Sleep deprivation exhibits age-dependent effect on fMRI-based glymphatic measures

2346 

Abstract Submission 

Yibing Yang¹, Yutong Mao¹, Tianxin Mao², Weiwei Zhao², Yao Deng², Xiaocui Zhang², Yifan Yang¹, Hengyi Rao^2,3, Xiao Liu^1,4

¹Department of Biomedical Engineering, The Pennsylvania State University, University Park, PA, 16802, USA, ²Center for Functional Neuroimaging, Department of Neurology, University of Pennsylvania, Philadelphia, PA, USA, ³Chronobiology and Sleep Institute, Department of Psychiatry, University of Pennsylvania, Philadelphia, PA, USA, ⁴Institute for Computational and Data Sciences, The Pennsylvania State University, University Park, PA, 16802, USA

Yibing Yang  
Department of Biomedical Engineering, The Pennsylvania State University, University Park
PA, 16802, USA

Yutong Mao  
Department of Biomedical Engineering, The Pennsylvania State University, University Park
PA, 16802, USA

Tianxin Mao  
Center for Functional Neuroimaging, Department of Neurology, University of Pennsylvania
Philadelphia, PA, USA

Weiwei Zhao  
Center for Functional Neuroimaging, Department of Neurology, University of Pennsylvania
Philadelphia, PA, USA

Yao Deng  
Center for Functional Neuroimaging, Department of Neurology, University of Pennsylvania
Philadelphia, PA, USA

Xiaocui Zhang  
Center for Functional Neuroimaging, Department of Neurology, University of Pennsylvania
Philadelphia, PA, USA

Yifan Yang  
Department of Biomedical Engineering, The Pennsylvania State University, University Park
PA, 16802, USA

Hengyi Rao  
Center for Functional Neuroimaging, Department of Neurology, University of Pennsylvania|Chronobiology and Sleep Institute, Department of Psychiatry, University of Pennsylvania
Philadelphia, PA, USA|Philadelphia, PA, USA

Xiao Liu  
Department of Biomedical Engineering, The Pennsylvania State University, University Park|Institute for Computational and Data Sciences, The Pennsylvania State University, University Park
PA, 16802, USA|PA, 16802, USA

Sleep plays a crucial role in various brain functions and essential homeostatic processes, particularly in the clearance of toxic proteins out of the brain. The brain glymphatic system, a recently proposed pathway for brain waste clearance, exhibits heightened activity during sleep(1). Despite its significance, directly measuring glymphatic clearance in the human brain in vivo poses a considerable challenge. Recent research has unveiled a highly organized global brain activity at rest(2), manifested as prominent peaks in the global mean of resting-state fMRI (rsfMRI) BOLD signal (gBOLD), that is coupled to cerebrospinal fluid (CSF) movement and thus potentially linked to brain waste clearance(3). Consequently, the coupling strength of gBOLD and CSF has been proposed as a surrogate measure of brain glymphatic function. This gBOLD-CSF coupling index has been found to correlate with neurodegenerative disease pathologies, including amyloid-beta (Aβ) and tau accumulation(4–6). Given that sleep deprivation is known to impede waste clearance and increase Aβ and tau levels in the brain(7), it becomes intriguing to explore the impact of sleep deprivation on this fMRI-based glymphatic measure, particularly across different age groups with varying risk levels of Aβ and tau accumulation(8). Our study investigates these questions with rsfMRI data acquired in a rigorously controlled in-laboratory sleep deprivation study(9).

Seventy subjects were enrolled in a 5-day, 4-night experiment and three of them were excluded due to significant head motion, technical issues and/or missing data. Among the remaining 67 participants (aged: 34.1 ± 8.9 years old, 29 females), 51 of them underwent total sleep deprivation on Night 2. RsfMRI sessions and PVT task were conducted in the mornings of Days 2 (Scan1), 3 (Scan2), and 5 (Scan3) (Fig. 1A). Eye-tracking was employed to ensure that subjects stayed awake during all scans. The CSF signal was extracted from the bottom slices of fMRI data near the cerebellum base, as the previous study(2). The gBOLD was obtained by averaging fMRI signals within the entire gray matter regions (Fig. 1B). The gBOLD-CSF coupling was assessed at the negative peak (+4 seconds time lag) of their cross-correlation function (Fig. 1D), following the previous study(4).

Sleep deprivation had pronounced effects on the gBOLD-CSF coupling (Fig. 2A). Subjects after total sleep deprivation (N = 51) exhibited a large and significant increase (i.e., more negative) in the gBOLD-CSF coupling (p = 0.0071, paired t-test) in the Scan2 (Day3) right after the sleep deprivation, and then returned to baseline in Day5. In contrast, the gBOLD-CSF coupling remains stable and showed no significant changes (Scan2 vs. Scan1, p = 0.7823; Scan3 vs. Scan2, p = 0.8784, paired t-test) over the three scans in the control group.

The above analysis was stratified by age groups among the sleep-deprived subjects (younger, age 20-30, N1 = 20; middle-aged, age 30-40, N2 = 14; older, age 40-50, N3 = 17) (Fig. 2B). Both the younger and middle-aged groups displayed a significant (younger, p = 0.0189; middle-aged, p = 0.0085, paired t-test ) increase and then recovery (younger, p = 0.0132; middle-aged, p = 0.0037, paired t-test) of the gBOLD-CSF coupling, but this effect is much weaker and non-significant in the older group (Scan2 vs. Scan1, p = 0.8196; Scan3 vs. Scan2, p = 0.4410, paired t-test ). Consistent with this observation, the age-scan interaction effect was found significant (p = 0.0367, two-way ANOVA), suggesting distinct effects of sleep deprivation on this fMRI measure across different age groups.

Our results indicate a marked increase in gBOLD-CSF coupling following sleep deprivation, implying an enhanced clearance as a compensatory response to the lack of sleep. Yet, this compensatory mechanism seemed substantially diminished in participants of age 40-50, which might be a factor contributing to the age-related decline of brain waste clearance(8).

BOLD fMRI ¹

Sleep and Wakefulness ²

Cerebro Spinal Fluid (CSF)

Sleep

Other - glymphatic function, sleep deprivation, global brain activity