*2.1. Maximal Workload and Respiratory Variables*

The paired sample t-test showed that maximal workload (WRmax) decreased significantly (*p* < 0.001) by 16.3% in hypoxia (3000 m) compared to the initial measurements in normoxia (Figure 1). The same trend of changes was observed in VO2max values. The values of VO2max decreased significantly (*p* < 0.001) in hypoxia compared to normoxia respectively by 14.5% (Figure 1). Additionally, there were statistically significant changes in delta values of blood lactate concentration (ΔLA) after the incremental test between normoxia and hypoxia 3000 m. The Wilcoxon test showed that ΔLA increased significantly (*p* < 0.01) by 6.05% despite the significant reduction in WRmax in hypoxia 3000 m compared to the measurements in normoxia (Figure 2).

**Figure 1.** Maximal workload (WRmax), and maximal oxygen uptake (VO2max) during incremental test performed in different conditions. \*\*\* *p* < 0.001.

**Figure 2.** Delta values of blood lactate concentration (ΔLA) during incremental test performed in different conditions. \*\* *p* < 0.01.

#### *2.2. Choice Reaction Time and Number of Correct Reactions*

No significant interaction (condition x time of measure) effect was found on CRT but only a significant main effect of time of measurement (at rest, max and after 3 min of recovery) on CRT values (F = 21.88; *p* < 0.001) was observed. Additionally, there was a significant interaction (condition × time of measure) effect in the NCR (F = 3.44; *p* < 0.05) during the incremental test.

The post-hoc Tukey's test showed that CRT decreased significantly (*p* < 0.05) by 9.7% after 3 min of recovery after the incremental test (CRTafter 3 of recovery) compared to initial values observed at rest (CRTrest) in normoxia. Additionally, the values of CRTafter 3 of recovery decreased significantly in normoxia (*p* < 0.001) and hypoxia (*p* < 0.01) compared to CRT during maximal workload of the incremental test (CRTmax) respectively by 17.7 and 12.2% However, CRTmax increased significantly (*p* < 0.01) by 13.8% compared to CRTrest in hypoxic conditions (Figure 3).

**Figure 3.** Choice reaction time (CRT) at rest, during maximal effort (max) and after 3 min of the recovery period in normoxia and hypoxia (3000 m). \* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001.

The post-hoc Tukey's test showed that number of correct reactions (NCR) during maximal workload of the incremental test (NCRmax) decreased significantly (*p* < 0.05) by 9% in hypoxic conditions compared to normoxia. NCRmax values in hypoxia were significantly lower compared to NCR at rest (11.5%; *p* < 0.001) and NCR after 3 min of recovery after the incremental test (9.7%; *p* < 0.01) (Figure 4).

**Figure 4.** Number of correct reactions (NCR) at rest, during maximal effort (max) and after 3 min of the recovery period in normoxia and hypoxia (3000m). \* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001.

## *2.3. Brain-Derived Neurotrophic Factor and Selected Biochemical Variables*

There was a significant interaction (condition x time of measure) effect in the BDNF (F = 3.66; *p* < 0.05) serum concentrations.

The post-hoc Tukey's test showed that BDNF concentration increased significantly immediately after the incremental test (BDNFmax) in normoxia (by 29.3%; *p* < 0.01), as well as in hypoxia (by 50.0% *p* < 0.001). Additionally, BDNF concentration decreased significantly aftera1hrecovery period (BDNFrest) by 20.7% (*p* < 0.01) in normoxia and by 41.3% (*p* < 0.001) in hypoxia (Figure 5). There were no statistically significant differences in BDNF concentration between normoxia and hypoxia trials.

**Figure 5.** Brain-derived neurotrophic factor (BDNF) serum concentration at rest, during maximal effort (max) and after 1 h recovery of the recovery period in normoxia and hypoxia (3000 m). \*\* *p* < 0.01; \*\*\* *p* < 0.001.

The Friedman test showed a statistically significant effect of the time of measurement on selected biochemical variables such as NO2 −, C, adrenalin (A) and dopamine (DA) in both conditions (normoxia and hypoxia) (Table 1).

The post-hoc Friedman test showed that NO2 − concentration measured immediately after the incremental test was significantly higher (*p* < 0.05) compared to the NO2 − concentration observed at rest and 1 h after the incremental test in normoxia as well as hypoxia. However, the Wilcoxon test showed that were no significant differences between normoxia and hypoxia in NO2 − concentrations (Table 1).


*Int. J. Mol. Sci.* **2020**, *21*, 5569

326

0.001.

There was no significant effect of the time of measurement (at rest, max, 1 h after test) on C concentrations in both conditions. However, there was a significant effect of conditions (normoxia vs. hypoxia) on C concentration. The Wilcoxon test showed that C concentration measured immediately after the incremental test (Cmax) was significantly higher (*p* < 0.05) by 20.7% in normoxic conditions compared to hypoxia (Table 1).

Moreover, the post-hoc Friedman test showed that A concentration measured immediately after the incremental test (Amax) was significantly higher (*p* < 0.05) compared to the A concentration observed at rest (Arest; by 656.1%) and 1 h after the incremental test (A1h after; by 215,1%) in normoxia. However, in hypoxia, Amax and A1h after were significantly higher (*p* = 0.001) compared to Arest respectively by 505.9% and 86.4%. The Wilcoxon test showed that Arest, Amax and A1h after in hypoxia were significantly higher (*p* = 0.001) compared to these values in normoxia, respectively by 368.2%, 275% and 534.2% (Table 1).

The post-hoc Friedman test showed that DA concentration measured 1 h after the incremental test (DA1h after) significantly decreased (*p* < 0.05) by 26.9% compared to the DA concentration measured immediately after the incremental test (DAmax) in normoxia. However, in hypoxia, DAmax and DA1h after significantly increased (*p* < 0.05) compared to the DA concentration at rest, respectively by 97% and 104.5%. Additionally, the Wilcoxon test showed that DAmax and DA1h after in hypoxia were significantly higher (*p* < 0.01) compared to these values in normoxia, respectively by 69.2% and 140.3% (Table 1).

#### **3. Discussion**

Physical training is planned to improve physical fitness and performance, which confers numerous positive effects on the whole body function [69]. A pivotal role in regulation of these changes is assigned to the brain, particularly the prefrontal cortex, which takes part in regulation of many executive functions to prepare humans for situations demanding high levels of working memory, attention and cognitive flexibility [70]. On the other hand, altitude training, which nowadays has become a standard training protocol in many sports to increase exercise capacity [71–73], causes cognitive frailty [74]. The cognitive decline is most intensely manifested after acute hypoxic exposure and is more profound in athletes than non-trained individuals [75]. Therefore, the primary purpose of this study was to examine the impact of a single bout of exercise to volitional exhaustion during acute exposure well endurance-trained athletes to moderate hypoxia on psychomotor performance. We also measured serum BDNF, some selected hormones (C and catecholamines), the NO pathway metabolites (nitrite), as well as ET-1, as a possible candidate that may be involved in modulation of this phenomenon. We used measurements of CRT and NCR as indicators of psychomotor skills because these variables were demonstrated to be a dependable measure of cognition in different experimental approaches [22].

Results from this investigation revealed that exercise to volitional exhaustion extended CRT in both experimental conditions but only the impact of hypoxic conditions on this variable was statistically significant. A similar alteration, but with statistically significant changes in both cases, was seen in NCR (Figure 4). Additionally, NCR was significantly increased in response to exercise to volitional exhaustion performed in hypoxia as compared with normoxia. These results are in agreement with previous data reported by others which showed cognitive impairment of trained subjects at high altitude [75] and additionally suggested that NCR which specified response accuracy was a more sensitive tool for estimating cognitive state than CRT. Interestingly, the aforementioned exercise-induced effect was transient in both experimental conditions and studied variables returned to basal values within a few minutes after cessation of exercise.

Increasingly more evidence supports the action of BDNF as an underlying factor that elicits exercise/training-induced beneficial changes CNS [76–78]. Moreover, there are also suggestions that the negative impact on cognitive function in hypoxia can be at least partially explained by the simultaneous decrease of BDNF [79,80]. Studies with animal models have shown that this neurotrophin is produced, among other tissues, in brain by motor neurons [81] and intermittent hypoxia increases BDNF levels in

neurons of the primary motor cortex [82]. Also, central BDNF cannot be measured in living humans. It has been suggested that the brain is the main source of the increased BDNF in circulation [83,84]. If so, and considering the fact that in our study similar circulating levels of BDNF were seen in both experimental groups but cognitive performance was suppressed in hypoxia, one could conclude that acute elevation of BDNF did not compensate for hypoxia-induced cognition impairment. Previous evidence suggests that BDNF plays a key role in memory and learning [85] and is a vital regulator of neuronal function and plasticity [86,87]. However, these aforementioned effects appeared almost selectively in response to a repetitive or chronic stimulus or in studies in vitro. Results of the current research are in line with reports of some studies using cognitive tasks such as executive function or attention which were unrelated to changes in BDNF after acute exercise [88–90]. One reason which at least theoretically might be considered for the acute action of BDNF is its involvement in synaptic transmission [91]. The results from animal studies indicate that during hypoxia neurons can within minutes alter synaptic transmission [92,93]. Support for a link between exposure to hypoxia and run-down of synaptic transmission has been widely documented in in vitro studies [94,95] and in rats subjected to severe (6100 m) hypoxia [96]. The mechanism by which severe hypoxia induced cognitive impairment was accompanied by a decrease in Acetylcholine (Ach) level and increase in Acetylcholinesterase (AChE) in the cortex [96].

Apart from BDNF, C has been identified as a possible endocrinological mediator of exercise which may modulate brain function [97,98]. Both hypoxia and exercise stimulate the gland cortex to release C. While it is well recognized that elevation of C has been observed in response to acute exercise of higher than moderate intensities, data from hypoxic studies are less consistent. In response to this environmental stimulus, increases or lack of changes have been reported. In our study, when maximal effort was performed in moderate hypoxia the C level was significantly lower compared to normoxia. As this hormone impairs the prefrontal cortex [99,100], the region of the brain that controls more of our cognitive function, one can argue that its deteriorative impact on psychomotor performance, if any, was negligible. However, there is a study showing similar exercised increased plasma C in normoxia and acute hypobaric hypoxia (3000 m altitude) [101]. This discrepancy may be the result of different research designs between aforementioned and our studies (normobaric vs. hypobaric hypoxia as well as cyclists vs. cross-country skiers and ice hockey players).

Previous studies demonstrated that physically active persons have shorter CRT than sedentary ones [102] and the regular U-shaped curve was obtained in athletes when CRT values were plotted against A and noradrenaline (NA) during graded incremental exercise to volitional exhaustion [34,62]. The latter results indicate that CRT exceeds the resting values at exercise loads close to maximal, and catecholamines may play a role in this phenomenon. Although most studies agree that catecholamine levels increase at rest and exercise at high altitude, brief or moderate hypoxia does not always elevate their levels [103,104], and this is especially true for NA [105]. The present finding reflects this phenomenon in the case of NA and serotonin (5-HT) while indicating elevated A as a potential player in cognitive control in our experimental paradigm. In favor of such an assumption is the significantly elevated A level during maximal effort in hypoxia with simultaneous statistically significant extension of CRT. Previous studies on patients with psychological trauma indicate that A and DA are involved in the activation of the prefrontal cortex [106] and similarly stress and aggressive behavior were shown to increase turnover of both A and DA in this area in rodents [107]. Acute hypoxia is considered as a systemic stress factor and was also seen to reduce cognitive function in rats, which was associated with DA signaling in the prefrontal cortex [63]. This finding is additionally supported by increased errors in the cognitive test, which were associated with reduced DA signaling in the prefrontal cortex [108]. However, the lack of significant differences in DA between our investigated groups probably excludes DA participation in the modulation of cognitive function in our subjects. It is worth noting that there is also a study [66] showing that during a short episode of anoxia, an increase in A level can have a protective effect against its disruptive effects. In the present study, this effect did not occur in response to exercise either in normoxia or in moderate hypoxia despite an elevated level of

A in both cases. Collectively, the aforementioned results lead to the conclusion that A probably reveals biphasic action on cognitive function in hypoxic conditions, i.e., a positive effect during short-term and negative after prolonged actions.

It is well established that hypoxia releases a diffusible vasoconstrictor and vasodilator substance and this process can affect blood flow to the brain. Some evidence supports the view that an increase of blood flow to the brain may induce cognitive improvement by delivering elevated oxygenated hemoglobin [109]. In accordance with the above knowledge, we have measured in this research ET-1 and NO metabolites as the most potent representatives of endothelial released vasoconstrictors [110] and vasodilators [111], respectively. In humans, no changes in ET-1 response to maximal exercise in acute hypobaric hypoxia (3000) were noted as compared to normoxia [101]. In our study, both exercise and exercise in hypoxia also provoked no changes in ET-1, confirming occurrence of the above-mentioned effect in normobaric hypoxia, and suggesting that ET-1 had no important influence on blood low regulation in the acute response to moderate hypoxia.

Nitric oxide is mainly generated in the body by endothelial cells, but it is also produced in the CNS, where it is closely involved in neurotransmission and modulation of neuron metabolism [112]. A previous study conducted with diabetic patients indicated that increased bioavailability of NO was a factor that might enhance cognitive function [113]. However, the latest data questioned this possibility when psychomotor performance was tested in hypoxic conditions [114]. Our research conducted in a normobaric hypoxic chamber revealed an inconsiderable rise in serum NO2 − levels both in basal and exercise conditions as compared to normoxia. Since simultaneously cognitive performance was blunted it implied that NO production under these circumstances was likely too low to affect cognitive functions or NO was not a crucial player in this phenomenon. These assumptions are partially in line with recent findings which do not support a beneficial effect of NO3 − supplementation on cognitive function in sedentary males at moderate and very high simulated altitude [115]. However, it is clear that if during a profound reduction of absolute work under hypoxia the NO2 − level was higher than in normoxia then an additional exercise-independent system was responsible for NO2 − formation. This pathway is activated in parallel during exercise under hypoxic conditions, yet it needs to be identified.

Physical exercise involves markedly increased activity of many brain structures [116]. The metabolites produced in the muscles, which can diffuse into the CNS and can be utilized as a fuel to sustain increased energy requirements, may participate in this process [117]. Such possibility underscores the importance of the muscle produced lactate (LA) during exercise which can be transferred to neurons via monocarboxylate carriers and used in addition to LA delivered via astrocytes-neurons lactate shuttle [118] as energy fuel during neuronal activation as well. On the other hand, LA was recognized as a signaling molecule in the brain [119]. Among others, it can bind a receptor of the G protein coupled receptor family (GPRs) [120] and thereby cause a decrease of cyclic adenosine monophosphate (cAMP) level. This raises the possibility of interaction between LA and A in metabolism regulation on a subcellular level. If true, this phenomenon should have been more strongly connected with hypoxic conditions because A level during maximal effort in hypoxia was significantly elevated while lowering LA (as compared to controls). This was accompanied with simultaneous statistically significant extension of CRT.

#### **4. Materials and Methods**

#### *4.1. Participants*

Eleven cyclists (20 ± 1.4 years of age) were recruited for the study as volunteers. All participants had current valid medical examinations and showed no contraindications that would exclude them from the study. They declared that for at least one month before testing they did not take either medications or dietary supplements. Written informed consent was obtained prior to study commencement.

The basic anthropometric data of the volunteers (body height—BH, body mass—BM, fat content—FAT) are presented in Table 2. The experimental procedures involved, and the related

risks were explained to all the participants verbally, informed written consent was taken from each participant and they could withdraw at any time of the study. The research project was conducted according to the Helsinki Declaration and was approved (no. 5/2013, approval date: 26.06.2013) by the Ethics Committee for Scientific Research at the Jerzy Kukuczka Academy of Physical Education in Katowice, Poland.

**Table 2.** Mean values of body height (BH), body mass (BM) and fat content (FAT) of study participants (*n* = 11).


The subjects participating in the study were tested on two randomized occasions separated by 5 days duration in normoxic and hypoxic conditions. Participants were allocated to conditions using a computer-generated randomized list [121].

Hypoxic conditions were created using a normobaric hypoxia chamber (LOSA HYP/HYOP-2/3NU system, LOWOXYGEN SYSTEMS, Berlin, Germany) that is in use in the Laboratory of Hypoxia of the Jerzy Kukuczka Academy of Physical Education and the selected hypoxia was an equivalent of 3000 m altitude (FiO2 = 14.7%).

On each occasion, the participants were subject to two graded ergocycle tests each performed under normobaric normoxic and normobaric hypoxic conditions (3000 m asl). Before each test, body mass and body composition of each participant was determined using a model Inbody 720 (Biospace Co., Tokio, Japan) body composition analyzer using electrical impedance measurements.

#### *4.2. Ergocycle Graded Exercise Test*

The ergocycle tests were performed on a model Excalibur Sport (Lode BV, Groningen Netherlands) cycloergometer, beginning at a work load of 40 W, which was increased by 40 W every 3 min until volitional exhaustion. If a subject terminated the test before completing a given workload, then the maximum workload was calculated from the formula WRmax = WRk + (t/T × WRp) [122], where WRk—previous workload, t—exercise duration with the work-load until premature failure, T—duration of each workload, WRp—the amount of workload by which exercise intensity increased during the test.

During the tests, heart rate (HR), minute ventilation (VE), breathing frequency (BF), oxygen uptake (VO2) and carbon dioxide content in expired air (VCO2) were recorded in the subjects with the MetaMax 3B gas analyzer (Cortex, Leipzig, Germany). Fingertip capillary blood samples for the assessment of LA concentration (Biosen C-line Clinic, EKF-diagnostic GmbH, Barleben, Germany) were drawn at rest and at the end of each step of the test, as well as during the 3rd, 6th, 9th, and 12th minute of recovery. Additionally, capillary rest and post-exercise blood samples were used to determine acid-base equilibrium and oxygen saturation of hemoglobin (Rapid Lab 248, Siemens/Bayer Diagnostics, Erlangen, Germany).

### *4.3. Psychomotor Performance Determination*

The choice reaction time and NCR were selected as indices of psychomotor performance as described previously [26]. Briefly, the CRT console was mounted on the wall in front of the ergometer at eye level, 1.5 m away from the subject. The test included 15 positive (red light or a sound) and 15 negative (green and yellow lights) stimuli applied in a randomized order in 1 to 4 s intervals. The subjects were asked to press and then to release, as quickly as possible, the button of the switch devised kept in the right hand in response to the red light, the button in the left hand in response to the sound and not react to the negative stimuli. The total time for each CRT was 107 s. The stimuli and the subjects' responses were recorded using the reaction time measuring device (MRK 432, ZEAM, Zabrze, Poland). The reaction time was determined to the nearest 0.01 s. The results are presented at the mean

reaction time of 15 responses to positive stimuli. The subjects were familiarized with the procedure a week before the study by practicing the task both at rest and during cycling.
