Abstract

   Ketogenic diet (KD) has been proven to be an optional avenue in weight
   control. However, the impacts of KD on muscle strength and exercise
   endurance remain unclear. In this study, mice were randomly allocated
   to normal diet and KD groups to assess their exercise tolerance and
   transcriptomic changes of the gastrocnemius. KD suppressed body-weight
   and glucose levels and augmented blood ketone levels of mice. The total
   cholesterol, free fatty acids, and β-hydroxybutyric acid levels were
   higher and triglycerides and aspartate aminotransferase levels were
   lower in KD group. There was no notable difference in running
   distance/time and weight-bearing swimming time between the two groups.
   Furthermore, KD alleviated the protein levels of PGC-1α, p62, TnI FS,
   p-AMPKα, and p-Smad3, while advancing the LC3 II and TnI SS protein
   levels in the gastrocnemius tissues. RNA-sequencing found that 387
   differentially expressed genes were filtered, and Cpt1b, Acadl, Eci2,
   Mlycd, Pdk4, Ptprc, C1qa, Emr1, Fcgr3, and Ctss were considered to be
   the hub genes. Our findings suggest that KD effectively reduced body
   weight but did not affect skeletal muscle strength and exercise
   endurance via AMPK/PGC-1α, Smad3, and p62/LC3 signaling pathways and
   these hub genes could be potential targets for muscle function in
   KD-treated mice.

   Keywords: ketogenic diet, exercise tolerance, lipid metabolism,
   high-throughput sequencing

1. Introduction

   The classic definition of the ketogenic diet (KD) is a formula diet
   with high fat, low carbohydrate, and normal protein [[27]1]. Compared
   with an ordinary diet, the body mainly relies on the ketone body
   produced by fat oxidation for energy after consuming KD [[28]2]. The
   ketone body was composed of 78% β-hydroxybutyrate (β-HB), 20%
   acetoacetic acid, and 2% acetone [[29]3]. The ketone body is a normal
   intermediate product of fatty acid decomposition and oxidation in the
   liver, its molecule is small and easily soluble in water, and can pass
   through the capillary wall of muscle and the blood–brain barrier, which
   is an important source of energy in muscle, especially brain tissues
   [[30]4]. In case of long-term starvation or insufficient sugar supply,
   the ketone body could not only replace glucose (Glu) and become a
   significant energy source for metabolic tissues such as the brain and
   muscle but also participate in the modulation of various metabolic
   pathways of the body as signal molecules [[31]5,[32]6].

   The classic KD was first put forth in 1921 [[33]1]. Except for classic
   KD, there were numerous types of KDs such as using medium chain
   triglycerides, long chain triglycerides, and low glycemic index
   treatment [[34]1]. With the ongoing advancement of research, a previous
   report clarified that KD has been developed and successfully taken to
   treat epilepsy, especially to reduce seizures in children who do not
   respond to pharmacological interventions [[35]7]. On the other side,
   Marchiò et al. found that ghrelin and des-acyl ghrelin were
   downregulated in children affected by refractory epilepsy treated with
   the KD in a short term or even after 1 year, which may associate with
   growth retardation in children [[36]8,[37]9]. In addition, scientific
   research confirmed that KD relieved the hyperglycemia symptoms of type
   2 diabetes mellitus by producing ketone bodies and using ketone bodies
   for energy [[38]10]. Growing evidence has extended KD to metabolic
   diseases such as obesity, fatty liver, cardiovascular, and genetic
   kidney diseases, but also certain types of cancer [[39]11–[40]13]. It
   is widely known that this dietary method reduces body weight, the most
   famous of which is the “Atkins diet,” whose effectiveness in weight
   control has been confirmed [[41]14,[42]15]. High-carbohydrate (CHO)
   diets have long been used as a common dietary strategy to ensure that
   athletes have adequate muscle glycogen stores for better performance
   during training or pre-competition. However, the use of high-CHO
   dietary strategies exposes athletes to a higher risk of chronic
   diseases such as diabetes and cardiovascular disease. Research
   indicated that KD for a while contributes to converting the way the
   body uses energy, from carbohydrate-centered to fat-based energy
   production, i.e., “fat adaptation” [[43]16,[44]17]. KD could
   efficiently lower body weight, lower fasting blood Glu, minimize
   oxidative stress brought on by exercise, and promote adipose tissue
   oxidation when compared to a high-CHO diet [[45]15–[46]17].
   Additionally, recent research has demonstrated that KD has
   neuroprotective effects, ameliorates mitochondrial content and
   function, activates autophagy, and enhances antioxidant and
   anti-inflammatory effects [[47]18–[48]22]. However, as a special
   dietary pattern, whether KD affects skeletal muscle strength and
   exercise endurance has not been reported.

   As we know, KD has a better treatment effect for a variety of diseases,
   including tumors, diabetes, and asthma. The mechanism of action of KD
   is complex. For tumors, tumor growth can be inhibited by enhancing
   oxidative stress, cellular autophagy, and reducing blood Glu and
   glucose transporter 1 expression levels [[49]22]. In terms of
   neurological diseases, the protective effect of KD was associated with
   inhibiting the inflammatory response, enhancing mitochondrial function
   and improving oxidative stress. Generally, KD has important clinical
   applications as an adjunctive dietary treatment method. Although
   numerous clinical studies indicate the potential and advantages of KD
   in patients or athletes. Nevertheless, KD the participating molecules
   and events that determine the exercise endurance in athletic sports
   have not yet been fully elucidated.

   In this study, C57BL/6 mice were fed KD and a normal diet (ND) for 4
   weeks, respectively, to observe and measure the exercise endurance and
   lipid metabolism level of mice. Meanwhile, transcriptome sequencing was
   used to analyze the effect of KD on genomic changes in gastrocnemius of
   C57BL/6 mice. The present study aimed to determine the effect and
   potential mechanism on strength and exercise endurance of skeletal
   muscle in mice treated with a KD.

2. Materials and methods

2.1. Animals

   Twenty male C57BL/6 mice with SPF grade (6–8-week old, weight varies
   from 18 to 22 g) were introduced from Shanghai SLAC Laboratory Animal
   Co., Ltd (SCXK (Hu) 2013-0016, Shanghai, China). All mice were reared
   in an SPF laboratory animal barrier environment with temperature
   (20–25°C), humidity (40–60%), and light (12 h light/dark cycle).

   Ethical approval: The research related to animal use has been complied
   with all the relevant national regulations and institutional policies
   for the care and use of animals.

2.2. Experimental design

   After 1 week of adaptive feeding, mice were randomly divided into the
   ND and KD groups, with ten mice in each group. Among them, ketogenic
   feed was purchased from Beijing Botai Hongda Biological Co., Ltd
   ([50]D12358, Beijing, China) and standard feed was acquired from the
   Experimental Animal Center of Xi’an Jiaotong University (D12450B, Xian,
   China). Each animal was fed in a single cage and their body weight and
   food intake were recorded daily. Animals in the ND group were given a
   conventional diet ad libitum, whereas those in the KD group received a
   KD of equivalent energy based on the average of food consumption in the
   first 24 h in the ND group for a period of 4 weeks. All animals
   utilized in this experiment were allowed by the Ethics Committee of
   Hangzhou Eyong Biotechnological Co., Ltd Animal Experimental Center
   (SYXK (Zhe) 2021-0033). We spared no effort to mitigate the suffering
   of animals during the experiments.

2.3. Determination of blood Glu and blood ketone

   After 4 weeks of KD, we harvested blood samples from the tail vein of
   mice. Glu and blood ketone were determined via a blood ketone meter
   (Abbott Diabetes Care, USA) and blood glucose meter (580, Yuwell,
   China), respectively.

2.4. Assessment of blood biochemical indexes

   After 4 weeks of KD, the harvested blood samples underwent
   centrifugation (5,000 rpm, 20 min). The obtained supernatant was known
   as serum. Serum total cholesterol (TC) kit (CS0005), triglycerides (TG)
   kit (TR0100), free fatty acids (FFA) kit (MAK044), β-HB acid kit
   (MAK041), aspartate aminotransferase (AST) kit (MAK055), and alanine
   aminotransferase (ALT) kit (MAK052) were introduced from Sigma-Aldrich
   (USA). All experimental operations were done in strict accordance with
   the instructions of the corresponding kit to evaluate the content of
   blood biochemical indexes.

2.5. Endurance test

   Seven days before exhaustive exercise, each mouse ran on a treadmill
   for 10 min at a speed of 15 m/min. The endurance test was carried out
   on an open treadmill (Natsume, Kyoto, Japan). The endurance test
   started at 10 m/min for 15 min, followed by 15 and 20 m/min for 15 min
   each, and then 24 m/min and 7% grade till exhaustion. We recorded the
   running distance and running time.

2.6. Weight-bearing swimming experiment

   After 4 weeks of KD, mice were subjected to a weight-bearing swimming
   test to assess the endurance of mice [[51]23]. The homogeneous iron
   ring was placed 2 cm away from the tail tip of the mouse, and the
   effective weight bearing was 3% of the body weight. The water
   temperature and water depths were 25 ± 1°C and 50 cm, respectively.
   Depletion time was determined by recording each mouse from the time
   point at which it was unable to surface within 10 s of release to
   submersion. Furthermore, the levels of Glu and blood ketone in mice
   were detected before and within 3 min after weight-bearing swimming.

2.7. Hematoxylin–eosin (H&E) staining

   After the weight-loaded forced swimming test, each mouse was euthanized
   by inhaling excess CO[2]. The gastrocnemius was quickly removed. The
   gastrocnemius tissue fixation was done with the aid of 10% formalin
   (E672001, Sangon, China). Following dehydration and embedded in
   paraffin, the tissues were sectioned to 5 µm thick. After that, the
   slices were dewaxed and hydrated. After being stained with hematoxylin
   (BL700A, Biosharp, China) for 5 min, differentiation was performed with
   1% hydrochloric acid for 30 s. After counterstaining with eosin for
   3 min, the slices were dehydrated. We chose xylene to permeate the
   slices. Pathological changes of gastrocnemius in mice were monitored
   with the help of an optical microscope (ECLIPSE 80i, Nikon, Japan).

2.8. Western blotting

   The acquired gastrocnemius tissues were lysed with the assistance of a
   lysis buffer (abs9229, absin, China). Subsequently, we selected the BCA
   kit (BI-WB005, SBJBIO, China) to quantify the lysed protein.
   Thereafter, the quantified protein was electrophoresed for protein
   separation, which was then loaded onto the PVDF membrane (PW0034,
   Leagene, China). After being sealed in 5% bovine serum albumin (BL-082,
   SBJBIO, China) at 37°C for 60 min, the membrane underwent primary
   antibodies (4°C, overnight). Afterward, the bound antibodies were then
   exposed to anti-rabbit secondary antibody (31466, Invitrogen, USA) or
   anti-mouse secondary antibody (S0002, Affinity, USA) with the condition
   of 37°C for 60 min. Visualization of protein was conducted by applying
   an ECL reagent (GK10008, GlpBio, USA) on a gel imaging system (A44114,
   Invitrogen, USA). The primary antibodies of PGC-1α (1:1,000, ab191838),
   P62 (1:10,000, ab109012), LC3B (1:2,000, ab48394), AMPKα (1:5,000,
   ab32047), p-AMPKα (1:10,000, ab133448), Smad3 (1:1,000, ab208182),
   p-Smad3 (1:2,000, ab52903), and β-actin (1:5,000, ab8227) were provided
   by Abcam (UK).

2.9. RNA extraction and library construction for RNA-seq

   To check the role of KD on the genomic changes of mice, an RNA-seq
   assay was done on gastrocnemius tissues of ND mice and KD mice. Total
   RNA extraction was conducted with Trizol (10296010, Invitrogen, USA).
   For the detection of the quality and purity of total RNA, we applied
   NanoDrop 2000 (Thermo Scientific, USA). The integrity of RNA was
   measured by using Bioanalyzer 2100 (Agilent, CA, USA). Next, sequencing
   libraries were conducted by applying TruSeq RNA Library Prep Kit v2
   (RS-122-2001/2, Illumina, USA). Afterward, the library was paired-end
   sequenced in a Hiseq 2500 platform (Illumina, USA).

2.10. Raw data filtering

   The sequencing data contained some reads with adapters and low quality,
   which would cause a great interference to the subsequent information
   analysis. Therefore, the sequencing data should be further filtered.
   The criteria for data filtering majorly included the application of
   Cutadapt to remove sequences with adapters at the 3′ end and reads with
   an average quality score below Q20.

2.11. Differential expression analysis

   We adopted R package DESeq2 to analyze significant differences in
   gastrocnemius samples between ND group and KD group. The conditions for
   screening differentially expressed genes (DEGs) were expression
   difference fold |log 2^FoldChange| >1 and significant P-value <0.05.

2.12. Clustering analysis

   The Pheatmap software package in R language was applied for
   bidirectional clustering analysis of DEGs in ND and KD samples. Cluster
   analysis was conducted based on the expression level of the same gene
   in different specimens and the expression pattern of different genes in
   the same specimen. The Euclidean method was taken to calculate the
   distance, and the hierarchical clustering longest distance method was
   exploited for clustering.

2.13. Functional enrichment analysis and construction of protein–protein
interaction (PPI) network

   For identification of mRNAs in the ND group and KD group, Gene Ontology
   (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment
   analyses were done by applying Cytoscape plug-in ClueGO
   ([52]http://apps.cytoscape.org/apps/cluego). After that, the STRING
   database ([53]https://string-db.org/) was selected to analyze the
   protein interaction of the predicted target genes. The MCODE plug-in in
   Cytoscape was exploited to construct a PPI network.

2.14. Statistical analysis

   All quantitative data were described as mean ± standard deviation (SD).
   Data comparisons were done with the aid of a one-way analysis of
   variance for multiple time points with Tukey’s post hoc test in
   [54]Figure 1 by using SPSS 16.0 (IBM, SPSS, USA). Kruskal–Wallis H test
   was applied to the heterogeneity of variance. Statistical analysis
   between the two groups was tested by a two-tailed Student’s t-test.
   Statistically, P < 0.05 was meaningful.

Figure 1.

   [55]Figure 1
   [56]Open in a new tab

   KD suppressed body weight and Glu level and augmented blood ketone
   level of C57BL/6 mice. (a) Body weight trajectories of mice on KD for 4
   weeks. (b) Blood Glu level of mice on the KD for 4 weeks. (c) Blood
   ketone level of mice on the KD for 4 weeks. Data are expressed as mean
   ± SD (n = 10/group). Statistical significance was determined by one-way
   ANOVA. ^▲ P < 0.05, ^▲▲ P < 0.01 vs ND group. ND, normal diet; KD,
   ketogenic diet.

3. Results

3.1. KD suppressed body weight and Glu level and augmented blood ketone level
of C57BL/6 mice

   The results of body weight monitoring clarified that from the second
   week, the body weight of mice in the KD group was lower relative to the
   ND group ([57]Figure 1a, P < 0.05). We also discovered that KD led to a
   decrease in Glu level from the second week ([58]Figure 1b, P < 0.05).
   Meanwhile, compared to the ND group, the blood ketone level in the KD
   group was largely upregulated ([59]Figure 1c, P < 0.01).

3.2. Effect of KD on lipid metabolism levels in C57BL/6 mice

   By conducting an ELISA assay, we found that relative to the ND group,
   the results elucidated that the contents of serum TC, FFA, and β-HB
   were higher and the contents of serum TG and AST were lower in the KD
   group ([60]Figure 2a–e, P < 0.01). There was no obvious difference in
   serum ALT levels between the KD group and the ND group ([61]Figure 2f).

Figure 2.

   [62]Figure 2
   [63]Open in a new tab

   Effect of KD on lipid metabolism levels in C57BL/6 mice. Levels of (a)
   TC, (b) TG, (c) FFA, (d) β-HB, (e) AST, and (f) ALT in serum samples of
   mice treated with KD for 4 weeks. Data are expressed as mean ± SD.
   Differences of data in mice were assessed by two-tailed Student’s
   t-test. ^▲▲ P < 0.01 vs ND group. ND, normal diet; KD, ketogenic diet.

3.3. Effect of KD on exercise tolerance in C57BL/6 mice

   The endurance test results revealed that there was no notable
   difference in running distance and running time between KD group and ND
   group ([64]Figure 3a and b). Also, a weight-loaded forced swimming test
   result unveiled that the weight-bearing swimming time of the KD group
   and ND group mice had no prominent change ([65]Figure 3c). Furthermore,
   there was no evident change in Glu level before and after swimming in
   the KD group and ND group ([66]Figure 3d). The blood ketone level of
   mice in the ND group after weight-bearing swimming was higher than that
   before swimming ([67]Figure 3e, P < 0.01), while there was no notable
   change in blood ketone level in the KD group before and after
   weight-bearing swimming.

Figure 3.

   [68]Figure 3
   [69]Open in a new tab

   Effect of KD on exercise tolerance in C57BL/6 mice. (a) Running
   distance, (b) running time, (c) weight-bearing swimming time of mice
   treated with KD for 4 weeks. (d) Levels of blood Glu and (e) ketone in
   mice before and after swimming. Data are expressed as mean ± SD.
   Differences of data in mice were assessed by two-tailed Student’s
   t-test. ^▲▲ P < 0.01 vs ND group. ND, normal diet; KD, ketogenic diet.

3.4. Effect of KD on pathological changes, muscle fiber-, and
autophagy-related proteins of gastrocnemius tissues in mice

   In this part, we monitored the pathological changes of gastrocnemius
   tissues in the KD group and ND group by H&E staining. Our results
   demonstrated that there were no obvious pathological changes in the
   gastrocnemius of mice in the KD and ND groups ([70]Figure A1a). To
   further explore the relationship between exercise endurance and muscle
   fiber type, the autophagy, AMPK/PGC-1α, and Smad3 pathway-associated
   markers were measured. Western blot experiment documented that KD was
   found to alleviate the protein levels of PPARγ coactivator-1α (PGC-1α),
   sequestosome 1 (p62/SQSTM1), and troponin I-FS (TnI FS) and the ratios
   of p-AMPKα/AMPKα and p-Smad3/Smad3, while advancing the protein
   expressions of LC3II/LC3I and troponin I-SS (TnI SS) in the
   gastrocnemius tissues ([71]Figure A1(b and c), Appendix file 1).

3.5. Identification and functional enrichment analysis of DEGs of
gastrocnemius tissues in KD-treated mice

   The volcano plots of [72]Figure 4a exhibited the DEGs in two
   gastrocnemius samples. Relative to the ND group, 162 genes showed high
   expression and 225 genes displayed low expression in the KD group. The
   cluster analysis of DEGs in the results presented that the top 100
   genes with the lowest P value were applied for mapping, among them, red
   represented upregulated genes and green manifested downregulated genes
   ([73]Figure 4b). Afterward, GO and KEGG enrichment of DEGs was done.
   Biological processes, cellular components, and molecular functions
   presented the top ten signaling pathways, respectively ([74]Figure 5).
   The KEGG enrichment analysis scatter plot was based on the enrichment
   significance (P-value) of the top 20 pathways for plotting ([75]Figure
   6).

Figure 4.

   [76]Figure 4
   [77]Open in a new tab

   Identification of DEGs in gastrocnemius tissues in KD treated mice. (a)
   Volcano plot: two black vertical lines represent 1.5 times upregulation
   and 1.5 times downregulation, green horizontal lines indicate P-value
   of 0.05. (b) Heatmap showing the expressive pattern of DEGs. Red
   represents upregulated genes and green represents downregulated genes.

Figure 5.

   [78]Figure 5
   [79]Open in a new tab

   GO analysis of DEGs in mice after KD treatment.

Figure 6.

   [80]Figure 6
   [81]Open in a new tab

   KEGG pathway analysis of DEGs in mice after KD treatment.

3.6. Identification of hub genes of upregulated and downregulated in PPI
network

   We mapped upregulated DEGs to the STRING database to establish a PPI
   network, and then re-imported these interactive network data into
   Cytoscape software. The PPI network is composed of 97 nodes and 210
   edges. Hub genes acquired by the Degree algorithm, carnitine
   palmitoyltransferase 1B (Cpt1b), long-chain acyl-CoA dehydrogenase
   (Acadl), enoyl-CoA-(delta) isomerase 2 (Eci2), malonyl-CoA
   decarboxylase (Mlycd), and pyruvate dehydrogenase kinase 4 (Pdk4) were
   seen as key upregulated genes ([82]Figure 7). We also mapped the
   downregulated DEGs to the STRING database to build a PPI network and
   re-imported them into Cytoscape software. PPI network composed of 149
   nodes and 684 edges, and protein tyrosine phosphatase receptor type C
   (Ptprc), complement C1q (C1qa), epidermal-growth-factor-like-module
   receptor 1 (Emr1), Fc gamma receptor III (Fcgr3), and cathepsin S
   (Ctss) were regarded as key downregulated genes ([83]Figure 8).

Figure 7.

   [84]Figure 7
   [85]Open in a new tab

   PPI network of upregulated DEGs and identification of hub genes. We
   mapped upregulated DEGs to the STRING database to construct a PPI
   network, and then re-imported these interactive network data into
   Cytoscape software. PPI network composed of 97 nodes and 210 edges. Hub
   genes obtained by the Degree algorithm, Cpt1b, Acadl, Eci2, Mlycd, and
   Pdk4 were considered to be key upregulated genes.

Figure 8.

   [86]Figure 8
   [87]Open in a new tab

   PPI network of downregulated DEGs and identification of hub genes. We
   also mapped downregulated DEGs to the STRING database to construct a
   PPI network and re-imported them into Cytoscape software. PPI network
   composed of 149 nodes and 684 edges. Hub genes obtained by Degree
   algorithm, Ptprc, C1qa, Emr1, Fcgr3, and Ctss were considered to be key
   downregulated genes.

4. Discussion

   Mounting reports illustrated that KD could control body weight on the
   premise of providing sufficient energy for the body, so KD may have a
   certain application value in sports involving weight classes
   [[88]24,[89]25]. However, research on the relationship between KD and
   exercise is still very limited. In this study, C57BL/6 mice were used
   as the research objects. After 4 weeks of feeding, the results
   displayed that the weight of mice in the KD group was lower than the ND
   group from the second week, which was coincident with the results of
   other literatures [[90]26,[91]27]. These results illuminated that KD
   could lessen the body weight of mice under the condition of equal
   energy feeding.

   KD augmented the level of blood ketone and decreased the level of Glu
   in mice, suggesting that the increase of ketone bodies as energy supply
   substances is one of the main manifestations of the KD animal model
   [[92]4]. The ketone body is an intermediate product of fatty acid
   oxidation in the liver, and changes in the blood ketone body may have
   an impact on the energy metabolism of the body [[93]28]. A relevant
   study pointed out that the utilization of the ketone body by
   extrahepatic tissues and organs in ND mice after exercise is lower than
   the production level of the ketone body, so blood ketone level
   increases after exercise [[94]29]. Weight-bearing swimming is a
   considerable indicator for evaluating exercise endurance in mice
   [[95]23]. The results of this experiment found that there was no
   notable difference in weight-bearing swimming time and distance between
   the KD group and ND group, which was consistent with the results
   reported by Helge et al. [[96]30], and the possible reason was that a
   high-fat diet could change the utilization rate of metabolic substrates
   and increase the fat oxidation rate during exercise [[97]31]. In
   addition, there was no obvious change in blood Glu level before and
   after weight-bearing swimming in the KD group and ND group, suggesting
   that KD had no evident effect on the exercise tolerance of mice, and
   the blood Glu level remained relatively stable before and after
   exercise.

   We also assessed blood ketone levels in mice before and after
   weight-bearing swimming. In this study, the blood ketone level of mice
   in the ND group after swimming was effectively enhanced than those
   before swimming, which was in line with the increase in blood ketone
   level after exercise; while there was no distinct difference in the
   blood ketone level of mice in the KD group before and after swimming,
   indicating that feeding KD for 4 weeks changed the ability of mice in
   the KD group to utilize ketone bodies, namely “fat adaptation”
   [[98]14]. In addition, plasma metabolic parameters (TC, TG, FFA, β-HB,
   AST, and ALT) were measured in this study to characterize the effect of
   KD on metabolism. It has been reported in the literature that after 9
   weeks of administration of KD to mice, a significant decrease in body
   weight can be observed, plasma FFA and ketone body are notably
   increased, and Glu is evidently decreased [[99]32]. β-HB is mainly
   derived from the oxidation of fatty acids and exported to the
   surrounding tissues for energy [[100]4]. The concentration of β-HB was
   largely enhanced in the KD group, indicating the utilization of fat or
   ketone body [[101]33,[102]34]. By conducting ELISA analysis, the
   results illustrated that the contents of serum TC, FFA, and β-HB were
   higher and the contents of serum TG and AST were lower in the KD group
   when compared to the ND group. These results unveiled that KD
   transferred energy substrates from Glu to fat.

   The literature indicated that the changes in the expression levels of
   different functional signaling pathways, such as AMPKα, Smad3, and
   PGC-1α, have the function of regulating the composition of skeletal
   muscle fiber types [[103]35]. PGC-1α is a key factor in promoting
   mitochondrial biosynthesis [[104]36]. AMPK is a key sensor of skeletal
   muscle energy status and modulates Glu and fatty acid metabolism
   [[105]37]. AMPK can regulate PGC-1α through phosphorylation and
   deacetylation, and participate in mitochondrial function and
   biosynthesis [[106]38,[107]39]. High-fat diet feeding has been
   manifested to weaken the expression and activity of PGC-1 and AMPK as
   well as downstream signaling in skeletal muscle [[108]40]. The results
   by western blot illustrated that KD was found to alleviate the protein
   level of PGC-1α and the ratios of p-AMPKα/AMPKα and p-Smad3/Smad3 in
   the gastrocnemius tissues. In order to further verify the effect of KD
   on muscle fiber types, this experiment detected TnI SS, which can
   reflect the content of slow-twitch muscle fibers in skeletal muscle,
   and TnI FS, which can reflect the content of fast-twitch muscle fibers
   in skeletal muscle. The contents of TnI SS and TnI FS in different
   types of muscle fibers were detected by western blot, and it was found
   that the expression of TnI SS was increased and the expression of TnI
   FS was decreased in the KD group in the gastrocnemius, indicating that
   KD can effectively increase the expression of slow muscle fibers in
   mixed-type muscle fibers.

   In addition, autophagy is a universal life phenomenon of eukaryotic
   cells, responsible for the removal of damaged and dysfunctional
   organelles. Autophagy pathway activation is characterized by the
   formation of autophagosomes in LC3-containing cells. P62 is a selective
   substrate of LC3 and is continuously consumed during autophagic
   activation. The level of expression of p62 and LC 3 indicators can be
   used to assess the degree of autophagy pathway activation [[109]41]. Mu
   et al. found that KD intervention could mitigate pulmonary fibrosis in
   vivo via upregulating the expression of LC3 II/LC3 I and Beclin1 and
   downregulating the expression of P62 [[110]42]. This research found
   that KD effectively repressed the protein level of P62, while advancing
   the ratio of LC3II/LC3I in the gastrocnemius tissues, illuminating that
   KD triggered autophagy in C57BL/6 mice.

   With the development of RNA-seq technology, second-generation
   sequencing has been broadly utilized in genome research [[111]43]. To
   further analyze the effect of KD on genomic changes in C57BL/6 mice,
   RNA-seq analysis was performed on gastrocnemius samples from mice in
   the KD and ND groups. As a result, a total of 387 DEGs were filtered
   between ND and KD-treated groups. Importantly, the GO and KEGG pathway
   enrichment analysis found that DEGs participated in fatty acid
   elongation, fatty acid degradation, fatty acid metabolic process, etc.
   It is well known that fatty acids are closely related to skeletal
   muscle strength and exercise endurance [[112]44,[113]45]. In future
   research, this study will explore the effects of KD on skeletal muscle
   strength and exercise endurance through fatty acid metabolism-related
   pathways. Additionally, hub genes were obtained by using Cytoscape
   software, Cpt1b, Acadl, Eci2, Mlycd, Pdk4, Ptprc, C1qa, Emr1, Fcgr3,
   and Ctss were considered to be key genes.

   By searching the literature related to skeletal muscle strength and
   exercise endurance of the above ten hub genes, Cpt1b, Acadl, Eci2,
   Mlycd, Pdk4, Ptprc, C1qa, Emr1, Fcgr3, and Ctss are all closely
   correlated with skeletal muscle strength and athletic endurance. Cpt1b
   is a mitochondrial enzyme responsible for the formation of
   acylcarnitine, which helps to regulate fatty acids β-oxidation
   [[114]46,[115]47]. It was reported that Cpt1b is highly expressed in
   skeletal muscle, heart, and adipose tissues [[116]48]. Acadl is a key
   enzyme for mitochondrial fatty acid oxidation, which is notably lower
   by exercise in Nmrk2^−/− mice [[117]49,[118]50]. Eci2 is a peroxisomal
   and/or mitochondrial protein and Eci2 knockdown lessens Glu consumption
   [[119]51,52]. Mlycd is a considerable enzyme for fatty acid metabolism
   and acyl-CoA synthesis and metformin restrains lipid accumulation in
   obese adults via the upregulation of Mlycd [[120]53]. Pdk4 is a crucial
   regulator of cellular energetic metabolism, which is widely expressed
   in skeletal muscle tissues including humans and rodents [[121]54].
   Ptprc (also known as CD45) is an essential transmembrane glycoprotein
   on the cell surface of the hematology and immune system [[122]55]. It
   was reported that exogenous C1qa weakened skeletal muscle regeneration
   in young mice, while C1qa destruction restored aging-related muscle
   regeneration injury [[123]56]. Emr1, the adipocyte gene, is related to
   advanced age and diet-induced obesity [[124]57]. Fcgr is a protein
   family expressed by a variety of immune cells, among which Fcgr3 is one
   of the members of the Fcgr protein family [[125]58]. Ctss is a key
   member of the cysteine protease family, and its inhibitor could reduce
   fat production and liver fat accumulation in obese mice mediated by a
   high-fat diet [[126]59]. The expression and activity of Ctss were
   upregulated in the skeletal muscles of mice with Duchenne muscular
   dystrophy [[127]60]. Therefore, their correlations with skeletal muscle
   strength and exercise endurance warrant further investigation, as these
   genes are potential targets of KD.

5. Conclusions

   This study provides a reference for the dietary pattern of KD to lose
   weight, and maintain muscle strength and exercise capacity, and its
   mechanism is related to the regulation of skeletal muscle fiber
   type-related proteins and autophagy-related proteins. However, whether
   longer-term KD affects aspects such as skeletal muscle strength remains
   to be further investigated.

Appendix

Figure A1.

   [128]Figure A1
   [129]Open in a new tab

   Effect of KD on pathological changes, muscle fiber, and
   autophagy-related proteins of gastrocnemius tissues in mice. (a) HE
   staining of the gastrocnemius tissues showed that KD had no
   differential influence in mice. (b) Representative protein bands and
   (c) quantification data of PGC-1α, p62, LC3, TnI SS, TnI FS, p-AMPKα,
   AMPKα, p-Smad3, and Smad3 in gastrocnemius tissues of mice treated with
   KD for 4 weeks. Data are expressed as mean ± SD. Differences of data in
   mice were assessed by two-tailed Student’s t-test. ^▲ P < 0.05, ^▲▲ P <
   0.01 vs ND group. ND, normal diet; KD, ketogenic diet.

Footnotes

   Funding information: Authors state no funding involved.

   Author contributions: Jie Zhang and Bo Chen have given substantial
   contributions to the conception or the design of the manuscript, Jie
   Zhang and Ke Zou to acquisition, analysis, and interpretation of the
   data. All authors have participated to drafting the manuscript, author
   A revised it critically. All authors read and approved the final
   version of the manuscript.

   Conflict of interest: Authors state no conflict of interest.

   Data availability statement: The datasets generated during and/or
   analyzed during the current study are available from the corresponding
   author on reasonable request.

References