Muscle PGC‐1α Overexpression Drives Metabolite Secretion Boosting Subcutaneous Adipocyte Browning

ABSTRACT

Muscle and adipose tissue (AT) are in mutual interaction through the integration of endocrine and biochemical signals, thus regulating whole‐body function and physiology. Besides a traditional view of endocrine relationships that imply the release of cytokines and growth factors, it is becoming increasingly clear that a metabolic network involving metabolites as signal molecules also exists between the two tissues. By elevating the number and functionality of mitochondria, a key role in muscle metabolism is played by the master regulator of mitochondrial biogenesis peroxisome‐proliferator‐activated receptor‐γ coactivator‐1α (PGC‐1α), that induces a fiber type shift from glycolytic to oxidative myofibers. As a consequence, the upregulation of muscle respiratory rate might affect metabolite production and consumption. However, the underlying mechanisms have not yet been fully elucidated. Here, we used a muscle‐specific PGC‐1α overexpressing mouse model (MCK‐PGC‐1α) to analyze the metabolite secretion profile of serum and culture medium recovered from MCK‐PGC‐1α muscle fibers by NMR. We revealed modified levels of different metabolites that might be ascribed to the metabolic activation of the skeletal muscle fibers. Notably, the dysregulated levels of these metabolites affected adipocyte differentiation, as well as the browning process in vitro and in vivo. Interestingly such effect was exacerbated in the subcutaneous WAT, while only barely present in the visceral WAT. Our data confirm a prominent role of PGC‐1α as a trigger of mitochondrial function in skeletal muscle and propose a novel function of this master regulator gene in modulating the metabolite production in turn affecting the activation of WAT and its conversion toward the browning.

1. Introduction

Skeletal muscle (SkM) and adipose tissue are two distinct tissues in the body, each with unique functions and characteristics. SkM is the organ of voluntary movement and posture maintenance with a key role in metabolic processes and in regulating energy expenditure (Mukund and Subramaniam 2020). Adipose tissue includes two divisions: white adipose tissue (WAT), primarily involved in energy storage and insulation, and brown adipose tissue (BAT) which is specialized in heat generation through a process called non‐shivering thermogenesis (Park 2014).

Although anatomically separated, adipose tissue and SkM are strictly and functionally interconnected in the regulation of metabolic health. SkM is the major site of glucose uptake and utilization and thus plays a crucial role in insulin sensitivity (Merz and Thurmond 2020). Adipose tissue also influences systemic metabolism and can contribute to metabolic disorders when its function is impaired, as seen in obesity and type 2 diabetes (Harvey, Boudreau, and Stephens 2020). A clear proof of the muscle‐adipose tissue cross talk is the effect of physical exercise on both tissues. Physical activity can increase SkM mass and improve muscle function, while also influencing the distribution and function of adipose tissue (Pedersen and Febbraio 2008; Boström et al. 2012; Seldin and Wong 2012; Rao et al. 2014; Roberts, West, et al. 2014; Dong et al. 2016). Also, nutrition and lifestyle play significant roles in determining the overall health and function of both tissues (Atakan et al. 2021). Both SkM and adipose tissue exert an endocrine function. They secrete soluble factors that act in an endocrine manner to facilitate tissue‐to‐tissue communication and crosstalk, likely working together to improve overall metabolic health (Stanford and Goodyear 2018). The endocrine function of adipose tissue is largely known, through the secretion of hormones like leptin, which regulates energy balance and appetite, and adiponectin, which plays a role in glucose regulation and fatty acid oxidation (Luo and Liu 2016). Some studies have recently shown that also SkM can play an endocrine function: contracting SkMs release myokines and factors that function to alter the phenotype of WAT, including WAT “beiging” (Stanford and Goodyear 2018). Prolonged cold exposure or treatment with β‐adrenergic compounds convert certain WAT depots into a “brown‐like” state (Lončar 1991; Nedergaard and Cannon 2010). Browning of fat deposits is also induced by other molecules and hormones, for example, thyroid hormones and their derivatives (Huang et al. 2024; Senese et al. 2019). In addition, an increasing amount of data reinforce the concept of metabolic networking among tissues, in which metabolic intermediates drive the emergence of inducible brown adipocytes in WAT. Both white and brown adipose cells express metabolite receptors and transporters, which may act as sensors of their metabolic environment. Proton‐linked monocarboxylate transporters (MCTs) (Halestrap 2013), which drive lactate, pyruvate, and ketone body transport across plasma membranes, are present on the surface of rodent (Hajduch et al. 2000; Iwanaga, Kuchiiwa, and Saito 2009) and human (Pérez de Heredia, Wood, and Trayhurn 2010) adipocytes. Notably, the MCT1 isoform, mainly involved in lactate import into cells, strongly increases in BAT after exercise (De Matteis et al. 2013).

During physical exercise, when an increase in energy demand occurs, multiorgan sensing and adaptation to the enhanced metabolic rate are required (Cariati et al. 2021; Cicatiello et al. 2022; Sagliocchi 2024; Miro et al. 2023; Nappi 2024). The induction of the mitochondrial biogenesis master regulator Peroxisome‐proliferator‐activated receptor‐γ coactivator‐1α (PGC‐1α) is the first molecular hint following physical exercise (Kelly and Scarpulla 2004; Puigserver and Spiegelman 2003; Wu et al. 1999). Indeed, the overexpression of PGC‐1α modulates several metabolic pathways to meet the enhanced energy needs of different metabolically relevant tissues (Baar et al. 2002; Koves et al. 2005; Norrbom et al. 2004). In addition, PGC‐1α activation is essential for brown fat thermogenesis (Puigserver et al. 1998), hepatic gluconeogenesis (Koo et al. 2004), cardiac homeostasis (Arany et al. 2005), and brain axonal integrity (Lin et al. 2004). In SkM, sustained PGC‐1α‐dependent mitochondrial biogenesis and increased respiratory activity are the preeminent processes among the biological adaptations often triggered by physical exercise (Menshikova et al. 2006). Moreover, SkM PGC‐1α is required for exercise‐induced angiogenesis (Arany et al. 2008; Chinsomboon et al. 2009; Rowe, Safdar, and Arany 2014). Several in vivo studies used transgenic mice with PGC‐1α expression driven by a muscle creatine kinase (MCK) promoter (hereafter, MCK‐PGC1‐α) to elucidate its role specifically in muscle physiology. PGC‐1α has been demonstrated to be the principal factor in regulating fiber type determination (Lin et al. 2002), in improving muscle endurance, mitochondrial remodeling, and in maintaining balance and motor coordination (Gill et al. 2018).

Moreover, a consistent body of literature highlights that PGC‐1α influences the expression of biologically relevant myokines in humans and rodents. For example, Irisin was initially identified in transgenic mice with increased SkM PGC‐1α expression (Boström et al. 2012). These mice had beige adipocytes within their subcutaneous WAT (scWAT), an effect that was further amplified when mice were exercise‐trained (Boström et al. 2012). Also, the myokine Metrnl was identified from a screen analyzing factors secreted from primary myotubes overexpressing PGC‐1α (Rao et al. 2014), a splice form of PGC‐1α that regulates SkM hypertrophy and energy expenditure (Ruas et al. 2012). Moreover, BAIBA (Roberts, Boström, et al. 2014) is another myokine that has been recently identified to have a role in the beiging of scWAT after exercise and that was identified by liquid chromatography–mass spectrometry (LC‐MS) metabolic profiling in PGC‐1α overexpressing human myocytes.

Despite the clear evidence of an effect of muscle PGC‐1α on adipose phenotype, the influence of the “metabolic environment” on the browning process is to date poorly understood. Investigating the distinct and interrelated roles of the SkM and adipose tissue is key to comprehending broader physiological and metabolic processes in the body.

To resolve these unanswered questions, in the present study, we used an ex vivo cellular model of primary SkM fibers and adipocytes obtained from the MCK‐PGC‐1α mouse and investigated the effects of metabolites secreted from muscle on the browning process.

2. Materials and Methods

2.1. Mouse Models

The MCK‐PGC‐1α mouse model expressing PGC‐1α in SkM used in this study was previously described (Lin et al. 2002). The mouse model was kindly provided by Prof. Marco Sandri, University of Padova, Italy. Mice were in a C57BL/6 background, maintained in fresh bedding cages under controlled conditions, namely 20°C–24°C, 50%–60% humidity, and 14:10 light‐dark cycle and fed regular chow ad libitum. At necropsy, serum sampling has been performed using standard procedures, and tissues were collected for further analysis. In this study, we used male mice in all the described experiments. All mouse experiments were performed according to national and European community guidelines, and procedures received the approval of the Institutional Animal Care and Use Committee (IACUC) (protocol n. 354/2019‐PR and n. 1/2024‐PR).

2.2. Isolation of Single SkM Fibers

Single myofibers (SMFs) were obtained from the extensor digitorum longus (EDL) muscles as previously described (Nappi et al. 2022). EDL were obtained from 8‐months old male MCK‐PGC‐1α and the corresponding littermate wild‐type (WT) mice. Briefly, EDL muscles were removed and placed into 0.1% type 1 collagenase (C0130, Sigma‐Aldrich Co. St. Louis, MO) solution in DMEM at 37°C for 1 h. Fiber bundles that had not been released during the incubation were separated using a wide‐bore glass pipette. The obtained SMFs were washed four times in fresh culture medium and plated onto 60‐mm dishes in the medium for satellite cells (50% DMEM, 50% MCDB, 20% fetal bovine serum [FBS], 1% Ultroser G, 2 mM glutamine, 50 i.u. penicillin, and 50 μg/mL streptomycin). Fibers were incubated for 48 h at 37°C, 5% CO₂. At the end of the experiments, the supernatant from muscle fibers was collected and analyzed by NMR metabolic profiling or used for subsequent analysis, while SMFs were lysed to quantify the genomic DNA, through standard techniques.

2.3. NMR Spectroscopy of Conditioned Media and Serum Samples

The samples for NMR spectroscopy were prepared adding 0.06 mL of ²H₂O to 0.54 mL of conditioned media or adding 0.030 mL of ²H₂O to 0.27 mL of murine sera. All the spectra were acquired with a Bruker Avance NMR spectrometer operating at 600 MHz ¹H Larmor frequency, equipped with a cryogenically cooled probe. 1D ¹H‐NMR spectra were recorded at 25°C with a “zgesgp” pulse‐program (a gradient‐based excitation sculpting using 180° water‐selective pulses), including the following parameters: 64 scans, spectral width 18 ppm, delay 5 s, and 64 k points. NMR data were processed by TOPSPIN 4.0.7 software packages (Bruker Biospin Gmbh, Rheinstetten, Germany). NMR results are expressed as area under curve (AUC). For conditioned media analysis, metabolites intensity is shown as AUC over fiber DNA (µg DNA) and normalized with the control medium alone.

2.4. In Vitro Isolation of Primary White Adipocytes From Stromal Vascular Fraction (SVF) and Browning Induction

Primary white adipocytes were obtained as described previously (Villanueva‐Carmona et al. 2023). Briefly, sWAT and visceral WAT (vWAT) were dissected from 8 weeks WT or MCK‐PGC‐1α mice, cleaned from adherent circumstantial tissues and then sequentially digested in collagenase solution. The collagenase concentration was adjusted based on the tissue weight. For 1 g of collected WAT adipose tissue, we used 5 mL of PBS containing 1% BSA and 0.2% collagenase and incubated for 30 min h at 37°C on an orbital shaker set at 200 rpm. This allows for the collagenase enzyme to digest the tissue.

Upon completion of the digestion process, the subsequent steps focus on segregating the SVF from fat depots and tissue remnants by centrifugating the digested samples at 300 g for 10 min at 22°C. The pellet was resuspended in fresh complete medium and then plated in 6‐well plate in basal medium, Dulbecco essential medium (DMEM) supplemented with 10% FBS (GIBCO Thermo Fisher Scientific, Waltham, MA, USA), 2 mM glutamine and 1% P/S (Sigma‐Aldrich) (basal medium). Cells were cultured and expanded by incubation at 37°C and 5% CO₂ and the medium was replaced every 2 days. Cells were then used for subsequent experiments up to passage 3.

For the induction of browning, WT vWAT and sWAT‐derived adipocytes were grown in basal medium in 24‐well plates for 48 h, until the reaching of 90% confluence, and then induced to differentiate into beige adipocytes as previously described (Cicatiello et al. 2024). Briefly, brown adipogenesis was induced in DMEM supplemented with 10% FBS, 1% l‐glutamine and 1% penicillin/streptomycin, 0.25 μM dexa, 0.5 mM 3‐isobutyl‐1‐methylxanthine (IBMX) and 10 μg/ml insulin. After 48 h, the dexa was removed from the medium and 1 μM rosiglitazone and 50 nM T3 were also added to the medium and the cells were harvested at 7 days post‐induction. Media, cell sera, antibiotics, IBMX, dexa and T3 were provided by Sigma‐Aldrich (MO, USA), rosiglitazone by Enso Life Sciences (NY, USA) and insulin by Life Technologies (MA, USA).

Mouse 3T3L1 fibroblasts were grown in DMEM supplemented with 10% FBS, 1% l‐glutamine and 1% penicillin/streptomycin solution at 37°C in a humidified atmosphere containing 95% air and 5% CO₂ and induced to differentiate as above described. When required, 1 mM lactate, 1 mM sodium pyruvate, 1mM l‐glutamine were added exogenously to differentiation medium and cells harvested at 7 days post‐induction.

2.5. Oil Red O Staining

vWAT and sWAT derived adipocytes and 3T3‐L1 were fixed with 4% paraformaldehyde for 15 min and then washed with PBS1X. Oil Red O (Sigma‐Aldrich) staining was performed for 30 min at room temperature. After removal of excess stain, cells were photographed under a microscope at × 40 magnification using an inverted microscope Leica DMi1 (Leica Microsystems, Wetzlar, Germany). To measure the accumulation of cellular lipid droplets, Oil Red O was dissolved in 100% isopropanol and the absorbance was measured at 490 nm using the VictorTM X4 (Perkin Elmer, CT, USA). At least 5 fields per sample were photographed and analyzed.

2.6. SMFs Conditioned Medium (CM) Treatments and Exogenous Lactate, Pyruvate and Glutamine Administration in Browning Induction

For the CM experiments, WT vWAT and sWAT‐derived adipocytes were cultured in a differentiation medium in the presence of a 20% volume of CM derived from SMFs WT or MCK‐PGC‐1α for 7 days. The induction of browning was evaluated by molecular, biochemical and histological analysis.

To evaluate the contribution of intermediate metabolites arise from NMR analysis of SMFs supernatants we exogenously added the metabolites at a concentration that mimics those measured by NMR. In detail, 1 mM lactate, 1 mM sodium pyruvate, 1mM l‐glutamine were added at the beginning of browning induction for 7 days, then the cells were harvested at 7 days post‐induction and analyzed as above described.

2.7. Real‐Time PCR

Messenger RNAs were extracted with Trizol reagent (Life Technologies) from GC muscles, and vWAT and sWAT isolated from 8‐month‐old male MCK‐PGC‐1α and WT mice. Complementary DNAs were prepared with Vilo reverse transcriptase (Life Technologies) as indicated by the manufacturer. The cDNAs were amplified by PCR in an iQ5 Multicolor Real Time Detector System (BioRad) with the fluorescent double‐stranded DNA‐binding dye SYBR Green (BioRad). Specific primers for each gene were designed to work under the same cycling conditions (95°C for 10 min followed by 40 cycles at 95°C for 15 s and 60°C for 1 min), thereby generating products of about 200 bp for each amplification. Primer sequences are reported in Supporting Information S1: Table 1. All samples were run in triplicate. The template concentration was calculated as previously described (Sagliocchi et al. 2019).

2.8. Western Blots

Proteins were isolated from GC muscles and vWAT and sWAT isolated from MCK‐PGC‐1α and WT mice. The lysis was carried out using RIPA buffer (VWR 97063‐270) supplemented with protease inhibitor (11836153001 ROCHE) and phosphatase inhibitors. Thirty micrograms of protein extracts were separated on a 10% SDS‐PAGE, transferred to a PVDF blotting membrane, and probed overnight at 4°C using primary antibodies against LDHA, LDHB, PKM1, PKM2, SMYD1 and PGC‐1α. Anti‐tubulin and anti‐actin‐specific antibodies were used as loading controls. Membranes were then washed, probed with secondary antibodies conjugated with horseradish peroxidase and developed using the Immobilon Western Kit (Millipore, Watford, UK). Images were captured using the Chemi‐Doc MP Imaging System equipped with Image Lab Software (Bio‐Rad). The primary and secondary antibodies and dilutions used are indicated in Supporting Information S1: Table 2. All Western blots were run in triplicate, and bands were quantified with Image J software (NIH Image, Bethesda, MD, USA).

2.9. Mitotracker Staining and Flow Cytometry Analysis

Primary adipocytes were stained for 30 min at 37°C with 200 μM of a red fluorescent dye, MitoTracker Red FM (#M7512, Invitrogen), which passively diffuses across the plasma membrane and accumulates in active mitochondria. After centrifugation, the pellet was resuspended in serum‐free DMEM because the probe is sensitive to potential oxidases in serum, and mitochondrial content was measured by FACS (BD FACSCantoTM flow cytometer, CA, USA). Results were analyzed using BD FACS Diva software (Duke University Flow Cytometry Shared Facility) and expressed as mean fluorescence intensity (%) of CM SMFs PGC‐1α cells compared to CM SMFs WT. In parallel, primary adipocytes seeded on slides were stained for immunofluorescence analysis.

2.10. Statistical Analyses

Statistical analysis was performed using t test when comparing two groups. One‐way ANOVA with Tukey's posttest was used for multiple comparisons (GraphPad Prism 9.0; GraphPad Software Inc., La Jolla, CA, USA). Statistical significance was considered when p < 0.05 (*p < 0.05, **p < 0 .01, ***p < 0.001). All data are presented as mean ± SD.

3. Results

3.1. PGC‐1α Overexpression in SkM Modifies the Profile of Secreted Metabolites

To detect a broad range of muscle‐secreted compounds mediating the networking between SkM and the adipose tissue, we first evaluated the metabolomic profile by NMR of the serum samples obtained from WT and MCK‐PGC‐1α mice (Figure 1A, upper). While no differences were found in the circulating levels of glucose, acetate, alanine, and lactate (Supporting Information S1: Table 3), NMR analysis showed an increase of pyruvate circulating levels in MCK‐PGC‐1α as compared to WT (Figure 1B). Considering the unmodified lactate level and the 1.7 ± 0.31‐fold increase of pyruvate concentrations in MCK‐PGC‐1α, an alteration of systemic lactate/pyruvate ratio (L/P) is likely present.

Figure 1.

Metabolic profiling of serum and supernatants isolated from SkM PGC‐1α versus WT muscle fibers. (A) Schematic overview of the experimental plan. In detail, we performed NMR analysis on serum samples collected from 8‐month MCK‐WT and MCK‐PGC‐1α mice and on CM derived from SMFs WT and SMFs PGC‐1α cultured for 48 h. (B) Lactate and pyruvate levels in the serum of MCK‐WT and MCK‐PGC‐1α mice were measured by NMR analysis and expressed as area under curve (AUC). (C) Metabolite secretion levels in the CM SMFs WT and CM SMFs PGC‐1α was measured by NMR and expressed as normalized intensity (AUC/μg DNA. (D) Succinate and pyruvate uptake levels in the CM SMFs WT and CM SMFs PGC‐1α was measured by NMR and expressed as normalized intensity (AUC/μg DNA. Metabolite signals were normalized using internal standards. Comparisons and differences were analyzed for statistical significance using the two‐way ANOVA test and Bonferroni posttest analysis. Data represent the mean ± standard deviation of the mean of three replicates. *p < 0.05, **p < 0.01, ***p < 0.001. CM, conditioned‐medium; MCK, muscle creatine kinase promoter; SMFs, single myofibers.

To verify whether the disbalanced circulating L/P ratio in MCK‐PGC‐1α mice could be ascribed specifically to muscle, we evaluated the metabolomic profile of the CM for 48 h from the EDL‐derived SMFs, collected from WT (SMFs WT) and MCK‐PGC‐1α (SMFs PGC1α) mice (Figure 1A, below). Compared to the SMFs WT, the NMR analysis of CM SMFs PGC‐1α revealed interesting changes in the concentration of some metabolic intermediates involved in glycolysis, amino acids and derivatives metabolism related to tricarboxylic acid cycle (TCA). In detail, five different metabolites were identified in the SMFs secretome by the NMR (Figure 1C). We observed a significant increase of lactate, acetate, glutamate, glutamine, and creatinine in the CM SMFs PGC‐1α as compared to CM SMFs WT, suggesting an enhanced secretion rate of these molecules (Figure 1C). Moreover, we observed a different uptake ability in SMFs PGC1α with higher levels of succinate uptake and lower levels of pyruvate uptake from the culture medium (Figure 1D) as compared to SMFs WT. These data confirm a rearrangement of the profile of secreted metabolites by PGC‐1α overexpressing SkM, which might result from a disbalance between metabolite secretion and uptake.

3.2. PGC‐1α Overexpressing Muscles Exhibit an Altered Expression of Enzymes Related to Glucose Metabolism

To get better insight into the mechanisms by which PGC‐1α overexpression in muscle impacts the metabolic status of myofibers, we measured the expression of genes encoding several enzymes involved in glucose metabolism. Interestingly, compared to controls, PGC‐1α overexpressing muscles showed a significant decrease of Pkm1 and Ldha mRNA levels, encoding for pyruvate kinase M1 and lactate dehydrogenases A, respectively, and an increase of Mpc1, encoding for the mitochondrial pyruvate carrier 1 (Figure 2A). Accordingly, the Ldha/Ldhb and Pkm1/Pkm2 ratios were reduced in the PGC‐1α overexpressing as compared to WT muscles (Figure 2B). We also observed a differential protein expression of LDH isoforms, characterized by a strong shift from LDHA, the lactate‐producing isoform, to LDHB, the lactate‐utilizing enzyme (Figure 2C,D).

Figure 2.

Metabolic reprogramming of MCK‐PGC‐1a muscle. (A) mRNA expression levels of key genes involved in glycolysis (Pkm1, Pkm2, Ldha, Ldhb, Pdha1, Pdk3, Idh1, and Mpc1) were measured by real‐time PCR in GC of 8‐month‐old male MCK‐PGC‐1α compared to WT mice. (B) The ratio of Ldha/Ldhb and Pkm1/Pkm2 mRNA expression levels in the same samples as shown in (A). (C) Western blot analysis of LDHA and LDHB proteins in the GC muscles of MCK‐PGC‐1α compared to WT mice. Tubulin was used as an internal loading control. (D) Quantification of LDHA/LDHB ratio versus Tubulin levels. (E) mRNA expression levels of key genes involved in the OXPHOS pathway (Ucp3, Fabp4, Cox5a, and Atp5o) were measured by real‐time PCR in GC of MCK‐PGC‐1α compared to WT mice. Cyclophilin‐A was used as an internal control. Normalized copies of the indicated genes in CTR mice were set as 1. (F) Western blot analysis of the OXPHOS Complex I‐V proteins in same GC samples as in C. (G) Quantification of OXPHOS complex I–V proteins versus tubulin levels shown in (C). Comparisons and differences were analyzed for statistical significance by the two‐way ANOVA test and Bonferroni posttest analysis. Data are shown as mean ± SD from at least three separate experiments. **p < 0.01, ***p < 0.001, ****p < 0.0001. GC PGC‐1α, MCK‐PGC‐1α gastrocnemius; GC WT, WT gastrocnemius.

As expected, we found high expression levels of Ucp3, Fabp4, and Atp5o, which are well‐known PGC‐1α target genes (Lima et al. 2019; Supruniuk, Mikłosz, and Chabowski 2017) (Figure 2E) and a significant increase of oxidative phosphorylation (OXPHOS) complexes in PGC‐1α overexpressing muscles compared to WT (Figure 2F,G).

These data confirm that PGC‐1α overexpression modifies the metabolic behavior of muscle fibers, promoting the conversion from glycolytic‐to‐oxidative myofibers and suggest that the observed changes in metabolite secretion and uptake might be associated with an enzymatic reassessment that alters the intracellular metabolite intermediates processing.

3.3. The Altered Metabolites Secretion From PGC‐1α SMFs Potentiates the Browning of Subcutaneous White Adipocytes

To directly assess the effect of soluble factors produced by MCK‐PGC‐1α SkM fibers on adipocyte biology, we performed browning induction of WT vWAT and sWAT derived adipocytes in the presence of CM obtained from WT and PGC‐1α SMFs, hereinafter referred as CM SMFs WT and PGC‐1α (Figure 3A).

Figure 3.

The conditioned medium from SMFs‐PGC‐1α promotes subcutaneous adipocyte browning. (A) Schematic overview of the experimental plan. The SMFs‐conditioned medium was utilized for the browning induction of adipocytes derived from the subcutaneous and visceral WAT of WT mice. (B) Representative images for Oil Red O staining of sBAd and vBAd treated with CM SMFs PGC‐1α compared to controls (scale bar = 400 mm). (C) Quantification of red oil O (ORO) staining in (B). (D) mRNA levels of metabolic markers PPARγ, UCP1, LIPE, SMYD1, PGC‐1α, MCT1, LDHA, and LDHB were measured by real‐time PCR in sBAd treated with CM SMFs PGC‐1α compared to controls (n = 3 per group). (E) LdhA and LdhB ratio in the same samples as shown in (D). (F) mRNA levels of metabolic markers PPARγ, UCP1, LIPE, SMYD1, PGC‐1α, MCT1, LDHA, and LDHB were measured by real‐time PCR in vBAd treated with CM SMFs PGC‐1α compared to controls (n = 3 per group). (G) LdhA and LdhB ratio in the same samples as shown in (F). Cyclophilin‐A was used as internal CTR. Comparisons and differences were analyzed for statistical significance by the two‐way ANOVA test and Bonferroni posttest analysis. (H) Representative images of Red MitoTracker staining of sBAd and vBAd treated with CM SMFs PGC‐1α compared to controls. (I, J) The relative MitoTracker red fluorescence intensity and median fluorescence intensity (MFI) in the same samples as shown in (H). Data represent the mean ± standard deviation of the mean of three replicates. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. CM, conditioned‐medium; sBAd, subcutaneous brown adipocytes; SMFs, single myofibers; vBAd, visceral brown adipocytes.

The differentiation of vWAT and sWAT derived white adipocytes into beige adipocytes in the presence of CM SMFs PGC1α resulted in a higher accumulation of lipid droplets compared to those cultured in the presence of CM SMFs WT, as shown by staining with Oil Red O (Figure 3B). On light microscopy, the lipid droplets in brown adipocytes derived from sWAT in the presence of CM SMFs PGC1α appeared markedly shrunken compared with that exposed to CM SMFs WT. Moreover, we observed that the CM SMFs PGC‐1α induced brown adipocytes with lipid droplets more uniformly distributed than observed in the presence of CM SMFs WT and this difference was remarkably evident in sWAT‐derived adipocytes (sBAd) respect to vWAT‐derived adipocytes (vBAd) (Figure 3C).

Also, we observed that CM SMFs PGC‐1α induced thermogenic gene expression in white subcutaneous adipose cells. Indeed, the 7 day treatment of differentiated adipocytes isolated from sWAT with CM SMFs PGC‐1α resulted in a robust increase of Ucp1 mRNA levels together with an upregulation of the expression of additional key thermogenic genes such as SmyD1 (Cicatiello et al. 2024) and the master regulator of mitochondrial biogenesis Pgc‐1α (Figure 3D). CM SMFs PGC‐1α treatment significantly also increased the expression of the nuclear receptor Pparγ, a master regulator of both white and brown adipogenesis, as well as the fatty acid oxidation marker LipE (Figure 3D). While no effect in the expression of key thermogenic genes was observed in CM SMFs PGC‐1α treated vWAT derived adipocytes (Figure 3F). The only significant change observed in both vWAT and sWAT derived adipocytes was the upregulation of LdhB isoform. However, the increase was prominent in sBAd, as such as the ratio LdhA/LdhB was significantly reduced only in sBAd (Figure 3E) and not in vBAd (Figure 3G). Consistently, the elevation of LdhB expression was associated with the upregulation of the lactate transporter Mct1 in sBAd treated with CM SMFs PGC‐1α. The enhanced thermogenic gene expression by CM SMFs PGC‐1α was associated with an increase of mitochondrial activity as shown by immunofluorescence and FACS analysis (Figure 3H–J), which showed a higher mitochondrial fluorescence intensity in sBAd treated with CM SMFs PGC‐1α respect to CM SMFs WT treated control cells (Figure 3I,J, left). While no difference was observed in vBAd (Figure 3I,J, right). Together, these data reveal that the CM SMFs PGC‐1α is a strong inducer of brown adipose gene expression in white adipose cells, and this effect is not shared in vWAT and sWAT but is of exclusive pertinence of sWAT. Since the sWAT‐derived cells express high level of Mct1, our hypothesis is that the alteration of muscle‐secreted metabolite may be sensed only from subcutaneous white adipocytes that express the selective transporter.

3.4. Muscle‐Specific PGC‐1α Overexpression Remodels WATs in Adult Mice

To evaluate whether the overexpression of PGC‐1α in SkM induces significant metabolic alterations in vivo that impact on adipose tissue phenotype, we analyzed the body fat distribution and visceral and subcutaneous adiposity in the adult mice (Figure 4A). We observed a slightly reduced total body weight in MCK‐PGC‐1α mice (Figure 4A), which was associated to a diminished fat mass in front of an unchanged lean mass, as indicated by the measurements of the vWAT, sWAT, BAT and gastrocnemius (GC) weights (Figure 4B). The decreased adiposity was specifically located in sWAT depots (Figure 4B). On light microscopy, all the MCK‐PGC‐1α animals presented sWAT morphology with smaller adipocytes than the correspondent WT mice (Figure 4C). The diameter of the adipose cells was smaller than the control, suggesting that a process of beiging could be activated in MCK‐PGC‐1α sWAT. Interestingly, the analysis of gene expression of key thermogenic and metabolic factors revealed molecular changes similar to those observed in ex vivo cultures, with an upregulation of PPARγ, LipE, and LdHB in sWAT but not in vWAT (Figure 4D–G). The increase of LdhB mRNA levels in sWAT of MCK‐PGC‐1α mice was associated with an increase in protein content as shown by Western blot (Figure 4H,I). The analysis of protein levels also evidenced a shift from PKM1 to PKM2, this is of relevance since PKM2 has been reported to be enriched in brown adipose compared to WAT (Marie S. Isidor FEBS 2020). Moreover, we observed a strong expression of SMYD1 and PGC‐1α well‐known browning markers (Cicatiello et al. 2024) (Figure 4H,I).

Figure 4.

Skeletal muscle‐specific PGC‐1α overexpression slightly induces browning of subcutaneous WAT. (A) Whole body weights expressed in grams (g) were measured in 8‐month‐old male MCK‐WT and MCK‐PGC‐1α mice. (B) The weights of sWAT, vWAT, BAT, and GC tissues expressed in milligrams (mg) were measured in the same mice as in (A). (C) Representative images of hematoxylin and eosin staining of subcutaneous and visceral WAT (sWAT and vWAT, respectively) derived from MCK‐WT and MCK‐PGC‐1α mice (scale bar = 400 mm). (D) mRNA expression analysis of metabolic marker genes in sWAT MCK‐WT and sWAT MCK‐PGC‐1α tissues (n = 3 per group). (E) LdhA and LdhB mRNA ratio in the same samples as shown in (B). (F) mRNA expression analysis of metabolic marker genes in vWAT MCK‐WT and vWAT MCK‐PGC‐1α tissues (n = 3 per group). (G) LdhA and LdhB mRNA ratio in the same samples as shown in (D). Cyclophilin‐A was used as internal CTR. Comparisons and differences were analyzed for statistical significance by the two‐way ANOVA test and Bonferroni posttest analysis. Data represent the mean ± standard deviation of the mean of three replicates. (H) Western blot analysis for LDHA, LDHB, PKM1, PKM2, SMYD1, and PGC‐1α as in A. β‐Actin was used as an internal loading control. (I) Histograms represent the relative protein quantification versus β‐Actin levels in the same samples as shown in (A). (J) Representative images for Oil Red O staining of 3T3 cells treated with lactate, pyruvate, and glutamine, separately (scale bar = 400 mm). (K) Quantification of red oil O (ORO) staining in (H). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. sWAT MCK‐WT or MCK‐PGC‐1α, subcutaneous WAT derived from MCK‐WT or MCK‐PGC‐1α; vWAT MCK‐WT or MCK‐PGC‐1α, visceral WAT derived from MCK‐WT or MCK‐PGC‐1α.

To further confirm that the muscle‐produced metabolites are key determinants of the browning, we performed an in vitro addition experiment in which 3T3L1 cells were induced to differentiate in brown adipocytes in the presence of physiological amounts of the relevant metabolites by NMR (Figure 1C). We treated 3T3L1 cells for 7 days with 1 mM lactate, 1 mM pyruvate (Yoon et al. 2022) and 1 mM l‐glutamine (Cicatiello et al. 2022), recapitulating the doses observed in CM and in vivo conditions. We observed that all the three tested metabolites enhanced lipid droplets with an increased Oil Red O staining as compared to control condition (Figure 4J,K), suggesting that the stimulation of the browning of subcutaneous white adipocytes observed in CM SMFs PGC‐1α experiments and in the tissues of MCK‐PGC‐1α mice was linked to the presence of the tested metabolites.

Altogether, these data, in agreement with the previous in vitro experiments, point to a metabolic remodeling in the adipose tissue of MCK‐PGC‐1α mice that, although potentially “normal” harboring a WT genomic DNA, exhibits an activation of pathways known to induce the browning of white adipocyte with a conversion towards a beige‐like phenotype.

4. Discussion

Adipose tissue and skeletal muscle are highly plastic and closely related tissues. Both tissues secrete molecules capable of modulating local and systemic metabolism, thus affecting energetic homeostasis. Indeed, myokines and adipokines secreted from the corresponding tissues have a key role in the modulation of body composition and metabolism. Besides the well‐known list of myokines and adipokines involved in SkM‐adipose tissue cross talk, a large number of molecules such as miRNAs, exosomes or metabolites are secreted by both tissues and act as signal molecules. In the network of molecules produced by muscle and adipose tissue, metabolites play a central role, also considering that SkM and AT are metabolically active organs (Cicatiello, Di Girolamo, and Dentice 2018). Metabolites can induce macromolecule activity and control phenotypes, thus reshaping the conventional thinking about the molecular linear “central dogma” that states that the transfer of information from DNA to RNA and protein, is possible, but the transfer from protein to protein or protein to nucleic acid is impossible (Guijas et al. 2018). From this perspective, SkM and AT can be considered endocrine organs involved in the regulation of several metabolic processes by the production of metabolites. Although the relationship between SkM and AT is largely explored, there is a lack of knowledge of the molecular basis of a such network and its adaptations that could be relevant in both physiological and pathological conditions.

Considering the high metabolic rate of SkM fibers, it is likely that muscle‐secreted metabolites can be relevant for adipose tissue physiology.

Here, we investigated the metabolites secreted by SkM and their potential role in SkM‐adipose tissue cross talk. In this study, we used mice with muscle‐specific overexpression of PGC‐1α, a transcriptional coactivator that promotes mitochondrial biogenesis, to determine whether increased oxidative potential impacts on the phenotype and metabolic proprieties of adipose tissue.

The rational choice of MCK‐PGC‐1α as the study model is based on the observation that the muscle‐specific overexpression of this coactivator augments mitochondrial biogenesis, fat oxidation, and energetic metabolism, therefore representing a tool to activate metabolic responses in muscle and potentiate metabolite production and processing. Moreover, since PGC‐1α expression and activity increase with exercise training, this mouse model is also a credited paradigm of exercise. We showed that the highly oxidative activity of PGC‐1α overexpressing muscles causes a modified secretion and an altered uptake capacity of metabolites as compared to WT mice. This led to an overall altered balance of the extracellular metabolites, including lactate, pyruvate and glutamine.

The NMR analysis of serum from MCK‐PGC‐1α mice showed that the reduced L/P ratio was due to increased pyruvate circulating level and, in agreement, the CM derived from PGC‐α1a SMFs (CM SMFs PGC‐1α) showed a drastic reduction of pyruvate uptake, also leading to a reduction of extracellular L/P.

A working muscle both produces and utilizes lactate as fuel energy, with much more lactate generated by glycolytic fibers and, instead, more lactate oxidized in oxidative fibers. The net result of trained muscle that promotes mitochondriogenesis is an enhanced lactate clearance (Bergman et al. 1999; Dubouchaud et al. 2000). In our cellular model, the increased lactate secretion and the reduced pyruvate uptake in CM SMFs PGC‐1α, can be explained by the high metabolically activity and by the switch to oxidative metabolism. The PGC‐1α overexpression driven by MCK promoter induces the shift of type II fibers in mitochondria‐enriched muscle fibers, as established elsewhere (Lin et al. 2002; Yang et al. 2020) and it is associated to an enhanced expression of mitochondrial enzyme as confirmed in our study.

The clear shift from LDHA to LDHB isoform suggests that in muscle the lactate is converted in its oxidized form, the pyruvate, probably to replenish the TCA cycle. This elevated lactate‐deriving pyruvate, eventually, renders the muscle fibers metabolically autonomous, leading to a reduced pyruvate uptake from culture medium, with an overall consequent decrease of extracellular L/P ratio. We isolated fast‐twitch muscle fibers from EDL to study the effect of forcing PGC‐1α‐dependent metabolism in a muscle that poorly depends on mitochondria for its energetics and, therefore, in which the effects of mitochondrial activation are more appreciable than in slow‐twitch fibers, such as the soleus, which is an oxidative muscle per se. The EDL is the typical white glycolytic muscle, in which the pyruvate consumption is principally accompanied by an abundant lactate production (Khattri et al. 2022). However, in our mouse model that constitutively expresses PGC‐1α in muscle, EDL switches from glycolytic to oxidative metabolism, acquiring a hypermetabolic mitochondrial metabolism. As a consequence, the MCK‐PGC‐1α muscles, although still maintaining high levels of lactate production, exhibit an increased intracellular generation and oxidative utilization of pyruvate.

Due to the extensive changes in metabolite secretion and uptake observed in both sera and supernatants of SMFs isolated from MCK‐PGC‐1α, and also inspired by the literature indicating a tight relationship between SkM and AT, we explored whether this disbalance in muscle‐derived metabolites could have an effect on adipose phenotype. In particular, we asked if the perturbed metabolites profile influenced the thermogenic activity and metabolism of white adipose cells in vitro, and WAT phenotype in vivo.

We found a global hyperactivation of cellular metabolism with an acceleration of glycolytic pathway and an increased mitochondrial pyruvate utilization in white subcutaneous adipose cells in presence of muscle‐derived metabolites. Indeed, white subcutaneous adipose cells exposed to CM SMFs PGC‐1α exhibited a strong shift from LDHA to LDHB and from PKM1 to PKM2 isoforms. These observations pointed to an altered glycolytic and oxidative metabolism that are known to be fundamental in browning. This is supported by the robust induction of Ucp1, PPARγ and SmyD1, well‐known browning markers, specifically in white subcutaneous adipose cells (Cicatiello et al. 2024). Accordingly, when we analyzed the adult phenotype of MCK‐PGC‐1α mice, we found a reduction of sWAT depots and a propension to beige‐like phenotype compared to control mice. Regarding to lean mass, we did not observe differences in MCK‐PGC‐1α mice. This is in agreement with previous literature data showing an induction of metabolic switch from glycolytic to oxidative muscle fibers in MCK‐PGC‐1α mice without a net effect on fiber size and muscle weight (Lin et al. 2002), a phenomena referred as “fiber type–fiber size paradox” that describes an inverse relationship between striated muscle fiber size and its oxidative capacity: the high oxidative fibers remain relatively small (van Wessel et al. 2010).

To establish whether the stimulation of browning in sWAT was due to the muscle‐produced metabolites, we performed in vitro addition experiments in which selected metabolites were added in the culture media. The addition experiments demonstrated that the stimulation of the browning in the subcutaneous white adipocytes conditioned by PGC‐1α SMFs was linked to the presence of the lactate, pyruvate and glutamine in the extracellular environment. Similar results were obtained in tissues of MCK‐PGC‐1α mice.

Also, our data highlight an important dogma of cellular physiology, whereby a cell is sensitive to a signal, independently of its nature if hormonal or metabolic, if it exposes the transporters or receptors for a such signal. In this regard, the different behavior between vWAT and sWAT could be explained by the higher sensitivity of sWAT to the metabolic changes caused by the PGC‐1α overexpression in SkM due to enhanced expression of specific transporters and the enzymes for processing of muscle‐derived metabolites.

Overall, our study contributes to better understanding of the previously described SkM and AT network in MCK‐PGC‐1α mice, adding a further layer of complexity to the direct muscle–adipose tissue metabolic signaling, in which muscle‐derived metabolites signalize to adipose tissue regulating biological processes and conferring adaptive responses to environmental changes.

5. Conclusions

Metabolite secretion by oxidative SkM, due to constitutive PGC‐1α activation, plays a role in regulating adipose tissue metabolism and phenotype. Specifically, lactate, pyruvate and glutamine altered secretion by hyperactivated SkM fibers boosts browning of subcutaneous white adipocytes. The communication between SkM and adipose tissue through metabolite secretion represents a complex interplay between these tissues and might be relevant for maintaining overall metabolism. Future research focused on the study of the networking molecules produced by muscle and adipose tissue under physiological and pathological conditions is needed to better understand this intricate communication and to elucidate metabolites precise mechanisms of action. The investigations of the newly identified molecules produced by muscle and adipose tissue and elucidating their roles in SkM and AT crosstalk may provide novel insights on the effect of physical activity on metabolic adaptations in healthy subjects and also pave the way to innovative strategies for the prevention and treatment of metabolic disorders including obesity.

Author Contributions

C. Miro performed the muscle molecular studies, including isolation of single muscle fibers. L.A., A.N., S.S., and F.R. carried out qPCR and Western blot analysis on muscle cells and tissues. C. Menale performed the adipocytes isolation and differentiation. S.T. performed molecular analysis on adipocytes. M.S. performed the nuclear magnetic resonance (NMR) based metabolites profiling. A.G.C. designed the experiments, analyzed the results, validated the data reproducibility, and wrote the paper. M.D. supervised the research activity and execution, and revised the paper. All authors read and agreed on the manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting Information.

Data Availability Statement

The original contributions presented in the study are included in the article/Supporting Information, further inquiries can be directed to the corresponding author.