Sleep restriction impairs blood-brain barrier function

Abstract

The blood-brain barrier (BBB) is a large regulatory and exchange interface between the brain and peripheral circulation. We propose that changes of the BBB contribute to many pathophysiological processes in the brain of subjects with chronic sleep restriction (CSR). To achieve CSR that mimics a common pattern of human sleep loss, we quantified a new procedure of sleep disruption in mice by a week of consecutive sleep recording. We then tested the hypothesis that CSR compromises microvascular function. CSR not only diminished endothelial and inducible nitric oxide synthase, endothelin1, and glucose transporter expression in cerebral microvessels of the BBB, but it also decreased 2-deoxy-glucose uptake by the brain. The expression of several tight junction proteins also was decreased, whereas the level of cyclooxygenase-2 increased. This coincided with an increase of paracellular permeability of the BBB to the small tracers sodium fluorescein and biotin. CSR for 6 d was sufficient to impair BBB structure and function, although the increase of paracellular permeability returned to baseline after 24 h of recovery sleep. This merits attention not only in neuroscience research but also in public health policy and clinical practice.

Keywords: BBB; glucose metabolism; sleep duration; sleep restriction; transport.

Copyright © 2014 the authors 0270-6474/14/3414697-10$15.00/0.

Figures

Figure 1.

Effect of CSR on sleep architecture. A, The time within a 24 h light/dark cycle that mice spent in each stage is shown across the days of CSR. There was an overall increase of wake and reduction of both NREM and REM in comparison with their respective baseline on day 0. However, NREM rebound was seen after 2 d with a corresponding decrease of wake time. B, The number of REM bouts was decreased across day 1–6 of CSR, whereas wake and NREM bouts did not show significant changes in comparison with day 0 baseline. C, The duration of NREM bouts decreased on days 3–6 of CSR in comparison with day 0. The duration of wake and REM bouts did not change; *p < 0.05, **p < 0.01, ***p < 0.005, compared with baseline 24 h recording before CSR.

Figure 2.

A, EEG and EMG tracings of three typical sleep states and their transitions, shown in 5 min time spans for baseline recording (top) and after CSR (bottom). B, Hypnograms of a representative mouse across days of CSR, where increased transitions from NREM to wake are seen.

Figure 3.

Effect of CSR on GLUT1 expression in enriched cerebral microvessels. A, The mRNA for GLUT1 was decreased by CSR (n = 7) in comparison with the control (n = 6). B, Enriched microvessels had a high level of 45 kDa GLUT1 isoform and less abundant 55 kDa GLUT1 isoform. CSR reduced the level of expression of both after 6 d. C, The significant reduction of both GLUT1 isoforms in WB is shown by densitometric analysis (n = 5 for control and n = 6 for CSR); *p < 0.05.

Figure 4.

Effect of CSR on glucose transport across the BBB. A, Under anesthetized conditions, compared with the control mice, the uptake of 3H-2DG 45 min after intraperitoneal injection was reduced in the cerebral cortex and subcortical regions, although significant changes were not detected in the brainstem and cerebellum in the CSR mice (n = 4/group). B, Without any anesthesia, compared with the control mice, the uptake of 3H-2DG 45 min after intraperitoneal injection was reduced in most of the CNS region of CSR mice, but not the sleep recovery mice; *p < 0.05.

Figure 5.

Effect of CSR on genes related to vascular function. A, In comparison with their respective controls (n = 6/group), enriched cerebral microvessels from the CSR mice had reduced mRNA for eNOS, iNOS, and ET-1, but increased mRNA for COX-2; *p < 0.05, **p < 0.01, ***p < 0.005. B, ICC of fresh cortical microvessels from cerebral cortex. The immunofluorescence of COX-2 was increased in the CSR group in comparison with the naive control or the recovery mice. Scale bar, 5 μm.

Figure 6.

Effect of CSR on TJ protein expression. A, Several TJ proteins showed decreased mRNA expression in the CSR group, including occludin, claudin-1, claudin-5, and ZO-2; **p < 0.01, ***p < 0.005. B, The immunofluorescence of occludin in freshly isolated cerebral microvessels showed line-like paracellular distribution in the naive group, was decreased in the CSR group, and returned to the control level after 24 h of recovery sleep. Scale bar, 5 μm.

Figure 7.

Effect of CSR on BBB permeability. A, Sodium fluorescein uptake 10 min after intravenous injection in anesthetized mice was increased in cerebral cortex and brainstem in the CSR group (n = 5) in comparison with the naive control (n = 7). B, In the brain, there was regional variation of the distribution of biotin signal (shown by AlexaFluor 488-streptoavidin staining). Coronal section across the thalamus is shown for a naive mouse (top). CSR was associated with an increase of background fluorescence, and somewhat increased diameter of microvessels suggestive of increased capillary filling resulting from hyperemia (bottom). C, In the brainstem, there was also regional variation of biotin distribution in horizontal section across upper medulla at the level of the caudal fourth ventricle. Although the control section shows variation of biotin distribution in different areas, functional hyperemia is most apparent in the medial vestibular nucleus, and dye extravasation is seen in the parapyramidal nucleus in the ventral brainstem (arrows). D, Sodium fluorescein uptake 45 min after intraperitoneal injection in unanaesthetized mice was increased in all brain regions and spinal cord in the CSR mice (n = 7/group) compared with the naive control (n = 8) or those after 24 h of recovery sleep following 6 d of CSR (n = 5); *p < 0.05, **p < 0.01, ***p < 0.005. Scale bar, 100 μm.

Figure 8.

Blood concentrations of CRP (A) and leptin (B). The CSR group showed an increase of serum CRP (*p < 0.05 from control), whereas the 24 h recovery group showed an increase of serum leptin from the control when comparison was made by unpaired t test.