Chronic exposure of rats to noise: relationship between long-term memory deficits and slow wave sleep disturbances.

Noise is now recognized as a serious health problem in our modern societies. Although its deleterious and direct effects on cognitive tasks (long-term memory, mental arithmetic activity, visual tasks, etc.) are clearly admitted, no studies have determined a delayed indirect effect of noise on cognitive processes. Furthermore, the link between sleep disturbances related to environmental noise (EN) exposure and these indirect deteriorations of human performances has never been demonstrated. This could be due to inappropriate evaluation of sleep as well as to uncontrolled and confounding factors such as sex, age, and also inter-individual vulnerability. Based on a recently validated animal model [Rabat A, Bouyer JJ, Aran JM, Le Moal M, Mayo W. Chronic exposure to an environmental noise permanently disturbs sleep in rats: inter-individual vulnerability. Brain Res 2005;1059:72-82], aims of the present study were (i) to determine long-term memory (LTM) deficits following a chronic exposure to EN and (ii) to link these behavioral problems to sleep disturbances related to EN. For this purpose in a first experiment, LTM performances were evaluated before and following 9 days of EN. Results show LTM deficits following a chronic exposure to EN with inter-individual vulnerability. Vulnerability profile was related to the psychobiological profile of rats. Results of the second experiment show LTM deficits correlated to both debt of slow wave sleep (SWS) and to daily decrease of SWS bout duration. Our results demonstrate that chronic exposure to noise indirectly disturbs LTM possibly through SWS disturbances and suggest a possible role of the stress hormonal axis in these biological effects of noise.

Chronic exposure to an environmental noise permanently disturbs sleep in rats: inter-individual vulnerability.

Chronic exposure to an environmental noise (EN) induces sleep disturbances. However, discrepancies exist in the literature since many contradictory conclusions have been reported. These disagreements are largely due to inappropriate evaluation of sleep and also to uncontrolled and confounding factors such as sex, age and also inter-individual vulnerability. Based on a recently validated animal model, aims of the present study were (i) to determine the effects of a chronic exposure to EN on sleep and (ii) to evaluate the inter-individual vulnerability of sleep to EN. For this purpose, rats were exposed during 9 days to EN. Results show that a chronic exposure to EN restricts continually amounts of slow wave sleep (SWS) and paradoxical sleep (PS) and fragments these two sleep stages with no habituation effect. Results also evidence the existence of subpopulations of rats that are either resistant or vulnerable to these deleterious effects of EN on sleep and especially on SWS amounts, bouts number and bout duration. Furthermore, importance of SWS debt and daily decrease of SWS bout duration are correlated to each others and both correlate to the amplitude of the locomotor reactivity to novelty, a behavioral measure of reactivity to stress. This last result suggests that this psychobiological profile of subjects, known to induce profound differences in neural and endocrine systems, could be responsible for their SWS vulnerability under a chronic EN exposure.

Deleterious effects of an environmental noise on sleep and contribution of its physical components in a rat model.

Sleep disturbances induced by environmental noise (EN) exposure are now well admitted. However, many contradictory conclusions and discrepancies have been reported, resulting from uncontrolled human factors or the use of artificial noises (pure tone). Thus, the development of an animal model appears to be a useful strategy for determining whether EN is deleterious to sleep. The aims of this study were: (i) to confirm the effects of noise on sleep in a rat model; and (ii) to determine the most deleterious physical component of noise regarding sleep structure. For this purpose, rats were exposed during 24 h either to EN or to artificial broad-band noises [either continuous broad-band noise (CBBN) or intermittent broad-band noise (IBBN)]. All the noises decrease both slow wave sleep (SWS) and paradoxical sleep (PS) amounts during the first hours of exposure. However, CBBN acts indirectly on PS through a reduction of SWS bout duration, whereas IBBN and EN disturb directly and more strongly both SWS and PS. Finally, EN fragments SWS and decreases PS amount during the dark period, whereas IBBN only fragments PS. These results demonstrate the validity and suitability of a rodent model for studying the effects of noise on sleep and definitively show that sleep is disturbed by EN exposure. Two physical factors seem to be implicated: the intermittency and the frequency spectrum of the noise events, which both induce long-lasting sleep disturbances. An additive effect of frequency spectrum to intermittency tends to abolish all possible adaptations to EN exposure. Since sleep is involved in cognitive processes, such disturbances could lead to cognitive deficits.

The effect of restraint stress on paradoxical sleep is influenced by the circadian cycle.

It is well known that the physiological impact imposed by events or behaviors displayed during the waking period determines the way organisms sleep. Among the situations known to affect sleep both in its duration and quality, stress has been widely studied and it is now admitted that its effects on sleep architecture depend on several factors specific to the stressor or the individual itself. Although numerous reports have highlighted the prominent role of the circadian cycle in the physiological, endocrine and behavioral consequences of restraint stress, a possible circadian influence in the effects of stress on the sleep-wake cycle has never been studied. Thus the present study was designed to compare the effects on sleep of a 1 h-lasting restraint stress applied at light onset to those observed after the same stressor was applied at light offset. We report that in both conditions stress induced a marked paradoxical sleep increase, whereas wakefulness displayed a moderate decrease and slow wave sleep a moderate augmentation. Although the effects of stress at lights on were of similar magnitude than those of stress at lights off, important differences in the sleep rebound latencies were observed: whatever the time of day the stress was applied, its effects on sleep always occurred during the dark period. This result thus shows that restraint stress could be efficiently used to study the interaction between the circadian and homeostatic components of sleep regulation.

Infusion of neurosteroids into the rat nucleus basalis affects paradoxical sleep in accordance with their memory modulating properties.

The neurosteroids pregnenolone sulfate and allopregnanolone affect memory processes in an opposite manner, pregnenolone sulfate acts as a potent memory-enhancer whereas allopregnanolone impairs memory performance. The mechanisms underlying these memory modulating properties have yet to be elucidated. We have previously reported that infusions of either neurosteroid into the nucleus basalis magnocellularis, one of the main forebrain cholinergic nuclei, differentially affect spatial memory in rats. The relationships between memory performance and paradoxical sleep are well documented, therefore we investigated whether neurosteroids infused into the nucleus basalis magnocellularis affected the sleep-wakefulness cycle in rats, measured by electroencephalographic recordings. Results show that pregnenolone sulfate (5 ng) increased by 12%, whereas allopregnanolone (2 ng) decreased by 24%, the duration of paradoxical sleep in the 24 h interval following injection compared to control recordings. Pregnenolone sulfate inhibits GABA(A) receptors whereas allopregnanolone stimulates them. Since cholinergic neurons of the nucleus basalis magnocellularis are GABA-modulated, it may be postulated that these neurosteroids modify paradoxical sleep by acting on the cholinergic transmission. This may account, at least in part, for the memory modulating properties of these compounds.

The promnesic neurosteroid pregnenolone sulfate increases paradoxical sleep in rats.

The effect of systemic administration of the neurosteroid pregnenolone sulfate (PREG-S) on sleep-wakefulness cycle and on spatial memory performances was investigated in male Sprague-Dawley rats. In the first experiment, the effect of PREG-S administration (saline, 4.75, 47.5 mg/kg, i.p.) on 24 h EEG recording was evaluated. In the second experiment, spatial memory performance in a two-trial memory task was evaluated after post-acquisition administration of similar doses of PREG-S as in the first experiment. Results show that PREG-S increases paradoxical sleep and improves the performance on the memory task yielding similar dose response curves. Starting 4 h after administration of 47.5 mg/kg PREG-S, paradoxical sleep is increased for 10 h. The PREG-S effect on spatial memory lasts for at least 24 h after injection. These results suggest that an enhancement of paradoxical sleep may be involved in the promnesic effects of this neurosteroid.

Reaction of sleep-wakefulness cycle to stress is related to differences in hypothalamo-pituitary-adrenal axis reactivity in rat.

Acute stress is known to modify sleep-wakefulness cycle, although with considerable interindividual differences. The origin of these individual differences remains unknown. One possibility is an involvement of the hypothalamo-pituitary-adrenal axis (HPA), as its reactivity is correlated with an individual's behavioral reactivity to stress, and it is known to influence the sleep-wakefulness cycle. The present study was designed to analyze relationships between natural differences in behavioral reactivity to stress associated with differential HPA reactivity and stress-induced changes in sleep-wakefulness. Adult rats were classified into two sub-groups according to their locomotor reactivity to a mild stress (novel environment): the 'low responders (LR)' and the 'high responders (HR)' animals exhibited different glucocorticoid secretion in response to stress. We show that immobilization stress induced an increase in wakefulness in LR animals and a decrease in wakefulness in HR animals. On the other hand, paradoxical sleep was increased in both LR and HR animals. Moreover, we observed that LR animals slept more than the HR animals, whereas the two groups had similar levels of paradoxical sleep. These results indicate that the response of the sleep-wakefulness cycle to stress is related to the behavioral reactivity to stress, in turn governed by the individual's reactivity of the HPA axis. The involvement of dopaminergic mechanisms is discussed.

Inter-individual differences in the effects of acute stress on the sleep-wakefulness cycle in the rat.

It has been described that an acute immobilization stress (IS) modifies subsequent paradoxical sleep (PS). However, its effects are complex because some subjects remain unaffected. This discrepancy might result from constitutive inter-individual psychobiological differences. In order to test this hypothesis, an inter-individual analysis of sleep patterns and their modifications after 60 min IS has been performed. Even though global analysis showed a PS increase after IS, inter-individual analysis evidenced different PS reactivity; subjects which had the least PS during control recordings were those with the largest PS increase. Unlike global analysis, an inter-individual study evidenced different modifications of wakefulness and slow wave sleep according to individuals. Subjects presenting the highest amount of wakefulness in control conditions (the lowest amount of slow wave sleep) decreased their wakefulness amount, while subjects with the lowest amount of wakefulness increased it. Thus, individual characteristics of the sleep-wakefulness cycle should be considered while studying its modifications induced by different treatments.

Effects of locus coeruleus lesions on vigilance and attentive behaviour in cat.

Previous data have suggested that in the cat, expectancy behaviour (waiting for a target to appear) and associated electrocortical, focal, synchronized activity ('mu' rhythms) are modulated by a noradrenergic system possibly originating from the locus coeruleus (LC). To test the latter hypothesis, we have examined the behavioural and ECoG changes induced after bilateral LC lesions. Our results demonstrated that destruction of the anterior 3/4th of the LC (A6 noradrenergic cell group) resulted in a considerable increase of mu rhythms and expectancy behaviour, without episodes of drowsiness that normally occur. Destruction of the posterior fourth of LC (A4 noradrenergic group) only increased the duration of slow sleep. Extending the A6 lesion to include the dorsal ascending noradrenergic bundle also increased the expectancy behaviour and mu rhythms. Finally, when the nucleus subcoeruleus was also involved, the duration of slow sleep and the frequency of paradoxical sleep episodes increased. These findings indicate that the LC exerts an inhibitory effect on structures involved in the induction and persistence of expectancy behaviour with accompanying mu rhythms.

Effects of mediodorsalis thalamic nucleus lesions on vigilance and attentive behaviour in cats.

In this study performed on 10 cats we analysed the effects of limited lesions of nucleus medialis dorsalis of the thalamus (MD) on behaviour and on some specific electrocorticographic (ECoG) patterns, known from previous works to accompany various states of waking attentiveness. The animals were tested during 90 min in three distinct behavioural situations; a neutral one (NS) where they simply explored their environment and then usually went to sleep; a second one (FA) where they could watch a mouse (but not catch it), which favoured 'focussed' attentiveness accompanied by the development of rhythmic ECoG activities in the anterior frontoparietal cortex ('beta rhythms' at 40 Hz); a third one (EX) with a hidden mouse whose appearance the cat was waiting for ('expectancy') which rather than favouring the development of beta rhythms elicited the appearance of another rhythmic activity dominating in somatic area SI, 'mu rhythms' at 14 Hz. The duration of each waking behaviour and its ECoG concomitant as well as that of slow wave and of paradoxical sleep were compared in each cat before and after lesion. Our results showed that MD lesions situated in the posterior part of the nucleus tended to increase the time occupied by focussed attentive behaviour and the accompanying beta rhythms. On the other hand, anteriorly located MD lesions elicited a concomitant reduction of both manifestations, behavioural and electrocortical. No such contrasting effects could be systematically noticed for the durations of expectancy, of slow sleep and of paradoxical sleep. These findings were discussed considering previous data showing that at least two systems project upon MD, that play distinct--in a way even antagonistic--roles in focussed attention upon a target. Both originate from the ventral tegmental mesencephalic area; one reaches MD through the ventral striatum (nucleus accumbens), the other one through the amygdala. The MD nucleus thus receiving contrasting information may participate in a final adjustment of the attentive state of the animal to its environment.

Noradrenaline-like terminals in the cat nucleus ventralis posterior of the thalamus.

Noradrenaline-like immunoreactivity in the cat nucleus ventralis posterior of the thalamus was investigated using an indirect immunocytochemical technique. Specific antinoradrenaline antibodies, raised in rabbits, were used. It was first verified that these antibodies recognize noradrenaline cells bodies of the locus coeruleus and their ascending axons in the ascending noradrenergic tract. In the nucleus ventralis posterior itself, noradrenaline-like fibers were observed. They were either randomly distributed or grouped around nonlabeled cell bodies. These neurons were generally oblong and measured 60-80 microns. With electron microscopy, preliminary results showed immunoreactive fibers in close apposition to unlabeled cell bodies or dendrites. The precise nature of these profiles was sometimes difficult to ascertain, since experiments were done in presence of detergent. In some cases symmetric synapses might be observed between immunoreactive axon terminals and unlabeled dendrites. The specificity of the reaction is discussed in the light of several control experiments.

Action of tianeptine on focalization of attention in cat.

Tianeptine, an antidepressant substance devoid of sedative action, was investigated for its effects on focalization of attention in cats, using combined behavioural and electrocorticographic (ECoG) observations. The ECoG index was the presence of 40 Hz frontoparietal rhythmic cortical activities, developing while the animal displayed a behaviour suggesting focused attention. Cats were observed in two conditions: a "neutral" one, with no specific target, and another one, "of focused attention" where a live mouse placed in a transparent box was present in the recording room, each test lasting for 90 min. After treatment the animals displayed increased attention even in the neutral situation, where the cat, instead of sleeping like during control sessions, payed sustained attention to its environment. Conversely, at corresponding doses, amitriptyline, another antidepressant drug chosen as a reference, induced uninterrupted slow sleep even in the situation of focused attention.

Effect of DSP4, a neurotoxic agent, on attentive behaviour and related electrocortical activity in cat.

Six behaving cats were administered N-(2-chloroethyl)-N-ethyl-2-bromobenzylamine (DSP4) (i.p. 10 mg/kg), a neurotoxic agent known to destroy noradrenergic endings in the CNS. They were placed, both before (control) and after treatment in 3 different situations, each time for 90 min: (1) a 'neutral' one (N) with no significant stimuli; (2) another one eliciting focused attention (F); (3) a third one, creating a situation of 'expectancy of an event to occur' (E), with the animal usually displaying a posture of quiet waking. Simultaneously, the electrocorticogram (ECoG) was recorded from the sensorimotor and parietal cortex through implanted electrodes. We already knew and were able to confirm that the normal animals (i.e. before treatment) displayed distinct behavioural patterns depending on the situation and distinct accompanying parietofrontal ECoG activities, with a dominance of drowsiness and sleep during N, that of a 36-Hz 'beta' rhythms in condition F, and of 14-Hz 'mu' rhythms in condition E. It was shown that the prevailing attitude of the animals after treatment was now, in all 3 situations, that of 'quiet waking and/or expectancy-like watching', with a large if not exclusive dominance of only one ECoG pattern, namely mu. These changes were considered as due to a release of the mu system from a noradrenergic modulatory blockade, in accordance with some of our previous data. An immunohistochemical study with anti-tyrosine hydroxylase antibody was also performed; it confirmed that after DSP4 treatment there were substantial alterations in the immunoreactivity of locus coeruleus cells, the structure which is likely to be involved in this NAergic control of the mu rhythms and of its concomitant behaviour.

[Regulation pathways of expectancy behavior in the cat: histologic study]. / Voies de régulation du comportement d'attente chez le chat: étude histologique.

In the behaving cat, motion expectancy of an event to occur (for a prey to appear) is accompanied by the development of 14 Hz electrocortical mu rhythms in the hand subarea of cortical somatic area SI. Our first aim here was to identify subcortical sites projecting to this cortical mu focus, using localised retrograde HRP marking. The only site thus labelled was the thalamic zone well known to project to the cortical mu area, and to act as a generator for the mu rhythms (ventral posterior nucleus, VP); no other deep structure could be identified, that could have been considered as a putative zone for control of cortical mu. We then injected minute amounts of HRP into the thalamic mu zone; labelled neurones were located (apart from those expected in the relays of the somatic pathway) in locus coeruleus (bilaterally) and ipsilaterally in the thalamic nuclei anteroventralis and laterodorsalis. In brief then, it seems that the regulation of the VP-SI mu channel (that we could previously demonstrate), by other deep structures is exerted upon the thalamic side.

[DSP 4, a neurotoxic agent that destroys noradrenergic endings, selectively increases "expectancy" in the cat]. / Le DSP 4, neurotoxique destructeur de terminaisons noradrénergiques, favorise électivement le comportement "d'attente" chez le chat.

Cats were treated with DSP 4, a neurotoxic agent known to destroy central noradrenergic endings. A significant increase was subsequently noticed in the amount of time spent by the treated animals in an attitude of "expectancy", i.e. of motionless waiting for an "event to occur". They even developed this attitude when no such real situation existed. Concomitantly, an increase was noticed in the power of the 14 Hz electrocortical rhythms recorded over the somatic sensory cortex. These patterns, designated as "mu" rhythms, had previously been shown to characterize this particular type of attentive state. The present data tends to confirm our previous hypothesis, that immobile expectancy and its accompanying electrocortical pattern are under a noradrenergic inhibitory control.

Anatomical localization of cortical beta rhythms in cat.

Beta electrocorticographic rhythms (40 Hz) develop during motionless focused attention in two distinct cortical foci in cats. A cytoarchitectonic study was performed to determine the precise location of these foci. Electrode tips recording beta rhythms were found: (i) in motor areas 4 gamma and 6a beta, in a band extending from the postcruciate cortex to the walls of the presylvian sulcus, crossing the frontal pole (anterior beta focus); (ii) in the posterior parietal associative area 5a, along the divisions of the ansate sulcus, extending to the mesial aspect of the hemispheres (posterior beta focus). The two foci are separated by areas 3, 2 and 1, where beta rhythms were never recorded. The fact that both these areas, containing giant pyramidal cells, develop a specific type of activity during immobility may have a functional meaning: area 5 may be involved in the cat as it is in the monkey in the control of motor behaviour.

Compulsive attentive behavior after lesion of the ventral striatum in the cat: a behavioral and electrophysiological study.

Bilateral lesions of the nucleus accumbens in cat elicited the following changes: preservation in tests requiring focused attention, with difficulty to shift to other targets; paucity of movements, animals displaying moderate hypokinesia and loss of reaction to changes in the environment. These symptoms were accompanied by a significant increase in the amount of beta rhythms, an activity that has been shown to be concomitant with the development of focused attentive behavior. The observed behavioral and electrocortical modifications are opposite those that have been previously obtained in the same species after lesions of the ventral tegmental area.

[Cytoarchitectonic localization of foci of electrocortical activity accompanying focused attention in cats]. / Délimitation cytoarchitectonique des foyers d'activité électrocorticale accompagnant l'attention focalisée chez le chat.

Beta electrocorticographic rhythms (30-45 Hz) develop during focused immobile attention within two distinct foci in cats. A multiple electrode exploration was performed, followed by post-mortem histological analysis, to determine the precise localization of these foci. Electrode tips recording beta rhythms in the waking attentive cat were located: in motor areas (Brodmann's areas 4 and 6), in a band extending from the postcruciate cortex to the walls of the presylvian sulcus, crossing the frontal pole (anterior beta focus); in the posterior parietal associative area 5a, along the divisions of the ansate sulcus (posterior beta focus). The two foci are separated by somatic areas 3, 2 and 1, where beta rhythms were never recorded. The location of the posterior focus may suggest that area 5 is, in the cat as it is in the monkey, involved in motor control.

Ultrastructural localization of tyrosine hydroxylase in rat nucleus accumbens.

Immunocytochemical localization of tyrosine hydroxylase (TH) was used to determine the ultrastructural morphology and synaptic associations of catecholaminergic terminals in the nucleus accumbens of the rat. The brains were fixed by vascular perfusion with 4% paraformaldehyde and 0.2% glutaraldehyde. Coronal sections cut with a vibrating microtome were incubated with rabbit antiserum to TH then immunocytochemically labeled by the peroxidase-antiperoxidase method. Immunoreactivity for the enzyme was found within unmyelinated axons and axon terminals. These terminals contained either all small clear or combined small clear and large dense core vesicles. Approximately 40% of the labeled terminals formed symmetric synapses with unlabeled proximal or distal dendritic shafts. The dendrites showed a spare distribution of spines. Axosomatic synapses and axonal associations of the TH-containing terminals also were detected. The recipient perikarya were usually 10-20 micrometers in diameter and contained an indented nucleus and abundant cytoplasm. The content of large dense vesicles and synaptic associations with somata and proximal dendrites suggest that a certain proportion of the TH-containing terminals within the nucleus accumbens are morphologically distinct from catecholaminergic terminals within the dorsal striatum. These differences are discussed in relation to neuropeptides and functions of the dopaminergic mesolimbic and nigrostriatal pathways.

Chemical and structural analysis of the relation between cortical inputs and tyrosine hydroxylase-containing terminals in rat neostriatum.

Levels of tyrosine hydroxylase (TH) and the ultrastructural relation between axons from cerebral cortex and TH containing, predominantly dopaminergic terminals were examined in the adult rat neostriatum at 2 and 12 days following unilateral decortication. The caudate nuclei from the unlesioned and lesioned hemispheres were biochemically assayed for TH processed for light or electron microscopic localization of the enzyme. At both time intervals examined, there was no statistically significant alteration in TH activity or apparent change in the intensity of reactive labeling visualized by light microscopy. However, electron microscopic examination of the caudate nucleus homolateral to the decortication at two days following surgery revealed the presence of numerous small, osmiophilic boutons which were much less frequently seen on the contralateral side. Further ultrastructural examination showed that the osmiophilic boutons formed predominantly asymmetric, axodendritic synapses. In sections containing both degenerating and TH labeled terminals, two patterns of connectivity could be discovered. First and most commonly, the degenerating and TH-labeled terminals formed synapses with the same dendrite or dendritic spine. Less frequently, the two types of terminals were in direct contact with each other. In this axo-axonic relation, the outer membranes between the terminals were in apposition but usually failed to exhibit pre- or postsynaptic specializations. These findings indicate that the cortical and dopaminergic nigral efferents have actions on common recipient neurons in the rat caudate nucleus and provide support for a possible direct axonal interrelationship between these two primary inputs.