Abstract
Impairment of memory is one of the most frequently reported symptoms during sudden hypoxia exposure in human. Cortical atrophy has been linked to the impaired memory function and is suggested to occur with chronic high-altitude exposure. However, the precise molecular mechanism(s) of hypoxia-induced memory impairment remains an enigma. In this work, we review hypoxia-induced learning and memory deficit in human and rat studies. Based on data from rat studies using different protocols of continuous hypoxia, we try to elicit potential mechanisms of hypobaric hypoxia–induced memory deficit.
Introduction
Hypoxia is defined as inadequate oxygen supply to the cells and tissues of the body. In this review, the term ‘hypoxia’ refers to ‘hypoxic hypoxia’, which is also the most common type of hypoxia and is due to low arterial oxygen concentration or hypoxemia. Low arterial oxygen concentration is first detected by sensory receptors in the carotid or aortic body and impulses are sent to cardiorespiratory centre in the medulla. 1 The body responds to normalize the decrease in the oxygen arterial concentration by activation of the sympathetic nervous system which leads to increased heart rate, blood pressure, ventilation and the production of stress hormones. 2 –4 However, in the case of very low arterial oxygen concentration, the body will fail to normalize.
Hypoxia has long been known to diminish brain function in humans and animals. 5 –7 In humans, the degree of hypoxia required to impair performance, and the task most sensitive to it, is controversial. 7 It has been reported that the earliest cognitive signs and symptoms of hypoxia are typically experienced around 15–16% of oxygen concentration, 8 and complex information processing and learning are among the most sensitive processes to be affected. 9
Impairment of memory is one of the most frequently reported symptoms during sudden hypoxia exposure, for instance, during hypoxia-awareness training of aircrews or after an in-flight hypoxic incident. 10 However, the effects of acute hypoxia on memory have only been sporadically studied in laboratory-controlled conditions, mainly for military aviation 10 and even less for civilian aviation. With more severe hypoxia, critical judgement declines, and this proceeds to stupor, coma and, finally, death if the hypoxia is not reversed. 11
Materials and methods
All articles indexed to MEDLINE and books were searched using the following key words: hypoxia, hypobaric, normobaric, learning and memory. This review mainly aims to look into how different protocols of continuous hypoxia induced by hypobaric/normobaric hypoxic chamber affect the learning and memory of adult rats. Results from brain hypoxia induced by invasive procedures such as ligation of arterial supply to the brain 12 or by chemical induction such as carbon monoxide hypoxia 13 are not included. The effects of different hypoxia protocols on learning and memory in rats are summarized in Table 1.
Comparison of hypoxic protocols and major cognitive findings from rat studies.a
MWM: Morris water maze; RAM: radial arm maze; NORT: novel object recognition test.
aThe duration of hypoxic exposure was continuous for the stipulated period apart from a 10- to 30-min interval each day for replenishment of food and water, drug administration and changing the cages housing the animals except for entries marked with superscript alphabet ‘b’.
Hypoxia and memory impairment in human
Studies on the effects of hypoxia in life sciences are generally performed either by decreasing barometric pressure (PB), leading to hypobaric hypoxia as shown by Bert, 39 or by decreasing oxygen fraction (FIO2) without changing PB as shown by Barcroft 40 in his ‘Glass House’ experiment (normobaric hypoxia (NH)). It has been shown that exposure to hypobaric hypoxia (FIO2 < 20.9%; PB < 760 mmHg) and NH (FIO2 < 20.9%; PB = 760 mmHg) may or may not result in different physiological responses in human subjects. 41 –43
In a study by Shukitt-Hale et al., 44 23 non-acclimatized males were exposed to 500, 4200 or 4700 m in an altitude chamber for 4–5 h. The study demonstrated that the higher the altitude and the longer the duration of exposure, the more severe mood and memory impairment. A similar study was later conducted by Du et al. 45 involving 18 healthy young male subjects exposed to 300 m (control group), 2800, 3600 and 4400 m altitude in hypobaric chamber. Compared to control group, after exposure to 2800 m for 1 h, only the performance of continuous recognition memory decreased significantly. After exposure to 3600 m, total reaction time in all tests increased significantly and performance decreased, but the error rates in memory scanning and space memory test were unchanged. During exposure to 4400 m, performance of memory test decreased further and error rates also increased significantly. It was concluded that the performance of human short-term memory deteriorates after exposure to acute mild and moderate hypobaric hypoxia for 1 h, and these effects are aggravated with the increase in altitude as seen in another study. 46 In a study by Malle et al. 10 involving a bigger number of participants, 28 subjects in the experimental group were exposed to a simulated altitude level of 10,000 m (31,000 ft) in a hypobaric chamber, while 29 subjects in the control group stayed at sea level. The short-term (working) memory assessed using paced auditory serial addition test (PASAT) showed that performance was strongly impaired in the hypoxic group with increased mean error frequency rate. While working memory performance decreased linearly with hypoxemia, peripheral oxygen saturation (SpO2) was found to be a weak predictor of PASAT performance and vice versa.
Structural brain alterations following exposure to high altitude may include brain swelling, 47 increase in the number of white matter hyperintensities 48 and grey and white matter atrophy, 49 –51 contributing to a reduction in total brain volume. Hemosiderin deposits (microhaemorrhages) have also been reported in subjects who have experienced high-altitude cerebral edema. 52 Cortical atrophy is linked to impaired cognitive function and has been suggested to occur with chronic high-altitude exposure. 53 However, human studies shed little light on the precise molecular mechanism(s) of hypoxia-induced memory impairment.
Animal studies related to hypoxia
It has been established that the brain is highly sensitive to hypoxia and that some areas, such as the hippocampus, are especially vulnerable to hypoxic damage. 54 Compared to human brain, the brain of smaller animal (such as rat) is more resistant to hypoxia because of its higher capillary density, that is, much more severe hypoxic condition must be applied to overtly damage its brain tissues relative to human. 17 Male rats are more susceptible to hypoxia compared to female rats; hence, most hypobaric hypoxia studies were conducted on male rats to obtain better and more conclusive results. 55–56
Depending on the objectives of the study, smaller animals are either exposed to continuous or intermittent hypoxia. Continuous hypoxia with or without low PB rat model is commonly used to study the effect of acute mountain sickness, diseases associated with limited oxygen supply to the brain such as chronic obstructive pulmonary disease and acute respiratory distress syndrome, medication and drugs, concussion or changes in air quality, for example, in a nuclear power plant. Intermittent hypoxia rat model, on the other hand, is commonly used to study the effects of obstructive sleep apnoea. Furthermore, understanding how hypoxia alters brain function has implications for understanding other metabolic encephalopathies as well as aging and age-related disorders, such as Alzheimer’s disease. 57
Hypoxia and memory impairment in rats
Groo et al. 15 used spontaneously hypertensive rats (SHRs) to study avoidance response to hypoxia as they were reported to be more sensitive to stressful stimuli 58,59 and showed greater hypoxic vulnerability than normotensive Wistar rats. The SHRs were exposed to either NH (6% oxygen) or normoxic conditions. Lowering the oxygen contents of inspired air to 6% impaired acquisition of the avoidance response, and the difference between the performance (in percentage of conditioned avoidance response) of animals kept under normoxic and hypoxic conditions was significant on day 3 (at 69.2% and 38.0%, respectively). Hypoxia was also found to increase the incidence of escape failures. These findings are in agreement with earlier study findings by Saligaut et al., 14 where the acquisition of a conditioned avoidance response was also impaired in hypobaric hypoxia at 300 Torr (7180 m; equivalent to 8% oxygen content).
Hippocampal damage
Direct correlation between working memory impairment and level of altitude was evidenced in a study involving Fischer male rats exposed to various altitudes equivalent to sea levels 5500, 5950 and 6400 m, for 2 and 6 h. 16 In a later study by the same group of researchers, two groups of rats were exposed to 6400 m altitude for either 72 or 144 h. It was found that the longer the time of exposure, the more noticeable the hippocampal damage (delayed neurotoxicity) as shown by 78% neuronal damage after 144 h compared to 50% after 72 h of exposure. 53 Titus et al. 19 showed that hippocampal-dependent spatial learning in rats was affected marginally following 2 days of exposure to simulated hypobaric hypoxia at 6000 m, while 7 days of exposure severely affected learning of partially baited radial-arm maze (RAM) task. The study also found that exposure for 2 days to hypobaric hypoxia resulted in minimal deleterious effects on the CA1 pyramidal neurons, while exposure for 7 days caused a significant decrease in the number of branching points, intersections and dendritic length. Unlike the CA1 pyramidal neurons, CA3 neurons exhibited dendritic atrophy following both 2 and 7 days of hypobaric hypoxia exposure. Thus, CA3 neurons are more vulnerable to hypoxic insult compared to CA1 neurons. 23,60,61 Findings by Titus et al. are in agreement with the observation of neuronal degeneration in rat hippocampus exposed to hypobaric hypoxia in previous studies 23,53 as well as a later study by Prasad et al. 33
Maiti et al. 24 investigated spatial memory functions and dendritic changes in CA1, CA3 and entorhinal cortex of hippocampus, and layer II of prefrontal cortex (PFC) in rats exposed to simulated hypobaric hypoxia at 6100 m, but with different durations of hypobaric hypoxia exposure, that is, 3, 7, 14 and 21 days. There was impairment of spatial memory after 3 and 7 days, but slight improvement of spatial memory was noted after 14 and 21 days of exposure. The study suggested that hypobaric hypoxia induces dendritic plasticity of PFC and hippocampal pyramidal neurons of rat brain, which may be associated with improvement of spatial memory function after 21 days of hypobaric hypoxia exposure. These findings are in contrast with a recently published data on the adverse effects of hypobaric hypoxia on the brain which became more severe after 4 weeks compared to 2 days of exposure. The total brain weight and oxidative stress in all three brain regions (striatum, hippocampus and cortex) were significantly increased after 4 weeks compared to after 2 days of hypobaric hypoxia. The study, however, did not assess cognitive function. 62 Further studies are necessary to attest to the possibility of compensatory stage after 2–3 weeks of hypobaric hypoxia exposure followed by a decompensated stage after 4 weeks.
Brain oxidative stress
Liu et al. 63 reported that oxidative damage to hippocampus impairs spatial memory of rats. Rats showed impairment of working memory but no change in reference memory after 3 days of exposure to simulated hypobaric hypoxia at 6100 m. 18,20 The study revealed a significant decrease in reduced glutathione (GSH) levels with concomitant increase in oxidized glutathione (GSSG) in the hypoxic rats. The increased generation of free radicals might have resulted in increased utilization of GSH, thus leading to increased GSSG synthesis. The increase in GSH utilization is further accompanied by decreased GSH synthesis as evidenced by decreased GSH reductase activity, thus depleting antioxidant status. This might have been triggered through low levels of NADPH, which is a cofactor for GSH reductase to convert GSSG to GSH. Increased generation of free radicals during hypobaric hypoxia is due to low oxygen availability, and this may be explained by leakage of free electrons triggering a chain reaction resulting in the formation of hydrogen peroxide and reactive hydroxyl radicals. 64 These reactive oxygen species (ROS) have a high affinity for membrane lipids, 65 especially in the brain, as it is rich in polyunsaturated fatty acids. The ROS leads to lipid peroxidation and membrane damage. 66 In addition, there is deterioration in the antioxidant defence mechanism that under normal physiological conditions scavenges the free radicals produced in the cell as by-products of various metabolic pathways. Similar findings were noted by Hota et al. 21,22 in studies on hypobaric hypoxia and oxidative stress. In these studies, however, acclimatization to oxidative stress occurred after prolonged hypobaric hypoxia exposure, that is, 14 days. The study also pointed towards the crucial role that glutamate might play in causing the hypobaric hypoxia–induced oxidative stress. Since numerous biochemical and molecular pathways are involved in the stress response of the cells, the precise mechanism responsible for the onset of acclimatization to oxidative stress still remains to be elucidated.
A study by Shi et al. 31 supported earlier findings on the oxidative stress and apoptosis associated with memory impairment in rats exposed to hypobaric hypoxia. The study showed that exposure to simulated hypobaric hypoxia at 6000 m for 5 days caused spatial memory impairment as well as oxidative stress (increased lactate dehydrogenase activity, decreased GSH level, decreased superoxide dismutase (SOD) level, inhibition of GSH synthesis and greater utilization of GSH for detoxification of hypoxia-induced free radicals, thus leading to increased GSSG synthesis) and apoptosis in different regions of the brain in rats. They also found that the hippocampus is more susceptible to hypoxia when compared with the cortex.
A later study by Baitharu et al. 34,35 found that hypobaric hypoxia–induced memory impairment was associated with neurodegeneration along with alteration in nitric oxide (NO), glucocorticoid, corticosterone, oxidative stress and acetylcholinesterase (AChE) activity in the hippocampal region. In this study, rats were exposed to a simulated altitude of 7600 m in a specially designed animal decompression chamber for 7 days. Another study showed that a shorter and lower simulated altitude exposure also resulted in spatial memory impairment. 36 Exposure of rats to an altitude of 5500 m (375 mmHg) for 3 days was found to induce oxidative stress as evidenced by significant increase in malondialdehyde and reduction in GSH in serum and brain tissue, and neuronal death associated with reactive astrogliosis in hippocampus and superjacent cortex.
Signalling pathways involved in oxidative stress
Barhwal et al. 26,27 examined the signalling cascades involved in mediating oxidative stress neuronal damage in hypobaric hypoxia. The study revealed increased thioredoxin (Trx-1) expression and increased extracellular-signal-regulated kinase (ERK) phosphorylation in rats exposed to hypobaric hypoxia for 14 days. 67 Trx-1 increases hypoxia-inducible factor-1α (HIF-1α) expression in both normoxic and hypoxic conditions. 68 The HIF-1α activates many genes involved in erythropoiesis, angiogenesis, energy metabolism, proliferation/cell survival, apoptosis and generation of NO. 69 –71 Alteration of NO production plays an important role in brain injury in conditions of hypoxia/reoxygenation. 67 NO produced in excess in the brain under the action of hypoxia-induced neuronal nitric oxide synthases rapidly reacts with superoxide anion to form peroxynitrite (ONOO−), a more toxic metabolite that causes brain injury. 72 The peroxynitrite is responsible for protein denaturation, lipid peroxidation, DNA damage and depletion of antioxidant defences. 25
On the other hand, ERK pathway contributes to nuclear factor (erythroid-derived 2)-like 2 (Nrf2) protein stabilization and may result in the increased translocation of Nrf2 into the nucleus. 73 Nrf2 is the central transcription factor involved in regulating the expression of antioxidant enzymes like GSH S-transferase and SOD that are important in protecting the cells against oxidative damage. However, despite increased nuclear translocation of Nrf2, there is increased free radical generation, protein oxidation and lipid peroxidation in hypobaric hypoxia rats. 27
Calcium overload-induced oxidative stress
Calcium overload in the neurons is previously reported to generate ROS by activating phospholipase A2, xanthine oxidase and monoamine oxidase. Release of cytochrome c from the mitochondria due to calcium sequesteration is also known to generate ROS and trigger apoptotic cell death. 26
The elevation of intracellular calcium concentrations and its L-type calcium channel expression, as well as increased calpain expression, have been demonstrated in rats exposed to hypobaric hypoxia. 26 Calpain and L-type calcium channel expressions are maximal on day 7 of hypobaric hypoxia exposure and decline on day 14. Calcium is known to activate calpain that mediates proteolysis of selective proteins and significantly contributes to neuronal damage 74 in several hypoxic and ischemic models. 75,76 Calcium is also known to be sequestered into the mitochondria in excitotoxic conditions, thus resulting in release of cytochrome c that triggers mitochondria-mediated apoptotic cascades. 77,78 Increased cytosolic cytochrome c on day 7 is also demonstrated followed by a marginal decline on day 14 of hypoxic exposure. Expression of active caspase 3, however, shows a different trend with its maximal level on day 14 of hypoxic exposure. This is, however, in accordance with another previous study that reported progressive increase in caspase 3 activity and neurodegeneration with increased duration of exposure. 33 This anomaly in caspase 3 activity may be due to existence of other signalling cascades on day 14 of chronic hypoxic exposure 21 and warrants further investigation.
Glutamate excitotoxicity
The synthesis and release of neurotransmitters are particularly sensitive to hypoxia. Hypobaric hypoxia induces increased release of excitatory amino acid such as glutamate
79
and upregulation of N-methyl-
Alteration in cholinergic and adrenergic systems
Apart from the crucial role of oxidative stress and glutamate excitotoxicity in mediating cognitive deficits following exposure to hypobaric hypoxia, cholinergic systems are also known to be involved. 92 Previous reports have discussed that decreased choline acetyltransferase, 93 increased AChE and decreased α7 nicotinic and muscarinic M1 acetylcholine (ACh) receptors 94 are closely associated with cognitive deficits. Muthuraju et al. 28,29 revealed that impairment in relearning ability and memory retrieval in rats exposed to hypobaric hypoxia was associated with decreased ACh and increased AChE levels which then led to morphological damage in cortical and hippocampal neurons. Administration of AChE inhibitors, such as physostigmine and galantamine, resulted in amelioration of the hypobaric hypoxia-induced neuronal morphological damage in cortex and hippocampus. These AChE inhibitors improved ACh level, decreased AChE activity and increased ACh synthesis by increasing acetyltransferase activity. 95
The greatest amount of nerve growth factor (NGF) is produced in the cortex, hippocampus and pituitary gland, although significant quantities are also produced in other areas, including the basal ganglia, thalamus, spinal cord and in the retina. 96 The NGF plays a pivotal role in the survival and function of cholinergic neurons of the basal forebrain complex 97 ; such functions include attention, arousal, motivation, memory and consciousness.
Other studies associate memory impairment with adrenergic dysregulation and neuronal damage in medial PFC. In these studies, rats were habituated at simulated altitude of 4572 m for 1 day followed by exposure to simulated altitude of 7620 m for a further 7 days. 37,38 It was suggested that norepinephrine (NE) dysregulation under hypobaric hypoxia might have been one of the possible underlying mechanisms leading to cognitive deficits and associated morphological damage. This is due to the fact that working memory is regulated by the PFC and is created by networks of PFC neurons engaged in recurrent excitation generating persistent activity. 98 If the PFC recurrent excitatory firing is profoundly altered by the NE dysfunction, working memory functions will also be affected. 98,99
Brain-derived neurotrophic factor
Jain et al. 32 examined the cellular and molecular pathways related to hypobaric hypoxia–induced neuronal cell death. Adult male Sprague Dawley rats exposed to hypobaric hypoxia equivalent to 7620 m for 7 days resulted in spatial memory impairment and neurodegeneration that was related to low brain-derived neurotrophic factor (BDNF). The low BDNF inhibits PI3K/AKT pathway resulting in activation of GSK3β and caspase 3 which further augments neuronal apoptosis and memory impairment.
cAMP response element binding protein (CREB) is a key transcription factor involved in several critical functions of the brain including learning, neuronal plasticity and cell survival. 100 CREB has been shown to be the key mediator for BDNF-mediated cell survival as studies showed that silencing the transcriptional activity of CREB impaired BDNF protection. 101 CREB can be activated by various kinases including ERK, AKT and GSK3β. 102
Although there was increased CREB phosphorylation in hypoxia as observed from the total CREB to phospho CREB ratio, total CREB was decreased when compared to the normoxic group, thus resulting in its decreased availability for phosphorylation. The decreased CREB expression in hypobaric hypoxia is probably due to the excitotoxic neuronal loss and free radical–mediated protein degradation as indicated by increased protein carbonyls. 22
Conclusion
Hypobaric hypoxia studies conducted in rats provide some insight into the underlying mechanism of hypobaric hypoxia–induced memory loss. Exposure to hypobaric hypoxia in rats induces glutamate excitotoxicity and increases influx of calcium ion by NR1 subunit of NMDA receptor and L-type calcium channel upregulation. 21,22,26 It also induces oxidative stress 18,20 –22,34 via apoptotic signalling pathways leading to hippocampal cell apoptosis. 26,27,32 Hypobaric hypoxia exposure also impairs cholinergic 28,29 and adrenergic 37,38 systems and lowers BDNF level which are important for memory function. All of these mechanisms may be related to learning and memory deficit in human exposed to hypobaric hypoxia.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research received funding from USM Short-term Grant (304/PPSP/61313086).
