Cerebrospinal fluid levels of orexin-A and histamine, and sleep profile within the Alzheimer process

Abstract

To better understand how sleep/wake dysregulation affects Alzheimer's disease (AD), we compared the cerebrospinal fluid (CSF) orexin and histamine/tele-methylhistamine (HA/t-MHA) levels of 82 patients (41 probable-AD-high level of evidence, 41 mild cognitive impairment MCI-due-to-AD), 24 other neurologic disorders (OND) and 24 controls. We determined the relationships between these biomarkers, the CSF AD biomarkers concentrations, and the clinical sleep profile. CSF orexin-A but not HA/t-MHA levels were higher in MCI and AD than OND and controls. CSF orexin-A is correlated to CSF amyloid-β₄₂in MCI and AD, independently of age, gender, MMSE, total-tau/phosphorylated-tau, HA or sleep parameters. Nighttime sleep duration was longer in MCI and AD patients than controls. In MCI, nighttime sleep duration negatively correlated with CSF amyloid-β₄₂ and MMSE. To conclude, CSF orexin-A but not HA/t-HMA was upregulated in AD and correlated with amyloid-β₄₂ level. Our data suggested a change in the sleep-wake pattern at an early stage of the disease that needs further investigation to deeply explain the mechanistic interplay between sleep and Alzheimer.

Introduction

Alzheimer's disease (AD) is a multifactorial and complex illness that seems to progress for many years before the first clinical symptoms (Bateman et al., 2012, Braak et al., 1998). Recent growing evidences suggest that changes in the sleep-wake cycle play a role in AD pathophysiology (Ju et al., 2014; Mander et al., 2016). Daytime sleepiness, nighttime awakenings, and sleep fragmentation have been observed in patients with AD, even at an early disease stage (Dauvilliers, 2007, Ju et al., 2014). A correlation between changes in sleep quantity and quality and amyloid β (Aβ) brain burden has been found in community dwelling older adults without apparent cognitive troubles (Spira et al., 2013). Specifically, shorter sleep duration (≤6 hours) was associated with greater cortical and precuneus Aβ deposition, measured using Pittsburgh compound B positron emission tomography. On the other hand, poor sleep quality was linked only to precuneus Aβ deposition (Spira et al., 2013). Cognitively healthy elderly subjects with daytime sleepiness have greater Aβ burden in the precuneus, angular and cingulate gyrus, and frontal medial orbital cortex (Sprecher et al., 2015). Similarly, in cognitively normal subjects at risk of AD, Aβ deposition (based on a cerebrospinal fluid [CSF] Aβ₄₂ level ≤500 pg/mL) was associated with poor sleep efficiency and frequent napping (Ju et al., 2013).

Among the many neurotransmitters involved in the regulation of the sleep/wake cycle, orexin-A/hypocretin-1, which promotes wakefulness but also energy homeostasis, stress adaptation, and reward behaviors (Espana and Scammell, 2011), may have a direct link with the amyloid (Dauvilliers et al., 2014, Ju et al., 2014, Kang et al., 2009, Slats et al., 2013) and tau pathways (Deuschle et al., 2014, Liguori et al., 2014). The influence of orexin-A signaling on Aβ metabolism in animals and humans was recently highlighted (Bateman et al., 2007, Dauvilliers et al., 2014, Kang et al., 2009, Liguori et al., 2014, Slats et al., 2013). In rats, orexin-A release shows a 24-h fluctuation similar to that of brain interstitial fluid Aβ (Yoshida et al., 2001). In transgenic mice that overexpress amyloid precursor protein (APP), brain ISF Aβ concentration increases during wakefulness and after orexin-A infusion. Conversely, it decreases during sleep and after infusion of an orexin-A receptor antagonist (Kang et al., 2009). In transgenic mice that overexpress APP/presenilin1 (PS1), in which the orexin gene is knocked out, a marked decrease in the amount of Aβ pathology in the brain was found together with an increase in sleep time (Roh et al., 2014). In addition, sleep deprivation by rescue of orexinergic neurons in these mice increased the amount of brain Aβ pathology (Roh et al., 2014). Conversely, in humans, results are more conflicting, especially concerning CSF orexin-A levels in AD patients. In a postmortem analysis, the number of orexin-positive neurons in the hypothalamus and the concentration of orexin in ventricular CSF were reduced in patients with AD compared with controls (Fronczek et al., 2012). Another study showed a gender-dependence of CSF orexin-A levels with elevations in females compared to males, without any difference between AD and controls (Schmidt et al., 2013). Other authors found higher CSF orexin-A levels in patients with AD than that in controls (Dauvilliers et al., 2014, Liguori et al., 2014, Wennstrom et al., 2012), possibly related to sleep deterioration and neurodegeneration (Liguori et al., 2014). Thus, it is not clear whether the observed changes in orexin-A levels are linked to the underlying neurodegenerative process (Aβ deposition, tau protein) or secondary to sleep/wake cycle alterations. Besides orexin-A, other wakefulness-promoting systems may be involved, particularly the histamine (HA) pathway. One recent study found neurofibrillary degeneration in the tuberomammillary nucleus, which consists mainly of histaminergic neurons, and a slight decrease of CSF tele-methylhistamine (t-MHA, the major HA metabolite) at the latest stages of AD (Motawaj et al., 2010). However, the activity of the histaminergic system and its involvement in sleep/wake cycle alterations remain largely unknown especially, at early AD stages.

Therefore, the aims of this study were (1) to assess, the CSF levels of wakefulness-promoting biomarkers (orexin, HA, and t-MHA) and sleep profile based on clinical interview and sleep-validated questionnaires in a population of cognitively impaired patients (AD, mild cognitive impairment [MCI] due to AD based on the NIA diagnosis criteria guidelines and other neurological disorders [OND]) and in controls, and (2) to investigate the correlations between CSF AD biomarkers, wakefulness-promoting biomarkers, and sleep profile at different AD stages.