Abstract
In recent years, ischemic preconditioning (IPC) has garnered
significant attention in sports research. While IPC has demonstrated
positive effects in high-intensity sports such as judo and swimming,
its potential benefits for enhancing the performance of Taekwondo
athletes have not been extensively studied. This study aimed to
investigate the effects of IPC on taekwondo performance and to observe
the metabolic characteristics associated with enhancing sports
performance via LC‒MS/MS-based plasma metabolomics. Seventeen
participants underwent the repeated frequency speed of kick test (FSKT)
after IPC, along with pre- and post-exercise plasma metabolite
analysis. Differential abundance metabolite analysis, enriched pathway
analysis, and weighted gene coexpression network analysis (WGNCA) were
employed to delve into metabolic characteristics. The findings
highlighted a significant enhancement in FSKT performance in the
experimental group. Metabolomic analysis revealed 109 differentially
abundant metabolites, including Dl-lactate, hypoxanthine,
acetylcarnitine, and acetylsalicylic acid. Enriched pathway analysis
revealed pathways such as pentose and glucuronic acid interconversion,
ascorbic acid and aldonic acid metabolism, the pentose phosphate
pathway (PPP), and the Warburg effect. In conclusion, IPC can
significantly increase the specific athletic abilities of Taekwondo
athletes, with enhancements linked to anaerobic metabolism, PPP
utilization, the Warburg effect for energy production, redox system
stability, reduced muscle fatigue, and pain alleviation.
Keywords: Metabolic characteristics, Ischaemic preconditioning, Sports
performance, LC‒MS/MS-based plasma metabolomics, Taekwondo athletes
Subject terms: Biochemistry, Metabolomics
Introduction
IPC was first discovered by MURRY and has been shown to have
significant effects on protecting tissues from long-term ischemic
damage, particularly myocardial ischemic damage^[34]1. Recent studies
have demonstrated that IPC can enhance the athletic performance of
athletes in various sports. For example, IPC was shown to increase
athletes’ maximum oxygen uptake by 3% in incremental load power bicycle
tests^[35]2. Additionally, in studies using constant maximum load
bicycle tests, IPC was found to reduce subjective fatigue, prolong the
time to exhaustion, and increase maximum oxygen uptake^[36]3. These
findings highlight the positive impact of IPC on aerobic exercise
capacity. Furthermore, research has indicated that IPC can also improve
anaerobic endurance performance. In the Wingate anaerobic power test,
IPC was found to increase the average output power of participants in
lower limb tests^[37]4. Similar results were observed in basketball
athletes subjected to the same test^[38]5. Notably, IPC has been shown
to be beneficial for enhancing the performance of judo athletes in
swimming^[39]6,[40]7. However, conflicting studies have suggested that
IPC may not universally enhance exercise performance. Some research has
indicated that IPC does not improve performance in repeated sprints
during all-out sprint phases^[41]8. Cyclists did not experience
improvements in aerobic capacity or 4-km racing performance after IPC
interventions^[42]9. Moreover, while IPC has been found to enhance
swimmers’ 100-m performance^[43]6, it may not have the same effect on
athletes competing in 200-m speed swimming races^[44]10. Therefore,
further investigation into the mechanisms underlying the effects of IPC
on sports performance is crucial for its optimal application in
training regimens. Notably, the impact of IPC on Taekwondo performance
remains unexplored.
Research indicates that IPC can produce various biologically active
substances, such as adenosine, NO, catecholamines, and opioids^[45]11.
IPC is known to mitigate ischemia‒reperfusion injury by acting on
selective mitochondrial ATP-sensitive potassium channels (mK[ATP]) and
safeguarding tissues from ischemia-induced damage. These mechanisms
play crucial roles in modulating vascular tone and facilitating muscle
contraction^[46]2,[47]12–[48]14. Notably, the mode of action of IPC
involves regulating energy metabolism and combating oxidative stress.
By enhancing the mitochondrial uptake of acetyl-CoA (a glycolysis
byproduct), IPC helps maintain lactate levels within a metabolically
suitable range, which is vital for enhancing anaerobic exercise
performance^[49]15,[50]16. Furthermore, IPC can increase mK[ATP] and
adenosine levels, leading to improved vasodilation, increased muscle
blood supply, and increased oxygen transport efficiency^[51]2. Some
studies have suggested that IPC can increase muscle fatigue
resistance^[52]17, potentially contributing to delayed exhaustion.
While there is abundant research on the protective mechanisms of IPC,
there remains a need for a more comprehensive and thorough
investigation into how IPC enhances exercise performance.
Metabolomics has made significant advancements in understanding
metabolic changes and identifying potential biomarkers in recent years.
Nontargeted metabolomics technology provides a comprehensive view of
various metabolites in samples^[53]18. A recent study conducted a
10-day experiment in which IPC was applied to subjects and utilized
metabolomics to establish a connection between IPC and cholesterol
metabolism^[54]19. However, the use of metabolomics in acute IPC,
especially in sports research, remains limited. This study aimed to
investigate the effects of acute IPC on the performance of taekwondo
athletes and the metabolic characteristics associated with enhanced
sports performance via LC‒MS/MS-based metabolomics. The goal of this
research is to provide scientific insights into the application of IPC
in taekwondo training and competitions and to lay the groundwork for
utilizing IPC to improve sports performance.
Methods
Subjects
Seventeen adult male Taekwondo athletes participated in the study. The
exclusion criteria included acute or chronic diseases such as anxiety,
depression, cardiovascular system diseases, sports, and metabolic
diseases, as well as creatine supplementation, alcohol or caffeine
intake within 24 h prior to the experiment, and strenuous physical
activity within the same timeframe. The inclusion criteria were adult
males over 18 years old, Taekwondo professional athletes with a
national level of two or more athletes, and those who had maintained
regular training in the past three months.
All participants were fully informed of the experimental process and
provided signed informed consent. The research protocol received
approval from the Ethics Review Committee of Guangzhou Institute of
Physical Education (ID Number: 2023LCL-81) and adhered to the
guidelines of the Declaration of Helsinki.
Experimental protocol of the study
The experiment utilized a self-parallel paired design in which
participants were tasked with completing two experiments, one under
220 mmHg pressure (experimental group) and the other under 20 mmHg
pressure (control group), with a minimum of one week between the two
experiments. The order in which the participants were subjected to the
experimental and control conditions was randomized.
Upon arrival at the test site, the subjects rested quietly for 30 min
before their blood pressure was monitored. They then underwent a body
composition test, wore a heart rate monitor, reported their rating of
perceived exertion (RPE), and had their blood lactate levels measured
from fingertip blood samples. Following this, a 40-min compressive
intervention was administered, with an immediate RPE test upon
completion, a blood lactate test 3 min later, and blood collection from
the median cubital vein via an EDTA-K2 anticoagulant tube 5 min later.
The warm-up activities consisted of 10 min of jogging, kicking, and
other exercises, followed by flexibility and stretching exercises. RPE
was assessed after the warm-up, followed by a period of rest and finger
prick blood sampling. The Taekwondo-specific test involved three sets
of FSKT tests with a 1-min rest interval, with the RPE recorded between
sets. After the test, the RPE, blood lactate levels, and blood samples
were collected, and the heart rate was monitored for 15 min after
exercise. The experimental process is illustrated in Fig. [55]1.
Fig. 1.
[56]Fig. 1
[57]Open in a new tab
Experimental flowchart.
In addition, only 8 randomly selected athletes in the experimental
group underwent venous blood collection and metabolomic analysis.
Basic indicator measurement methods
Heart rate was monitored via a heart rate monitoring device (Polar H10;
Finland). Blood pressure was measured via an oscillometric
method^[58]20 with a blood pressure monitor (Omron, HEM-1020). Body
composition analysis was performed via an InBody 370 body composition
analyser (Korea). Blood samples were obtained via finger prick tests,
and blood lactate concentrations were analysed via a Biosen C-line
analyser (Biosen C-line, EKF Diagnostics). Fatigue levels were
subjectively assessed through the Rating of Perceived Exertion
(RPE)^[59]21.
IPC
Participants underwent IPC intervention while in the supine position
and breathing normal oxygen. A blood pressure cuff was positioned on
the middle and upper thirds of both thighs, with the right leg cuff
inflated to 220 mmHg and the left leg cuff at 0 mmHg for 5 min.
Subsequently, the right leg cuff was deflated to 0 mmHg, while the left
leg cuff was raised to 220 mmHg for another 5 min, completing one
cycle^[60]5,[61]22,[62]23. This cycle was iterated 4 times, totaling
40 min.
Taekwondo performance testing protocol
The FSKT is a method used to evaluate the specific sports performance
of Taekwondo athletes^[63]24–[64]26. In this study, participants were
instructed to complete three sets of tests, with a 1-min break between
sets, mimicking the structure of a formal Taekwondo competition. During
each set of tests, participants were asked to use kicking scoring
equipment to kick at full speed with both legs, which was continuously
striking for 10 s followed by a 10-s rest. This pattern was repeated
for 5 rounds of kicking and 4 rounds of rest, totaling 90 s. After each
set, there was a 1-min break before the next set began. The number of
kicks was recorded via the World Taekwondo Grand Slam competition
electronic protective gear (professional version, China) and the
corresponding competition system. The total number of kicks per round
(10 s), the total number of kicks in five rounds (90 s), and the Kick
Decrement Index (KDI) were used to assess the participants’ sports
performance. Additionally, the Male Classificatory Table for the
Frequency Speed of Kick Test was used to evaluate athletic
performance^[65]26.
Methods for the metabolism experiments
Sample collection and preparation
After venous blood samples were collected, they were centrifuged at
1100 × g for 15 min at 4 °C within 30 min. The plasma was subsequently
separated and stored at − 80 °C.
LC‒MS/MS analysis
Following slow thawing at 4 °C, an appropriate amount of the sample was
mixed with a precooled methanol/acetonitrile/water mixture (2:2:1,
v/v), vortexed, subjected to low-temperature ultrasonication for
30 min, and then allowed to stand at − 20 °C for 10 min. The sample was
then centrifuged at 14,000 × g for 20 min at 4 °C, after which the
supernatant was subsequently dried under vacuum. For mass spectrometry
analysis, the dried sample was reconstituted by adding 100 μL of
acetonitrile aqueous mixture (acetonitrile:water = 1:1, v/v), followed
by vortexing and centrifugation at 14,000 × g for 15 min at 4 °C. The
supernatant was then used for sample analysis. QC samples were prepared
by extracting 10 μL from each sample, with QC analysis conducted after
every 5 samples.
Mass spectrometry analysis was performed using an AB Triple TOF 6600
mass spectrometer. The ESI source conditions for post-HILIC
chromatographic separation were as follows: Ion Source Gas1 (Gas1): 60,
Ion Source Gas2 (Gas2): 60, Curtain gas (CUR): 30, source temperature:
600 °C, IonSapary Voltage Floating (ISVF) ± 5500 V (positive and
negative modes), TOF MS scan m/z range: 60–1000 Da, product ion scan
m/z range: 25–1000 Da, TOF MS scan accumulation time 0.20 s/spectra,
and product ion scan accumulation time 0.05 s/spectra. The secondary
mass spectrum was acquired via information-dependent acquisition (IDA)
in high-sensitivity mode, with a declustering potential (DP) of ± 60 V
(positive and negative modes), a collision energy of 35 ± 15 eV, and
IDA settings to exclude isotopes within 4 Da. Candidate ions to monitor
per cycle: 10.
Data analysis
SPSS 25 was used to conduct the statistical tests. The distribution of
variables, such as exercise intensity and performance of the
experimental subjects, was assessed via the Shapiro‒Wilk test. Some
data did not follow a normal distribution, leading to the expression of
exercise performance data in terms of the median and interquartile
range. Intergroup comparisons of sports performance were carried out
via the Mann‒Whitney test for exercise intensity (a nonparametric
statistical analysis), with P values less than 0.05 considered to
indicate statistical significance.
Metabolites were detected via positive ion mode (POS) and negative ion
mode (NEG), followed by systematic cluster analysis of the data^[66]27.
The resulting dendrogram was calculated via average linkage, and
hierarchical clustering was performed via the R package
pheatmap^[67]28.
Principal component analysis (PCA) was conducted via the R language
gModels (v2.18.1), whereas orthogonal partial least squares
discriminant analysis (OPLS-DA) was performed via the R software
package model.
The differentially abundant metabolite screening criteria included a
variable importance for projection (VIP) of OPLS-DA ≥ 1 and a p
value < 0.05 according to a single-factor t test. Fold changes (FCs)
between the two groups were calculated, and a volcano plot was
generated. The VIP score from OPLS-DA was used to create a chart
displaying the 15 metabolites with the highest scores^[68]29. The
Pearson correlation coefficient and p value were calculated via the R
functions cor and cor.test to assess the similarity in metabolic
abundance. A correlation was deemed significant when p ≤ 0.05^[69]30.
The R corrploy package was used to produce correlation heatmaps^[70]31.
The abundance of differentially abundant metabolites was normalized via
the z score formula: z = (x—μ)/σ, where x is the specific score, μ is
the mean, and σ is the standard deviation. Hierarchical clustering was
performed via the R package pheatmap^[71]28 to create a cluster heatmap
showing the cumulative difference between the two groups in terms of
differentially abundant metabolites and samples.
The following method was used for Kyoto Encyclopedia of Genes and
Genomes (KEGG) enrichment analysis of the differentially abundant
metabolites:
[MATH: P=1-∑i=0
m-1M
iN-Mn-i<
mfenced close=")"
open="(">N
n :MATH]
N represents the total number of metabolites annotated by KEGG, n
represents the number of differentially abundant metabolites within N,
M represents the total number of metabolites annotated to a specific
pathway, and m represents the number of differentially abundant
metabolites within M. The p value calculated was FDR corrected, with a
threshold of FDR ≤ 0.05. Pathways that met these conditions were
considered to be significantly enriched in differentially abundant
metabolites.
Metabolite set enrichment analysis (MSEA)^[72]32 was conducted via the
MetaboAnalyst module and the small molecule pathway database library.
Fisher’s exact test was used for overrepresentation analysis via the R
package MSEAp ([73]https://rdrr.io/github/afukushima/MSEAp/).
WGCNA^[74]33,[75]34 was also performed, with a soft threshold β value
of 12, a similarity of 0.8, and a minimum of 50 metabolites in each
module.
Results
Subjects
Seventeen Taekwondo athletes participated in this experiment. The basic
characteristics of the experimental subjects are shown in Table [76]1.
Table 1.
Basic information of the experimental subjects and the exercise
performance table.
[MATH:
X¯±S
:MATH]
Median (25%, 75% IR) P[1] P[2]
20 mmHg 220 mmHg 20 mmHg 220 mmHg
Age (Y) 19.65
[MATH: ± :MATH]
1.41
Height (m) 179.06
[MATH: ± :MATH]
5.71
Weight (kg) 71.41
[MATH: ± :MATH]
7.95
Years of practice (Y) 7.00
[MATH: ± :MATH]
2.00
BMI (kg/m^2) 22.26
[MATH: ± :MATH]
2.19
Body fat percentage (%) 13.08
[MATH: ± :MATH]
4.76
FSKT[1-1] 19.7
[MATH: ± :MATH]
3.16 21.1
[MATH: ± :MATH]
1.87 20 (18,20.5) 20 (20,22) 0.021 0.031
FSKT[1-2] 19.2
[MATH: ± :MATH]
1.09 20.4
[MATH: ± :MATH]
1.41 19 (18.5,20) 20 (19.5,21.5) 0.008 0.035
FSKT[1-3] 18.7
[MATH: ± :MATH]
1.49 19.6
[MATH: ± :MATH]
1.77 19 (18,19.5) 20 (18.5,20.5) 0.021 0.058
FSKT[1-4] 18
[MATH: ± :MATH]
1.27 19
[MATH: ± :MATH]
1.90 18 (17.5,19) 19 (17.5, 20.5) 0.042 0.015
FSKT[1-5] 17.4
[MATH: ± :MATH]
1.66 18.7
[MATH: ± :MATH]
1.79 17 (17,18.5) 19 (17.5,19.5) 0.001 0.014
FSKT[total-1] 93.1
[MATH: ± :MATH]
6.86 98.9
[MATH: ± :MATH]
6.36 94 (90.5,96) 101 (94,103.5) 0.002 0.002
FSKT[KDI-1] (%) 8.56
[MATH: ± :MATH]
5.93 8.91
[MATH: ± :MATH]
1.42 0.653 1.000
FSKT[2-1] 18.7
[MATH: ± :MATH]
2.44 20.1
[MATH: ± :MATH]
2.09 19 (18,20) 20 (18,19) 0.042 0.273
FSKT[2-2] 18
[MATH: ± :MATH]
1.90 19.2
[MATH: ± :MATH]
1.85 18 (17,19) 19 (19,20) 0.083 0.398
FSKT[2-3] 17
[MATH: ± :MATH]
2.15 18.6
[MATH: ± :MATH]
1.70 17 (16,19) 19 (18,19.5) 0.004 0.112
FSKT[2-4] 17
[MATH: ± :MATH]
1.70 18.6
[MATH: ± :MATH]
1.70 17 (16,18.5) 19 (17.5,20) 0.001 0.004
FSKT[2-5] 16.7
[MATH: ± :MATH]
1.96 18.1
[MATH: ± :MATH]
1.41 17 (16,18) 18 (17.5, 19) 0.001 0.014
FSKT[total-2] 87.4
[MATH: ± :MATH]
8.18 94.6
[MATH: ± :MATH]
8.02 87 (84.5,92.5) 96 (90.5,99.5) 0.002 0.052
FSKT[KDI-2] (%) 9.67
[MATH: ± :MATH]
6.07 6.64
[MATH: ± :MATH]
4.02 0.055 0.119
FSKT[3-1] 18.3
[MATH: ± :MATH]
2.37 19.6
[MATH: ± :MATH]
1.70 19 (17.5,19.5) 20 (18,20.5) 0.007 0.011
FSKT[3-2] 17.1
[MATH: ± :MATH]
1.93 18.9
[MATH: ± :MATH]
1.95 18 (16,18) 19 (18,20.5) 0.002 0.039
FSKT[3-3] 16.6
[MATH: ± :MATH]
2.23 18.4
[MATH: ± :MATH]
1.84 17 (15.5,18) 19 (17,19.5) 0.004 0.046
FSKT[3-4] 16.3
[MATH: ± :MATH]
2.69 18.1
[MATH: ± :MATH]
1.60 17 (15.5, 18) 19 (16.5,19) 0.006 0.006
FSKT[3-5] 16.9
[MATH: ± :MATH]
1.93 18.6
[MATH: ± :MATH]
1.87 17 (15.5,18) 18 (17,20) 0.002 0.004
FSKT[total-3] 85.2
[MATH: ± :MATH]
10.05 93.6
[MATH: ± :MATH]
7.56 87 (80.5,90) 94 (89,99.5) 0.001 0.005
FSKT[KDI-3] (%) 8.28
[MATH: ± :MATH]
6.52 6.82
[MATH: ± :MATH]
5.13 0.21 0.166
[77]Open in a new tab
Blood lactate, heart rate and RPE results at each stage
A normality test was conducted for each variable, indicating that some
variables did not adhere to a normal distribution. Consequently, we
employed the nonparametric paired rank sum test to examine blood
lactate levels, heart rate, and the RPE. The results depicted in
Fig. [78]2a show no significant differences in blood lactate levels
between the control and experimental groups at different time points,
including before and after intervention, following warm-up, and after
the exercise test (P > 0.05). Similarly, there was no statistically
significant difference in the average heart rate between the groups
during various phases, such as the resting, jogging warm-up, kicking
warm-up, and exercise test phases (P > 0.05) (Fig. [79]2b). However, a
notable difference in the RPE was observed after FSKT between the first
and third groups (P > 0.05), with the experimental group exhibiting a
lower mean RPE than the control group did (Fig. [80]2c). The average
peak heart rates during FSKT for the experimental group were 180 bpm,
184 bpm, and 187 bpm, whereas for the control group, they were 184 bpm,
185 bpm, and 185 bpm, respectively. Both groups achieved high-intensity
exercise on the basis of the American College of Sports Medicine
guidelines for exercise testing and prescription^[81]35.
Fig. 2.
[82]Fig. 2
[83]Open in a new tab
Statistical analysis of blood lactate levels, heart rate and RPE at
each stage. a: BLA1-4 represent blood lactate before intervention,
after intervention, after warm-up and after the exercise test,
respectively. b: HR1–4 represent the average heart rate during the
resting, jogging warm-up, kicking warm-up and exercise tests,
respectively. c: RPE-1–3 represent the RPE before intervention, after
intervention, and after warm-up, respectively. RPE-4, RPE-5, and RPE-6
indicate the RPE after the FSKT test in the different groups. *
indicates a significant difference between groups (P < 0.05). The
broken line shows the change in the mean RPE at each stage.
Taekwondo-specific sports performance test results
A normality test was conducted on the data, revealing that some data
did not follow a normal distribution. The nonlinear paired rank sum
test was subsequently used to compare sports performance between the
experimental and control groups. The results revealed significant
differences in the number of kicks per round of FSKT in the first group
(P < 0.05). In the second group, all rounds except the second had
significant differences (P < 0.05). Similarly, the third FSKT group
showed significant differences in exercise performance between the
experimental and control groups in each round (P < 0.05). Significant
differences were also found in the total number of FSKT kicks between
groups (P < 0.05), whereas no significant difference was found in the
KDI of FSKT. The results, including the means, standard deviations, and
P values, are presented in Table [84]1.
The exercise performance of the experimental subjects was assessed via
the Male Classificatory Table for the Frequency Speed of Kick Test. A
comparison between the control and experimental groups revealed no
significant difference in the evaluation of kicking in the third round
of the first FSKT group or the first, second, or third rounds of the
second FSKT group. However, significant differences were found in the
intergroup comparisons for the remaining rounds (P < 0.05), indicating
a significant improvement in sports performance levels among the
experimental subjects in the experimental group. The results, including
medians, 25th and 75th percentile quartiles, and P values, are shown in
Table [85]1.
Following notable variations in exercise performance metrics (while
blood lactate levels remained consistent), a thorough examination of
the metabolic profiles of participants undergoing specific tests after
IPC was conducted via a metabolomic approach to explore the potential
benefits of IPC in enhancing Taekwondo performance.
Table [86]1 FSKT[x-y]: x represents which group of FSKT, y represents
which round; FSKT[total-x]: x represents which group of FSKT; IR
represents the interquartile range; P[1] and P[2] represent the P
values of the rank sum test between groups for the number of kicks and
exercise level, respectively.
Metabolite statistical results
Blood samples collected after IPC were assigned to Group A, whereas
blood samples taken after specific exercise testing were assigned to
Group B, resulting in a total of 16 samples. A total of 20,138
metabolites, comprising both positive and negative ion species, were
detected in the comprehensive analysis. This data set included 2,257
known metabolites and 17,881 unknown metabolites, representing the
complete range of metabolic features identified. Principal component
analysis (PCA) was conducted on the metabolites from Group A, Group B,
and the quality control samples, leading to the generation of positive
and negative ion PCA plots (Fig. [87]3a). The quality control results
demonstrated the high stability, excellent data quality, and
reliability of the experimental data. Furthermore, a heatmap
illustrating the correlations between samples in positive and negative
ion modes was created on the basis of Pearson correlation coefficients
(Fig. [88]3b). Notably, in positive ion mode, the B5 sample presented
low Pearson correlation values with the other samples, indicating that
it was an outlier. To ensure the accuracy of subsequent metabolomic
analyses, the B5 sample was excluded.
Fig. 3.
[89]Fig. 3
[90]Open in a new tab
Data quality control chart. (a): The PC1 coordinate represents the
first principal component, with the percentage in parentheses
indicating its contribution to the sample difference. Similarly, the
PC2 coordinate represents the second principal component, with its
corresponding percentage showing the contribution to the sample
difference. The colored points in the figure represent individual
samples, where closer proximity indicates better repeatability within
the same group. (b): Each row and column in the figure represents a
sample, with the value in each cell representing the Pearson
correlation coefficient between the two samples. A higher value and
darker color indicate a stronger correlation between the samples.
Identification of 109 differential metabolites through nontargeted
metabolomics analysis
LC‒MS/MS analysis was conducted in both positive and negative ion modes
to screen for differentially abundant metabolites simultaneously. PCA
was performed in both positive and negative ion modes, as shown in
Fig. [91]3a, revealing a tendency for samples in group B to separate
from those in group A along the second principal component. The OPLS-DA
model was then utilized to identify the differentially abundant
metabolites between groups A and B, resulting in positive and negative
ion OPLS-DA score plots (Fig. [92]4a). A permutation test was
subsequently carried out on the OPLS-DA model (Fig. [93]4b). The
permutation test model results confirmed the reliability of the OPLS-DA
model predictions for both positive and negative ions. This study
highlights the notable variations in metabolite levels observed between
the two sample groups. Metabolites that showed differential abundance
were identified on the basis of the criteria of OPLS-DA VIP ≥ 1 and
univariate t test P < 0.05. A total of 109 differentially abundant
metabolites were identified, with 49 showing increased levels and 60
showing decreased levels.
Fig. 4.
[94]Fig. 4
[95]Open in a new tab
Differentially abundant metabolite screening chart. (a): This diagram
depicts an OPLS-DA, with the blue circle representing samples from
Group A and the yellow circle representing samples from Group A.
Circles represent samples from group B. (b): If all blue Q2 points are
lower than the original blue Q2 point on the right or if the
intersection of the Q2 point regression line on the ordinate is less
than or equal to 0, this indicates reliable model prediction results.
(c): The VIP value is shown on the x-axis, representing the top 15
differentially abundant metabolites on the y-axis. Metabolite abundance
is averaged per group and analysed via z scores, indicated by the color
scale on the right. The upregulated metabolites are marked in yellow,
and the downregulated metabolites are marked in blue. Metabolites with
a VIP greater than 1 were considered to have significant differences,
with larger VIP values indicating greater contributions to
distinguishing samples. (d): The x-axis represents the log2-transformed
fold change in metabolite abundance between comparison groups, whereas
the y-axis represents the -log10-transformed P value from the T test.
The vertical dashed line on the y-axis indicates the threshold for
screening differentially abundant metabolites on the basis of P values.
Yellow dots indicate upregulated metabolites (fold change > 1), whereas
blue dots indicate downregulated metabolites (fold change < -1). Larger
points correspond to higher VIP values for the metabolites. (e): Yellow
indicates a positive correlation between changes in differentially
abundant metabolites, and blue indicates a negative correlation. (f):
Each row in the figure represents a metabolite, and each column
represents a sample. The intensity of the yellow color indicates a
greater abundance of the metabolite, whereas a bluer hue signifies a
lower abundance.
The VIP values of the top 15 differentially abundant metabolites in
positive and negative ion modes were calculated and used to generate
the VIP statistical chart of OPLS-DA (Fig. [96]4c). The results
indicated that Dl-lactate had the highest VIP value (VIP = 50.24),
followed by acetylcarnitine (VIP = 28.69). Additionally, a volcano plot
of the differentially abundant metabolites based on the fold difference
(FC), VIP, and P value was created (Fig. [97]4d) to visualize the up-
and downregulation changes in metabolite abundance. Pearson correlation
coefficient analysis was conducted to examine the relationships between
differentially abundant metabolites, and a heatmap illustrating the
correlations among these metabolites was generated (Fig. [98]4e).
Furthermore, Z scores were calculated for the differentially abundant
metabolites, and a heatmap was generated for differentially abundant
metabolite clustering (Fig. [99]4f), revealing notable variations in
expression levels and clustering patterns among the groups.
Receiver operating characteristic (ROC) curve analysis was conducted.
Among the nine differentially abundant metabolites, swainsonine,
L-deprenyl, Dl-lactate, cholesteryl sulfate, 1-hydroxyanthraquinone,
aminomethylphosphonic acid, Ponceau 6, taurine, and Dl-a-hydroxybutyric
acid had the highest AUC values (AUC = 1). The next most common
metabolites were xylitol, hypoxanthine, acetylsalicylic acid, and
pyruvate, with AUC values of 0.9821 each.
Key metabolic pathways identified through KEGG enrichment analysis of
differential metabolites
Kyoto Encyclopedia of Genes and Genomes^[100]36 pathway enrichment
analysis was conducted on the differentially abundant metabolites,
leading to the identification of 53 candidate differentially abundant
metabolites with pathway annotations. Some of these metabolites include
pyruvate, succinate, xylitol, D-ribulose 5-phosphate, D-arabinose,
Dl-a-hydroxybutyric acid, taurine, acetylcarnitine, D-arabinonic acid,
L-gulono-1,4-lactone, acetol, argininosuccinic acid, hydrocinnamic
acid, deoxyadenosine, and phosphoric acid. Additionally, a total of 97
enriched pathways were identified in this study. These pathways
included pentose and glucuronate interconversions (P < 0.01, involving
hits such as pyruvate, D-ribulose 5-phosphate, D-arabinose, and
xylitol), ascorbate and aldarate metabolism (P < 0.01, involving hits
such as pyruvate, D-arabinose, D-arabinonic acid, and
L-gulono-1,4-lactone), propanoate metabolism (P < 0.05, involving hits
such as succinate, acetol, and Dl-a-hydroxybutyric acid), alanine,
aspartate, and glutamate metabolism (P < 0.05, including hits such as
pyruvate, succinate, succinic semialdehyde, and argininosuccinic acid),
oxidative phosphorylation (P < 0.05, involving hits such as phosphoric
acid and succinate), phenylalanine metabolism (P < 0.05, including hits
such as pyruvate, succinate, and hydrocinnamic acid), and butanoate
metabolism (P < 0.05, including hits such as pyruvate, succinate, and
succinic semialdehyde), among others. Statistical analysis was also
conducted on the KEGG enrichment pathways of the top 20 differentially
abundant metabolites (Fig. [101]5a). A circle diagram illustrating the
enrichment of the top 20 differentially abundant metabolites was
subsequently generated on the basis of the P value (Fig. [102]5b).
Furthermore, an enrichment circle plot representing the KEGG enrichment
pathways of the top 20 differentially abundant metabolites was created
(Fig. [103]5c). Notably, pentose and glucuronate interconversions
presented the lowest P value among all differentially abundant
metabolite KEGG enrichment pathways.
Fig. 5.
[104]Fig. 5
[105]Open in a new tab
Enriched pathway analysis plot. (a): The KEGG enrichment bar chart was
generated using the top 20 pathways with the smallest Q values. The
y-axis represents the pathways, whereas the x-axis displays the
percentage of each pathway compared with all the differentially
abundant metabolites. Darker shades of blue in the figure indicate
smaller Q values. Each bar on the chart represents the number of paths
and their respective Q values. (b): The KEGG enrichment bubble chart
displays the top 20 pathways with the smallest Q values. The y-axis
represents the pathways, whereas the x-axis represents the enrichment
factor (the ratio of differentially abundant metabolites in the pathway
to all metabolites in the pathway). The size of each bubble corresponds
to the quantity. A smaller Q value is indicated by a yellower color.
(c): The first circle displays the top 20 enriched pathways, with the
number of differentially abundant metabolites represented outside the
circle. Different colors indicate different classes. In the second
circle, the number of pathways and Q values are shown against the
background of the differentially abundant metabolites. A longer bar
indicates a greater number of differentially abundant metabolites,
whereas a bluer color signifies a smaller Q value. The third circle
presents a bar chart showing the proportion of up- and downregulated
differentially abundant metabolites, with dark purple indicating
upregulated metabolites and light purple indicating downregulated
metabolites. The fourth circle displays the RichFactor value for each
pathway, which is calculated as the number of differentially abundant
metabolites in the pathway divided by the total quantity in the
pathway. Background grid lines are included, with each grid
representing 0.1. (d): The enrichment pathway is shown on the left side
of the figure. The length of each column corresponds to the degree of
enrichment, and the color represents the p value. Paths are sorted from
small to large p values.
Comprehensive metabolic pathways and potential key pathways revealed through
metabolite set enrichment analysis
MSEA was conducted, with positive ion MSEA results indicating
enrichment in pathways such as purine metabolism, caffeine metabolism,
PPP, oxidation of branched chain fatty acids, beta oxidation of very
long-chain fatty acids, lysine degradation, alanine metabolism, and the
glucose‒alanine cycle, totaling 85 pathways. Negative ion MSEA revealed
enrichment in 95 pathways, including ketone body metabolism,
peroxisomal oxidation of phytanate, carnitine synthesis, acetyltransfer
into mitochondria, butyrate metabolism, carnitine synthesis, glutamate
metabolism, gluconeogenesis, glycolysis, pyruvate metabolism, alanine
metabolism, and cysteine metabolism. Statistical diagrams for MSEA in
positive and negative ion modes were obtained (Fig. [106]5d).
Enrichment of pathways such as pentose and glucuronate interconversions
revealed in WGNCA
Metabolites were subjected to WGNCA to generate a module-level
clustering diagram (Fig. [107]6a). A total of 13 modules were
identified, and a histogram displaying the number of metabolites in
each module was created (Fig. [108]6b). We subsequently utilized module
characteristic values to perform correlation analysis with the
phenotypic data of groups A and B. The results revealed that the blue
module presented the strongest correlation with the phenotypes of
groups A and B (P < 0.001), with correlation coefficients of -0.94 and
0.94, respectively. Additionally, the green‒yellow module and the blue
module displayed notable negative correlations (Fig. [109]6c).
Furthermore, the average gene significance (GS) of metabolites within
each module was calculated and is represented in a column chart
(Fig. [110]6d), which shows that the blue module had the highest GS
value, followed by the green‒yellow module. A heatmap was generated to
visualize the expression of metabolites within the modules
(Fig. [111]6e), indicating an increasing trend in the blue module and a
decreasing trend in the green‒yellow module. Finally, KEGG Orthology
(KO) enrichment analysis of the module metabolites revealed enrichment
of pathways such as pentose and glucuronate interconversions; alanine,
aspartate and glutamate metabolism; ascorbate and aldarate metabolism;
and propanoate metabolism within the blue module. Pentose and
glucuronate interconversions were significantly enriched (P < 0.001,
Q < 0.05), and bubble and bar charts illustrating the top 20 enriched
KEGG pathways in the blue module were generated (Fig. [112]6f).
Additionally, the quantity of differentially abundant metabolites
annotated by the enriched pathways was determined (Fig. [113]6g).
Fig. 6.
[114]Fig. 6
[115]Open in a new tab
The figures of WGNCA. (a): Metabolites exhibiting similar expression
patterns were clustered together within modules. The clustering tree
branches were cut to create distinct modules, where each color
represents a module and gray indicates metabolites that do not align
with any specific module. (b): The abscissa represents each module, and
the ordinate represents the number of metabolites. (c): The abscissa
represents the trait, the ordinate represents the module, and the
Pearson correlation coefficient is used for plotting. Positive
correlations are depicted in yellow, whereas negative correlations are
shown in blue. Darker colors signify stronger correlations. The numbers
in parentheses below indicate significant P values. (d): Histogram of
the GS of each module and trait. The abscissa is represented by
different colours, and the ordinate is the GS value. (e): Heatmap
displaying metabolite expression patterns in the blue and green‒yellow
modules. The upper panel presents a heatmap displaying metabolite
expression across samples within a specific module, where red denotes
up-regulation and green signifies down-regulation. The lower panel
depicts the module eigenvalues for various samples. (f): The y-axis
represents various metabolic pathways, while the x-axis indicates the
ratio of differentially abundant metabolites within each pathway
relative to the total number of metabolites present. The size of each
bubble corresponds to the quantity of metabolites, with smaller Q
values resulting in a more intense yellow color. (g): The y-axis
represents the various pathways, while the x-axis illustrates the
percentage of each pathway relative to the total number of
differentially abundant metabolites. Darker blue shades indicate lower
Q values. The numbers displayed after the bars correspond to the number
of pathways and their respective Q values.
Discussion
In this study, IPC had a significant positive effect on the
sport-specific performance of Taekwondo athletes. The results of the
plasma metabolomic analysis indicated that anaerobic metabolism is
crucial for Taekwondo-specific tests, whereas aerobic metabolism plays
a supportive role. Furthermore, the PPP and the Warburg effect are
enhanced by IPC, facilitating energy replenishment. Interestingly, the
differentially abundant metabolites identified not only improved the
body’s resistance to oxidative stress during intense exercise but also
displayed analgesic and excitatory properties that may ultimately
benefit athletes’ overall performance. Notably, previous studies on
high-intensity exercise metabolism have not reported functional
characteristics related to improved antioxidative stress capacity and
analgesic excitability.
This study presents compelling evidence that IPC enhances performance
in Taekwondo athletes. The intense and frequent kicks in Taekwondo
competitions rely heavily on anaerobic metabolism for energy^[116]37.
High-level athletes tend to exhibit faster kicking speeds^[117]38. In
this study, IPC significantly improved the total kicking performance of
Taekwondo athletes in the first group of FSKTs, indicating that IPC can
enhance the specific anaerobic endurance of these athletes^[118]26. We
also observed similar characteristics in the metabolomic results. The
findings revealed that the levels of dl-lactate, pyruvate, and
hypoxanthine were significantly increased, with the most pronounced
change observed in the level of dl-lactate. Concurrently, ROC curve
analysis underscored the importance of dl-lactate in anaerobic
glycolytic metabolism. Furthermore, our results revealed enrichment of
the glycolysis and purine metabolism pathways in MSEA, suggesting that
increased pyruvate production during high-intensity FSKT exercise
accelerates lactate synthesis and ATP resynthesis^[119]39,[120]40.
Hypoxanthine levels reflect muscle metabolism under anaerobic
conditions, and purine metabolism significantly responds to
high-intensity anaerobic exercise^[121]41. The increase in hypoxanthine
further indicates enhanced glycolysis, increased consumption and
mobilization of energy substrates, and decreased ATP flux^[122]42.
These metabolic patterns align with energy metabolism profiles
typically observed during high-intensity exercise^[123]43,[124]44.
Additionally, research indicates that kickboxing athletes require a
robust glycolytic energy supply. Therefore, we assert that IPC can
significantly increase the specific anaerobic endurance of Taekwondo
athletes, with glycolytic metabolism playing a crucial role in the
specific test process.
As the duration of the test increased, aerobic metabolism progressively
intensified^[125]45. Compared with the control group, the second and
third FSKT groups within the experimental group demonstrated enhanced
exercise performance, which may be attributed to improvements in
aerobic metabolism. Among the differentially abundant metabolites,
succinate and taurine, both of which are essential for energy
metabolism, were significantly elevated^[126]46,[127]47. KEGG
enrichment analysis revealed enrichment of the TCA cycle and oxidative
phosphorylation pathway, with the latter showing significant enrichment
(i.e., P < 0.05, Q < 0.05). Taurine release has been shown to
contribute to lipolysis, glucose uptake, and the regulation of
glycolytic flux^[128]48,[129]49. Succinic acid was found to increase
the activity of mitochondrial complex enzymes, increase creatine kinase
activity, and increase ATP production^[130]50. Under fast exercise
conditions, succinate can aid in lactate consumption and accelerate
oxidative phosphorylation through succinate receptor 1, thereby
increasing energy production efficiency^[131]51. The activation of the
oxidative phosphorylation metabolic pathway observed in our study
contrasts with findings from a previous study on acute high-intensity
exercise^[132]52. Concurrently, several studies have indicated that IPC
positively influences aerobic exercise performance^[133]9,[134]53.
Additionally, fatty acid-derived hydrogen peroxide in peroxisomes is
crucial for maintaining reduced NAD + and ATP levels in highly hypoxic
environments, and there is a significant substrate-level interaction
between peroxisomes and mitochondrial metabolism. This interaction,
along with fatty acid breakdown, aids in the delivery of oxygen to
muscles and tissues, facilitating metabolic adaptations to hypoxic
conditions^[135]54. Taekwondo athletes may exhibit similar adaptations
after IPC. Our research revealed that pathways enriched in MSEA include
ketone body metabolism, branched-chain fatty acid oxidation,
peroxisomal oxidation of folate esters, β-oxidation of very long-chain
fatty acids, carnitine synthesis, and acetyl transport into
mitochondria. These pathways contribute to improved fatty acid
breakdown and oxygen transport during high-intensity exercise, leading
to an increased energy supply and improved exercise performance.
Importantly, however, LC‒MS/MS-based plasma metabolomics alone cannot
definitively assess the specific impact of IPC on mitochondrial ATPase
function and fat remodelling.
In addition to providing anaerobic and aerobic metabolic energy, our
study revealed potential beneficial impacts of the PPP and the Warburg
effect on energy metabolism following high-intensity exercise post-IPC.
The pentose phosphate pathway, a glucose oxidation pathway that runs
parallel to upper glycolysis^[136]55, primarily involves the production
of nicotinamide adenine nucleotide phosphate (NADPH) and ribose
5-phosphate (R5P) while releasing energy^[137]56–[138]58.
Interestingly, our KEGG enrichment analysis of differentially abundant
metabolites and MSEA highlighted the enrichment of the PPP, a pathway
that has rarely been discussed in previous exercise metabolomics
studies. These findings suggest that PPP activation due to IPC may
influence the energy metabolism of Taekwondo athletes during subsequent
specialized tests. Furthermore, we observed enrichment of the Warburg
effect pathway via MSEA. Some studies suggest that the body enhances
the Warburg effect (aerobic glycolysis) to increase bioenergetic ATP
circulation in high metabolic states, despite less efficient glucose
consumption. Additionally, transient ischemic conditions may also
trigger the activation of the Warburg effect pathway^[139]59.
Further investigation revealed certain differentially abundant
metabolites and enriched pathways with functional characteristics that
may improve the body’s capacity to resist oxidative stress.
High-intensity training is known to disrupt the redox balance of
skeletal muscle^[140]60, leading to oxidative stress and subsequent
muscle fatigue^[141]61,[142]62. However, in this study, despite
rigorous testing, several differentially abundant metabolites, such as
7-hydroxyflavanone, acetyl-L-carnitine, and acetylsalicylic acid, were
notably upregulated. These upregulated metabolites have antioxidant
properties that can increase the body’s resistance to oxidative stress
and maintain redox balance during high-intensity
exercise^[143]63–[144]65. Moreover, the levels of argininosuccinic acid
and tetradinone notably decreased. Argininosuccinic acid diminishes
antioxidant defenses in the cerebral cortex, induces oxidative stress
in the striatum, and elevates reactive oxygen species in the cerebral
cortex^[145]66. Moreover, a decrease in tetradinone levels hinders
mutations and safeguards the antioxidant system from
harm^[146]67,[147]68. A decrease in these metabolites alleviates
oxidative stress and contributes to maintaining redox balance.
According to the KEGG enrichment analysis of the differentially
abundant metabolites and WGCNA, the ascorbate and aldarate metabolism
pathways were enriched. These findings indicate that these pathways
were activated after high-intensity special testing following IPC.
Ascorbate and aldarate metabolism are vital carbohydrate pathways that
safeguard cells from oxidative damage^[148]69. Previous studies have
demonstrated that IPC can increase the body’s antioxidant capacity and
mitigate inflammatory responses^[149]70. These results suggest that
after performing three sets of FSKTs following IPC, Taekwondo athletes
may still maintain a relatively good and stable redox state, delaying
muscle fatigue and enabling higher output during exercise, ultimately
leading to improved sports performance.
In addition, we found that acetylsalicylic acid and caffeine were
unexpectedly significantly upregulated. Acetylsalicylic acid can
increase pain tolerance, reduce pain and injury-induced inflammation,
and improve endurance and neuromuscular performance^[150]71,[151]72.
Caffeine acts on the central nervous system by blocking adenosine
receptors, leading to increased neurotransmitter release, motor unit
firing rates, and pain relief^[152]73,[153]74. It also increases
dopamine levels in brain regions associated with attention and enhances
sodium/potassium pump activity for muscle contraction^[154]75. Previous
studies have demonstrated that IPC reduces pain sensitivity during
painful cold stimulation, which aligns with our findings^[155]76.
Therefore, the increased endogenous production of acetylsalicylic acid
and caffeine might enhance the athletic performance of taekwondo
athletes post-IPC because of their central analgesic and stimulant
effects. This could explain why the experimental group showed improved
performance during intense exercise while reporting reduced subjective
fatigue.
Conclusion
IPC significantly enhances the specific sports performance of taekwondo
athletes. Plasma metabolomics revealed the upregulation of Dl-lactate,
pyruvate, and hypoxanthine in the athletes’ bodies, along with the
enrichment of the glycolysis and purine metabolism pathways, suggesting
that anaerobic metabolism plays a crucial role in FSKT. Additionally,
the upregulation of succinic acid and taurine, along with the
enrichment of the TCA cycle and oxidative phosphorylation pathways,
suggested a supporting role for aerobic metabolism. The PPP and the
Warburg effect may also provide supplementary energy for high-intensity
exercise. Notably, after IPC and high-intensity exercise, the
expression of 7-hydroxyflavanone, acetyl-L-carnitine, acetylsalicylic
acid, and caffeine increased, whereas the expression of
argininosuccinic acid and tetradinone decreased, leading to significant
enrichment of ascorbate and aldarate metabolism. These metabolic
alterations may enhance the body’s ability to combat oxidative stress
during exercise, improve muscle fatigue resistance, and provide
analgesic and stimulating effects, thereby enhancing exercise
performance.
However, this study has certain limitations, as it focused only on
metabolic changes in the experimental group without a non-IPC metabolic
group as a control. This limits our ability to fully comprehend the
specific impact and magnitude of IPC on the Taekwondo exercise.
Additionally, owing to limited research methods, the dissolved O[2] in
blood and deutenomics arguments were not determined, preventing the
assessment of IPC’s effect on mitochondrial ATPase functions and Acyl-
and acetyl-carnitine profiles. Future research will focus on targeted
testing based on mechanisms identified through metabolomics to
elucidate how IPC enhances sports performance.
Acknowledgements