Humans have an amazing ability to survive in extreme environments and to acclimatize to varying environmental conditions ranging from humid tropical forests to polar deserts (Burtscher et al., 2018; Ilardo and Nielsen, 2018). Stress hormones and other stress activated systems play an important role as mediators of acclimatization responses to changes in the environment (Dhabhar, 2018). To measure these hormones and evaluate the function of the corresponding and interacting physiological systems, helps to better understand such acclimatization processes and thus offers to explore options to prevent and counteract possible negative responses with detrimental effects on human physiological integrity.

It was demonstrated that catecholamine and cortisol responses to physical exercise differ under conditions of normoxia or hypoxia with higher concentrations under hypoxia (Mazzeo et al., 1994; Woods et al., 2017). However, the question whether normobaric and hypobaric hypoxia elicit the same reactions remains contradictory (Girard et al., 2012; Millet et al., 2012; Mounier and Brugniaux, 2012). Furthermore, to date, most studies investigated the consequences of acute or intermittent hypoxia (Xie et al., 2001; Calbet, 2003; Lusina et al., 2006; Sander, 2016) but data and knowledge about the effect of prolonged and chronic exposure to hypoxic conditions are rather scarce (Dhar et al., 2014).

Moreover, the exposure often compasses not only one environmental stressor (e.g., hypoxia) but demands the acclimatization to several interacting and combined stressors (e.g., hypobaric hypoxia and cold) (Burtscher et al., 2018). In this context, the acclimatization to one stressor was found to be able to modify and influence the response to the other, which is called cross-adaptation (Chauhan et al., 2015). It was evidenced for cold/heat and hypoxia (Launay et al., 2006; Lunt et al., 2010; Keramidas et al., 2015; Gibson et al., 2017) which modified their interaction but broad data in this field of research is still missing due to difficulties of the measure so that a general conclusion cannot be drawn to date. The same accounts for the acclimatization response of each individual that seems to be affected by distinct differences (Bartone et al., 2018).

Besides the sympathoadrenal and glucocorticoid system, the endocannabinoid system (ECS) plays an important role in coping with such stress reactions. Its lipid mediators the endocannabinoids (ECs) and chemically related N-acylethanolamides (NAEs) are very related with acclimatization processes at several physiological lines (e.g., psychological, metabolic, peripheral, and central nervous system) in response to environmental factors such as hypobaric hypoxia and temperature to reach physiological homeostasis (Campolongo et al., 2009, 2013; Richard et al., 2009; Chouker et al., 2010; Dlugos et al., 2012; Feuerecker et al., 2012; Hauer et al., 2013, 2014; Morena et al., 2014; Neumeister et al., 2015; Hanlon et al., 2016).

We investigated a cohort of healthy male individuals over 12 months including a 9-month overwintering period at two Antarctic research stations: Neumayer III near the Antarctic coast in the Queen Maud land and Concordia in Inner East Antarctica.

In general, the Antarctic environment is characterized by extreme shifts in daylight from 24 h light during the Antarctic summer to complete 24 h darkness during the winter period as well as massive changes in outside temperatures at onset of the Antarctic winter requiring complete confinement of the crew to the protective research stations. Thus, expeditioners suffer from sensory deprivation by a monotone environment with lack of stimuli, restricted food variation, and a distinct social narrowing with high potential for conflicts (Das et al., 2018). Furthermore, a disruptive circadian rhythm may influence the hormonal stress response (Chen et al., 2016; Vitale et al., 2018).

In detail, living conditions at Concordia are more severe and thus ensue more rigorous changes than at Neumayer III due to its inland location. Additionally, in contrast to Neumayer III which is situated at sea level, Concordia is situated at 3,233 m above sea level inducing an environment of hypobaric hypoxia to acclimatize to.

We hypothesized that (i) the extreme environmental conditions would lead to an increase of stress-related hormones and that (ii) this increase would be aggravated and sustained in the crew of Concordia Station as of their exposition to long-term, chronic hypobaric hypoxia.

The aim of our study was to investigate and precise the modulations of the stress-related ECS in humans exposed to harsh and extreme environmental living conditions as represented in Antarctica. The main focus hereby was set on the changes induced by chronic hypobaric hypoxia.

To confirm our hypothesis, we therefore determined hormonal/metabolic stress parameters (ECs, catecholamines, and glucocorticoids) and evaluated the psychological stress level in three expedition crews at Neumayer III (sea level) and in two at Concordia (high altitude), respectively.

The present study investigated humans during long-term acclimatization to the highly aversive environment of two Antarctic research stations. Main focus of the research was to investigate the effects of hypobaric hypoxia on stress-related metabolites. Interestingly and contrary to expectations, a massive increase of these metabolites was observed under normoxic conditions (Neumayer III, sea level) while concentrations under hypobaric hypoxia (Concordia, high altitude) stayed low during the isolation period but enhanced significantly at the end. Conversely, catecholamines showed an increase at the beginning of exposure to high altitude being prolonged for norepinephrine whereas cortisol showed no changes.

It seems that the exposure of humans from a physiologically familiar to an unknown and physically very challenging environment triggers this enhanced EC response and that hypobaric hypoxia modulates this response in terms of a downregulation. Downregulated expression or lack of increased expression of ECs in the Concordia (high altitude) crews may originate in hypoxia signaling and EC metabolism. In the tumor microenvironment, hypoxia is known to upregulate the expression of cyclooxygenase-2 (COX-2) via hypoxia-inducible factor-1 (HIF-1) signaling (Greenhough et al., 2009). COX-2 is one of the enzymes responsible for metabolism of ECs (Sugimoto et al., 2017) with the main substrates being AEA and 2-AG (Chiurchiu et al., 2015). Therefore, missing upregulation of ECs at Concordia (high altitude) might be due to hypoxia-induced upregulation of COX-2 (and possibly other enzymes) which in turn oxidizes ECs into compounds (e.g., prostaglandin-ethanolamides, hydroxy-AEAs) (Chiurchiu et al., 2015) that were beyond the scope of our investigations. Furthermore, other dysregulations of degrading and metabolizing systems of the ECs [e.g., fatty acid amide hydrolase (FAAH), intracellular transporters] that were not focus of our study might also have contributed to the stated downregulation of the ECS.

However, HIF-1α expression under long-term hypobaric hypoxia seems to be time-dependent. A short-lived transient activation of HIF-1α-dependent pathways was observed at the beginning of exposition to high altitude (Bigham and Lee, 2014; Bigham, 2016) but seems to elapse gradually over time (Petousi et al., 2014). Thus, this regulation pathway seems to be less significant in acclimatization processes upon long-term exposition to hypobaric hypoxia (Baze et al., 2010). This is in good accordance with findings of Feuerecker et al. (2018) who observed in a former study at Concordia a downregulation of HIF-1α expression under chronic hypobaric hypoxia. Goyal and Longo (2014) observed similar results albeit in an animal model. Against the background of the assumed COX-2 influence on EC metabolism, one might speculate that this mechanism might have also contributed to the rising EC levels at Concordia (high altitude) at the end.

The hypothesis that an unknown and physically challenging environment triggers enhanced EC responses is in good accordance with recent studies which demonstrated that the consequences of strenuous physical conditions (e.g., sleep restrictions, physical exercise) are able to induce a heightened EC response (Cedernaes et al., 2016; Hanlon et al., 2016).

In the light of such stressful conditions, we also expected enhanced catecholamines and cortisol levels due to an elevated sympathetic tone but surprisingly the stress hormone answers stayed normal in the Neumayer III (sea level) expeditioners. One possible explanation for these unexpected observations derives from the interaction between the EC and the sympathetic nervous system and the glucocorticoids. During an acute stress model realized by parabolic flights, subjects who experienced no motion sickness showed similar results to ours presenting high EC concentrations and simultaneously low cortisol levels. By contrast, in subjects with motion sickness, the hormones’ balance was inversed (Chouker et al., 2010). Increasing EC levels were also described in a chronic stress model when examining astronauts on the International Space Station (Strewe et al., 2012).

On the other hand, Yi et al. (2016) reported during a 520 days lasting confinement in the course of a simulated mission to Mars a reduction in circulating 2-AG but a significant increase in urinary norepinephrine and saliva cortisol. The chronic stressor of confinement resulted here in an interaction of these systems that is in contrast to our findings albeit it also displays its apparent inverse character. Moreover, the hypothesis of an inverse interaction is also supported by Ishac et al. (1996) who stated an inhibitory effect of presynaptic cannabinoid CB₁ receptors on noradrenaline release in peripheral sympathetic nerves and Surkin et al. (2018) who evidenced that pharmacological augmentation of EC signaling reduces the neuroendocrine response to stress. Niederhoffer and others also found this inhibitory effect of ECs on noradrenaline to be the cause of cardiovascular depression (Wagner et al., 1998; Niederhoffer et al., 2003; Pfitzer et al., 2005). However, opposite reactions were also described dependent on the acting EC (Kurihara et al., 2001). Furthermore, the way of interaction seems to vary dependent on the physiological system that is examined. Indeed, Simkins et al. (2016) described a positive correlation when they detected a reduced noradrenergic signaling in the spleen capsule in the absence of cannabinoid receptors. Additionally, cannabinoid receptor agonists were shown to induce peripheral antinociception by activating the noradrenergic system (Romero et al., 2013) and they increased norepinephrine efflux in the frontal cortex in an animal model (Page et al., 2008).

By contrast, when analyzing the data of the Concordia (high altitude) crew, we stated that low EC levels were associated with significantly increased norepinephrine concentrations and increased cortisol levels albeit not significant. A former study of our research group in the same Antarctic environment already evidenced similar results with elevated catecholamine levels and assumed their association with stated immune alterations (Feuerecker et al., 2014). Furthermore, several other studies in (simulated) high altitude investigated well the impact of hypobaric hypoxia on sympathoadrenal and adrenocortical stress responses and found a positive correlation (Calbet, 2003; Simeoni et al., 2011; Dhar et al., 2014; Aliyev et al., 2017; Woods et al., 2017). Woods et al. (2017) showed that simulated altitude under normobaric or hypobaric hypoxia (equivalent to 3,375 m) appears to induce similar effects in humans than genuine high altitude. Nevertheless, these effects of hypoxia are also variable when combined with other co-(stress) factors such as, e.g., immobilization in bed rest. Thus, we demonstrated in a former study exposing subjects to 21 days of normobaric hypoxia (simulated altitude ∼ 4,000 m; 14% O₂) and/or bed rest a reduced sympathoadrenal answer after exposition to this combination of stressors. Normobaric hypoxia alone did not induce any significant changes in the catecholamine or cortisol answer (Strewe et al., 2017). However, the results of this study suggest a different mechanism of regulation due to the hypobaric hypoxic environment at Concordia. This mechanism seems to be based on an inverse interaction of the sympathoadrenal with the EC response. Hypobaric hypoxia would therefore either downregulate the EC response or upregulate the sympathoadrenal answer with subsequently raised norepinephrine levels which, in turn, would downregulate and suppress the other system, respectively. The opposite action could be stated in the Neumayer III (sea level) crew.

The hypothesis of an upregulated sympathoadrenal answer is supported by the findings while the mission approaches its end. In the course of time, the sympathoadrenal response declined gradually and stagnated on a low level at the end. Here, interestingly, the EC response started to increase and concentrations of, e.g., AEA and 2-AG were significantly higher than at BDC.

The increase of ECs and NAEs at Concordia (high altitude) at the end with the adjustment of the physiological answers is in line with the hypothesis of a recent study reported by Alarcon-Yaquetto et al. (2017) who demonstrated significant higher NAE concentrations in natives of high altitude (3,830 m) with higher hemoglobin concentration and lower pulse oxygen saturation. Apparently, the time course and the long-term impact of the physical conditions seem to play an important role in the development of the physiological answer. In addition, Feuerecker et al. (2012) evidenced no changes in the human ECS under short-term hypobaric hypoxia (altitude of 3,196 m) but only in its combination with physical effort. The opposite effect of an enhanced EC response was probably due to the study design that displayed an acute stress model. Furthermore, Strewe et al. (2017) found no changes in the ECS when exposing humans for 21 days to normobaric hypoxia (14% O₂ and a simulated altitude of ∼4000 m) either in combination with bed rest or in an ambulatory setting. Here, the different results may be explained by the different physical stressors and the shorter time frame. Furthermore, differences in reaction and regulation patterns of the ECS may also vary dependent of the status and type of cell that is examined (tumor versus healthy cells as mentioned above).

Additionally, in the Concordia (high altitude) crew, great SEMs for norepinephrine (especially in the first months of exposition to hypobaric hypoxia), epinephrine, and cortisol displayed a great inter-individual variability in the concentrations measured that was not found in the Neumayer III (sea level) crew. This might express the individual responses of acclimatization to high altitude in the subjects (Siques et al., 2009; Hermand et al., 2015; Richalet and Lhuissier, 2015).

In summary, our findings at both Antarctic stations evidence that human exposure to a physically challenging and physiologically new and unknown environment with extreme conditions generates a neuroendocrine reaction pattern of interacting physiological systems. Herein, high altitude exposure evokes a significant increase of norepinephrine at the beginning in contrast to other neuroendocrine parameters. Furthermore, there seems to be a great variability in the inter-individual acclimatization responses to high altitude.

As seen by the results in contrast to the baseline data, the reaction pattern seems to display and express a long-term acclimatization process of the human body and physiology to these conditions rather than a pathological process as its intensity and efficiency diminishes over time to finally return to a nearly “normal” status.

Furthermore, our results attribute and relate these reactions to the physical challenge in this environment and do not support an explanation based on increased psychological stress. As all inquired stress questionnaires on both stations negated its presence. However, as a limiting factor, it must be considered that all participants in such expeditions are highly motivated and well prepared which might be a reason for the absence of a distinct psychological effect. Furthermore, such psychological evaluation tests depend on the motivation and honesty (manifested self-reflection) of the subject. If one wants to mask his real feelings and psychological state of mind, any emotional impairment might just be negated. Moreover, a constant higher score level in these tests in the Neumayer III (sea level) crew, albeit in the normal range, might also just display national differences in carrying out the tests or depend on the way how the tests were explained to the subjects in the first place.

The raw data supporting the conclusions of this manuscript will be made available by the authors, without undue reservation, to any qualified researcher.

MF, AC, and GS designed the work. CS, MF, AC, CD, FvdB, BF, HB, and PG-L collected the data. All authors (CS, DT, CD, BF, FvdB, HB, AS, PG-L, PC, DM, RQ, MM, SB, EK, GS, AC, and MF) performed data analysis and interpretation as well as drafting the article and critical revision for important intellectual content. All authors gave final approval of the version to be submitted and any revised version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.