Abstract

   The benefits of physical exercise on human health make it desirable to
   identify new approaches that would mimic or potentiate the effects of
   exercise to treat metabolic diseases. However, whether far-infrared
   (FIR) hyperthermia therapy could be used as exercise mimetic to realize
   wide-ranging metabolic regulation, and its underling mechanisms remain
   unclear. Here, a specific far-infrared (FIR) rays generated from
   graphene-based hyperthermia devices might promote exercise capacity and
   metabolisms. The material characterization showed that the graphene
   synthesized by chemical vapour deposition (CVD) was different from
   carbon fiber, with single-layer structure and high electrothermal
   transform efficiency. The emission spectra generated by graphene-FIR
   device would maximize matching those adsorbed by tissues. Graphene-FIR
   enhanced both core and epidermal temperatures, leading to increased
   blood flow in the femoral muscle and the abdominal region. The
   combination of microbiomic and metabolomic analysis revealed that
   graphene-FIR modulates the metabolism of the gut-muscle axis. This
   modulation was characterized by an increased abundance of short-chain
   fatty acids (SCFA)-producing bacteria and AMP, while lactic acid levels
   decreased. Furthermore, the principal routes involved in glucose
   metabolism, such as glycolysis and gluconeogenesis, were found to be
   altered. Graphene-FIR managed to stimulate AMPK activity by activating
   GPR43, thus enhancing muscle glucose uptake. Furthermore, a microbiota
   disorder model also demonstrated that the graphene-FIR effectively
   restore the exercise endurance with enhanced p-AMPK and GLUT4. Our
   results provided convincing evidence that graphene-based FIR therapy
   promoted exercise capacity and glucose metabolism via AMPK in
   gut-muscle axis. These novel findings regarding the therapeutic effects
   of graphene-FIR suggested its potential utility as a mimetic agent in
   clinical management of metabolic disorders.

   Keywords: Graphene, Far-infrared, AMPK, Glucose metabolism, Gut-muscle
   axis

   Subject terms: Physiology, Health care, Medical research, Molecular
   medicine

Introduction

   The human body benefits from exercise training, which enhances the
   cardiovascular system and regulates metabolism. Obesity, diabetes,
   cancer, and the burden of heart disease appear to be correlated with
   sedentary lifestyles^[44]1. According to reports, inactive individuals
   have shorter life expectancies than physically active
   individuals^[45]2. Many individuals and patients with these illnesses,
   however, are unable or unwilling to engage in physical activity.
   Metabolic modulators like PPAR agonist [46]GW501516 and AMP-activated
   protein kinase (AMPK) agonist, termed "exercise mimetics", have been
   explored for their potential to emulate the favorable effects of
   aerobic exercise. These agents show promise in aiding the management of
   obesity and diabetes and in enhancing endurance, independent of
   physical training^[47]3.

   According to a previous study, aldometanib, an AMPK agonist, increases
   the lifespan of mice^[48]4. However, due to abuse issues, there is
   significant discussion about the use of AMPK activators. AICAR, as an
   AMPK activator, is included in the prohibited list of substances by the
   World Anti-Doping Agency^[49]5. Additionally, a single pharmacological
   substance would not be adequate to simulate the complicated and
   extensive effects of physical activity because exercise has
   broad-ranging impacts on numerous types of cells, tissues, and organs.

   The whole-body far-infrared ray (FIR) hyperthermia, also known as Waon
   therapy, has been employed in the management of diverse pathological
   conditions and symptomatic relief^[50]6. Previous studies demonstrated
   that FIR hyperthermia therapy has anti-ischemic condition,
   anti-inflammatory, anti-depression activity based on the biological
   activities of radiation^[51]7,[52]8. Impressively, hyperthermia therapy
   contributes to the prevention and treatment of obesity through the
   activation of beige fat, mediated by the HSF1-A2B1 transcriptional
   axis^[53]9. It has also been reported that hyperthermia prevented bone
   loss by influencing the gut microbiota^[54]10. However, it is not clear
   whether far-infrared (FIR) hyperthermia therapy could be used as
   exercise mimetic to realize wide-ranging metabolic regulation.

   FIR radiation (3–1000 µm) transmits energy solely as radiant heat that
   can reach up to 4 cm below the skin. In addition to resonating at
   cellular frequencies, FIR energy can induce rotational and vibrational
   motions in cellular molecules, particularly affecting water
   molecules^[55]11. The human skin can efficiently absorb the efficient
   infrared radiation produced by FIR emitters, which resonates with the
   human body and accelerates metabolism and blood flow^[56]12.
   Non-absorbed infrared radiation may build up on the body's surface and
   cause skin abrasion. The FIR devices receive energy from the human
   body, acting as "absorbers" and keeping their temperature high enough
   to re-emit the FIR back to the body^[57]6.

   The efficacy of FIR hyperthermia has been augmented with the advent of
   improved methods for generating pure FIR emissions. However, FIR
   hyperthermia equipment with high resonance effect and electrical
   stability is absent in practical applications. Moreover, achieving an
   optimal balance between therapeutic efficiency and minimizing adverse
   effects poses a significant challenge. Resonance adsorption would
   increase adsorption efficiency and reduce side effects in living
   tissue, making it desirable to have a safe FIR generating source whose
   emission spectra match those adsorbed by tissues. Previous studies have
   shown that the blackbody limit in the far-field may be eluded by the
   radiative heat transfer rate at nanogaps^[58]13–[59]15. Carbon-based
   nanomaterials, such as carbon nanotube and graphene, have atomically
   restricted lattice structures and can support a significantly higher
   current density than traditional metals, making them particularly
   potential thermal emitters^[60]16,[61]17. Graphene film has a
   remarkable FIR emissivity over 90%, a large portion of electronic
   energy can be converted into infrared radiation when a bias voltage is
   applied through the graphene layer. The infrared spectrum
   characteristics of graphene do not change with radiation
   temperature^[62]18. However, the implications of graphene-FIR
   hyperthermia on enhancing physical activity and maintaining metabolic
   equilibrium as an exercise mimetic warrant further investigation.

   It is now widely documented that the gut microbiota plays an important
   role in regulating host metabolism^[63]19. Recent research has
   demonstrated that the gut microbiota has a variety of effects on
   skeletal muscle bioenergetics. For example, micronutrients and
   metabolites derived from the gut microbiota affect energy homeostasis
   in muscle^[64]20. Probiotic supplementation in mice decreases muscular
   atrophy but increases skeletal muscle strength^[65]21,[66]22. A current
   study investigates the induction of microbiome depletion in mice
   through the administration of a broad-spectrum antibiotic mixture (ABX)
   to explore the role of gut microbiota. Microbiota depletion in mice
   exhibit decreases in skeletal muscle mass, strength and mitochondrial
   function^[67]20. The notion of "gut-muscle axis" was proposed to
   describe how gut microbes influence muscle function and endurance
   performance^[68]23. It was reported that accelerated nitrogen recycling
   via gut symbionts may benefit for muscle preservation in humans^[69]24.
   The richness abundances of Veillonella and Lactobacillus salivarius in
   microbiomes could enhance exercise performance and decrease
   fatigue^[70]25,[71]26. The skeletal muscle responds to endurance
   exercise by activating AMPK and remodeling metabolism. Narkar et al.
   had an intriguing reported that activating AMPK could mimic the
   benefits of exercise without training^[72]27. Previous studies
   suggested that AMPK play a key role in the regulation of skeletal
   muscle energy metabolism promoting glucose uptake and fatty acid
   oxidation^[73]28.

   In the present study, we have demonstrated that graphene-FIR therapy
   different from carbon fiber-FIR, exhibiting enhanced electrothermal
   stability and a more pronounced resonance effect in FIR hyperthermia
   treatment. We found that graphene-FIR hyperthermia could promote
   exercise capacity and glucose metabolism via AMPK. We further providing
   a mechanistic understanding of the role of a gut-muscle interaction in
   hyperthermia therapy.

Materials and methods

Animals

   All the animal experiment protocols were approved by the Committee on
   the Ethics of Animal Experiments of Academy of Military Medical
   Sciences and reported in accordance with the ARRIVE guidelines.
   Approval number: IACUC-DWZX-2022–632. We confirmed that all experiments
   were performed in accordance with relevant guidelines and regulations.
   The mice were euthanized by carbon dioxide. The C57BL/6 J male mice
   (Sibeifu, Beijing, China) utilized in the experiments were 8–14 weeks
   old. They were subjected to a 12-h light/12-h dark cycle and were kept
   in a temperature and humidity-controlled environment (24 ± 2 °C,
   55 ± 5% relative humidity). The grouping information of animal
   experiments were provided in the Supporting Information Table 1.
   ABX-mediated microbiota perturbations model was established according
   to previously protocols (details are provided in Supplementary data).

Materials

   The details of materials are provided in Supplementary Data (including
   manufacturers and product numbers).

Infrared spectrum analysis

   The details of FIR spectrum detection and analysis can be found in
   Supplementary data.

FIR irradiation and exposure

   The mice were exposed to FIR radiation in a rectangular plastic chamber
   (40 cm × 25 cm × 20 cm). Each chamber contains 16 FIR films with series
   connection, the single films resistance value is 27 Ω. There are four
   films above, on the left, and on the right of the chamber separately,
   and two films in front and behind. The FIR films were installed in the
   surface of chamber and powered by an independent variable voltage
   power. During the FIR irradiation experiment, the mice were placed in
   the center of chamber and allow to free move. Except for the feet, the
   entire body of a mouse can receive FIR radiation. The temperature was
   approximately (24.0 ± 1) °C in the control cages and (30.0 ± 1) °C in
   the FIR chamber. The experiment room temperature was controlled at
   25 °C. The input power of the graphene-based device and carbon
   fiber-based device were all set at 30 W. The animals were radiated for
   two consecutive weeks, 40 min daily.

Exercise endurance test

   Exercise endurance was determined using a treadmill (Sansbio, Jiangsu,
   China) running tests as previously described with minor
   modified^[74]29. The low and high-intensity exercise test were defined
   by setting different running time. The details of experimental can be
   found in Supplementary data.

Locomotor activity

   To examine the effect of FIR on the locomotor activity, 30 min or 24 h
   after the last FIR radiation (day 14), locomotor activity was examined
   in a 15 min monitoring period. The movement tracking was recording by
   using an automated video tracking system, as described previously.
   Eight test boxes (40 cm × 40 cm × 30 cm high) were operated
   simultaneously by a compatible computer.

Core and surface body temperature

   Body temperature of mice was monitored using Anipill loggers (BodyCap,
   Paris, France) to get temperature data from animals after Anipill
   transmitter implantation. Isoflurane was used to sedate mice before
   surgery. After inserting the Anipill logger, the skin was stitched
   using surgical suture. Prior to the studies, the animals were returned
   to the cage for four days of rest.

   The surface temperature of subjects (mice and FIR film) was tested by
   the infrared imaging. IR imaging tested in a room with a steady
   temperature and humidity. Thermographic camera (UTi320E, China) was
   recorded temperature in mice or FIR film.

Blood flow measurements/laser speckle contrast imaging (LSCI)

   To prevent the carryover effects of differing acclimation temperatures,
   all mice were placed at 25 °C for 1 h before to research. The mice were
   placed in an RFLSI Pro (RWD life sciences) and kept at 1.2% isoflurane.
   A CCD camera detects blood flow and then captures the image. Regions of
   interest (ROIs) were chosen and assess speckle contrast.

Western blot analysis

   As previously reported, SDS-PAGE electrophoresis was used to separate
   an identical amount of protein, which was then transferred to a
   nitrocellulose membrane. 5% skim milk was used to block membranes for
   two hours. To maximize efficiency in isolating the target proteins, the
   strips will be cut post-blocking in alignment with the protein ladder.
   After primary antibody incubation for 12 h at 4 °C, secondary antibody
   was incubated for 2 h. Using a gel imaging analyzer (AlphaImager, USA),
   the optical density of the appropriate immunoreactive band was
   measured.

AMPK, AMP and ATP assay

   The fresh muscle samples were homogenized. The supernatant was
   incubated with the appropriate reaction solution and substrate solution
   in accordance with the kit's instructions. A luciferase reaction was
   used to measure the levels of AMPK, AMP, and ATP using a microplate
   reader (Envision 2104). A standard curve was used to connect the
   luminescence signal to the concentrations of AMPK, AMP, or ATP.

Fluorescence imaging analysis

   In vivo fluorescence imaging was detected and quantified by AniView600
   imaging system (Guangzhou Biolight Biotechnology Co., Ltd). To reduce
   autofluorescence, animals were fasted for 12 h and hair was removed
   from joints. Animals underwent caudal vein injection solution of
   Cy5.5-glucose (2.5 mg/kg). As recommended by the system for image
   acquisition, the excitation and emission filters were set at
   675 nm/Cy5.5 and 740 nm/Cy5.5, respectively. Animals were scanned
   before injection and at 3 h after injection. We calculated average
   Cy5.5-glucose signal in skeletal muscle.

Metabolomics and microbiome analysis

   The details of experimental protocols and analysis methods were
   provided in Supplementary data.

Statistical analysis

   The results were presented as the mean ± SEM. For comparisons involving
   three or more groups, one-way ANOVA was employed, followed by the
   Bonferroni test to determine significance. Body temperature data were
   statistically analyzed using the Holm-Sidak method. The correlation
   between metabolites and running distance was assessed using Pearson’s
   correlation test. P < 0.05 was regarded as statistically significant in
   each case. GraphPad Prism 8.01 was used to conduct statistical
   analysis.

Results

The emission spectrum of FIR rays by graphene-based device

   First, we identified the material characterization of graphene film and
   carbon fiber. The single-layer graphene was grown using CVD on a copper
   foil and to a polyethylene terephthalate (PET) membrane (Fig. [75]1A),
   and the carbon fiber-based device was commercially available
   (Fig. [76]1D). The SEM images showing the morphology of graphene film
   and carbon fiber were different. The graphene film was two-dimensional
   membrane (Fig. [77]1B), and the carbon fiber was existed in the fibrous
   form of graphitized structure (Fig. [78]1E). Raman spectroscopy
   verified the single-layer nature of the graphene film (Fig. [79]1C),
   and the multi-layer nature of the carbon fiber (Fig. [80]1F). In the
   electrothermal characteristic detection, compared to the carbon
   fiber-based device, the power of the graphene-based device was more
   stable, and the heating rate is faster when under the same power
   condition (Fig. [81]S1, Supporting Information).

Figure 1.

   [82]Figure 1
   [83]Open in a new tab

   Specific FIR rays generated by graphene-based devices. (A) Photograph
   of a graphene film. (B) SEM image of graphene film. (C) Raman spectrum
   of the graphene FIR. (D) Photograph of a carbon fiber film. (E) SEM
   image of carbon fiber film. (F) Raman spectrum of the carbon fiber FIR.
   (G) FIR emission spectra of graphene and carbon fiber (graphene: 30 W,
   carbon fiber: 30 W). (H) The FIR emission peak of the graphene was
   approximately 8 µm, the carbon fiber was approximately 7.5 µm. (I) The
   absorption peak of the human body surface was approximately 8.0 µm.

   The emission peak of the single-layer graphene device was different
   from carbon fiber device in the same input powers (Fig. [84]1G–H). We
   also assessed the human body's absorption spectra (F[85]ig. [86]1I).
   The emission peak position of the graphene device (8.0 µm) coincided
   with the characteristic absorption peaks of the human body surface,
   which were around 8.0 µm. For the same power input, the carbon-fiber
   devices' emission peak wavelength was lower than 7.7 µm (Fig. [87]1H).
   These data showed that the FIR generated by graphene may have more
   efficiently resonated with epidermal molecules.

Graphene-FIR hyperthermia increased exercise endurance in mice

   The diagram of FIR devices by graphene or carbon fiber and the
   experimental design were shown in Fig. [88]2A,B. Firstly, the body
   weight, food intake and water intake of mice were not changed by
   graphene or carbon fiber-FIR hyperthermia (Fig. [89]S2A–C, Supporting
   Information). We examined the difference in exercise capacity of naive
   mice between control and FIR treatment groups. We tested the treadmill
   performance of mice under 2 different protocols: low and high
   intensities that generally correspond to 50% and 100% of their maximal
   aerobic capacity, respectively (Fig. [90]2C). In low-intensity exercise
   protocol, the graphene group showed greater running endurance than
   control mice, while the running distance of carbon fiber group was
   slightly increased but with no significant difference (Fig. [91]2D). As
   shown in high-intensity protocol (Fig. [92]2E), the graphene group also
   performed substantially better than the control group, and the running
   distance of graphene group was obviously higher than the carbon fiber
   group (p < 0.05). Neither control nor FIR-exposed animals showed any
   significant in spontaneous locomotor activity test after 30 min or 24 h
   of last FIR-hyperthermia, indicating that FIR-hyperthermia will not
   induce locomotor patterns parallel those of heat stress or psychomotor
   stimulant (Fig. [93]2F).

Figure 2.

   [94]Figure 2
   [95]Open in a new tab

   Graphene-FIR hyperthermia increased exercise endurance in mice. (A)
   Diagram of the generation of FIR rays by graphene-based and carbon
   fiber-based devices. (B) Experimental procedure. Mice received
   40 min/day FIR radiation for 14 days. The endurance test was conducted
   on day 15. (C) Schematic depiction of the different treadmill exercise
   protocols. (D–E) Running distance on the treadmill for low and high
   exercise protocols. Distance until exhaustion on the treadmill after
   FIR irradiation. (F) The distance of locomotor activity 1 h or 24 h
   after the last FIR radiation. Data are expressed as mean ± SEM; n = 8
   in each group, *p < 0.05 compared to Control group, ^#p < 0.05 compared
   to Carbon fiber group, one-way ANOVA followed by Bonferroni test.

Graphene-FIR hyperthermia improved body temperature and blood flow in mice

   In present study, as showed similar spectral peaks in graphene-FIR and
   skin surface, whether FIR energy of graphene can improve body
   temperature more effectively. To test this hypothesis, core temperature
   of mice was recorded by using body capsules (Fig. [96]3A). The
   comparison of the data obtained from the two FIR chambers was firstly
   performed during FIR hyperthermia, there was no significant difference
   between graphene group and carbon fiber group (Fig. [97]3B). We then
   analyzed the core temperature of mice in different groups. During the
   FIR hyperthermia, core temperature of the graphene group exhibited
   significant rises compared to the carbon fiber group (Fig. [98]3C). In
   addition, temperature profiles of the graphene group exhibited similar
   significant rises near 38 °C at several time points immediately after
   the FIR hyperthermia. Thermographic images showed that the surface
   temperature in graphene group was significantly higher than carbon
   fiber group after FIR hyperthermia (Fig. [99]3D). We also observed the
   high level of the epidermal temperature will resist nearly 50 min,
   which consisted with the core temperature (Fig. [100]3E). Notably, our
   findings indicated that serum NE levels remained unchanged in mice
   following FIR hyperthermia (Fig. [101]S3, Supporting Information),
   thereby ruling out the possibility of sympathetic activation due to
   heat stress and increased thermogenesis.

Figure 3.

   [102]Figure 3
   [103]Open in a new tab

   Graphene-FIR hyperthermia improved body temperature and blood flow in
   mice. (A) Body capsules and temperature monitor. (B) The ambient
   temperature in FIR chamber before and after radiation. (C) The core
   temperature of mice before and after radiation. (D) The representative
   infrared thermal images after FIR hyperthermia. (E) The body surface
   temperature of mice before and after radiation. (F) The representative
   blood flow images. (G) The quantitation of blood perfusion in skeletal
   muscle and abdomen. Data are expressed as mean ± SEM; n = 8 in each
   group, *p < 0.05 compared to Carbon fiber group, Tukey's multiple
   comparisons test, **p < 0.01 and ***p < 0.01 compared to Control group,
   ^#p < 0.05 and ^##p < 0.01 compared to Carbon fiber group, one-way
   ANOVA followed by Bonferroni test.

   Using the RFLSI system, we further detected the blood flow in skeletal
   muscle and abdomen (Fig. [104]3F). Following continuous FIR
   hyperthermia, the blood flow was measured without acute hyperthermia.
   As shown in Fig. [105]3G, the HR-LDI signal increased significantly
   throughout the abdomen region in both graphene and carbon fiber groups.
   And the blood flow of graphene-FIR mice was higher than carbon fiber
   mice. We also found that the femoral muscle blood flow in graphene-FIR
   group was significantly higher than that of carbon fiber group. Taken
   together, these data suggested that graphene-FIR may promoted
   physiological changes in a metabolic regulation fashion.

Graphene-FIR hyperthermia altered the structure of the gut microbiome

   Recent evidence suggested an interaction between the gut microbiota and
   exercise endurance. To investigate whether graphene-FIR hyperthermia
   can alter the microbiota composition, we performed 16S ribosomal DNA
   analysis of microbiota in colon after the endurance test. Diversity
   analyses reveal that there are variations between the graphene and
   control groups in the α- and β-diversities of the gut flora
   (Fig. [106]4A,B). The Sobs and Chao index of graphene group were
   significantly higher than these of control and carbon fiber group. The
   PCoA plot (Fig. [107]4B) shows that at baseline, the graphene group and
   control group were separated along the PC2 axis, however, the carbon
   fiber group and control group were converged to the same position. Venn
   diagram based on genera (Fig. [108]4C). The three groups have 496
   shared genera, with 45 unique genera in the graphene group, 17 unique
   genera in the carbon fiber group and 9 unique genera in the control
   group. These findings showed that the community diversity of gut
   microbiota was increased by the graphene treatment.

Figure 4.

   [109]Figure 4
   [110]Open in a new tab

   Graphene-FIR hyperthermia altered the structure of the gut microbiome.
   (A) α-diversity (Sobs and Chao) comparison of the gut microbiota. (B)
   Unweighted UniFrac distances plotted using principal component analysis
   (PCoA) and the relative abundance of OTUs. (C) Venn diagram analysis of
   OTU compositions in different groups. (D) Relative gut microbiota
   abundance at the phylum level in the different groups. (E) Relative
   abundance rates of Bacteroidetes/Firmicutes, Blautia, Allobaculum and
   Anaerostipes in control, graphene and carbon fiber mice. (F)
   Correlation analysis: regression plot between the abundance of Blautia
   or Anaerostipes and running distance. Data are expressed as mean ± SEM;
   n = 6 in each group, *p < 0.05 compared to Control group, one-way ANOVA
   followed by Bonferroni test.

   In Fig. [111]4D, Bacteroidetes and Firmicutes dominated the gut
   microbiota composition at the phylum level, contributing to 62.2 and
   28% of the gut microbiota in the control group, 35.6 and 58.3% in the
   graphene group, 69.7 and 27.5% in the carbon fiber group, respectively.
   The Bacteroidetes/Firmicutes (B/F) ratio was significantly reduced by
   graphene-FIR (Fig. [112]4E). We further studied the group differences
   at the genus level. Compared with the control group, SCFAs-producing
   bacteria including Anaerostipes (belonging to the butyrate producing
   bacteria) Allobaculum and Blautia, were induced by graphene-FIR
   hyperthermia in mice (Fig. [113]4E). They are genus of anaerobic
   bacteria with probiotic characteristics. We further analyzed the
   relationship between the gut microbiota and motor functions using
   pearson’s correlation (Fig. [114]4F). We found that the running
   distance of mice was significantly positively correlated with
   Anaerostipes (R = 0.4705, P = 0.0203), Blautia (R = 0.4481,
   P = 0.0365), respectively.

   Fecal metabolic profiles of FIR treated group and control group were
   acquired by LC–MS. As shown in PLS-DA plots (Fig. [115]5A), a clear
   separation between the graphene-FIR group and other groups was found,
   suggesting that graphene-FIR hyperthermia led to significant
   biochemical changes. In volcano plot of the metabolites, x-axis denoted
   the fold changes of the metabolites in different groups and y-axis
   denoted the significance of the changes for the metabolites
   (Fig. [116]5B). The variation tendencies of metabolites were described
   by a heat map in Fig. [117]S4A. 18 metabolites were significantly
   decreased and 32 were significantly increased in graphene-FIR mice
   compared with controls. Interestingly, the abundance of adenosine
   monophosphate (AMP) in graphene group was significantly higher than
   control and carbon fiber group (Fig. [118]5C). In addition, the running
   endurance of mice has been positively correlated with the abundance of
   AMP in gut microbiota (Fig. [119]5D). The related metabolites were
   enriched in different pathways, including PI3K-Akt signaling pathway,
   mTOR signaling pathway, FoXO signaling pathway, glycerophospholipid
   metabolism etc. On the basis of their p value, KEGG annotation was
   performed (Fig. [120]S4B). Therefore, untargeted metabolomics support
   the global changes in the composition of the microbiome.

Figure 5.

   [121]Figure 5
   [122]Open in a new tab

   Graphene-FIR hyperthermia changes the metabolomic profile of the gut
   microbiome. (A) Partial least squares-discriminant analysis (PLS-DA).
   (B) A volcano plot of the metabolites. (C) Relative abundance rates of
   AMP in control, graphene and carbon fiber group. (D) Correlation
   analysis between the abundance of AMP and running distance. n = 6 in
   each group, *p < 0.05, **p < 0.01 and ***p < 0.001 compared to Con
   group or carbon fiber group, one-way ANOVA followed by Bonferroni test.

Graphene-FIR hyperthermia increased phosphorylation of AMPK via GPR 43

   To elucidate the potential roles of microbial metabolism in skeletal
   muscle function and exercise capacity, we firstly determined the
   protein expression of GPR43 and p-AMPK Thr172. Compared with control
   group, GPR43 protein expression increased significantly both in resting
   and exercise state mice after graphene-FIR treatment, whereas it was
   not obviously in carbon fiber group (Fig. [123]6A,C). Previous studies
   have demonstrated the ability of SCFAs to induce the phosphorylation of
   AMPK, which is known to be the main regulator of exercise-mediated
   responses in the skeletal muscle^[124]30. Therefore, we observed that
   the expression of p-AMPK (Thr 172) in the graphene group significantly
   increased compared with the control group, while there was no
   difference between carbon fiber group and control group
   (Fig. [125]6B,D). The AMPK activity of graphene group was significantly
   higher than the carbon fiber group (Fig. [126]6E). During exercise the
   most important glucose transporter is GLUT4, which is rate limiting for
   glucose uptake into muscle. We found that the expression level of GLUT4
   in graphene group was higher than that in controls (Fig. [127]6B,D). We
   also measured AMP and ATP levels of muscle in exercise state. The AMP
   level and AMP/ATP ratio increased in graphene group compared with the
   respective controls (Fig. [128]6F). Moreover, the level of AMP and
   AMP/ATP ratio were decreased in carbon fiber groups compared with the
   graphene groups. These results indicate that AMPK activated by
   graphene-FIR might be a key mechanism through which SCFAs exert changes
   in skeletal muscle metabolism.

Figure 6.

   [129]Figure 6
   [130]Open in a new tab

   Graphene-FIR hyperthermia increased phosphorylation of AMPK by
   activated GPR43 in muscle. Exercised and sedentary mice were
   sacrificed; the gastrocnemius was dissected for Western blot. (A)
   Representative Western blot and quantitation of GPR 43 level in
   sedentary mice. (B) Representative Western blot and quantitation of
   p-AMPK-Thr172 and GLUT4 level in sedentary mice. (C) Representative
   Western blot and quantitation of GPR 43 level in exercise mice. (D)
   Representative Western blot and quantitation of p-AMPK-Thr172 and GLUT4
   level in exercise mice. (E) AMPK activity. (F) The levels of ATP, AMP
   and AMP/ATP. Data are expressed as mean ± SEM, n = 3, *p < 0.05,
   **p < 0.01 and ***p < 0.001 compared to Control group, ^#p < 0.05 and
   ^##p < 0.01 compared to Carbon fiber group, one-way ANOVA followed by
   Bonferroni test.

Graphene-FIR hyperthermia mimics exercise by enhanced muscle glucose uptake
and blood flow via AMPK

   To further verify the effects of graphene-FIR on glucose metabolism,
   fluorescence imaging was applied to assess the uptake of glucose
   located in abdomen and skeletal muscle. In the resting state, the
   glucose uptake ratio of skeletal muscle in graphene group was
   significantly increased compared to the control group (Fig. [131]7A,C),
   and graphene group showed an increasing tendency of glucose uptake in
   abdomen region (Fig. [132]7B). However, in the exercise state,
   graphene-FIR could significantly increase the accumulation of glucose
   in both muscle and abdomen compared with the control group
   (Fig. [133]7D–E). Notably, we found that graphene-FIR induced increased
   of glucose uptake was blocked by ComC (10 mg/kg, i.p.), but not by
   propranolol (nonselective β-adrenergic antagonist, 10 mg/kg, i.p.)
   treatment (Fig. [134]7F). Interestingly, we also found that
   graphene-FIR induced increased of blood flow was inhibited by ComC (a
   specific inhibitor of AMPK, 10 mg/kg, i.p.) systematically
   administration (F[135]ig. [136]7G–I).

Figure 7.

   [137]Figure 7
   [138]Open in a new tab

   Graphene-FIR hyperthermia mimics exercise by enhanced muscle glucose
   uptake and blood flow via AMPK. (A) Representative fluorescence images
   showing the glucose uptake rate of mice in abdomen and skeletal. (B)
   Quantitation of relative fluorescence intensity of resting mice in
   abdomen. (C) Quantitation of relative fluorescence intensity of resting
   mice in muscle. (D) Quantitation of relative fluorescence intensity of
   endurance exercise mice in abdomen. (E) Quantitation of relative
   fluorescence intensity of endurance exercise mice in muscle. (F)
   Quantitation of relative fluorescence intensity of endurance exercise
   mice in muscle. (G) The representative blood flow images. (H, I) The
   quantitation of blood perfusion of endurance exercise mice in skeletal
   abdomen and muscle. Data are expressed as mean ± SEM, n = 5–8,
   *p < 0.05, **p < 0.05 and ***p < 0.001 compared to Control group,
   ^#p < 0.05 compared to Graphene group, one-way ANOVA followed by
   Bonferroni test.

Graphene-FIR hyperthermia changed the metabolomic profile of the skeletal
muscle

   To explore the potential regulatory roles of graphene-FIR on skeletal
   muscle metabolism, we consider the metabolomic analyses for providing a
   quantitative profile of metabolites in biological systems. Skeletal
   muscle samples from exercise mice were analyzed by non-targeted
   metabolomics. Principal component analysis (PCA) revealed a distinct
   separation of the graphene group and control group (PC1 22.7%), as
   shown in Fig. [139]8A. The Fig. [140]8B also exhibited obvious
   separation between the control group and graphene group, indicating
   that the metabolic profiles of the two groups were completely
   different. The upregulated metabolites were shown in red in
   Fig. [141]8C, whereas the downregulated metabolites were shown in
   green. Notably, the graphene group's metabolites were significantly
   altered, with 23 increased (e.g. creatinine and phosphate) and 4
   decreased (e.g. lactic acid).

Figure 8.

   [142]Figure 8
   [143]Open in a new tab

   Graphene-FIR hyperthermia changed the metabolomic profile of the
   skeletal muscle. (A) PCA plots of OTUs for different groups. (B) PCA
   plots between the graphene and control groups. (C) A heatmap of
   metabolites in the gastrocnemius of control and graphene groups, red
   and green represent higher and reduced concentrations of metabolites.

   We next investigated the impact of the metabolic pathways involved. As
   shown in Fig. [144]S5A, multiple pathways including galactose
   metabolism, glycolysis, gluconeogenesis, fatty acid biosynthesis,
   butanoate metabolism, and pyruvate metabolism, were shown to be
   involved in the pathway analysis of the 25 ANOVA-significant
   metabolites. Pathway enrichment analysis identified a nearly fivefold
   enrichment in galactose metabolism, fatty acid biosynthesis, pyruvate
   metabolism, glycolysis/gluconeogenesis, butanoate metabolism,
   propanoate metabolism, and alanine metabolism pathways with p
   values ~ 1 × 10^−3 (Fig. [145]S5B).

Graphene-FIR hyperthermia rescued the running endurance of ABX mice

   To further demonstrate the role of the microbiome in endurance-exercise
   promoted by graphene-FIR hyperthermia, we administered an antibiotic
   cocktail (ABX) in the drinking water for 4 weeks (Fig. [146]9A). The
   body weight and food intake were not changed in different groups (Fig.
   [147]S6A,B, Supporting Information). Microbiota depletion significantly
   changed the morphology of the small intestine. The small intestine of
   ABX mice was significantly longer than control mice (Fig. [148]S6C,
   Supporting Information). The morphology of the fecal pellets was also
   significantly changed (Fig. [149]S6D, Supporting Information).
   Graphene-FIR significantly decreased the length of small intestine, and
   reduced the weight of fecal pellets compared to the ABX mice.
   Histological analysis showed that the intestinal mucosa of the mice in
   the ABX group was atrophied, and the crypts of the intestinal mucosa
   were shallower, muscular layer were thinner. The damaged morphology of
   intestinal were partly improved by graphene-FIR hyperthermia
   (Fig. [150]9C).

Figure 9.

   [151]Figure 9
   [152]Open in a new tab

   Graphene-FIR hyperthermia rescued the running endurance of ABX mice.
   (A) Experimental procedure. (B) Running distance until exhaustion on
   the treadmill for control, ABX, ABX + Gra and ABX + Car groups (n = 8).
   (C) H&E staining images of colon tissues were presented. The morphology
   changes of the intestinal mucosa, crypts of the intestinal mucosa and
   muscular layer were indicated by arrows. (n = 3–5, scale
   bar = 200 μm, × 40). (D) Representative Western blot and quantitation
   of p-AMPK-Thr172 and GLUT4 level in gastrocnemius (n = 3). Data are
   expressed as mean ± SEM, *p < 0.05, **p < 0.01 compared to Con group or
   ABX + Car group, ^#p < 0.05, ^##p < 0.01 and ^###p < 0.001 compared to
   ABX group, one-way ANOVA followed by Bonferroni test.

   Moreover, treadmill running time was lower in the ABX group than the
   control group, graphene-FIR hyperthermia significantly increased
   treadmill running time, while the carbon-fiber group only have an
   increasing trend (Fig. [153]9B). As shown in Fig. [154]9D, the protein
   level of p-AMPK was significantly lower in ABX mice in skeletal muscle,
   which is consistent with their impaired ability to maintain endurance.
   Graphene-FIR was significantly up-regulated p-AMPK and GLUT4 expression
   compared to the ABX group, and significant difference was also observed
   between carbon fiber and graphene-FIR mice. These data indicated that
   the effect of graphene-FIR hyperthermia on exercise endurance may
   attribute to rescue the microbiota depletion and AMPK activity.

Discussion

   With the advancing innovation of technology, specialty pure FIR
   emitting, have become safe, effective, and widely used in clinical
   treatment. However, FIR application in physical activity and metabolism
   requires understanding the potential mechanisms of biological effects
   due to FIR radiation at specific range. In our study, we introduced a
   graphene-based flexible device tailored for systemic FIR hyperthermia
   therapy. The graphene-based device showed higher electrothermal
   stability and resonance effect in hyperthermia therapy compared to the
   carbon fiber-based device. We hypothesized that graphene-FIR
   hyperthermia could promote exercise capacity as exercise mimetic. Our
   research provided a mechanistic insight into the biological effects of
   graphene-FIR radiation within a specific wavelength range. The
   resonance effects generated by graphene-FIR improved the body
   temperature and glucose uptake which associated with the metabolic
   benefits. We suggested that the application of graphene-FIR stimulates
   the metabolism homeostasis-related AMPK/GLUT4 signaling pathway through
   gut-muscle interaction. The findings of the present study provided
   evidence that improving physical activity and metabolisms with
   graphene-FIR hyperthermia as exercise mimic, and the possible future
   applications in medical are wide ranging.

   In recent years, graphene and its unique properties have been widely
   used in biomedical field. In the present study, owing to its high
   carrier mobility, conductivity, flexibility and optical transparency, a
   manufactured low-voltage, flexible graphene film was utilized as a FIR
   emitter. We demonstrated that exposure to graphene-FIR promoting
   running endurance in animal experiments. Graphene-FIR hyperthermia
   significantly improved the running distance of mice both in the low and
   high exercise intensity. We excluded for the possible
   psycho-stimulatory effect after long-term graphene-FIR hyperthermia, in
   according to the locomotor activity test.

   However, the question that arises is: how the graphene-based
   photo-biomodulations work in the enhancement of running endurance? To
   investigate the spectral properties and metabolic regulatory effects of
   graphene-based FIR, a carbon fiber-based device was employed as a
   control. We found that the emission peak wavelength of the
   graphene-based device aligned more closely with the human body's
   absorption peak compared to the carbon-fiber device. Graphene-FIR
   hyperthermia was effective in raising both core and epidermal
   temperatures in mice. Remarkably, these thermal effects persisted for
   almost 1 h after the graphene-FIR heater was removed. Resulting
   epidermal temperature was higher when the skin was irradiated with
   graphene-FIR than if similar thermal loads from shorter wavelengths
   were used. It indicated that graphene-FIR hyperthermia promoted
   metabolism processing. Graphene-FIR hyperthermia increased blood flow
   in femoral muscle and abdomen region. The microvascular blood flow
   plays an important role in muscle metabolism and exercise capacity.
   Blood flow to the working muscles is crucial for muscle function during
   exercise^[155]31. Skeletal muscle microvascular blood flow is
   stimulated by muscle contraction to aid in the transport of nutrients
   and hormones^[156]32. The efficient transport of radiant energy to the
   tissues may have been made possible by the graphene-FIR's ability to
   more effectively resonance with cellular frequencies. Notably, exercise
   enhances endurance capacity in humans, improves heart function, and
   stimulates peripheral blood flow. Additionally, AMPK activation in
   muscles increases muscle vasodilation and microvascular
   perfusion^[157]33,[158]34. Our results demonstrated that graphene-FIR
   can improve blood flow in femoral muscle and abdomen, and these effects
   could be inhibited by AMPK inhibitor. These provided an important
   foundation for the biological effects by graphene-based FIR
   hyperthermia except for its thermo-effect. However, whether
   graphene-FIR hyperthermia can play a similar role as exercise-mimetics
   to improve exercise capacity and metabolisms by AMPK activation, and
   the fundamental mechanism is still unknown.

   The composition and diversity of gut microbiota have impact on
   intestinal function and health status^[159]35. Previous studies have
   demonstrated that endurance training can influence the composition and
   diversity of the gut microbiota in mice^[160]36. Human microbiome
   studies have also shown that the gut microbiota of athletes is more
   diverse as compared with that of controls^[161]37. In the present
   study, we found that graphene-FIR hyperthermia significantly increased
   the α- and β-diversities index of gut microbiota compared with the
   control group. The normal gut microbiota comprises of two major phyla,
   Bacteroidetes and Firmicutes, comprise 90% of the average gut
   microbiota^[162]38. A recent study showed that compared to
   non-athletes, athletes exhibited higher levels of firmicutes and lower
   levels of bacteroidetes^[163]39. The microbiota of exercise mice
   presented a greater abundance of firmicutes and lower bacteroides, with
   related changes highly correlated with VO[2] max^[164]37,[165]40. We
   found that the Bacteroidetes/Firmicutes ratio was significantly
   decreased in graphene-FIR group compared with the controls.
   Furthermore, our findings demonstrated a significant increase in
   several potentially beneficial bacterial genera in mice exposed to
   graphene-FIR radiation, specifically Blautia, Allobaculum, and
   Anaerostipes^[166]41–[167]43. Allobaculum and Blautia are known
   producers of short-chain fatty acids (SCFAs), which are conducive to a
   healthier gut environment^[168]41. It has been reported that
   Allobaculum were enriched in the exercise animals^[169]41,[170]42.
   Dietary inositol could converse into propionate and acetate by
   Anaerostipes, which have a potential beneficial role for promoting host
   health^[171]43. Our results also indicated that the running endurance
   of mice was significantly positively correlated with the abundance of
   Anaerostipes and Blautia. Graphene-FIR hyperthermia led to significant
   changes of fecal metabolic phenotype. The abundance of AMP was
   increased in graphene group, which was strongly correlated with running
   endurance. These results indicated that graphene-FIR hyperthermia can
   positively alter the amount and structure of the gut and enriched the
   SCFAs-producing bacteria, which was associated with improving the
   exercise endurance.

   Most current studies generally believe that moderate exercise training
   is beneficial to gut microbiota^[172]37. In contrast to light exercise,
   strenuous exercise cause changes in immune response by reducing the
   gastrointestinal blood flow while increasing permeability of the
   gastrointestinal epithelial wall and the destruction of gut mucous
   thickness^[173]44. In our study, we found that graphene-FIR increased
   the blood flow and the abundance of intestinal probiotics, exerted an
   intestinal protective effect. These protective effects of gut
   microbiota may help delay the fatigue symptoms in endurance
   performances. Therefore, the graphene-FIR group showed greater exercise
   endurance in acute strenuous exercise tests.

   Numerous studies have demonstrated that gut microbiota impact the
   function of various organs through bacterial metabolites, such as SCFAs
   and bile acids^[174]45. Growing evidence suggests that there is
   bidirectional contact between the gut microbiota and muscle, and some
   researchers have proposed the existence of a "gut-muscle
   axis"^[175]23,[176]46,[177]47. It has been documented that SCFAs may
   stimulate AMPK to phosphorylate in skeletal muscle^[178]28. It is
   significant to note that AMPK activation also plays a role in energy
   homeostasis through the processes of fatty acid oxidation, glucose
   absorption, glycogenesis, and glycolysis^[179]48. Studies conducted in
   vitro have shown that butyrate treatment improves fatty acid absorption
   in muscle cells^[180]49. The glucose transporter protein, GLUT4, is
   essential for the uptake and metabolism of glucose in skeletal
   muscle^[181]50. It is suggested that the influence of SCFAs on GLUT4
   expression may promote glycogen restoration and glucose uptake^[182]51.
   Notably, the administration of acetate has been shown to accelerate the
   rate of skeletal muscle glycogen replenishment in rats following
   exercise^[183]52. In present study, graphene-FIR stimulated the
   phosphorylation of AMPK and GLUT4 expression, and increased the AMP/ATP
   ratio in gastrocnemius muscle, probably because of the ability of GPR43
   activation to increase AMP concentrations within skeletal muscle
   tissue. We also found that graphene-FIR induced increased of glucose
   uptake was blocked by the AMPK inhibitor, but not by the β-adrenergic
   antagonist. It was suggested that graphene-FIR promotes glucose uptake
   by activating AMPK.

   AMPK is activated by numbers of activators, including the AMP mimicking
   AICAR. Previous study demonstrated that AICAR can induce metabolic
   changes and improve running endurance without exercise
   training^[184]27. In our study, we speculated that graphene-FIR could
   restore a more favourable energy balance in gut-muscle axis and
   improved the running endurance through AMPK activation. Metabolomic
   analyses metabolites were significantly altered after graphene-FIR
   hyperthermia. We found the abundance of AMP in gut microbiota was
   increased in the graphene group and it has positively correlated with
   running endurance. In muscle metabolomics analysis, graphene-FIR
   hyperthermia decreased the lactic acid and increased the creatinine and
   phosphate. Metabolic pathways analyses revealed that galactose
   metabolism, glycolysis, gluconeogenesis, fatty acid biosynthesis,
   butanoate metabolism and pyruvate metabolism were involved in
   graphene-FIR hyperthermia. Therefore, AMPK phosphorylation may be a key
   mechanism by which graphene-FIR exert changes in gut-muscle axis
   metabolism.

   To further verified the impact of graphene-FIR on exercise capacity and
   glucose metabolism through the modulation of microbiota homeostasis and
   AMPK activation, we utilized ABX mice as an animal model for gut
   microbial dysbiosis. It has been reported that gut microbiota depletion
   via antibiotics impaired endurance exercise performance and skeletal
   muscle contractile function in mice. Natural reseeding of the mice
   improved skeletal muscle function and restored the gut
   microbiota^[185]20,[186]53,[187]54. Our results indicated that the
   decreased of running endurance in ABX mice was reversed by graphene-FIR
   hyperthermia. We also observed that the expression level of p-AMPK and
   GLUT4 was up-regulated in graphene-FIR group compared to ABX group.
   Therefore, the effect of the graphene-FIR on skeletal muscle metabolism
   and especially on the endurance capacity may attribute to regulation of
   energy homeostasis in the gut microbiota.

Conclusions

   Collectively, our findings demonstrated that the graphene-based
   flexible device was different from carbon fiber materials, with
   single-layer structure and unique electrothermal stability and spectral
   properties. Graphene-FIR hyperthermia could promote exercise capacity
   and glucose metabolism via gut-muscle interaction. The potential
   mechanism may involve the modulation of microbiota homeostasis and
   activating AMPK. We present evidence supporting a novel application
   that graphene-based FIR hyperthermia can be used to as exercise mimetic
   in the future.

Supplementary Information

   [188]Supplementary Information.^ (4.7MB, docx)

Acknowledgements