Next Article in Journal
Heritability of Urinary Amines, Organic Acids, and Steroid Hormones in Children
Previous Article in Journal
Characterizing the Flavor Precursors and Liberation Mechanisms of Various Dry-Aging Methods in Cull Beef Loins Using Metabolomics and Microbiome Approaches
Previous Article in Special Issue
Gut Microbiota and Serum Metabolome in Elite Cross-Country Skiers: A Controlled Study
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Metabolites
Volume 12
Issue 6
10.3390/metabo12060473
metabolites-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Exercise and Interorgan Communication: Short-Term Exercise Training Blunts Differences in Consecutive Daily Urine 1H-NMR Metabolomic Signatures between Physically Active and Inactive Individuals

by
Leon Deutsch
1,
Alexandros Sotiridis
2,3,
Boštjan Murovec
4,
Janez Plavec
5,
Igor Mekjavic
3,
Tadej Debevec
3,6 and
Blaž Stres
1,3,7,*
1
Biotechnical Faculty, Department of Animal Science, University of Ljubljana, SI-1000 Ljubljana, Slovenia
2
Section of Sport Medicine and Biology of Exercise, School of Physical Education and Sport Science, National and Kapodistrian University of Athens, 17237 Athens, Greece
3
Department of Automation, Biocybernetics and Robotics, Jožef Stefan Institute, SI-1000 Ljubljana, Slovenia
4
Faculty of Electrical Engineering, University of Ljubljana, Jamova 2, SI-1000 Ljubljana, Slovenia
5
NMR Center, National Institute of Chemistry, SI-1000 Ljubljana, Slovenia
6
Faculty of Sports, University of Ljubljana, SI-1000 Ljubljana, Slovenia
7
Faculty of Civil and Geodetic Engineering, Institute of Sanitary Engineering, University of Ljubljana, SI-1000 Ljubljana, Slovenia
*
Author to whom correspondence should be addressed.
Metabolites 2022, 12(6), 473; https://doi.org/10.3390/metabo12060473
Submission received: 6 April 2022 / Revised: 20 May 2022 / Accepted: 20 May 2022 / Published: 24 May 2022
(This article belongs to the Special Issue Effect of Exercise, Physical Activity, and Performance on Human Gut Microbes and Microbial Metabolome)
Browse Figures
Review Reports Versions Notes

Abstract

:
Physical inactivity is a worldwide health problem, an important risk for global mortality and is associated with chronic noncommunicable diseases. The aim of this study was to explore the differences in systemic urine 1H-NMR metabolomes between physically active and inactive healthy young males enrolled in the X-Adapt project in response to controlled exercise (before and after the 3-day exercise testing and 10-day training protocol) in normoxic (21% O2), normobaric (~1000 hPa) and normal-temperature (23 °C) conditions at 1 h of 50% maximal pedaling power output (Wpeak) per day. Interrogation of the exercise database established from past X-Adapt results showed that significant multivariate differences existed in physiological traits between trained and untrained groups before and after training sessions and were mirrored in significant differences in urine pH, salinity, total dissolved solids and conductivity. Cholate, tartrate, cadaverine, lysine and N6-acetyllisine were the most important metabolites distinguishing trained and untrained groups. The relatively little effort of 1 h 50% Wpeak per day invested by the untrained effectively modified their resting urine metabolome into one indistinguishable from the trained group, which hence provides a good basis for the planning of future recommendations for health maintenance in adults, irrespective of the starting fitness value. Finally, the 3-day sessions of morning urine samples represent a good candidate biological matrix for future delineations of active and inactive lifestyles detecting differences unobservable by single-day sampling due to day-to-day variability.

Graphical Abstract

1. Introduction

Physical inactivity is a worldwide health problem ranking as the fourth most important risk for global mortality [1]. The efforts undertaken by the World Health Organization (WHO) to minimize the time spent sedentary [2] are directed at decreasing the risks for more than twenty chronic noncommunicable diseases (e.g., coronary heart disease, stroke, type 2 diabetes, obesity, metabolic syndrome, glucose insensitivity) next to mental health and neurological problems such as depression and dementia [1]. As physical activity in the form of various types of exercise promotes wellbeing and increased quality of life, understanding the biological mechanisms through which it impacts is of central importance. Although genetics, lifestyle and environment are likely the most important parameters, their relative contributions and interactions are not well-understood.
Exercise-related stress alters the chemical steady state of the internal biochemical environment. The net result is modifications in the rate of production and consumption of various metabolites within biochemical network affecting the systemic levels of metabolites relative to the exercise intensity, muscle damage or the extent of the exercise as part of the lifelong history [3,4,5].
In respect to the progressively emerging picture of metabolic states characteristic of various noncommunicable diseases, a number of metabolomic studies have clearly shown that physical activity results in modifications of hundreds of metabolites associated with fatty-acid mobilization, lipolysis and metabolism, the TCA cycle, glycolysis, amino-acid metabolism, carnitine, purine and cholesterol metabolism and insulin sensitivity [1]. Based on these results, it has become obvious that while the metabolomics patterns may differ slightly between groups, it was the overall volume of exercise acting as the most important driver of the metabolomics makeup [6,7,8,9], even irrespective of hypoxia [4]. This points to the multifactorial dose–response relationship between activity (intensity, frequency, time frame (exposure measured in hours, days, weeks, years, lifelong)) and metabolomic signatures [1,4].
Metabolomics has become a technology-driven discipline focusing on improved high-throughput and large-scale data collection, analysis and interpretation. Metabolomes were characterized utilizing proton nuclear magnetic resonance (1H-NMR) in minimally user-invasive biomaterial—the first morning urine. The approach of 1H-NMR was utilized in this study as it is nondestructive, quantitative, cost-effective, reproducible and requires no sample derivatization [10]. Although the approach captures a modest number of metabolites (n > 350 in our past studies [4,5,11]) it enables identification of unknown novel compounds in complex biological matrices such as serum, saliva, urine or feces [4,10] and has been frequently utilized (>40% of studies) in observational and experimental studies next to short-term (<1 week) or long-term (>1 week) interventions [1]. The use of consecutive three-day urine 1H-NMR data points was first tested recently in the form of a 3-day sliding window within the PlanHab project and showed promising results for delineation of systemic differences between groups [3,4,5].
The aim of this study was to explore the differences in systemic urine 1H-NMR metabolomic signatures between groups of physically active and inactive individuals before and after a 10-day training protocol in normoxic (21% O2) and normal-temperature conditions (23 °C) of the campaign number 4 within the X-Adapt: Cross-adaptation between heat and hypoxia project (Figure S1) [12]. The aim of the X-Adapt study itself was to investigate the effects of a 10-day exercise protocol on aerobic performance in young males. The X-Adapt training sessions were composed of controlled 60 min normoxic and normobaric (~1000 hPa) exercise [12] utilizing prescreened participants (graded exercise test on a cycle ergometer to determine their normoxic VO2max and maximal power output (Wpeak—the highest workload sustained by incremental exercise until exhaustion). In short, aerobic fitness was defined using maximal oxygen uptake (VO2max) values (untrained VO2max < 45 mL·kg−1·min−1; trained VO2max > 55 mL·kg−1·min−1) [13,14]. Untrained participants were also required to not participate in organized sports, while minimal cycling and walking for commuting to work were allowed. In contrast, trained participants performed endurance-type activities (running, cycling, swimming) several times per week.
The X-Adapt urine-sample collection produced by the original project outline described before [12,15,16] (i.e.,) was augmented by including two additional urine-sampling periods, extending the project outline and resulting in the extended sample collection (Figure 1 and Figure S1). As a result of these extended urine-sampling periods there was no effect on human physiology or exercise approaches utilized in the X-Adapt project. The extended sample collection included the additional three-day baseline urine samples before the actual start of the X-Adapt campaign 4 [12] and samples collected during the last three days of the 10-day exercise session. In total, the time span between the two sampling periods of extended sample collection contained 3 days testing, 1 day rest and 10 days exercise, amounting to almost 14 days of exercise [12]. This enabled us to capture the daily variability between the trained and untrained groups before the actual onset of the X-Adapt campaign 4 and to observe the actual systemic differences in response to the almost 14-day concerted exercise between the trained and untrained groups. In addition, this enabled us to perform additional comparisons between the various sections based on 1H-NMR urine metabolomes collected uniquely over three consecutive days (Figure S1).
As there is a lack of data and understanding on the differences between healthy trained and untrained young males and the progressive changes in human metabolomics responses coupled to introduction of exercise, we first compiled and performed a multivariate analysis of the exercise dataset [12,15,16] and hypothesized that (i) significant differences existed in the exercise dataset between trained and untrained groups; (ii) the 2-week experimental setup would enable us to detect overall change in resting urinary metabolome 3-day sequences; (iii) significant differences existed between the trained and untrained group’s urinary metabolomes despite the nonsynchronized diet of participants; (iv) the introduction of scheduled 2-week physical exercise would significantly change urine 1H-NMR metabolomes in the untrained group at least; (v) discriminant metabolites could be identified between the trained and untrained groups; (vi) the extended X-Adapt experiment utilized in this study provided insight into the significantly different metabolic pathways between the trained and untrained experimental variants, signifying the importance of the training history of participants for responses in human metabolomes that were also linked to the VO2max values (the maximal rate of oxygen consumption).

2. Results and Discussion

Twenty male participants 23.5 ± 2.5 years old were recruited for this study and were divided into two groups (10 participants per group) based on their physical performance (trained and untrained group). Table S3 represents their baseline characteristics. Participants in the trained group were 23 ± 2 years old, 180 ± 5 cm tall, weighed 74 ± 3 kg and had a body surface area of 1.96 ± 0.08 m2 and body fat of 9.2 ± 2.3%. On the other side, untrained participants were 25 ± 3 years old, 179 ± 3 cm tall, weighed 85 ± 14 kg, had a body surface area of 2.05 ± 0.17 m2 and body fat of 16.3 ± 4.9%. VO2peak, Wpeak and Wpeak per kg were significantly different between the untrained and trained group. The untrained group had a lower VO2peak (42 ± 5 mL/kg × x min in untrained and 58 ± 6 mL/kg × x min in trained group), lower Wpeak performance (309 ± 46 W in untrained and 364 ± 35 W in trained group) and lower Wpeak per kg (3.6 ± 0.4 W/kg in untrained and 4.9 ± 0.5 in trained group) [12,15,16,17].

2.1. Integrated Analysis of Exercise Data and the X-Adapt Urine-Sample Collection

In this study, exercise data reported before [17] and 1H-NMR metabolomic data obtained in this study were explored. The previously reported physiological data [12,15,16,17] relevant for metabolomic analyses within the same 3-day series of X-Adapt pre/post-testing were analyzed. Their integrated analysis in this study showed that significant multivariate differences existed between the trained and untrained groups at pretesting (Figure 1) before the onset of the 10-day 50% Wpeak training session and after the training (PERMANOVA; F = 7.304; p(same) = 0.0001; npermutations = 5000). In addition, nearly significant differences (p = 0.054) existed between the pre-exercise untrained and postexercise untrained groups, suggesting a larger magnitude of changes in human exercise-related characteristics than in those leading active lifestyles.
The nonmetric multidimensional scaling (nmMDS) results also showed significant groupings separating trained from untrained (Figure 2) showing that significant differences at the level of human exercise data also remained detectable after the 10-day training period. A heatmap (Figure 3 and Figure S2) of the measured exercise parameters shows large differences in measured parameters between trained and untrained groups, but also reflects significant interpersonal variability within each of the measured parameter. This suggests that although significant differences in the multivariate description of exercise states can be reported for the trained and untrained groups before and after the training sessions, the rate of change within the 10-day training at 50% Wpeak was significantly higher for the untrained group, as reported before [12]. This observation is further supported by detailed analyses of the exercise parameters contributing most to differences between trained and untrained groups, as VO2max values in fact decreased 3.2% and increased for 9.2% in trained and untrained groups, respectively. This observation is in line with past observations showing that the pretraining VO2peak and percentage change in VO2peak with training were inversely correlated, showing that the rate of adaptation is largest in less physically prepared participants [18]. In addition, the integrated exercise data reported in this study showed that trained and untrained groups responded differently, as VO2max of the trained group could not be sustained by 50% Wpeak training in comparison to a further increase in the untrained group in response to 50% Wpeak training. Taken together, these results show that during the 10-day 50% Wpeak training, the trained and untrained groups were becoming more synchronized in terms of measured exercise parameters, as also suggested before [12,15,16,17]. A two-way PERMANOVA confirmed that participant status (trained or untrained) and 50% Wpeak training exercise (pre- or post-training) were significantly associated with the underlying multivariate exercise data (F = 13.07; F = 2.57 and p(same) = 0.0001; p(same) = 0.038), respectively), while interaction between participant status (trained or untrained) and time of training exercise (pre-or post-training) was not significant (status x exercise; F = 0.47; p(same) = 0.79), suggesting that the response of the two groups to the application of exercise was not uniform.
In contrast to results from physiological measurements, our 1H-NMR analyses of the X-Adapt urine-sample collection (i.e., the 3-day morning urine samples taken within the same timeframes of X-Adapt pretraining and post-training test sessions) did not identify any significant difference between any groups (PERMANOVA; p > 0.3; npermutations = 5000) (Figure 1). This is in line with past observations that metabolomes at rest (e.g., systemic morning urine samples) cannot be indicative of physical status and capacity due to their gradual return to baseline within 24 h after exercise [19]. In addition, these results point to the potentially homogenizing short-term responses in trained and untrained individuals to the standardized pre- and post-testing conducted on three consecutive days utilized in the X-Adapt study [12]. The normoxic, temperature and hypoxic tests utilized in X-Adapt were described in detail before [12,15,16,17]. Moreover, additional in-depth tests of statistical significance between 1H-NMR metabolomes from trained and untrained groups on a day-to-day basis also did not produce significant differences (PERMANOVA; p > 0.05; npermutations = 5000). These results show the lack of significant differences between the trained and untrained groups on the level of urine 1H-NMR metabolomes in response to the X-Adapt pretesting and post-testing trials (Figure 1).

2.2. Differences in Urine 1H-NMR Metabolomes between the Trained and Untrained Groups: The Extended Urine-Sample Collection

In order to elucidate the potentially homogenizing responses in trained and untrained individuals to the X-Adapt training regimen the extended urine-sample collection (Figure 1) was analyzed by 1H-NMR. The 1H-NMR fingerprints of trained and untrained groups were compared to identify the existence of internaldata-structure characteristics for the two groups of participants. The results of one-way and two-way PERMANOVA showed that significant differences existed between metabolomes of trained and untrained participants (p < 0.01). This was also confirmed by the two-way PERMANOVA test, showing that activity (trained/untrained) was the only parameter significantly associated with the two groups (p = 0.0001). Regime (pre- or post-test) and the interaction between regime and activity was insignificant (p > 0.05). This was also confirmed by the nonsignificant change in the number of metabolites present and the sum of their concentrations in all sampled groups (Table S1).
The Mann–Whitney test showed that significant differences in distributions between trained and untrained group existed in physical characteristics of urine such as pH, salinity, total dissolved solids (TDS) and conductivity. Salinity, conductivity and TDS were significantly higher in the untrained group than in trained, while pH was slightly more alkaline in trained (Figure S3).
Based on the nonparametric approaches described below, we used statistical methods implemented in MetaboAnalyst 5.0 [20,21,22]. According to the partial least-squares-discriminant analysis (PLSDA) of variable importance in the projection (VIP) scores, differences existed between the trained and untrained groups of participants at the level of cholate, tartrate, cadaverine, lysine and N6-acetyllisine (HMDB0000206) as the most distinguishing metabolites to differentiate the trained and untrained groups (Figure 4). The first three metabolites were all present at higher concentrations in the untrained group while concentrations of lysine and N6-acetilysine were higher in the trained group. Primary bile acid synthesis, glutathione metabolism, aminoacyl-tRNA biosynthesis and lysine degradation pathways were enriched in the untrained group (Figure 5).
In addition to the multivariate analyses, we also performed extensive machine-learning modeling using Just Add Data Bio (JADBIO) [23] to investigate the importance of metabolites and physicochemical parameters in urine samples. A total of 181,020 models were trained using extensive tuning effort. The most interpretable model was logistic ridge regression with the penalty hyperparameter lambda of 10−4 and an area under the curve (AUC) value of 0.748. In addition to AUC (Figure 6a); all other thresholds were also statistically significantly different from baseline. Data were preprocessed and standardized by imputation of means and removal of constants. Features were selected based on the test-budgeted statistically equivalent signature (SES) algorithm with the following hyperparameters: maxK = 2, alpha = 0.1, and budget = 3 × nvars. PCA plot (Figure 6b) shows that differentiation based on modeled data is not complete, which means that larger groups should be formed in the future. 12 metabolites and pH were selected as the most important features for distinguishing the trained from the untrained group based on urine. Table S2 lists all the important metabolites. The major metabolite selected by JADBIO was tartrate. The power of the model obtained by using only tartrate was 73.8% (with 95% CI from 69.9% to 77.6%) (Figure S4). We applied the trained model to the test portion of our data (30% of our total dataset) and achieved validation performance with an AUC of 0.647.
Some metabolites (cholate, tartrate, methanol, N-acetylglucosamine, butanone, caprate) were selected as the top 25 metabolites using the PLSDA approach in MetaboAnalyst. Both tartrate and cholate were elevated in the untrained group, which could be related to their diet. The diet of athletes is much more constant, and the diet was not standardized in the X-Adapt project. However, the decreased tartrate levels may suggest that tartrate supplementation is needed in the trained group to reduce metabolic stress, minimize muscle damage, improve hormone receptor levels, and promote recovery after resistance exercise [24,25]. L-carnitine L-tartrate supplementation increases carbohydrate oxidation rates. Endurance athletes in particular have higher carnitine uptake in skeletal muscle [24,25]. Tartrate is a nonhuman metabolite found in grapes, wine, and as an additive in foods [26]. Increased consumption of tartrate-containing foods and beverages also lowers cardiovascular risk factors such as LDL cholesterol [27,28]. Tartrate is part of glyoxylate and dicarboxylate metabolism, which was also observed in the enrichment analysis of metabolic pathways and enriched in the untrained group. Glyoxylate and dicarboxylate metabolic pathways were observed in young patients with major depressive disorder. Improving physical activity improved patients with major depressive disorder and additionally reduced other complications of cardiovascular disease [29,30]. Inactivity in the bed rest study (e.g., PlanHab) also led to the development of psychiatric problems after one week of bed rest, showing possible associations between inactivity, metabolism and mental health problems [4,5,31,32,33,34,35,36,37,38,39,40,41,42,43,44].
Cholate, on the other hand, is one of the primary bile acids that may be involved in the development of an atrophic state in myotubes [45] and in the invasion of human colon cancer cells [46], which can be observed in less active and untrained individuals. Bile acids in general have also been associated with obesity [47], higher BMI, elevated blood glucose levels [48], liver dysfunction [49] and cardiovascular health [50]. Bile acids in urine can be used for diagnostic purposes, as it has already been shown that bile acids in urine have lower variability and higher stability than bile acids in serum [51]. Elevated cholate concentrations have also been observed in patients with gastric cancer. In our work, increased concentrations of cholate were observed in untrained individuals. A meta-analysis has previously shown that regular physical activity can prevent gastric cancer [52,53,54]. A single training run in amateur runners resulted in a significant decrease in circulating bile acids. Recent studies have also shown that bile-acid concentrations were higher in less fit women than in fit women [55,56].
Polyamines such as lysine and cadaverine, which were also detected in our study, have also been associated with the development of various diseases described by the common term “metabolic syndrome”. It has already been shown that elevated cadaverine concentrations may correlate with intestinal disease or colon and liver cancer. Cadaverine was also elevated in the untrained group and is part of the glutathione metabolism previously described in men with type 1 diabetes [57]. Metabolic syndrome develops mainly due to inactivity or lack of exercise [58,59,60]. Lysine is involved in aminoacyl-tRNA biosynthesis and was increased in trained group. Aminoacyl-tRNA biosynthesis was associated with higher physical activity, a less sedentary lifestyle and high-intensity interval training [61,62,63]. Using metabolomes in stool and serum, the same metabolic pathway was identified as altered in endurance cross-country athletes, reflecting modifications in protein synthesis [64].
In contrast, 2-hydroxy-3-methyl- valerate was identified only with machine learning and was decreased in the trained group, confirming that it may also be involved in affecting physical function through peroxisome proliferator-activated receptor alpha (PPAR-α) activation, which is associated with microbial metabolism and insulin sensitivity [65]. PPAR-α is a hormone-receptor transcription factor involved in energy metabolism. Untrained participants in X-Adapt are less physically active and have increased levels of 2-hydroxy-3-methyl valerate, leading to possible activation of PPAR-α, as shown in functionally impaired older adults [65,66]. N6-acetyl-L-lysine is an acetylated amino acid that is increased in the trained group and plays an important role in regulating gene transcription, cell-cycle progression, apoptosis, DNA repair and cytoskeletal organization, also decreasing chances of Alzheimer’s disease shown on rats. Physical activity has previously been shown to reduce the risk of age-related Alzheimer’s disease [67,68] and metabolic syndrome [69].
We also observed that an increased pH increased the chance of classifying participants into a trained group. A lower urine pH was associated with chronic kidney disease [70], chronic heart failure [71] and metabolic syndrome [69,72,73].
Our analyses of the same 3-day-series data on a daily basis did not produce interpretable patterns of significant differences between the daily metabolome groups of the same 3-day sampling campaign after the correction for multiple comparisons (PERMANOVA; p > 0.05; npermutations = 5000) and were not reported. This corroborates our past observation [5] on the higher resolution of 3-day series of 1H-NMR metabolomes in contrast to single-day sampling.

2.3. Differences between Trained and Untrained Groups before and after Synchronizing Normoxic Training Campaign: The Extended Urine-Sample Collection

Our last analysis focused on the exploration of the extended urine-sample collection between trained and untrained (Figure 1B) to identify differences in morning urine metabolomes as a result of their original lifestyle and almost 2 weeks of 1h training at 50% Wpeak (i.e., 3-day exercise tests, 1 day rest, 10 days 1 h training at 50% Wpeak; Figure 1A). The results of PERMANOVA (p(same) = 0.003; npermutations = 5000; Figure S5) showed that in the trained group, an active lifestyle supported significantly different metabolomic fingerprints in comparison to the untrained group (Figure S5). The differences between the trained and untrained groups were no longer significant at the end of training (p = 0.226), while shared metabolomics features were present within each of the groups on the relation between pre- and post-training states (horizontal lines; Figure S5) as the significant differences persisted in relation to pretrained vs. post-untrained and pre-untrained vs. post-trained (diagonal lines; Figure S5). The results of this study suggest that exercise introduced changes in trained and untrained groups, making their endpoints not significantly different, and was accompanied by the concomitant decrease in the VO2max values (−3.2%) in trained and increase (+9.2%) in untrained groups [12,15,16,18].
When all eight groups of metabolomes (Figure 1B) were analyzed, it became apparent that the first introduction of controlled exercise at pre-exercise tests generated rather similar resting morning metabolomic urine makeup (i.e., short-term multivariate phenotype) in the two physiologically significantly different groups, while measurable changes within the exercise parameters (long-term multivariate phenotype (Figure 2 and Figure S2) were detected much later. Consequently, the frequency of these training bouts (i.e., life-long exercise) is in fact a crucial parameter for maintaining a healthy metabolomic phenotype and VO2max next to other exercise-related parameters. In contrast to WHO’s proposed 75 min to 150 min of vigorous- to moderate-intensity training, respectively, for adults per week [1,2], our study showed that a 5 times larger exercise input was effective at bringing the urine metabolomics makeup and VO2max values closer to the trained group, while obviously for the maintenance of an active lifestyle pursued by the trained group, much higher efforts would need to be invested. This finding is also in-line with the past observations on the difficulties in observing differences between training regimes [74], the effects of which subsided within 3 h after exercise, even in clinical populations [19]. Putting it simply, long-term exercise makes us rather similar in health, but a lack of it makes us different in disease. X-Adapt findings presented in this study on homogenizing effects of exercise are mirroring our past results from the PlanHab project on negative effects of inactivity [4,5,33,34].
To conclude, morning urine, especially as utilized in the form of 3-day sessions, has been shown to represent a good candidate biological matrix for delineation of active and inactive lifestyles in this study, detecting differences unobservable by single-day sampling. Resting morning urine metabolomes as a result of 1 h 50% Wpeak daily activity provided a good basis for planning future recommendations for the maintenance of health in adults, irrespective of the starting fitness value. The maintenance of systemic homeostasis and the response to nutritional and environmental challenges require the coordination of multiple organs and tissues. To respond to various metabolic demands, the human body integrates and builds upon a system of interorgan communication through which one tissue can affect metabolic pathways in a distant tissue. Dysregulation of these lines of communication through lack of exercise (sedentary lifestyle) and of highly energetic diets contribute to human pathologies, including obesity, diabetes, liver disease and atherosclerosis. Increasing exercise levels in the untrained apparently has the capacity to significantly reconstitute the interorgan communication towards the levels observed in the healthy trained cohort. In addition, recent technical advances such as data-driven bioinformatics on layers of information (microbiome, proteome, metabolome) expanded our understanding of the complexity of systemic metabolic crosstalk and its underlying mechanisms [75].

3. Materials and Methods

3.1. Project Description

In this study, the fourth campaign of the X-Adapt: Cross-adaptation between heat and hypoxia–novel strategy for performance and work-ability enhancement in various environments project (ARRS research project J5-9350) was utilized as source of exercise data and urine samples for 1H-NMR metabolomics analyses (Figure S1).
The main objective of the X-Adapt project was to determine the metabolic differences between trained and untrained individuals and the effects of 10 days of training on metabolism utilizing urinary metabolomics.
During the prescreening procedure, participants completed a graded exercise test on a cycle ergometer to determine their normoxic (environment with normal O2 concentrations (e.g., 21%)) maximal-rate oxygen consumption (VO2max) and maximal power output (Wpeak). Wpeak is defined as the highest workload sustained by incremental exercise until exhaustion. Aerobic fitness was defined using VO2max values. A VO2max of less than 45 mL·kg−1·min−1 or greater than 55 mL·kg−1·min−1 was considered a requirement for participation in the lower-fitness (untrained) or higher-fitness (trained) group, respectively, consistent with values reported in previous studies [13,14]. To further ensure that VO2max reflected participants’ true cardiorespiratory fitness levels, untrained participants were also required to not participate in organized sports. Cycling and walking for commuting to work were allowed. Accordingly, trained participants performed endurance-type activities (running, cycling, swimming) several times per week. Participants were informed that the aim of the study was to investigate the effects of a 10-day exercise protocol on aerobic performance in young males [12,15,16,17].
Twenty healthy young male volunteers were recruited to participate in the study. Inclusion criteria included males between the ages of 18 and 30, nonsmokers, and unmedicated. All participants lived near the sea and had not been exposed to altitudes > 1500 m or temperatures > 30 °C for at least 1 month before the start of the study, which took place in November and December 2018. None of the participants had a history of cardiorespiratory or hematologic disease. Participants were instructed to abstain from caffeine and alcohol consumption throughout the study. They were given detailed information about the study protocol and potential risks.
The study consisted of three parts: pretraining exercise testing, a 10-day exercise training program, followed by the post-training exercise testing (Figure 1).
During the pre-exercise tests, participants completed the same maximal exercise performance test on three consecutive but separate days under thermoneutral normoxic, thermoneutral hypoxic and hot normoxic conditions as described before [12,15,16,17]. The order of exercise tests was randomized and counterbalanced between participants. All tests were performed at the same time of day for a given participant (±1 h). Exercise training sessions took place in the morning hours (9:00–12:00). Participants were given a 24 h rest period before and after the 10-day training to minimize the contribution of fatigue during the exercise tests [12,15,16,17].
During the 10-day training session, all participants completed 60 min supervised cycling sessions daily for 10 days. Exercise was performed on a cycle ergometer (Daum, Electronic, Furth, Germany). During training, each participant pedalled at a preferred cadence (between 60 and 90 rpm), which they maintained throughout the experiment via visual and verbal feedback. Exercise intensity was relatively similar for all participants and was set at 50% of the Wpeak calculated from the individual Wpeak achieved during the preparatory graded normoxic exercise test. Participants were only informed of the time remaining until the end of the exercise session and were allowed to drink ad libitum during each exercise session. Heart rate and SpO2 were measured with a finger pulse oximeter (Wristox 3100 Nonin, Plymouth, MN, USA) at 5 min intervals. Ratings of perceived exertion (RPE; 6–20) was also recorded at 5 min intervals. Ambient temperature was maintained at 24 °C. The training room was well-ventilated so that normoxic and normocapnic (normal arterial carbon dioxide pressure) conditions prevailed during training. Participants completed all exercise sessions at the same time of day. No other exercise training was allowed during the study. Sessions were supervised by at least two researchers to record exercise data and ensure that all participants maintained the desired workload at all times [12,15,16,17].
After the completion of 10-day training session, postexercise tests were performed. All pre- and postexercise tests were performed in a laboratory 300 m above sea level (Ljubljana, Slovenia). Trials were performed on a cycle ergometer (Daum, Electronic, Furth, Germany) and included two phases: a 30 min steady-state workout immediately followed by incremental training to exhaustion. Before (pre) and after (post) the 10-day training protocol, participants performed three trials on three consecutive days. At normal temperature and normoxic conditions (NOR), participants breathed room air (pre: partial pressure of oxygen in the inspired air (PiO2) = 143.7 ± 0.8 mmHg, post: PiO2 = 143.4 ± 0.7 mmHg) and exercised under thermoneutral conditions (pre: Ta = 23.2 ± 0.7 °C and relative humidity (RH) = 47.2 ± 2.2%, post: Ta = 23.2 ± 0.5 °C and RH = 46.6 ± 5.9%). In the hypoxic condition (HYP), they inspired a hypoxic gas mixture (pre: PiO2 = 92.2 ± 1.5 mmHg, post: PiO2 = 93.2 ± 1.2 mmHg) and exercised in thermoneutral conditions (pre: Ta = 22.8 ± 0.5 °C and RH = 51.2 ± 1.2%, post: Ta = 22.5 ± 0.6 °C and RH = 51.5 ± 1.3%). In the hot condition (HE), the participants inspired room air (pre: PiO2 = 142.6 ± 1.7 mmHg, post: PiO2 = 142.7 ± 1.8 mmHg), but exercised in a hot environment (pre: Ta = 34.1 ± 0.9 °C, RH = 48.1 ± 4.2%, post: Ta = 34.1 ± 1.1 °C and RH = 49.8 ± 3.0%) [12,15,16,17].

3.2. Sample Collection

Urine samples were collected in four sessions for 3 consecutive days to form 3-day series of urine samples for all participants (Figure 1 and Figure S1): (i) 3-day baseline data of participants before the start of X-Adapt campaign, (ii) 3-day pre-exercise testing before 10-day 50% Wpeak training, (iii) 3-day sampling of the last days of 10-day exercise; and (iv) 3-day postexercise testing after the 10-day 50% Wpeak training. All obtained samples were frozen at −20 °C for further analysis as described before [4,5,76]. For simplicity, the X-Adapt urine-sample collection was used to denote samples collected during the X-Adapt pre-exercise and postexercise testing periods. The extended sample collection encompasses the urine samples collected before the actual onset of the campaign (baseline) and during the last three days of the 10-day 50% Wpeak training.

3.3. NMR Metabolomics

All collected samples were centrifuged (1.5 mL) at 10,000× g for 30 min to remove fine particles. Then, 600 µL of supernatant was mixed with 300 µL 1H-NMR buffer as described before [77] and stored at −25 °C until analysis. Before analysis, samples were thawed at room temperature and transferred into a 5 mm NMR tube. TSP was used as an internal standard for quantification, as described before [77].
A Bruker Avance NEO 600 MHz spectrometer equipped with a 24-sample SampleCase autosampler and a 5 mm HCN Cold probe was used for the acquisition of NMR spectra at 25 °C. The 1H NMR spectra of the samples were recorded with a spectral width of 9.0 kHz, relaxation delay of 2.0 s, 32 scans and 32 K data points. A double-pulsed field gradient spin echo (DPFGSE) pulse sequence was used for water suppression. Total correlated spectrum (TOCSY) was measured with 1H spectral widths of 7.0 kHz, 4096 complex points, a relaxation delay of 1.5 s, 32 transients and 144 time increments. An exponential and cosine-squared function were used for apodization. Zeros were filled before Fourier transform. TopSpin v. 4.0.9 software (Bruker, Billerica, MA, USA) was used for processing urine NMR spectra [4,5,76,78]. AlpsNMR R package was used for the visualization of example spectra [79].

3.4. Physicochemical Parameters of Urine Samples

Urine samples were thawed at room temperature, homogenized. Additional physical chemical parameters were recorded such as pH, conductivity, total dissolved solids and salinity using Pocket pro+ Multimeter 2 (Hach Company, Loveland, CO, USA).

3.5. Statistical Analysis and Machine Learning

The resulting spectra were consequently analyzed using targeted quantitative metabolomics using Chenomx NMR Suite version 8.6 (Chenomx, Inc., Edmonton, AB, Canada). For the latter, all spectra were randomly ordered for spectral fitting using the ChenomX profiler and the Human Metabolome Database (https://hmdb.ca/ (accessed on 24 April 2022)) compound names were used [80]. In this study, spectral deconvolution utilizing Chenomx and HMDB was used instead of the binning approaches with extensive normalization as described before [81,82]. An ensemble approach to data analysis was utilized, employing three different approaches to asymmetric sparse matrix data analysis, establishing significant differences between tested groups as follows: nonparametric MANOVA (PERMANOVA) [83], MetaboAnalyst [20,21,22], and JADBIO [23]. Heatmap of measured physiological parameters was generated using gplots R package. Data were normalized with scale function.
First, for PERMANOVA, each compound concentration obtained was analyzed as described before [4,11]. Box–Cox transformation was used. The significance of metabolic differences between various groups of samples was tested using 1-way and 2-way PERMANOVA, and expressed as an overlap in nonmetric multidimensional scaling (nm-MDS) trait space (using Euclidean distance measures). The stress function was used to select the dimensionality reduction, whereas Shepard’s plots were used to describe the correspondence between the target and obtained ranks. Benjamini–Hochberg significance correction for multiple comparisons was used as described before [4,5].
Second, for MetaboAnalyst, log- or cube-root transformation in connection to Mean or Pareto scaling was utilized as implemented in MetaboAnalyst, followed by supervised classification using partial least-squares-discriminant analysis (PLSDA) method and random forest (RF). Statistical power for the identification of significant differences before and after treatment was also calculated using MetaboAnalyst Statistical Power module.
Metabolite Set Enrichment (MSEA) was used to identify biologically significant patterns between quantitative metabolome data from different groups. HMDB compound names were used to link to the KEGG database. Enrichment analysis was performed using the globaltest package implemented in MetaboAnalyst. The enrichment ratio was calculated by dividing observed hits and expected hits.
Finally, Just Add Data Bio (JADBIO), a web-based auto-machine-learning platform for analyzing potential biomarkers [23], was used. JADBIO 1.4.0 with extensive tuning effort and 6 CPU was used to model various dataset selections next to the overall 336 metabolites observed in urine samples in all groups (trained vs. untrained) by splitting the total urine metabolite data into a training set and a test set in a 70:30 ratio. The training set was used for model training and the test set was used for model evaluation.
The resulting model can be obtained as part of Supplementary Material (File S2) and run with Java executor for the classification of novel urine samples based on 1H-NMR in further explorations.

Supplementary Materials

The following Electronic Supplementary Materials are available online at https://www.mdpi.com/article/10.3390/metabo12060473/s1, File S1: Figure S1: Schematic overview of the X-Adapt project campaigns as described before; Figure S2: Physical characteristics of trained (T) and untrained (UT) participants involved in X-Adapt project pretesting (pre) and post-testing (post); Figure S3: X-Adapt nmMDS ordination; Figure S4: Predictive performance of the highest-scoring models generated in this study; Figure S5: Schematic representation of PERMANOVA; Table S1: Sum of concentrations and numbers of metabolites observed in all groups; Table S2: The most important features for discriminating trained from untrained group; Table S3: Group characteristics of individuals, Instructions for running a model on a local machine. File S2: The resulting classification model can be run with Java executor for classification of novel urine samples based on 1H-NMR. File S3: Example 1H-NMR spectra characteristic of the trained and untrained groups. File S4: Data table containing metabolite information in micromolar concentration.

Author Contributions

Conceptualization, T.D. and B.S.; methodology, L.D. and B.S.; conceptualization for metabolomic analysis, J.P. and B.S., formal analysis, L.D. and B.S.; data curation, B.M., L.D. and B.S; exercise database, T.D., A.S., I.M; writing—original draft preparation, L.D. and B.S.; visualization, L.D. and B.S.; supervision, B.S.; project administration, A.S., T.D., I.M., B.S.; funding acquisition, T.D., I.M., B.S. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Slovenian Research Agency (ARRS) grant J5-9350 X- ADAPT: Cross-adaptation between heat and hypoxia—novel strategy for performance and work-ability enhancement in various environments project to T.D. L.D. acknowledges the support of the Slovenian Research Agency (SRA R#51867).

Institutional Review Board Statement

The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the National Committee for Medical Ethics at the Ministry of Health of the Republic of Slovenia, No. 0120-494/2018/9).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The data presented in this study are available in article and Supplementary Material.

Acknowledgments

L.D. acknowledges the support of Slovenian Research Agency (ARRS research Agency (SRA R#51867; MR+ call awarded to B.S.). Sabina Kolbl Repinc, Institute of Sanitary Engineering, Faculty of Civil and Geodetic Engineering, University of Ljubljana is acknowledged for the support and understanding during sample preparation and measurements. Klemen Pečnik, National Institute of Chemistry, Ljubljana is acknowledged for organizing the NMR runtime slots for our measurements. The authors thank the five independent reviewers and editor for their constructive comments that helped improve the manuscript.

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. Kelly, R.S.; Kelly, M.P.; Kelly, P. Metabolomics, physical activity, exercise and health: A review of the current evidence. Biochim. Biophys. Acta. Mol. Basis Dis. 2020, 1866, 165936. [Google Scholar] [CrossRef]
  2. Sallis, J.F.; Bull, F.; Guthold, R.; Heath, G.W.; Inoue, S.; Kelly, P.; Oyeyemi, A.L.; Perez, L.G.; Richards, J.; Hallal, P.C.; et al. Progress in physical activity over the Olympic quadrennium. Lancet 2016, 388, 1325–1336. [Google Scholar] [CrossRef]
  3. Bongiovanni, T.; Genovesi, F.; Nemmer, M.; Carling, C.; Alberti, G.; Howatson, G. Nutritional interventions for reducing the signs and symptoms of exercise-induced muscle damage and accelerate recovery in athletes: Current knowledge, practical application and future perspectives. Eur. J. Appl. Physiol. 2020, 120, 1965–1996. [Google Scholar] [CrossRef]
  4. Sket, R.; Debevec, T.; Kublik, S.; Schloter, M.; Schoeller, A.; Murovec, B.; Mikus, K.V.; Makuc, D.; Pecnik, K.; Plavec, J.; et al. Intestinal Metagenomes and Metabolomes in Healthy Young Males: Inactivity and Hypoxia Generated Negative Physiological Symptoms Precede Microbial Dysbiosis. Front. Physiol. 2018, 9, 198. [Google Scholar] [CrossRef] [Green Version]
  5. Šket, R.; Deutsch, L.; Prevoršek, Z.; Mekjavić, I.B.; Plavec, J.; Rittweger, J.; Debevec, T.; Eiken, O.; Stres, B. Systems View of Deconditioning During Spaceflight Simulation in the PlanHab Project: The Departure of Urine 1 H-NMR Metabolomes From Healthy State in Young Males Subjected to Bedrest Inactivity and Hypoxia. Front. Physiol. 2020, 11, 532271. [Google Scholar] [CrossRef]
  6. Al-Khelaifi, F.; Diboun, I.; Donati, F.; Botrè, F.; Asayrafi, M.; Georgakopoulos, C.; Suhre, K.; Yousri, N.A.; Elrayess, M.A. A pilot study comparing the metabolic profiles of elite-level athletes from different sporting disciplines. Sports Med.-Open 2018, 4, 2. [Google Scholar] [CrossRef] [Green Version]
  7. Danaher, J.; Gerber, T.; Wellard, R.M.; Stathis, C.G.; Cooke, M.B. The use of metabolomics to monitor simultaneous changes in metabolic variables following supramaximal low volume high intensity exercise. Metabolomics 2015, 12, 7. [Google Scholar] [CrossRef] [Green Version]
  8. Zafeiridis, A.; Chatziioannou, A.C.; Sarivasiliou, H.; Kyparos, A.; Nikolaidis, M.G.; Vrabas, I.S.; Pechlivanis, A.; Zoumpoulakis, P.; Baskakis, C.; Dipla, K.; et al. Global Metabolic Stress of Isoeffort Continuous and High Intensity Interval Aerobic Exercise: A Comparative 1 H NMR Metabonomic Study. J. Proteome Res. 2016, 15, 4452–4463. [Google Scholar] [CrossRef] [Green Version]
  9. Xiao, Q.; Moore, S.C.; Keadle, S.K.; Xiang, Y.B.; Zheng, W.; Peters, T.M.; Leitzmann, M.F.; Ji, B.T.; Sampson, J.N.; Shu, X.O.; et al. Objectively measured physical activity and plasma metabolomics in the Shanghai Physical Activity Study. Int. J. Epidemiol. 2016, 45, 1433–1444. [Google Scholar] [CrossRef]
  10. Emwas, A.H.; Roy, R.; McKay, R.T.; Tenori, L.; Saccenti, E.; Gowda, G.A.N.; Raftery, D.; Alahmari, F.; Jaremko, L.; Jaremko, M.; et al. NMR Spectroscopy for Metabolomics Research. Metabolites 2019, 9, 123. [Google Scholar] [CrossRef] [Green Version]
  11. Murovec, B.; Makuc, D.; Kolbl Repinc, S.; Prevorsek, Z.; Zavec, D.; Sket, R.; Pecnik, K.; Plavec, J.; Stres, B. 1H NMR metabolomics of microbial metabolites in the four MW agricultural biogas plant reactors: A case study of inhibition mirroring the acute rumen acidosis symptoms. J. Environ. Manag. 2018, 222, 428–435. [Google Scholar] [CrossRef]
  12. Sotiridis, A.; Debevec, T.; Ciuha, U.; McDonnell, A.C.; Mlinar, T.; Royal, J.T.; Mekjavic, I.B. Aerobic but not thermoregulatory gains following a 10-day moderate-intensity training protocol are fitness level dependent: A cross-adaptation perspective. Physiol. Rep. 2020, 8, e14355. [Google Scholar] [CrossRef] [Green Version]
  13. Jay, O.; Bain, A.R.; Deren, T.M.; Sacheli, M.; Cramer, M.N. Large differences in peak oxygen uptake do not independently alter changes in core temperature and sweating during exercise. Am. J. Physiol. Regul. Integr. Comp. Physiol. 2011, 301, R832–R841. [Google Scholar] [CrossRef] [PubMed]
  14. Montero, D.; Lundby, C. Refuting the myth of non-response to exercise training: ‘non-responders’ do respond to higher dose of training. J. Physiol. 2017, 595, 3377–3387. [Google Scholar] [CrossRef] [PubMed]
  15. Sotiridis, A.; Debevec, T.; McDonnell, A.C.; Ciuha, U.; Eiken, O.; Mekjavic, I.B. Exercise cardiorespiratory and thermoregulatory responses in normoxic, hypoxic and hot environment following 10-day continuous hypoxic exposure. J. Appl. Physiol. 2018, 125, 1284–1295. [Google Scholar] [CrossRef] [PubMed]
  16. Sotiridis, A.; Debevec, T.; Ciuha, U.; Eiken, O.; Mekjavic, I.B. Heat acclimation does not affect maximal aerobic power in thermoneutral normoxic or hypoxic conditions. Exp. Physiol. 2019, 104, 345–358. [Google Scholar] [CrossRef] [PubMed]
  17. Sotiridis, A. Independent and Combined Effects of Heat and Hypoxic Acclimation on Exercise Performance in Humans: With Particular Reference to Cross-Adaption. Ph.D. Thesis, Jozef Stefan Institute, Ljubljana, Slovenia, 2019. [Google Scholar]
  18. Armstrong, N.; Barker, A.R. Endurance training and elite young athletes. Med. Sport Sci. 2011, 56, 59–83. [Google Scholar] [CrossRef]
  19. Siopi, A.; Deda, O.; Manou, V.; Kellis, S.; Kosmidis, I.; Komninou, D.; Raikos, N.; Christoulas, K.; Theodoridis, G.A.; Mougios, V. Effects of Different Exercise Modes on the Urinary Metabolic Fingerprint of Men with and without Metabolic Syndrome. Metabolites 2017, 7, 5. [Google Scholar] [CrossRef] [Green Version]
  20. Chong, J.; Soufan, O.; Li, C.; Caraus, I.; Li, S.; Bourque, G.; Wishart, D.S.; Xia, J. MetaboAnalyst 4.0: Towards more transparent and integrative metabolomics analysis. Nucleic Acids Res. 2018, 46, W486–W494. [Google Scholar] [CrossRef] [Green Version]
  21. Chong, J.; Wishart, D.S.; Xia, J. Using MetaboAnalyst 4.0 for Comprehensive and Integrative Metabolomics Data Analysis. Curr. Protoc. Bioinform. 2019, 68, e86. [Google Scholar] [CrossRef]
  22. Pang, Z.; Chong, J.; Zhou, G.; de Lima Morais, D.A.; Chang, L.; Barrette, M.; Gauthier, C.; Jacques, P.É.; Li, S.; Xia, J. MetaboAnalyst 5.0: Narrowing the gap between raw spectra and functional insights. Nucleic Acids Res. 2021, 49, W388–W396. [Google Scholar] [CrossRef] [PubMed]
  23. Tsamardinos, I.; Charonyktakis, P.; Lakiotaki, K.; Borboudakis, G.; Zenklusen, J.C.; Juhl, H.; Chatzaki, E.; Lagani, V. Just Add Data: Automated Predictive Modeling and BioSignature Discovery. bioRxiv 2020, 1–46. [Google Scholar] [CrossRef]
  24. Spiering, B.A.; Kraemer, W.J.; Hatfield, D.L.; Vingren, J.L.; Fragala, M.S.; Ho, J.Y.; Thomas, G.A.; Häkkinen, K.; Volek, J.S. Effects of L-carnitine L-tartrate supplementation on muscle oxygenation responses to resistance exercise. J. Strength Cond. Res. 2008, 22, 1130–1135. [Google Scholar] [CrossRef] [PubMed]
  25. Abramowicz, W.N.; Galloway, S.D. Effects of acute versus chronic L-carnitine L-tartrate supplementation on metabolic responses to steady state exercise in males and females. Int. J. Sport Nutr. Exerc. Metab. 2005, 15, 386–400. [Google Scholar] [CrossRef]
  26. Vázquez-Fresno, R.; Llorach, R.; Urpi-Sarda, M.; Khymenets, O.; Bulló, M.; Corella, D.; Fitó, M.; Martínez-González, M.A.; Estruch, R.; Andres-Lacueva, C. An NMR metabolomics approach reveals a combined-biomarkers model in a wine interventional trial with validation in free-living individuals of the PREDIMED study. Metabolomics 2014, 11, 797–806. [Google Scholar] [CrossRef] [Green Version]
  27. Garcia-Perez, I.; Posma, J.M.; Chambers, E.S.; Nicholson, J.K.; Mathers, J.C.; Beckmann, M.; Draper, J.; Holmes, E.; Frost, G. An Analytical Pipeline for Quantitative Characterization of Dietary Intake: Application To Assess Grape Intake. J. Agric. Food Chem. 2016, 64, 2423–2431. [Google Scholar] [CrossRef] [Green Version]
  28. Domínguez-López, I.; Parilli-Moser, I.; Arancibia-Riveros, C.; Tresserra-Rimbau, A.; Martínez-González, M.A.; Ortega-Azorín, C.; Salas-Salvadó, J.; Castañer, O.; Lapetra, J.; Arós, F.; et al. Urinary Tartaric Acid, a Biomarker of Wine Intake, Correlates with Lower Total and LDL Cholesterol. Nutrients 2021, 13, 2883. [Google Scholar] [CrossRef]
  29. Chen, J.J.; Xie, J.; Li, W.W.; Bai, S.J.; Wang, W.; Zheng, P.; Xie, P. Age-specific urinary metabolite signatures and functions in patients with major depressive disorder. Aging 2019, 11, 6626–6637. [Google Scholar] [CrossRef]
  30. Gerber, M.; Beck, J.; Brand, S.; Cody, R.; Donath, L.; Eckert, A.; Faude, O.; Fischer, X.; Hatzinger, M.; Holsboer-Trachsler, E.; et al. The impact of lifestyle Physical Activity Counselling in IN-PATients with major depressive disorders on physical activity, cardiorespiratory fitness, depression, and cardiovascular health risk markers: Study protocol for a randomized controlled trial. Trials 2019, 20, 367. [Google Scholar] [CrossRef] [Green Version]
  31. Debevec, T.; Bali, T.C.; Simpson, E.J.; Macdonald, I.A.; Eiken, O.; Mekjavic, I.B. Separate and combined effects of 21-day bed rest and hypoxic confinement on body composition. Eur. J. Appl. Physiol. 2014, 114, 2411–2425. [Google Scholar] [CrossRef]
  32. Debevec, T.; Simpson, E.J.; Mekjavic, I.B.; Eiken, O.; Macdonald, I.A. Effects of prolonged hypoxia and bed rest on appetite and appetite-related hormones. Appetite 2016, 107, 28–37. [Google Scholar] [CrossRef] [PubMed]
  33. Sket, R.; Treichel, N.; Debevec, T.; Eiken, O.; Mekjavic, I.; Schloter, M.; Vital, M.; Chandler, J.; Tiedje, J.M.; Murovec, B.; et al. Hypoxia and Inactivity Related Physiological Changes (Constipation, Inflammation) Are Not Reflected at the Level of Gut Metabolites and Butyrate Producing Microbial Community: The PlanHab Study. Front. Physiol. 2017, 8, 250. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  34. Sket, R.; Treichel, N.; Kublik, S.; Debevec, T.; Eiken, O.; Mekjavic, I.; Schloter, M.; Vital, M.; Chandler, J.; Tiedje, J.M.; et al. Hypoxia and inactivity related physiological changes precede or take place in absence of significant rearrangements in bacterial community structure: The PlanHab randomized trial pilot study. PLoS ONE 2017, 12, e0188556. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  35. Debevec, T.; Ganse, B.; Mittag, U.; Eiken, O.; Mekjavic, I.B.; Rittweger, J. Hypoxia Aggravates Inactivity-Related Muscle Wasting. Front. Physiol. 2018, 9, 494. [Google Scholar] [CrossRef] [Green Version]
  36. Stavrou, N.A.M.; Debevec, T.; Eiken, O.; Mekjavic, I.B. Hypoxia Exacerbates Negative Emotional State during Inactivity: The Effect of 21 Days Hypoxic Bed Rest and Confinement. Front. Physiol. 2018, 9, 26. [Google Scholar] [CrossRef] [Green Version]
  37. Stavrou, N.A.M.; Debevec, T.; Eiken, O.; Mekjavic, I.B. Hypoxia Worsens Affective Responses and Feeling of Fatigue During Prolonged Bed Rest. Front. Psychol. 2018, 9, 362. [Google Scholar] [CrossRef]
  38. Strewe, C.; Zeller, R.; Feuerecker, M.; Hoerl, M.; Kumprej, I.; Crispin, A.; Johannes, B.; Debevec, T.; Mekjavic, I.; Schelling, G.; et al. PlanHab study: Assessment of psycho-neuroendocrine function in male subjects during 21 d of normobaric hypoxia and bed rest. Stress 2017, 20, 131–139. [Google Scholar] [CrossRef]
  39. Strewe, C.; Zeller, R.; Feuerecker, M.; Hoerl, M.; Matzel, S.; Kumprej, I.; Crispin, A.; Johannes, B.; Debevec, T.; Mekjavic, I.B.; et al. PlanHab Study: Consequences of combined normobaric hypoxia and bed rest on adenosine kinetics. Sci. Rep. 2018, 8, 1762. [Google Scholar] [CrossRef] [Green Version]
  40. Keramidas, M.E.; Kolegard, R.; Mekjavic, I.B.; Eiken, O. PlanHab: Hypoxia exaggerates the bed-rest-induced reduction in peak oxygen uptake during upright cycle ergometry. Am. J. Physiol. Heart Circ. Physiol. 2016, 311, H453–H464. [Google Scholar] [CrossRef]
  41. Louwies, T.; Jaki Mekjavic, P.; Cox, B.; Eiken, O.; Mekjavic, I.B.; Kounalakis, S.; De Boever, P. Separate and Combined Effects of Hypoxia and Horizontal Bed Rest on Retinal Blood Vessel Diameters. Investig. Ophthalmol. Vis. Sci. 2016, 57, 4927–4932. [Google Scholar] [CrossRef] [Green Version]
  42. Simpson, E.J.; Debevec, T.; Eiken, O.; Mekjavic, I.; Macdonald, I.A. PlanHab: The combined and separate effects of 16 days of bed rest and normobaric hypoxic confinement on circulating lipids and indices of insulin sensitivity in healthy men. J. Appl. Physiol. 2016, 120, 947–955. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  43. Sarabon, N.; Mekjavic, I.B.; Eiken, O.; Babic, J. The Effect of Bed Rest and Hypoxic Environment on Postural Balance and Trunk Automatic (Re)Actions in Young Healthy Males. Front. Physiol. 2018, 9, 27. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  44. Rullman, E.; Mekjavic, I.B.; Fischer, H.; Eiken, O. PlanHab (Planetary Habitat Simulation): The combined and separate effects of 21 days bed rest and hypoxic confinement on human skeletal muscle miRNA expression. Physiol. Rep. 2016, 4, e12753. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  45. Abrigo, J.; Gonzalez, F.; Aguirre, F.; Tacchi, F.; Gonzalez, A.; Meza, M.P.; Simon, F.; Cabrera, D.; Arrese, M.; Karpen, S.; et al. Cholic acid and deoxycholic acid induce skeletal muscle atrophy through a mechanism dependent on TGR5 receptor. J. Cell. Physiol. 2021, 236, 260–272. [Google Scholar] [CrossRef] [PubMed]
  46. Li, S.; Ung, T.T.; Nguyen, T.T.; Sah, D.K.; Park, S.Y.; Jung, Y.D. Cholic Acid Stimulates MMP-9 in Human Colon Cancer Cells via Activation of MAPK, AP-1, and NF-κB Activity. Int. J. Mol. Sci. 2020, 21, 3420. [Google Scholar] [CrossRef]
  47. Mercer, K.E.; Maurer, A.; Pack, L.M.; Ono-Moore, K.; Spray, B.J.; Campbell, C.; Chandler, C.J.; Burnett, D.; Souza, E.; Casazza, G.; et al. Exercise training and diet-induced weight loss increase markers of hepatic bile acid (BA) synthesis and reduce serum total BA concentrations in obese women. Am. J. Physiol. Endocrinol. Metab. 2021, 320, E864–E873. [Google Scholar] [CrossRef]
  48. Zheng, X.; Chen, T.; Zhao, A.; Ning, Z.; Kuang, J.; Wang, S.; You, Y.; Bao, Y.; Ma, X.; Yu, H.; et al. Hyocholic acid species as novel biomarkers for metabolic disorders. Nat. Commun. 2021, 12, 1487. [Google Scholar] [CrossRef]
  49. Alamoudi, J.A.; Li, W.; Gautam, N.; Olivera, M.; Meza, J.; Mukherjee, S.; Alnouti, Y. Bile acid indices as biomarkers for liver diseases I: Diagnostic markers. World J. Hepatol. 2021, 13, 433–455. [Google Scholar] [CrossRef]
  50. Pushpass, R.G.; Alzoufairi, S.; Jackson, K.G.; Lovegrove, J.A. Circulating bile acids as a link between the gut microbiota and cardiovascular health: Impact of prebiotics, probiotics and polyphenol-rich foods. Nutr. Res. Rev. 2021, 1–20. [Google Scholar] [CrossRef]
  51. Bathena, S.P.; Thakare, R.; Gautam, N.; Mukherjee, S.; Olivera, M.; Meza, J.; Alnouti, Y. Urinary bile acids as biomarkers for liver diseases II. Signature profiles in patients. Toxicol. Sci. Off. J. Soc. Toxicol. 2015, 143, 308–318. [Google Scholar] [CrossRef] [Green Version]
  52. Lyu, J.; Li, H.; Yin, D.; Zhao, M.; Sun, Q.; Guo, M. Analysis of eight bile acids in urine of gastric cancer patients based on covalent organic framework enrichment coupled with liquid chromatography-tandem mass spectrometry. J. Chromatogr. A 2021, 1653, 462422. [Google Scholar] [CrossRef] [PubMed]
  53. Singh, S.; Edakkanambeth Varayil, J.; Devanna, S.; Murad, M.H.; Iyer, P.G. Physical activity is associated with reduced risk of gastric cancer: A systematic review and meta-analysis. Cancer Prev. Res. 2014, 7, 12–22. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  54. Abioye, A.I.; Odesanya, M.O.; Abioye, A.I.; Ibrahim, N.A. Physical activity and risk of gastric cancer: A meta-analysis of observational studies. Br. J. Sports Med. 2015, 49, 224–229. [Google Scholar] [CrossRef] [PubMed]
  55. Danese, E.; Salvagno, G.L.; Tarperi, C.; Negrini, D.; Montagnana, M.; Festa, L.; Sanchis-Gomar, F.; Schena, F.; Lippi, G. Middle-distance running acutely influences the concentration and composition of serum bile acids: Potential implications for cancer risk? Oncotarget 2017, 8, 52775–52782. [Google Scholar] [CrossRef] [Green Version]
  56. Maurer, A.; Ward, J.L.; Dean, K.; Billinger, S.A.; Lin, H.; Mercer, K.E.; Adams, S.H.; Thyfault, J.P. Divergence in aerobic capacity impacts bile acid metabolism in young women. J. Appl. Physiol. 2020, 129, 768–778. [Google Scholar] [CrossRef]
  57. Darenskaya, M.A.; Chugunova, E.V.; Kolesnikov, S.I.; Grebenkina, L.A.; Semenova, N.V.; Nikitina, O.A.; Kolesnikova, L.I. Content of Carbonyl Compounds and Parameters of Glutathione Metabolism in Men with Type 1 Diabetes Mellitus at Preclinical Stages of Diabetic Nephropathy. Bull. Exp. Biol. Med. 2021, 171, 592–595. [Google Scholar] [CrossRef]
  58. Liu, R.; Li, Q.; Ma, R.; Lin, X.; Xu, H.; Bi, K. Determination of polyamine metabolome in plasma and urine by ultrahigh performance liquid chromatography-tandem mass spectrometry method: Application to identify potential markers for human hepatic cancer. Anal. Chim. Acta 2013, 791, 36–45. [Google Scholar] [CrossRef]
  59. Maráková, K.; Piešťanský, J.; Zelinková, Z.; Mikuš, P. Simultaneous determination of twelve biogenic amines in human urine as potential biomarkers of inflammatory bowel diseases by capillary electrophoresis–tandem mass spectrometry. J. Pharm. Biomed. Anal. 2020, 186, 113294. [Google Scholar] [CrossRef]
  60. Venäläinen, M.K.; Roine, A.N.; Häkkinen, M.R.; Vepsäläinen, J.J.; Kumpulainen, P.S.; Kiviniemi, M.S.; Lehtimäki, T.; Oksala, N.K.; Rantanen, T.K. Altered Polyamine Profiles in Colorectal Cancer. Anticancer. Res. 2018, 38, 3601–3607. [Google Scholar] [CrossRef]
  61. Tian, Q.; Corkum, A.E.; Moaddel, R.; Ferrucci, L. Metabolomic profiles of being physically active and less sedentary: A critical review. Metab. Off. J. Metab. Soc. 2021, 17, 68. [Google Scholar] [CrossRef]
  62. Castro, A.; Duft, R.G.; Ferreira, M.L.V.; Andrade, A.L.L.; Gáspari, A.F.; Silva, L.M.; Oliveira-Nunes, S.G.; Cavaglieri, C.R.; Ghosh, S.; Bouchard, C.; et al. Association of skeletal muscle and serum metabolites with maximum power output gains in response to continuous endurance or high-intensity interval training programs: The TIMES study—A randomized controlled trial. PLoS ONE 2019, 14, e0212115. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  63. Robinson, M.M.; Dasari, S.; Konopka, A.R.; Johnson, M.L.; Manjunatha, S.; Esponda, R.R.; Carter, R.E.; Lanza, I.R.; Nair, K.S. Enhanced Protein Translation Underlies Improved Metabolic and Physical Adaptations to Different Exercise Training Modes in Young and Old Humans. Cell Metab. 2017, 25, 581–592. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  64. Tabone, M.; Bressa, C.; García-Merino, J.A.; Moreno-Pérez, D.; Van, E.C.; Castelli, F.A.; Fenaille, F.; Larrosa, M. The effect of acute moderate-intensity exercise on the serum and fecal metabolomes and the gut microbiota of cross-country endurance athletes. Sci. Rep. 2021, 11, 3558. [Google Scholar] [CrossRef] [PubMed]
  65. Lustgarten, M.S.; Price, L.L.; Chalé, A.; Fielding, R.A. Metabolites related to gut bacterial metabolism, peroxisome proliferator-activated receptor-alpha activation, and insulin sensitivity are associated with physical function in functionally-limited older adults. Aging Cell 2014, 13, 918–925. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  66. Coen, P.M.; Jubrias, S.A.; Distefano, G.; Amati, F.; Mackey, D.C.; Glynn, N.W.; Manini, T.M.; Wohlgemuth, S.E.; Leeuwenburgh, C.; Cummings, S.R.; et al. Skeletal muscle mitochondrial energetics are associated with maximal aerobic capacity and walking speed in older adults. J. Gerontol. Ser. A Biol. Sci. Med. Sci. 2013, 68, 447–455. [Google Scholar] [CrossRef]
  67. Liu, Y.; Liu, Z.; Wei, M.; Hu, M.; Yue, K.; Bi, R.; Zhai, S.; Pi, Z.; Song, F.; Liu, Z. Pharmacodynamic and urinary metabolomics studies on the mechanism of Schisandra polysaccharide in the treatment of Alzheimer’s disease. Food Funct. 2019, 10, 432–447. [Google Scholar] [CrossRef]
  68. Gronek, P.; Balko, S.; Gronek, J.; Zajac, A.; Maszczyk, A.; Celka, R.; Doberska, A.; Czarny, W.; Podstawski, R.; Clark, C.C.T.; et al. Physical Activity and Alzheimer’s Disease: A Narrative Review. Aging Dis. 2019, 10, 1282–1292. [Google Scholar] [CrossRef] [Green Version]
  69. Maalouf, N.M.; Cameron, M.A.; Moe, O.W.; Adams-Huet, B.; Sakhaee, K. Low urine pH: A novel feature of the metabolic syndrome. Clin. J. Am. Soc. Nephrol. CJASN 2007, 2, 883–888. [Google Scholar] [CrossRef]
  70. Kraut, J.A.; Madias, N.E. Metabolic Acidosis of CKD: An Update. Am. J. Kidney Dis. Off. J. Natl. Kidney Found. 2016, 67, 307–317. [Google Scholar] [CrossRef] [Green Version]
  71. Otaki, Y.; Watanabe, T.; Takahashi, H.; Hasegawa, H.; Honda, S.; Funayama, A.; Netsu, S.; Ishino, M.; Arimoto, T.; Shishido, T.; et al. Acidic urine is associated with poor prognosis in patients with chronic heart failure. Heart Vessel. 2013, 28, 735–741. [Google Scholar] [CrossRef]
  72. Shimodaira, M.; Okaniwa, S.; Nakayama, T. Fasting Single-Spot Urine pH Is Associated with Metabolic Syndrome in the Japanese Population. Med. Princ. Pract. Int. J. Kuwait Univ. Health Sci. Cent. 2017, 26, 433–437. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  73. Hara, S.; Tsuji, H.; Ohmoto, Y.; Amakawa, K.; Hsieh, S.D.; Arase, Y.; Nakajima, H. High serum uric acid level and low urine pH as predictors of metabolic syndrome: A retrospective cohort study in a Japanese urban population. Metab. Clin. Exp. 2012, 61, 281–288. [Google Scholar] [CrossRef] [PubMed]
  74. Kistner, S.; Rist, M.J.; Krüger, R.; Döring, M.; Schlechtweg, S.; Bub, A. High-Intensity Interval Training Decreases Resting Urinary Hypoxanthine Concentration in Young Active Men-A Metabolomic Approach. Metabolites 2019, 9, 137. [Google Scholar] [CrossRef] [Green Version]
  75. Di Liegro, C.M.; Schiera, G.; Proia, P.; Di Liegro, I. Physical Activity and Brain Health. Genes 2019, 10, 720. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  76. Deutsch, L.; Osredkar, D.; Plavec, J.; Stres, B. Spinal Muscular Atrophy after Nusinersen Therapy: Improved Physiology in Pediatric Patients with No Significant Change in Urine, Serum, and Liquor 1H-NMR Metabolomes in Comparison to an Age-Matched, Healthy Cohort. Metabolites 2021, 11, 206. [Google Scholar] [CrossRef]
  77. Beckonert, O.; Keun, H.C.; Ebbels, T.M.; Bundy, J.; Holmes, E.; Lindon, J.C.; Nicholson, J.K. Metabolic profiling, metabolomic and metabonomic procedures for NMR spectroscopy of urine, plasma, serum and tissue extracts. Nat. Protoc. 2007, 2, 2692–2703. [Google Scholar] [CrossRef] [PubMed]
  78. Deutsch, L.; Stres, B. The Importance of Objective Stool Classification in Fecal 1H-NMR Metabolomics: Exponential Increase in Stool Crosslinking Is Mirrored in Systemic Inflammation and Associated to Fecal Acetate and Methionine. Metabolites 2021, 11, 172. [Google Scholar] [CrossRef]
  79. Madrid-Gambin, F.; Oller-Moreno, S.; Fernandez, L.; Bartova, S.; Giner, M.P.; Joyce, C.; Ferraro, F.; Montoliu, I.; Moco, S.; Marco, S. AlpsNMR: An R package for signal processing of fully untargeted NMR-based metabolomics. Bioinformatics 2020, 36, 2943–2945. [Google Scholar] [CrossRef]
  80. Wishart, D.S.; Guo, A.; Oler, E.; Wang, F.; Anjum, A.; Peters, H.; Dizon, R.; Sayeeda, Z.; Tian, S.; Lee, B.L.; et al. HMDB 5.0: The Human Metabolome Database for 2022. Nucleic Acids Res. 2022, 50, D622–D631. [Google Scholar] [CrossRef]
  81. Emwas, A.H.; Saccenti, E.; Gao, X.; McKay, R.T.; Dos Santos, V.A.P.M.; Roy, R.; Wishart, D.S. Recommended strategies for spectral processing and post-processing of 1D 1 H-NMR data of biofluids with a particular focus on urine. Metab. Off. J. Metab. Soc. 2018, 14, 31. [Google Scholar] [CrossRef] [Green Version]
  82. Dieterle, F.; Ross, A.; Schlotterbeck, G.; Senn, H. Probabilistic quotient normalization as robust method to account for dilution of complex biological mixtures. Application in 1H NMR metabonomics. Anal. Chem. 2006, 78, 4281–4290. [Google Scholar] [CrossRef] [PubMed]
  83. Legendre, P.; Legendre, L.F.J. Numerical Ecology, 3rd ed.; Elsevier: Amsterdam, The Netherlands, 2012; Volume 24, p. 1006. [Google Scholar]
Metabolites 12 00473 g001a 550Metabolites 12 00473 g001b 550
Figure 1. A schematic outline of the X-Adapt project with the two sampling collections designated below: the X-Adapt urine-sample collection and the extended sample collection (A). The extended sampling was conducted at days −3, −2, −1 day before the start of the campaign and at days 13, 14, 15 of the X-Adapt campaign 4. T—trained; UT—untrained group of participants. Blue arrows indicate sampling days within each of the four 3-day urine-sampling series. The X-Adapt urine-sample collection thus encompasses samples collected during the X-Adapt pretesting and post-testing periods. The extended sample collection encompasses the urine samples collected before the actual onset of the campaign (baseline) and during the last three days of training (days 13, 14, 15 of the campaign). For simplicity, the collection days are linked by hyphens to mark the compatible datasets. (B) A schematic representation of the X-Adapt urine-sample collection and the extended sample-collection groups with their respective analyses and comparisons delineated with lines. Solid and dashed lines designate significant and not significant differences between the groups. Analyses were conducted on overall group, sample collection and daily basis separately.
Figure 1. A schematic outline of the X-Adapt project with the two sampling collections designated below: the X-Adapt urine-sample collection and the extended sample collection (A). The extended sampling was conducted at days −3, −2, −1 day before the start of the campaign and at days 13, 14, 15 of the X-Adapt campaign 4. T—trained; UT—untrained group of participants. Blue arrows indicate sampling days within each of the four 3-day urine-sampling series. The X-Adapt urine-sample collection thus encompasses samples collected during the X-Adapt pretesting and post-testing periods. The extended sample collection encompasses the urine samples collected before the actual onset of the campaign (baseline) and during the last three days of training (days 13, 14, 15 of the campaign). For simplicity, the collection days are linked by hyphens to mark the compatible datasets. (B) A schematic representation of the X-Adapt urine-sample collection and the extended sample-collection groups with their respective analyses and comparisons delineated with lines. Solid and dashed lines designate significant and not significant differences between the groups. Analyses were conducted on overall group, sample collection and daily basis separately.
Metabolites 12 00473 g001aMetabolites 12 00473 g001b
Metabolites 12 00473 g002 550
Figure 2. A nmMDS representation of physical parameters (n = 39) measured in X-Adapt project (UTpre (green)—untrained pre-exercise testing, UT-post (red)—untrained postexercise testing, Tpre (black)—trained pre-exercise testing, Tpost (blue)—trained postexercise testing). Stress value of nmMDS was 0.185. Please also see Figure S2 for more details.
Figure 2. A nmMDS representation of physical parameters (n = 39) measured in X-Adapt project (UTpre (green)—untrained pre-exercise testing, UT-post (red)—untrained postexercise testing, Tpre (black)—trained pre-exercise testing, Tpost (blue)—trained postexercise testing). Stress value of nmMDS was 0.185. Please also see Figure S2 for more details.
Metabolites 12 00473 g002
Metabolites 12 00473 g003 550
Figure 3. Heatmap representation of the underlying multivariate exercise physicochemical parameters measured in the X-Adapt project. Dendrogram clustering shows differences between the trained and untrained groups in these parameters before and after training. Higher-resolution heatmap can be found in electronic supplementary material. Abbreviations: body surface area (BSA), stroke volume (SV), hearth rate (HR) cardiac output (CO), rectal temperature (Tre), maximal power output (Wpeak), peak power output (PPO), gross mechanical efficiency (GME), respiratory quotient (RQ), cardiac index (CI). Please see Figure S2 for the heatmap on representation per sample basis of all participants before and after exercise performance.
Figure 3. Heatmap representation of the underlying multivariate exercise physicochemical parameters measured in the X-Adapt project. Dendrogram clustering shows differences between the trained and untrained groups in these parameters before and after training. Higher-resolution heatmap can be found in electronic supplementary material. Abbreviations: body surface area (BSA), stroke volume (SV), hearth rate (HR) cardiac output (CO), rectal temperature (Tre), maximal power output (Wpeak), peak power output (PPO), gross mechanical efficiency (GME), respiratory quotient (RQ), cardiac index (CI). Please see Figure S2 for the heatmap on representation per sample basis of all participants before and after exercise performance.
Metabolites 12 00473 g003
Metabolites 12 00473 g004 550
Figure 4. PLSDA ordination of metabolomics signatures present in the trained and untrained groups (a,b) VIP scores of the most important metabolites separating the two groups. The three-day series of urine samples of trained and untrained groups were analyzed with MetaboAnalyst. Prior to the PLSDA analysis, concentrations were transformed with Log10 normalization and scaled with Mean Centering approach. Each dot represents one sample of participant per day.
Figure 4. PLSDA ordination of metabolomics signatures present in the trained and untrained groups (a,b) VIP scores of the most important metabolites separating the two groups. The three-day series of urine samples of trained and untrained groups were analyzed with MetaboAnalyst. Prior to the PLSDA analysis, concentrations were transformed with Log10 normalization and scaled with Mean Centering approach. Each dot represents one sample of participant per day.
Metabolites 12 00473 g004
Metabolites 12 00473 g005 550
Figure 5. Primary bile-acid biosynthesis, glutathione metabolism and aminoacyl-tRNA biosynthesis were enriched in untrained group, based on increased levels of cholate (primary bile-acids biosynthesis), cadaverine (glutathione metabolism) and L-tryptophan and L-cysteine (aminoacyl-tRNA biosynthesis).
Figure 5. Primary bile-acid biosynthesis, glutathione metabolism and aminoacyl-tRNA biosynthesis were enriched in untrained group, based on increased levels of cholate (primary bile-acids biosynthesis), cadaverine (glutathione metabolism) and L-tryptophan and L-cysteine (aminoacyl-tRNA biosynthesis).
Metabolites 12 00473 g005
Metabolites 12 00473 g006 550
Figure 6. The receiver operator curve (ROC) (a) plot and PCA plot (b) of modeled data. The dimensionality reduction was performed within JADBio on a subset of the original dataset, keeping only the features included in the first signature. Features were standardized with statistical normalization ((x − µ)/σ). A total of 155 samples were included in this analysis for training the model from the entire dataset. A total of 72 samples belonged to trained and 82 samples belonged to untrained group.
Figure 6. The receiver operator curve (ROC) (a) plot and PCA plot (b) of modeled data. The dimensionality reduction was performed within JADBio on a subset of the original dataset, keeping only the features included in the first signature. Features were standardized with statistical normalization ((x − µ)/σ). A total of 155 samples were included in this analysis for training the model from the entire dataset. A total of 72 samples belonged to trained and 82 samples belonged to untrained group.
Metabolites 12 00473 g006
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Deutsch, L.; Sotiridis, A.; Murovec, B.; Plavec, J.; Mekjavic, I.; Debevec, T.; Stres, B. Exercise and Interorgan Communication: Short-Term Exercise Training Blunts Differences in Consecutive Daily Urine 1H-NMR Metabolomic Signatures between Physically Active and Inactive Individuals. Metabolites 2022, 12, 473. https://doi.org/10.3390/metabo12060473

AMA Style

Deutsch L, Sotiridis A, Murovec B, Plavec J, Mekjavic I, Debevec T, Stres B. Exercise and Interorgan Communication: Short-Term Exercise Training Blunts Differences in Consecutive Daily Urine 1H-NMR Metabolomic Signatures between Physically Active and Inactive Individuals. Metabolites. 2022; 12(6):473. https://doi.org/10.3390/metabo12060473

Chicago/Turabian Style

Deutsch, Leon, Alexandros Sotiridis, Boštjan Murovec, Janez Plavec, Igor Mekjavic, Tadej Debevec, and Blaž Stres. 2022. "Exercise and Interorgan Communication: Short-Term Exercise Training Blunts Differences in Consecutive Daily Urine 1H-NMR Metabolomic Signatures between Physically Active and Inactive Individuals" Metabolites 12, no. 6: 473. https://doi.org/10.3390/metabo12060473

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop