Comparison of the Metabolites of Water Polo Players before and after Competition by the Metabolomic Approach.

Background: The metabolic characteristics of body fluids of excellent water polo players before and after competition have not been reported. The purpose of the study was to compare the metabolites in the urine of water polo players before and after competition by 1H-NMR-based metabolomic approach.
Methods: Twenty-six young water polo players participated in the study voluntarily. The urine and blood samples of players were collected one week before competition (A), immediately after competition (B), and one week after competition (C). Metabolomic analysis was conducted on the urine samples. Urine routine items and biochemical indicators in blood samples were detected.
Results: Metabolomic results showed that the contents of eleven metabolites including lactic acid, acetoacetate, and succinic acid in the urine of the subjects increased and four metabolites such as dimethylamine, choline, and glucose decreased at stage B. Most metabolites at stage C had basically returned to the levels at stage A. Pyruvate metabolism, pantothenate and CoA biosynthesis, synthesis, and degradation of ketone bodies were mainly involved in the above process. Urine conventional analysis results showed that the urine pH decreased dramatically and the levels of PRO and URO significantly increased at stage B, and the three indicators had similar values between stages A and C. The other indicators did not have obvious difference among the three stages. Analysis of blood biochemical indicators showed that the levels of LDH, BUN, CK, and AST significantly increased at stage B and did not show an obvious difference between stages A and C. The results are helpful for coaches to arrange the athletes' diet reasonably and to conduct scientific training for athletes.

1. Introduction

Water polo is an intermittent and high-intensity team sport that requires endurance, strength, swimming speed, agility, tactical awareness, and specific technical skills, with the objective of scoring more goals than the opposite team [1]. Besides typically good muscle and taller height, water polo players require a combination of skills, a high level of physical fitness, and tactical skills. For water polo players, the major metabolic way is that aerobic metabolism alternates with anaerobic metabolism, mainly the latter [2]. During the competition, about 40–50% of players' energy comes from aerobic energy supply, 35–40% from glycolysis energy supply, and the others from phosphoric acid energy supply [3]. It is reported that there is a significant difference in the heart rate of female athletes between training and competition. There has been a lot of research on the head impact, body injury, and fatigue that water polo players may suffer [4, 5].

Metabolomics is a new research method after genomics and proteomics. It is a branch of system biology which aims at information modeling and system integration based on cluster index analysis, high-throughput detection, and data processing [6]. Metabolomics provides a comprehensive and simultaneous analysis of the metabolic profile of metabolite changes occurring in living organisms in response to pathophysiological stimuli and/or genetic modification [7]. At present, it has been applied in the scientific field of sports to analyze low molecular weight metabolites of athletes in a specific period of time and then systematically evaluate their rich changes [8, 9].

Now, the training monitoring of water polo mainly focuses on the technical and tactical analysis and the study of the influence of different training methods on the biochemical indexes such as lactate dehydrogenase (LDH) and creatine kinase (CK) in the athletes' blood by using conventional physiological and biochemical methods [10, 11]. However, the metabolic characteristics of body fluids of excellent water polo players before and after competition and during the recovery period have not been reported. Based on 1H-nuclear magnetic resonance (NMR) metabolomic technique combined with physiological and biochemical measurement, endogenous urine small molecule metabolites and urine and blood biochemical indexes of water polo players before and after a national competition were studied to systematically understand the metabolic characteristics and adaptation mechanism of water polo players in competition and thus provide a theoretical basis for scientific training.

2. Materials and Methods

2.1. Participants

Water polo athletes of Shanxi University Polo Team participated in the study, including 13 male and 13 female athletes, each of which has a first-class or higher sport level certificate. The team, respectively, won the champion of high-level B group of men and women in the 2nd National University Water Polo Championship organized in Kunming, Yunnan, China, on May 2018. All the subjects voluntarily participated in the study and personally signed the informed consent. The protocol of the study was reviewed and approved by the Ethical Committee of Shanxi University. All subjects were in good health without taking medicine, being sick, drinking, or staying up late within a month before the competition. The basic information of the players is listed in Table 1.

Table 1

Basic information of subjects.

Gender	Male (n=13)	Female (n=13)
Age (years)	20.36 ± 1.12	20.27 ± 0.79
Weight (kg)	73.13 ± 4.84	57.81 ± 3.37
Height (cm)	180.55 ± 5.68	166.73 ± 5.14
Training years (years)	14.18 ± 1.83	13.82 ± 1.40

2.2. Training Routine

One week before competition (A), immediately after competition (B), and one week after competition (C) were selected as experimental time points. The precompetition training was conducted according to the coach's training plan. Each player performed underwater training from the 1st to 5th day, land strength and flexibility training on the 6th day, and rest on the 7th day. The competition lasted three days.

2.3. Collection and Treatment of Urine and Blood Samples

The urine and blood samples of participants were collected in stages A, B, and C of the 2nd National University Water Polo Championship.

The urine samples were labeled in two tubes. One portion was used for routine urine detection, and the other one was centrifuged at 4°C with 13000 rpm for 10 min and then the supernatant was stored at −80°C for NMR detection.

Blood samples were taken from players with empty stomachs in stages A and C and blood samples were taken from players immediately after competition (stage B). The blood samples were immediately placed in anticoagulant tubes with heparin sodium and then labeled. The samples were centrifuged at 4°C with 3000 rpm for 10 min and then the supernatant was stored at −80°C for biochemical index determination.

2.4. Metabolomic Analysis on the Urine Samples

500 μL thawed urine samples and 200 μL phosphate buffers (pH 7.4, including H2O, 0.2 M Na2HPO4/NaH2PO4, 0.03% TSP) were well mixed and then centrifuged at 4°C with 13000 rpm for 20 min. 600 μL supernatant was transferred into a nuclear magnetic tube for NMR detection with a 600 MHz spectrometer (Bruker, AVANCE III HD, Germany).

The urine nuclear magnetic maps were processed using Mestrenova (Mestrelab Research, Santiago de Compostela) software. All maps were calibrated with TSP chemical displacement δ 0.00; then, the baseline adjustment and phase transfer were conducted, and water peak and urea peak δ 4.59–4.60 were manually carried out. After the interval of δ 0.00–9.00 was divided into segments, all the integral data were normalized. Finally, the data were imported into Excel for multivariate statistical analysis.

SIMCA-P 14.1 (Umetrics, Sweden) was used for data processing. By principal component analysis (PCA), the initial distribution of sample data may be intuitively seen, which reflected the inherent nature of the samples. The different metabolites among samples were identified by using partial least squares discriminant analysis (PLS-DA) combined with orthogonal partial least squares discriminant analysis (OPLS-DA).

2.5. Determination of Biochemical Indicators in Urine and Blood Samples

Urine protein (PRO), red blood cell (RBC), urobilinogen (URO), ketone body (KET), leucocyte (LEU), glucose (GLU), nitrite (NIT), bilirubin (BIL), and urine specific gravity (USG) of the subject urine at different stages were detected on the spectrophotometer (Beckman, AU5800, America) by using a kit that was purchased from Nanjing Jiancheng Institute of Bioengineering, China. Urine pH values were measured on the pH meter (Fangzhou, PHS-3C+, Chengdu, China).

The biochemical indexes of blood including LDH, blood urea nitrogen (BUN), CK, and aspartate aminotransferase (AST) were detected on the fully automatic biochemical analyser (Beckman, AU5800, America). The biochemical reagents were purchased from Nanjing Jiancheng Institute of Bioengineering, China.

2.6. Health Risk Assessment Model of Chemical Carcinogens

SPSS 22.0 was used for statistical analysis. Single-factor variance and multivariate statistical analysis were conducted among the basic data, biochemical indicators of blood, and differential metabolites. The data were expressed by means ± SD; P < 0.05 was a significant difference.

3. Results

3.1. Identification of Metabolites in Urine of Water Polo Players by 1H-NMR

Thirty metabolites, such as lactic acid, succinic acid, and glycine, as shown in Table 2, were identified from urine metabolic profiles (Figure 1) by referring to Human Metabolome Database (HMDB, http://www.hmdb.ca/) and Biological Magnetic Resonance Data Bank (BMRDB, http://www.bmrb.wisc.edu/), consulting relevant literature and comparing the corresponding chemical shift of metabolites.

Table 2

Identification of major metabolites in players' urine maps.

No.	Metabolites	δ (ppm)
1	3-Aminoisobutyrate	1.19(d), 1.20(d)
2	Lactic acid	1.34(d), 1.33(d)
3	N-Acetyl-glycoproteins	2.05(s)
4	Glutamine	2.13(s)
5	Acetoacetate	2.27(s)
6	Glutamate	2.35(s)
7	Citrate	2.55(s)
8	Citric acid	2.57(s)
9	Dimethylacetamide	2.68(s)
10	Dimethylamine	2.71(d), 2.73(d)
11	Succinimide	2.78(s)
12	Succinic acid	2.81(s)
13	Dimethylglycine	2.95(s)
14	Creatinine	3.02(s), 3.05(s), 4.06(s)
15	Diethyl malonate	3.12(s)
16	Choline	3.20(d), 3.21(d)
17	Carnitine	3.23(s)
18	Trimethylamine N-oxide	3.27(s)
19	Taurine acid	3.28(s)
20	Scyllo-inositol	3.37(s)
21	Glucose	3.45(d)
22	Glycine	3.57(s)
23	Phenylacetylglycine	3.67(s), 7.35(s), 7.37(q)
24	Ethanol	3.68(s)
25	β-Glucose	3.75(s)
26	Guanidinoacetic acid	3.81(s)
27	Pantothenic acid	3.96(s)
28	Hippurate	3.97∼3.98(d), 7.56(s), 7.57(s), 7.83(s), 7.85(q)
29	Guanine	7.68(s)
30	Hypoxanthine	8.13(d), 8.14(d)

Figure 1

Diagram of urine 1H-NMR metabolites in water polo players.

3.2. Multivariate Statistical Analysis of NMR Data

Multivariate statistical analysis was used to analyze the metabolites of the urine NMR spectra at the different time points, and to find out the differential metabolites. Firstly, PCA scatter plot was used to describe the data in general (data not shown). The contractility of sports practice is small, the variables are relatively complex, and the effects of irrelevant variables on experiments and the self-errors of samples cannot be ignored; therefore, data need to be further processed. PLS-DA was used to analyze the metabolites of A and B groups. As shown in Figure 2(a), the metabolites of A and B groups separated roughly. It can be seen from Figure 2(b) that the R2 and Q2 values produced by the permutation model verification were less than the original values (R2 of 0.987 and Q2 of 0.968), indicating that the model was reliable and effective. Here, R2 represents the interpretation ability of the model and Q2 represents the prediction ability of the model. These data indicated that there were different metabolic in players immediately after the competition compared with the player one week before the competition. Based on the correct PLS-DA model, OPLS-DA analysis was conducted to further determine the metabolic differences at different time points. The metabolites of A and B groups completely separated well (Figure 2(c)). The corresponding S-plot (Figure 2(d)) showed that levels of some components in urine increased and some decreased at B point, as compared with A point. The differences were also significant in terms of fitting both VIP > 1 and P < 0.05. In the search for metabolic biomarkers, multivariate discrimination models between two classes of subjects/samples are used. PLS-DA is one of the most used methods. If a statistically significant discrimination between two classes, for example, the cases and controls classes, can be found, then the model parameters can be interpreted for their discriminating power and metabolic biomarkers can be found.

Figure 2

(a) PLS-DA scatter plot. (b) PLS-DA model validation plot. (c) OPLS-DA scatter plot. (d) OPLS-DA S-plot between A group and B group.

3.3. One-Way ANOVA Analysis of Differential Metabolites

The results of one-way ANOVA analysis of the metabolites are shown in Figure 3. It can be easily found that fifteen metabolites were significantly different between time points of A and B. The levels of lactic acid, acetoacetate, carnitine, trimethylamine N-oxide, and ethanol in the urine of the subjects at B time point were extremely significantly increased (P < 0.001), phenylacetylglycine and guanine obviously increased (P < 0.01), and succinic acid, diethyl malonate, β-glucose, and pantothenic acid increased (P < 0.05), while the levels of dimethylamine and choline extremely significantly decreased (P < 0.001) and glucose and hippurate decreased (P < 0.05), compared to corresponding compounds of the A group. Compared with B time point, succinic acid, diethyl malonate, β-glucose, and hippurate levels did not visibly change at C time point; acetoacetate and ethanol levels extremely significantly decreased (P < 0.001); lactic acid level significantly decreased (P < 0.01); and carnitine, trimethylamine N-oxide, phenylacetylglycine, and guanine levels decreased (P < 0.05), while dimethylamine, choline, glucose, and pantothenic acid levels increased (P < 0.05). Compared with A time point, only the levels of pantothenic acid and guanine in the urine of the subjects at C time point increased (P < 0.05), the other metabolites had no significant change, indicating that most metabolites had basically returned to the levels at the week before the competition.

Figure 3

(a, b) Comparison of peak areas of different metabolites in urine 1H-NMR spectra of subjects at three time points (A, B, and C). P < 0.001, P < 0.01, and 0.01 < P < 0.05, compared with A group. ###P < 0.001, ##P < 0.01, and #0.01 < P < 0.05, compared with B group. (c, d) Comparison of peak areas of different metabolites in urine 1H-NMR spectra of subjects at three time points (A, B, and C). P < 0.001, P < 0.01, and 0.01 < P < 0.05, compared with A group. ##P < 0.001, ###P < 0.001, ##P < 0.01, and #0.01 < P < 0.05, compared with B group. (e, f) Comparison of peak areas of different metabolites in urine 1H-NMR spectra of subjects at three time points (A, B, and C). P < 0.001, P < 0.01, and 0.01 < P < 0.05, compared with A group. ##P < 0.001, ###P < 0.001, ##P < 0.01, and #0.01 < P < 0.05, compared with B group. (g, h) Comparison of peak areas of different metabolites in urine 1H-NMR spectra of subjects at three time points (A, B, and C). P < 0.001, P < 0.01, and 0.01 < P < 0.05, compared with A group. ##P < 0.001, ###P < 0.001, and ##P < 0.01, #0.01 < P < 0.05, compared with B group. (i, j) Comparison of peak areas of different metabolites in urine 1H-NMR spectra of subjects at three time points (A, B, and C). P < 0.001, P < 0.01, and 0.01 < P < 0.05, compared with A group. ##P < 0.001, ###P < 0.001, ##P < 0.01, and #0.01 < P < 0.05, compared with B group. (k, l) Comparison of peak areas of different metabolites in urine 1H-NMR spectra of subjects at three time points (A, B, and C). P < 0.001, P < 0.01, and 0.01 < P < 0.05, compared with A group. ##P < 0.001, ###P < 0.001, ##P < 0.01, and #0.01 < P < 0.05, compared with B group. (m, n) Comparison of peak areas of different metabolites in urine 1H-NMR spectra of subjects at three time points (A, B, and C). P < 0.001, P < 0.01, 0.01 < P < 0.05, compared with A group. ##P < 0.001, ###P < 0.001, ##P < 0.01, and #0.01 < P < 0.05, compared with B group. (o) Comparison of peak areas of different metabolites in urine 1H-NMR spectra of subjects at three time points (A, B, and C). P < 0.001, P < 0.01, and 0.01 < P < 0.05, compared with A group. ##P < 0.001, ###P < 0.001, ##P < 0.01, and #0.01 < P < 0.05, compared with B group.

The correlation among the different metabolites in athletes' urine is reflected in Figure 4. Phenylacetylglycine had a strong positive correlation with ethanol and guanine while a strong negative correlation was, respectively, found between pantothenic acid and dimethylamine and between diethyl malonate and hippuric acid.

Figure 4

Correlation analysis of differential metabolites.

3.4. Metabolic Pathways Involved in Differential Metabolites

The above 15 differential metabolites were introduced into MetaboAnalyst 4.0 (http://www.metaboanalyst.ca) for channel enrichment analysis. According to Holm P value, false discovery rate (FDR), and impact value, three important metabolic pathways were screened: pyruvate metabolism, pantothenate and CoA biosynthesis, synthesis, and degradation of ketone bodies. The detailed information is shown in Figure 5.

Figure 5

(a) Pathway analysis of metabolites. (b) Analysis of metabolic pathways.

3.5. Urine Routine Test

Except for the items that are listed in Table 3, the others of urine routine test were negative results. It was easily found that the urine pH at B time point decreased significantly (P < 0.05), and PRO (P < 0.05) and URO (P < 0.01) increased significantly, compared with A time point. The urine pH at C time point was significantly higher (P < 0.01), and PRO (P < 0.05) and URO (P < 0.01) were slower significantly than those at with B time point. The other indicators remained almost unchanged at the different time points.

Table 3

Urine routine indicators of subjects at different time points.

Items	One week before competition (A)	Immediately after competition (B)	One week after competition (C)
Urine pH value	6.16 ± 0.32	5.84 ± 0.39∗	6.34 ± 0.39##
PRO (%)	9.66 ± 1.02	18.45 ± 1.73∗	9.40 ± 0.82#
URO (%)	5.17 ± 0.84	36.25 ± 2.88∗∗	13.6 ± 0.21#

P < 0.05 and P < 0.01 compared with A group; #P < 0.05 and ##P < 0.01 compared with B group.

3.6. Changes of Blood Biochemical Indexes

Table 4 shows that the levels of LDH, CK, AST, and BUN in the blood of the subjects at the A stage were significantly higher than those at the B stage (P < 0.01), while the levels of all four components at the C stage were significantly lower than those at the B stage and did not show a difference compared to those at A stage, indicating that the levels of LDH, CK, AST, and BUN in the recovery period returned to normal levels the week before the competition.

Table 4

The biochemical index levels in the blood of the subjects at different time points.

Items	One week before competition (A)	Immediately after competition (B)	One week after competition (C)
LDH (U/L)	2428.87 ± 302.99	2843.65 ± 260.02∗∗	2539.24 ± 359.65#
BUN (mM)	4.70 ± 0.75	6.29 ± 0.99∗∗	5.00 ± 0.56#
CK (U/mL)	0.48 ± 0.27	1.20 ± 0.50∗∗	0.54 ± 0.28##
AST (U/L)	8.05 ± 1.74	11.12 ± 2.60∗∗	8.11 ± 2.00##

P < 0.01 compared with A group; #P < 0.05 and ##P < 0.01 compared with B group.

4. Discussion

Water polo is a rigorous, physically demanding contact sport that requires athletes to perform repeated overhead movements. It is rapidly growing and gaining popularity in many countries. Water polo players not only require a delicate balance of internal rotator strength for performance in throwing, swimming, and defending but also need sufficient external rotator strength to decelerate and dissipate force. Although sweat losses of water polo players are much less than those of land-based players, specific knowledge allows for appropriate planning of carbohydrate intake strategies for match play and training. Urinary metabolomics data showed that the metabolic characteristics of water polo players changed significantly immediately after the competition compared with the week before the competition. The levels of some organic acids and glucose increased, while the contents of choline and amine decreased, which participated in glucose metabolism, lipid metabolism, and amino acid metabolism [12].

Sugar is an important source of energy and carbon in activities of life, as well as in sports [13]. As the athlete's sympathetic nerves become excited during the competition, the secretion of adrenaline increases, and muscle glycogen and liver glycogen, two important sources of glycogen, are pumped out in large quantities and broken down into blood [14]. On the one hand, glucose concentration is affected by the glucose output rate of the liver [15], and on the other hand, it is affected by the glucose uptake and utilization by the tissue. Although gluconeogenesis may also play a supplementary role in blood glucose consumption [16]; its rate is far from that of glucose consumption during exercise [17]. Our data indeed indicated that the level of glucose in urine decreased after competition while the products of glucose metabolism (lactic acid and succinic acid) increased.

Fat is also the main energy source for long-term endurance exercise and provides energy to the body mainly by fatty acids produced by fat hydrolysis [18]. As a carrier of long-chain fatty acids, carnitine transports long-chain fatty acids into the mitochondrial matrix for oxidative decomposition [19]. In this study, the concentration of carnitine which was closely related to lipid metabolism significantly increased after the competition, suggesting that lipid metabolism was involved in the body's energy supply during the competition.

Choline, one of the important components of biomembrane, plays an important role in maintaining the integrity of cell membrane and lipid metabolism and directly or indirectly forms phospholipids, phosphocholine, and glycerophosphoryl choline to participate in lipid metabolism. Lack of choline may lead to fat accumulation and thus lipid metabolism disorders [20]. Acetoacetate is a ketone body which is produced by the accumulation of acetyl coenzyme A derived from the incomplete oxidation of fatty acids [21]. In the study, the level of choline decreased and acetoacetate increased after the competition, indicating that the lack of choline resulted in the incomplete oxidation of fatty acids and thus the formation of large amounts of ketones in the vigorous water polo competition. Choline also may be transformed into trimethylamine which is metabolized to dimethylamine and trimethylamine oxide in intestinal bacteria [22]. In the paper, the levels of choline and amines (dimethylamine, trimethylamine N-oxide) in urine decreased immediately after competition. Dimethylamine plays an important role in many human pathological processes. For example, the concentration of the dimethylamine metabolites in urine can reflect the metabolic status of human intestinal bacteria [23]. The changes of dimethylamine have been found in many acute exercise experiments, but the mechanism was not clear. Dimethylamine may provide protection for the biological activities of organisms by regulating the osmotic pressure and thus affecting the specific metabolic process [24]. So, the changes of dimethylamine may be a response to metabolic adaptation induced by exercise and may be used as an important biomarker of metabolic regulation during exercise.

Glycine participates in the synthesis of many important compounds in animals and detoxication in the human body. Glycine may bind with aromatic substances to form nontoxic substances which are oxidized in the liver and then excreted outside the body [25]. Hippurate is a component of urine produced by the reaction between benzoic acid and glycine under aminotransferase [26]. Our data showed that the levels of hippurate and phenylacetylglycine in urine increased significantly after the competition, indicating the detoxification function of glycine in water polo players' body.

Pantothenic acid, coenzyme A, is an important coenzyme in many reversible acetylation reactions during carbohydrate, fat, and amino acid metabolism [27]. In the study, the levels of ethanol and pantothenic acid in the urine of the participants increased significantly immediately after the competition compared with the week before the competition, as the former may be a metabolite of the altered intestinal bacteria of water polo players [28]. Studies have shown that elevated levels of ethanol can increase the consumption of pantothenic acid (precursor of acetyl coenzyme A), reduce acetyl coenzyme A in body, and then decrease acetylcholine [29-31]. The levels of ethanol and pantothenic acid simultaneously elevated in the study, which may be related to the dynamic balance between upstream and downstream metabolites of pantothenic acid that needed to be further studied. Comprehensive analysis of metabolic pathways showed that pyruvate metabolism, pantothenic acid and CoA biosynthesis, synthesis and degradation of ketone bodies, and other metabolic pathways changed at water polo players' bodies before and after competition, suggesting that water polo sports increased players' utilization of sugar and fat energy supply. The change of the metabolites reflected the metabolic characteristics and change rules of water polo players and an adaptive change of players' physical function to water polo sport. These results can provide an experimental basis for the selection of specific biomarkers in water polo players and give instruction to scientific training and diets.

5. Conclusions

In the work, the metabolites in the urine of water polo players before and after competition were compared by using 1H-NMR-based metabolomic approach. Twenty-six young players of the Shanxi University Polo Team voluntarily participated in the study. The urine and blood samples of players were collected one week before competition (A), immediately after competition (B), and one week after competition (C). Urine metabolomic results showed that the contents of eleven metabolites including lactic acid, acetoacetate, and succinic acid increased and four metabolites such as dimethylamine, choline, glucose, and hippurate decreased at stage B. For example, lactic acid at stage B increased to 0.0030 ± 0.0003 from 0.0013 ± 0.0006 at stage A, while dimethylamine decreased to 0.0037 ± 0.0007 from 0.0046 ± 0.0007. Most metabolites at stage C had basically returned to the levels at stage A. Pyruvate metabolism, pantothenate and CoA biosynthesis, synthesis, and degradation of ketone bodies were mainly involved in the above process. By detecting urine routine items and biochemical indicators in blood samples, it was seen that the urine pH decreased dramatically and the levels of PRO and URO significantly increased at stage B, and the three indicators had similar values between stages A and C. The other indicators did not show an obvious difference among the three stages. Blood biochemical indicator analysis showed that the levels of LDH, BUN, CK, and AST significantly increased at stage B and did not show an obvious difference between stages A and C. The results are helpful for coaches to arrange reasonably the players' diet and to conduct scientific training for players.