Sleep-disordered breathing (SDB) includes a wide range of conditions associated with respiratory disorders. According to the International Classification of Sleep Disorders, SDB is divided into obstructive sleep apnea (OSA), central sleep apnea (CSA), and sleep-associated hypoventilation and hypoxemia [1]. In addition to symptoms of sleep apnea and hypopnea, clinical signs of SDB include difficulty maintaining sleep, nocturia, morning hypertension, and excessive daytime sleepiness [2].
SDB is one of the most common types of sleep disorders. A. Benjafield et al. [3] conducted a meta-analysis of OSA prevalence in different countries and showed that OSA with an apnea-hypopnea index (AHI) of 5/h occurs in approximately 1 billion people aged 36–69 years, 45% of whom have a moderate OSA with an IAH of 15/h. In Russia, according to the Epidemiology of Cardiovascular Diseases and their Risk Factors in Regions of Russian Federation (ESSE-RF) population-based study, symptoms of snoring and sleep apnea were reported by 54 % and 9 % of the general population, respectively [4].
SDB is a risk factor for cardiovascular diseases and metabolic disorders, including arterial hypertension, heart failure, impaired glucose tolerance, myocardial infarction, and stroke [5]. SDB is also associated with brain damage, which may contribute to the development of cognitive dysfunction and neurodegenerative diseases [6], but the exact mechanisms of brain damage in SDB are not fully understood.
Gliocytes (astrocytes, microglial cells, oligodendrocytes, and NG2-glia) are thought to play an important role in brain damage in SDB. Thus, this review aims to address the role of glia in the pathogenesis of brain tissue damage in SDB.
Brain damage in SDB
In many clinical studies, brain damage has been reported in SDB and includes both structural and functional disorders.
In a recently published meta-analysis by X. Huang et al. [7] reviewed the results of 16 cross-sectional clinical studies comparing OSA patients versus controls. In OAS patients, the authors observed a bilateral decrease in grey matter volume of the orbitofrontal cortex, blunting of its functional response, and bilateral decrease in grey matter volume of the anterior cingulate cortex, hippocampus/parahippocampus, and left cerebellar hemisphere. In addition, the authors reported functional hypoactivation of the dorsolateral prefrontal cortex and hyperactivation of the insular cortex in patients with OSA compared to controls.
SDB can also be associated with changes in white matter integrity, including decreased fractional anisotropy and mean diffusivity of the superior longitudinal fasciculus fibers bilaterally, fractional anisotropy of the right arcuate fasciculus, the left uncinate fasciculus, and functional interrelationships of the posterior cingulate cortex with the left caudate nucleus and the left thalamus [8, 9].
Cognitive impairment in SDB is addressed in a recently published review by M. Olaithe et al. [10] and is characterized by disorders in almost all cognitive functions, including attention, memory, and executive functions. Cognitive impairment, defined as worse cognitive test results compared to reference values, occurs in 36–78% of patients with OSA compared to 2% of the gender- and age-matched control group without OSA [11, 12].
Of particular value are clinical studies describing the relationship between structural and functional abnormalities in patients with SDB. Some of the structural change zones observed in these studies and the associated functional impairments are shown in Figure. In a controlled study, N. Canessa et al. [13] reported a deficit in short- and long-term memory, attention, and executive functions in 17 patients with SDB versus 15 subjects in the control group matched by gender and age. The authors found a correlation between improvements in Digit-Span Forward (verbal short-term memory), Corsi (visual short-term memory), and Stroop (executive functions) with CPAP therapy (continuous positive airway pressure) with an increase in the gray matter volume of the left anterior parahippocampal gyrus (entorhinal cortex) [13]. Similarly, V. Castronovo et al. described the relationship between attention and executive function characteristics (Stroop tests, PASAT) with MRI features of the thalamus, caudate nucleus, and left corpus callosum; attention (Trial making A and B test) with MRI features of the arcuate fasciculus and superior and inferior longitudinal fasciculus on the right; short-term memory (Digit forward and Corsi tests) with MRI features of the left caudate nucleus, and long-term memory with MRI features of arcuate fasciculi [8]. In addition, Y. Xiong et al. [14] described the correlation of psychomotor vigilance test results with fractional anisotropy of the capsula interna on the left and mean diffusivity of the sagittal striatum on the left.
Brain damage in SDB. Structural brain changes and the related cognitive deficits.
These cognitive impairments may be due to excessive daytime sleepiness in SDB [13] and structural brain abnormalities in patients with OSA, indicating the need for early diagnosis and treatment of this condition. In several studies included in a meta-analysis by Y. Xia et al. [15], in patients with OSA, MR spectroscopy showed a change in the ratio of the following brain metabolites, especially in the hippocampus: choline, creatine, N-acetyl aspartate, glutamate, myo-inositol, which are also markers of glia state and functioning. According to the authors’ suggestion, an imbalance in the ratio of these metabolites in some brain regions (frontal, temporal, occipital lobes, thalamus, hippocampus, etc.) may correlate with some cognitive disorders.
Other studies in patients with OSA, using positron emission tomography of the brain with various radiopharmaceuticals, showed a more pronounced deposition of amyloid (in the posterior cingulate and temporal cortex) [16, 17] and tau-protein (in the inferior temporal lobe cortex and entorhinal area) [18], which may be associated with a higher risk of neurodegenerative diseases and dementia in this patient population.
However, it should be noted that most studies using neuroimaging techniques included relatively small samples of patients and showed relatively high heterogeneity of results.
The role of glia in the pathogenesis of brain damage in SDB
The described structural and cognitive disorders may be multifactorial due to the complex pathogenesis of SDB. A key element in the SDB pathogenesis is intermittent hypoxia due to episodes of recurrent airway collapse in OSA or decreased respiratory center sensitivity to hypoxia in CSA [19]. Chronic intermittent hypoxia causes chronic intermittent hypoxemia and hypercapnia, leading to the “pathogenetic triad” in SDB, including the closely interrelated oxidative stress, activation of the inflammatory response, and sympathetic activation. Sympathetic activation is also associated with sleep fragmentation, which has an adverse impact on neural tissue metabolism and exacerbates brain damage in SDB [20].
Effect of SDB therapy on reducing brain damage in SDB through glia function modulation
The standard of care for SDB is noninvasive ventilation, in particular CPAP therapy, which is continuous positive pressure generation in the upper airways, preventing their collapse [21].
There is evidence of a possible reduction in SDB-induced brain damage with CPAP therapy. Short-term CPAP therapy has a limited effect on brain damage in patients with SDB [13]. However, V. Castronovo et al. [8] reported white matter disruption in patients with severe SDB, which completely resolved after 12 months of CPAP therapy in compliant patients. In patients with OSA, C. Tonon et al. [22] showed a decrease in N-acetyl aspartate mediating communication between neurons and glia even after 6 months of CPAP therapy, which may indicate the limited CPAP effect on glia function in SDB.
Along with surgical treatments for SDB and lifestyle modification approaches, pharmacological therapy using antioxidants or apoptosis modulators is currently being studied to reduce SDB-induced brain damage. Many of these experimental studies suggest beneficial effects of glia function modulating agents on the outcome of brain damage in SDB (see Table).
Experimental studies which applied glia modulation approaches
|
Author |
Intermittent hypoxia model |
Damage assessment |
Agent |
Dosing regimen |
Mechanism of action |
Glia type |
|
Dong et al., 2018 [23]. |
Rats, 30 cycles/h (decrease in oxygen level from 21% to 5% in 50 sec, with rapid recovery to 21% in 40 sec) for 8 h, 4 weeks |
Assessment of spatial learning and memory using the Morris water maze, TNF-α and IL-1β levels in the hippocampus, and microglia activation |
Sevoflurane 2.6% |
Sevoflurane 2.6% with 30% humidified oxygen using a calibrated vaporizer for 4 h |
Anti-inflammatory: control of peroxisome proliferator-activated receptor gamma (PPARγ) expression and activity in the hippocampus by sevoflurane |
Microglia |
|
Lam et al., 2015 [24] |
Rats, decrease in oxygen concentration for 8 h from 21% to 5±0.5% every minute |
Cell death |
Boxthorn polysaccharide (goji berry, LBP) |
Oral administration of LBP solution (1 mg/kg) daily 2 h before exposure |
Anti-inflammatory (reduction of NF-kB signaling pathway activation) and antioxidant (reduction of malondialdehyde levels), reduction of endoplasmic reticulum stress, autophagy, external and internal caspase-dependent hypoxia-induced apoptosis in the hippocampus |
Microglia |
|
Yin et al., 2015 [25] |
Rats, reducing oxygen concentration to 6.5–7% after 25–30 sec by injecting 99.99% nitrogen for 30–35 sec. Then, with inhalation of 99.50% oxygen for the next 10 sec, increasing the O2 concentration to 21% for 50 sec |
Cell death |
Protocatechuic acid |
Intraperitoneal injection of protocatechuic acid at a daily dose of 15 mg/kg for 7 days and throughout the hypoxia exposure period |
Antioxidant: reduction of oxidative stress, apoptosis, glial proliferation, and IL-1β levels in the brain and increased expression of BDNF and SYN |
Microglia |
|
Al-Qahtani et al., 2014 [26] |
Rats, decreasing oxygen concentration to 5.7% and increasing to 21% every 90 sec for 12 h in the daytime |
Cell death |
Erythropoietin |
Intraperitoneal injection of erythropoietin 500 and 1000 IU/kg/day for 6 weeks |
Antioxidant: dose-dependent inhibition of hypoxia-induced increase in glial fibrillary acidic protein (GFAP), FAS receptor and caspase-3 levels in the hippocampus |
Astroglia |
|
Wang et al., 2018 [27]. |
Mice, hypoxia 8 h per day: cycles of 180 sec each: reduction of oxygen level to 5% for 50 sec, reoxygenation (21% O2 for 50 sec); for 10 weeks |
Cell death |
Curcumin |
Administration of curcumin 50, 100 or 200 mg/kg by oral gavage |
Anti-inflammatory: inhibition of aquaporin 4 and p38 mitogen-activated protein kinase |
Astroglia |
|
Deng et al., 2015 [28] |
Mice, 8 h from 9:00 a.m. to 5:00 p.m. decrease in oxygen levels from 21±1% to 6±1% every 90 sec |
Cell death |
Atorvastatin |
Atorvastatin, 5 mg/kg/day in 10% ethanol diluted in tap water, 200 µl |
Anti-inflammatory: reduction of nerve damage and decreased levels of TLR4, MyD88, TRIF, pro-inflammatory cytokines |
Astroglia and microglia |
|
Gong et al., 2020 [29]. |
Rats: reduction of oxygen levels from 21% to 5% every 60 sec for 8 hours from 8:00 a.m. to 4:00 p.m. |
Cell death |
Pinocembrin |
Intraperitoneal injection of 40 mg/kg pinocembrin every 2 days at 8:00 from day 1 to day 21 during the experiment |
Anti-inflammatory: inhibition of mitochondrial damage and cell apoptosis through functional activation of BNIP3 and blocking NLRP3-mediated inflammation |
Microglia |
|
Deng et al., 2015 [30] |
Rats: reduction of oxygen levels from 21% to 6% every 90 sec for 8 hours from 9:00 a.m. to 5:00 p.m. |
Cell death |
Selective P2X7R receptor antagonist Brilliant Blue G |
Intraperitoneal injection (50 mg/kg) of the drug diluted with phosphate-buffered saline 3 days before and during 4 weeks of hypoxia exposure |
Anti-inflammatory: inhibition of NF-kB transcription factor signaling pathway and NOX2 protein activation |
Microglia |
|
Liu et al., 2017 [31] |
In vitro: BV2 microglia cell culture exposed to hypoxia by cycles of oxygen level changes from 1% to 21% (400 sec/cycle) for 8 h |
Cell death |
Propofol |
Exposure to propofol at concentrations of 25, 50 and 100 μM 0.5 h before induction of intermittent hypoxia |
Anti-inflammatory: inhibition of NF-kB/p38 MAPK signaling pathway activation |
Microglia |
|
Burckhardt et al., 2008 [32] |
Rats, cycles of 240 sec, including 90 sec of 10% oxygen for 12 h |
Cell death |
Green tea catechins |
Oral administration of green tea catechin: 60% extract in potable water at a concentration of 0.05% for 3 days before and during exposure |
Anti-inflammatory: reduction of the hypoxia-induced increase in NADPH-oxidase gene expression, increase in GFAP expression in the large hemisphere cortex in the rat |
Microglia |
|
Yuan et al. 2015 [33] |
Rats, 8 weeks for 8 h cycles 10 min each: oxygen drop from 21% to 8% for 120 sec, hold for 120 sec, return to 21% oxygen in 50 sec, hold for 300 sec |
Cell death |
Telmisartan |
Oral administration of telmisartan (10 mg/kg) dissolved in double distilled water, once daily for 8 weeks before hypoxia exposure |
Antioxidant: reduction of apoptosis in the hippocampus due to regulation of nitric oxide synthase activity and inhibition of excess nitric oxide production, as well as reduction of lipid peroxidation and inflammatory reactions |
Astroglia and microglia |
Also, the direct glia function modulation effect of many agents in neuroprotection under hypoxia has not been shown. For instance, erythropoietin can reduce anxiety and spatial learning deficit in the model of intermittent hypoxia in mice by reducing the expression of NADPH oxidase and malondialdehyde 8-OHDG [26], suggesting the involvement of glia in erythropoietin’s protective effect. Erythropoietin expression increase during intermittent hypoxia in neurons and astrocytes has been shown [34]. Similarly, inhibition of HIF-1α with consistently high levels observed in intermittent hypoxia [35] also affects synaptic plasticity and cognitive deficits in SDB in rats [36]. Moreover, in an experimental model of intermittent hypoxia, lovastatin’s protective effects on brain tissue have been reported [37]. Lovastatin is known to reduce iNOS expression in astrocytes and nitric oxide production [38], which may contribute to the protective effect in intermittent hypoxia.
Conclusion
Thus, the available results of clinical studies confirm the experimental data on the significant role of glia in the development of brain damage in SDB. Considering that standard SDB therapies are associated with some challenges (low compliance and high intolerance rate of noninvasive ventilation and intraoral devices, their unavailability in many regions, etc.), alternative approaches aimed at reducing and preventing SDB-associated brain damage are of current interest. In this regard, pharmacological modulation of glia (microglia, astroglia, oligodendroglia, and other components can be used as a target) appears to be one of the promising directions, including its use in combination therapy.
This study was supported by the Russian Foundation for Basic Research grant 20-115-50512.
The authors declare no conflicts of interest.