Les altérations du sommeil et du cycle veille/sommeil au cours du vieillissement constituent des facteurs de risque de l’apparition d’un déclin cognitif et de l’évolution vers une démence. Pour autant, le lien entre ces altérations et les modifications cérébrales liées à l’âge reste encore peu connu. L’objectif de cette thèse a été de déterminer si les modifications du sommeil et du cycle veille/sommeil constituaient un marqueur de fragilité cérébrale pouvant être associé à l'apparition de troubles cognitifs chez des personnes âgées de la population AMImage. Nous avons d’abord montré que le désaccord existant entre les questionnaires de sommeil et les mesures objectives en actimétrie est dépendant de l’état émotionnel, renforçant ainsi l’intérêt d’utiliser une méthode de mesure objective du sommeil (Baillet et al., 2016). Puis, grâce à l’imagerie cérébrale, nous avons observé qu’une faible amplitude du cycle veille/sommeil est associée à une altération de la microstructure de la substance blanche, suggérant ainsi une origine vasculaire (Baillet et al., 2017). Nous avons également observé qu'une faible qualité de sommeil est associée à une charge amyloïde plus importante au sein des régions frontales, suggérant un déficit de clairance du peptide β-amyloïde lors du sommeil. Ainsi, les perturbations du sommeil et du cycle veille sommeil chez les personnes âgées seraient associées à deux processus physiopathologiques distincts aboutissant à une fragilité cérébrale. Considérés comme des facteurs de risque modifiables, des interventions visant à améliorer leur qualité offriraient une stratégie prometteuse afin de réduire le déclin cognitif au cours du vieillissement. / Sleep and activity/rest cycle disturbances represent risk factors for the development of cognitive decline and dementia in aging. However, the association between these disturbances and cerebral modifications during aging remains to be explored. The aim of this thesis was to determine if sleep and activity/rest cycle disturbances could constitute a cerebral frailty factor for the development of cognitive decline and dementia in older adults (AMImage cohort). First, we have shown that the discrepancy measured between self-reported sleep questionnaires and actigraphy - used as an objective technique - is influenced by the subject’s mood (Baillet et al., 2016). These results strengthen the use of an objective technique to measure sleep. Thanks to brain imaging, we observed that a reduced 24-h amplitude of the activity/rest cycle is associated with disruption of white matter structural integrity. Our results suggest that cerebral frailty associated with age-related activity/rest cycle dysfunction has a vascular origin (Baillet et al., 2017). Then, we observed that a poor sleep quality is associated with high cerebral amyloid burden, mainly circumscribed to frontal regions. This may be due to a deficit of amyloid-β peptide clearance from the brain during sleep. To conclude, sleep and activity/rest cycle disturbances in older adults may be associated with two separate physiopathological processes leading to cerebral frailty. As sleep and activity/rest cycle are modifiable risk factors, interventions to improve their quality could offer a potential useful strategy for reducing the burden of cognitive impairment and dementia in old age.

Made available in DSpace on 2014-12-17T15:37:17Z (GMT). No. of bitstreams: 1 CarolinaCA_DISSERT.pdf: 2842733 bytes, checksum: 3e54d35218351983a82f9dd10743303b (MD5) Previous issue date: 2013-04-12 / Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico / One of the main environmental cues for the adjustment of temporal organization of the animals is the light-dark cycle (LD), which undergoes changes in phase duration throughout the seasons. Photoperiod signaling by melatonin in mammals allows behavioral changes along the year, as in the activity-rest cycle, in mood states and in cognitive performance. The aim of this study was to investigate if common marmoset (Callithrix jacchus) exhibits behavioral changes under short and long photoperiods in a 24h cycle, assessing their individual behaviors, vocal repertoire, exploratory activity (EA), recognition memory (RM) and the circadian rhythm of locomotor activity (CRA). Eight adult marmosets were exposed to a light-dark cycle of 12:12; LD 08:16; LD 12:12 and LD 16:08, sequentially, for four weeks in each condition. Locomotor activity was recorded 24h/day by passive infrared motion detectors above the individual cages. A video camera system was programmed to record each animal, twice a week, on the first two light hours. From the videos, frequency of behaviors was registered as anxiety-like, grooming, alert, hanging position, staying in nest box and feeding using continuous focal animal sampling method. Simultaneously, the calls emitted in the experimental room were recorded by a single microphone centrally located and categorized as affiliative (whirr, chirp), contact (phee), long distance (loud shrill), agonistic (twitter) and alarm (tsik, seep, see). EA was assessed on the third hour after lights onset on the last week of each condition. In a first session, marmosets were exposed to one unfamiliar object during 15 min and 24h later, on the second session, a novel object was added to evaluate RM. Results showed that long days caused a decreased of amplitude and period variance of the CRA, but not short days. Short days decreased the total daily activity and active phase duration. On long days, active phase duration increased due to an advance of activity onset in relation to symmetric days. However, not all subjects started the activity earlier on long days. The activity offset was similar to symmetric days for the majority of marmosets. Results of EA showed that RM was not affected by short or long days, and that the marmosets exhibited a decreased in duration of EA on long days. Frequency and type of calls and frequency of anxiety-like behaviors, staying in nest box and grooming were lower on the first two light hours on long days. Considering the whole active phase of marmosets as we elucidate the results of vocalizations and behaviors, it is possible that these changes in the first two light hours are due to the shifting of temporal distribution of marmoset activities, since some animals did not advance the activity onset on long days. Consequently, the marmosets mean decreased because the sampling was not possible. In conclusion, marmosets synchronized the CRA to the tested photoperiods and as the phase angle varied a lot among marmosets it is suggested that they can use different strategies. Also, long days had an effect on activity-rest cycle and exploratory behaviors / Uma das principais pistas ambientais para o ajuste da organiza??o temporal dos animais ? o ciclo claro-escuro (CE), que sofre altera??o na dura??o das fases no decorrer das esta??es do ano. A sinaliza??o do fotoper?odo pelo horm?nio melatonina possibilita que os mam?feros se comportem de forma diferenciada ao longo do ano, como no ritmo de atividade-repouso, nos estados de humor e no desempenho cognitivo. Este estudo teve como objetivo avaliar se o sagui (Callithrix jacchus) exibe altera??es comportamentais em fotofases de diferentes dura??es em ciclo CE de 24h por meio da avalia??o dos repert?rios comportamental e vocal, da atividade explorat?ria (AE), da mem?ria de reconhecimento (MR) e do ritmo circadiano de atividade locomotora (RCA) em ambiente laboratorial. Oito saguis adultos, alojados individualmente, passaram 28 dias em cada etapa: dias sim?tricos (DS1 - CE 12h:12h), dias curtos (DC - CE 8h:16h), DS2 (CE 12h:12h) e dias longos (DL - CE 16h:8h). A atividade locomotora foi registrada continuamente por sensores de movimento acima das gaiolas. Um sistema de c?meras registrou os comportamentos individuais por 2 horas ap?s o in?cio do claro, duas vezes por semana. A partir das filmagens, foi amostrada a frequ?ncia dos comportamentos ansiosos, cata??o, alerta, pendurado, ficar na caixa ninho e comer pelo m?todo Animal Focal Cont?nuo. Ao mesmo tempo, um microfone registrou as vocaliza??es emitidas na sala de experimenta??o, categorizadas em afiliativa (whirr, chirp), contato (phee), longa dist?ncia (loud shrill), agon?stica (twitter) e alarme (tsik, seep, see). A AE foi avaliada na terceira hora ap?s o in?cio do claro em 2 sess?es de 15 min na ?ltima semana de cada ciclo. Na primeira apresenta??o, foi utilizado um objeto n?o familiar e 24 h depois, na segunda apresenta??o, foi adicionado um novo objeto para avaliar a MR. Os resultados mostraram que os DL diminu?ram a amplitude e a vari?ncia do per?odo do RCA, mas n?o os DC. Os DC causaram uma diminui??o do total di?rio de atividade e da dura??o da fase ativa. Nos DL, a dura??o da fase ativa aumentou devido a um avan?o no in?cio da atividade em rela??o aos dias sim?tricos. Contudo, nem todos os sujeitos iniciaram a atividade mais cedo nos dias longos. O hor?rio de t?rmino da atividade nos DL foi similar aos DS para a maioria dos sujeitos. Os resultados da AE mostrou que os DL e DC n?o afetaram a MR dos saguis e que os DL reduziram a dura??o da AE. Tamb?m foi observado nos DL uma redu??o da frequ?ncia e da diversidade das vocaliza??es, bem como da frequ?ncia dos comportamentos ansiosos, ficar na caixa ninho e cata??o nas duas horas ap?s in?cio do claro. Considerando toda a fase ativa na interpreta??o dos resultados dos comportamentos e vocaliza??es, ? poss?vel que essas altera??es nas duas horas iniciais tenham sido pelo deslocamento da distribui??o temporal da atividade do sagui e n?o um efeito genu?no dos DL, pois alguns animais n?o avan?aram o in?cio da atividade nos DL, causando a redu??o na m?dia do grupo por impossibilitar a amostragem. Em conclus?o, os saguis sincronizaram o RCA aos fotoper?odos e, pela varia??o individual nas rela??es de fase, parecem fazer uso de diferentes estrat?gias. Al?m disso, os DL afetaram o ciclo de atividade e repouso e as respostas explorat?rias