, 242 Vitkovic, J. 268 Vivaldi, E. 13, 41, 85 U von Lindeiner, H. 78

Uauy, R. 148 Ulfberg, J 206 W Upton, N. 33 Urade, Y. 177 Walker, AM. 281 Urade, Y. 104 Walters, AS 322 Ursin, R. 157, 224, 239 Washburn, T 322 Watanabe, T. 177 Watts, AM. 268 Wenning, G. 316 Wetter, TC. 78 Wild, J. 281 Wilkemeyer, M. 194 Wright, H. 134 Wu, MF. 31 336 Actas de Fisiología 7, 2001 X

Xi, M. 57, 173, 199

Yacoubi, ME. 155 Yamamoto, M. 179 Yamuy, J. 57, 173, 196 Yazdanparast, R. 129 Yoneda, H. 200 Yoshida, Y. 200 Yunhua, Y. 227

Zadra, SA 75 Zamboni, G. 14, 86, 154 Zanniello, GL. 243, 244, 245 Zeitlhofer, J. 11 Zhang, J. 57 Zoccoli, G. 282 Zorrilla, V 212 Zubia, S. 213 Züchner, S. 230 Zuloaga Soraya, Z 208 Zulley, J. 161 Zutter, D. 321