Blocking Ghrelin Receptors May Improve Muscle Function in Aging Male Mice

To understand the effects of GHSR‐1a deletion on sarcopenia in aged mice, we evaluated body composition, food intake, treadmill running time, grip strength, and hindlimb muscle mass in male GHSR‐1a WT and KO mice aged 6, 24 and 28 months. In the 6‐month groups, animals with GHSR‐1a deletion (KO) exhibited lower body weight and lean body mass (LBM) compared to WT mice (Figure 1A,B). Both genotypes exhibited increased body weight in 24‐month‐old mice, but not in 28‐month‐old mice, compared to 6‐month‐old mice (Figure 1A). A significant main effect of age was observed in LBM; however, the increase in LBM with age was only significant in KO mice (Figure 1B). As ghrelin may promote water retention via vasopressin release (Ishizaki et al. 2002), we assessed free water content (g) based on body composition analysis in young and very old mice. No significant differences were observed between WT and KO mice despite aging increasing free water content in both genotypes (Figure S1). Furthermore, 28‐month‐old mice displayed lower fat mass compared to 6‐ and 24‐month‐old mice in both genotypes (Figure 1C). No significant difference in daily food intake was found across groups (Figure 1D). Total hindlimb muscle mass was calculated as the combined weight of the bilateral soleus, gastrocnemius/plantaris complex (GAS/PL), tibialis anterior (TA), extensor digitorum longus (EDL), and quadriceps (Quad). It was lower in 6‐month‐old KO mice compared to same‐aged WT mice, it decreased with age in both genotypes, and no genotype difference was found in older groups (Figure 1E). Similarly, individual hindlimb muscles demonstrated age‐related atrophy in both genotypes, and KO mice showed reduced muscle mass at 6 months in soleus, GAS/PL, TA, and Quad (Figure 1F). Despite the absence of genotype differences in muscle mass at older ages, 24‐month‐old KO mice exhibited higher grip strength compared to age‐matched WT mice (Figure 1G). Six‐ and 28‐month‐old KO mice demonstrated 9.4% and 10.6% higher normalized grip strength (grip strength/LBM) compared to WT mice, respectively (Figure 1H); while this measurement declined with age in both genotypes. Twenty‐four‐ and 28‐month‐old KO mice showed 29.5% and 44.3% higher treadmill running times compared to WT mice, respectively, and KO mice also showed a delayed decrease in treadmill running time during aging compared to WT (Figure 1I). Additionally, we assessed sarcopenic status in mice aged 24 months or older based on our previously established definition in C57BL/6 mice, which incorporates measurements of grip strength, muscle mass, and treadmill running time (Kerr et al. 2024). Mice were categorized into four groups according to the number of deficits present across these three parameters (details in Methods): nonsarcopenic (NonS), probable sarcopenic (PS), sarcopenic (S), and severe sarcopenic (SS). Among aged mice, a significant genotype difference was observed between WT and KO, with KO mice exhibiting a higher proportion of NonS and S, and a lower proportion of PS and SS compared to the WT group (Figure 1J).

We examined fiber composition and cross‐sectional area (CSA) in plantaris (PL) muscles. In WT mice, aging significantly reduced the number of type IIB fibers, with 28‐month‐old mice exhibiting a lower fiber count compared to 6‐month‐old mice (Figure 2A). KO showed a similar pattern, though IIB fiber number did not decrease until 24 months. Also, KO mice exhibited fewer IIA fibers but more IIB fibers at 24 months (though the latter was not significant, p = 0.060; Figure 2A). In terms of the percentage of fibers, WT mice showed an increased proportion of IIA fibers, while a decrease in IIB fibers with aging (28 months vs. 6 months, Figure 2B). KO mice showed similar age‐related changes in proportions of IIA and IIB fibers at 28 months compared to both 6‐ and 24‐month‐old groups, though the difference in IIB fibers between 6‐ and 28‐month‐old KO mice was not statistically significant (p = 0.090). In addition, 24‐month‐old KO mice had a higher percentage of IIB fibers and a lower percentage of IIA fibers compared to WT mice (Figure 2B). Type IIB CSA decreased with age in 28‐month‐old mice of both genotypes (Figure 2C).

To understand the effects of GHSR‐1a deletion on muscle contractile function, we performed in situ muscle physiology in tibialis anterior (TA) muscles from 6‐ and 28‐month‐old WT and KO male mice. Force production of TA muscles decreased with aging during the force‐frequency test. In WT mice, the decline began at 100 Hz, whereas in KO mice, it started at 125 Hz (Figure 3A). Both genotypes exhibited age‐related decreases in peak force (Po, Figure 3B), but specific force remained unchanged (Figure 3C). TA muscle mass in the mice used for muscle physiology showed significant age‐related declines in both genotypes, while no genotype difference was detected (Figure 3D). During the fatigue test (150 Hz every 2 s), WT mice experienced age‐related declines in absolute force from 0 to 60 s, while KO mice showed a decline at 120 s. Compared to WT mice, KO mice were less fatigued, demonstrating higher absolute and normalized force at 30 and 60 s in 28‐month‐old mice, and higher absolute force at 120 s and normalized force at 120–240 s in 6‐month‐old mice (Figure 3E,F).

Declines in maximum mitochondrial respiration were observed in 28‐month‐old WT mice compared to their 6‐month‐old counterparts, including ADP‐stimulated and uncoupled maximum respiration stimulated by FCCP, in isolated mitochondria from PL muscles (Figure 4A). Similar patterns were also noted in baseline (p = 0.070), oligomycin (p = 0.054), and antimycin (p = 0.062) states in WT mice, though not significant. Age‐related declines in mitochondrial respiration were noted in GHSR‐1a KO mice; these were only significant in baseline and ADP and FCCP‐stimulated maximum respiration (Figure 4A). Notably, KO mice showed a significantly reduced delta change (old vs. young) in the FCCP state compared to the WT mice (Figure S2A). Age‐related decreases were also noted in oxidative phosphorylation (OXPHOS) complex IV content in WT mice (Figure 4B,C). We quantified citrate synthase (CS) activity in isolated mitochondria and derived whole‐muscle CS activity by multiplying mitochondrial CS activity (μmol/ml/min/mg) by the total protein content (mg, evaluated by Bicinchoninic acid assay) of mitochondria isolated from PL muscles. Notably, older KO mice displayed greater CS activity than age‐matched WT mice, and a significant age‐related reduction in CS activity was observed only in WT animals (Figure 4D). We also measured cytochrome C protein levels in isolated mitochondria to assess mitochondrial health; 6 months old KO mice showed higher levels of cytochrome C than the same‐aged WT or 28 months old KO (Figure S2B). In PL muscles, the number of succinate dehydrogenase (SDH)‐positive cells, an indicator of oxidative capacity, decreased with age in WT mice. KO mice showed a similar age‐related decline, though it was not statistically significant (Figure 4E,F). In WT mice, the mitochondrial biogenesis marker peroxisome proliferator‐activated receptor gamma coactivator (PGC)‐1α was lower in 28‐month‐old WT mice compared to 6‐month‐old mice, with a tendency towards lower levels compared to the 24‐month group (p = 0.059). In contrast, 28‐month‐old KO mice exhibited higher PGC‐1α levels than their younger peers (p < 0.05 vs. 6 months; p = 0.060 vs. 24 months) and age‐matched WT mice (Figure 4G,I). Young KO mice showed a higher level of Sirtuin 1 (Sirt1), a potential regulator of PGC‐1α, than same‐aged WT (Figure 4G). To assess the regulator of mitochondrial content, we measured mitochondrial DNA (mtDNA) copy number in PL muscles from 6‐ and 28‐month‐old WT and KO mice (Quiros et al. 2017). We observed a significant age‐related decline in relative mtDNA expression in WT mice, whereas this reduction was not evident in the KO group (Figure 4G). We further evaluated several mitophagy markers, as they can be served for mitochondrial quantity control. In WT mice, the protein levels of the mitophagy marker p62 showed a mild decrease in muscle tissues at 24 and 28 months of age compared to the 6‐month group (p = 0.072 and p = 0.088, respectively). In contrast, p62 levels in KO mice increased at 24 months and returned to baseline levels by 28 months. Notably, p62 levels were lower in 6‐month‐old KO mice but higher in 24‐month‐old KO mice relative to their age‐matched WT counterparts (Figure 4H,I). Also, gene expression of PINK1 (PTEN‐induced kinase 1), a key regulator of mitophagy, was significantly elevated in 28‐month‐old KO mice compared to WT mice of the same age (Figure 4H). In isolated mitochondria, 28‐month‐old KO mice displayed elevated protein levels of the mitophagy marker Parkin, normalized to voltage‐dependent anion channel 1 (VDAC1, Figure 4J), a marker of mitochondrial outer membrane. Consistently, 28‐month‐old KO mice also showed increased colocalization of autophagy marker microtubule‐associated proteins 1A/1B light chain 3B (LC3) and mitochondrial content marker Tomm20 compared to the same‐aged WT mice (Figure 4J).

To elucidate the molecular mechanisms underlying the progression of sarcopenia, we conducted proteomic analyses on GAS muscles of 6‐month‐old and 28‐month‐old WT and KO mice, in which we identified 3110 protein groups mapped from 24,483 detectable peptides. Individual protein abundances were quantified, and linear regression analyses were performed to assess associations between protein abundance and sarcopenia‐related measures, including grip strength, muscle mass, and treadmill running time. No significant correlations were detected at the individual protein level after multiple‐comparison adjustment (adjusted p < 0.05; data not shown).

To further identify pathways related to sarcopenia and evaluate how GHSR‐1a deletion affects these relationships differently in 6‐ and 28‐month‐old mice, we performed a competitive gene set test based on KEGG pathways (D. Wu and Smyth 2012). GHSR‐1a deletion had different effects on the correlations between KEGG pathways and sarcopenia measures in 6‐ and 28‐month‐old mice, with 5 pathways correlated with treadmill running time, 4 pathways correlated with grip strength, and 25 pathways correlated with muscle mass (Figure 5A and Table S1). Since KEGG pathway names do not always reflect the exact protein composition and several pathways share overlapping top‐ranked proteins, we listed the top five proteins from each pathway and mapped them to cellular locations using Gene Ontology (GO) Cellular Component terms (details in Table S2). As shown in Figure 5B, most of these proteins are located in mitochondria, followed by cytoplasm, plasma membrane (PM), and extracellular space (Figure 5B). Among the KEGG pathways, the complement and coagulation cascades pathway was affected by GHSR‐1a deletion differently in 6‐ and 28‐month‐old mice and was correlated with all three sarcopenia measurements, with its top‐ranked proteins located in the extracellular region (highlighted in green). Also, top‐ranked proteins correlated with treadmill performance and grip are primarily associated with the complement pathway and immune responses (IC1, CO8G, C1QB, C1QC, CO4B, and DAF), whereas those correlated with muscle mass are predominantly involved in coagulation and fibrinolysis (KLKB1, THRB, PLMN, A2AP, and PROC). Diabetic cardiomyopathy and thermogenesis pathways were found to be affected and correlated with treadmill and muscle mass, with most top‐ranked proteins being mitochondria‐based (highlighted in pink). Specifically, these proteins are primarily components of OXPHOS, including Complex I subunits (NDUB9, NDUB8, NDUA8, NDUAB, NDUS7, NDUB4, NDUB2, NDUF6) involved in ATP production, and Complex IV subunit COX6C, which facilitates the final step of mitochondrial respiration. Propanoate metabolism and valine, leucine, and isoleucine degradation pathways were affected in their correlations with grip strength and muscle mass. The top‐ranked proteins within these pathways are also mostly located in mitochondria, including those involved in amino acid metabolism (IVD, PCCA, MCCA, HIBCH), fatty acid beta‐oxidation (HCD2, DCMC, ACADS), the TCA cycle (ODBA, SUCB1, SCOT1), and glycolysis (LDHB).

As several of the top five proteins identified in the interaction term were localized to mitochondrion based on GO Cellular Component analysis, we further evaluated these proteins using GO Biological Process annotations. These mitochondrial proteins were predominantly related to mitochondrial respiration and ATP production, including aerobic respiration (GO:0009060), mitochondrial electron transport from NADH to ubiquinone (GO:0006120), proton motive force–driven mitochondrial ATP synthesis (GO:0042776), respiratory electron transport chain (GO:0022904), oxidative phosphorylation (GO:0006119), tricarboxylic acid cycle (GO:0006099) as well as mitochondrial ADP (GO:0140021) and ATP transmembrane transport (GO:1990544). While a subset of proteins was also associated with mitochondrial membrane function, oxidative stress responses, fatty acid and amino acid metabolism, and mitochondrial biogenesis, the majority were linked to mitochondrial respiration and energy metabolism (Table). S3

Additionally, as the second most populated top 5 protein location was extracellular, growing evidence shows its association with fibrosis and further leads to sarcopenia (Brack et al. 2007; Cai et al. 2023; Lieber and Ward 2013; Lofaro et al. 2021; Sanes 2003). We assessed muscle extracellular matrix (ECM) remodeling by quantifying collagen content in TA muscles using Masson's Trichrome staining, in which collagen is visualized in blue. The collagen‐positive area was quantified as a percentage of the total muscle cross‐sectional area. Aging significantly increased collagen content in both WT and KO mice; however, no significant genotype differences were observed (Figure S3A,B). To further explore the potential functional relevance of ECM alterations, we performed correlation analyses between collagen content and sarcopenia‐related outcomes, including muscle mass, grip strength, and treadmill running time, within the same age groups (Figure S3C). In young mice, collagen content was inversely correlated with grip strength (Spearman's correlation coefficient rho: −0.79, p = 0.036, N = 7). In contrast, in old mice, collagen content showed only a modest correlation with muscle mass though not significant (Spearman's correlation coefficient rho: −0.44, p = 0.057, N = 19), but no significant association with grip strength or treadmill performance.

We conducted a Kaplan–Meier survival analysis on 267 WT and 222 KO male mice. The median survival age was 126 weeks for WT mice and 125 weeks for KO mice, with no significant difference in survival between the two genotypes (Figure). S4

To investigate if pharmacologically targeting GHSR‐1a replicates the effects of GHSR‐1a deletion on muscle function and molecular markers, middle‐aged male mice (9–11 months) were treated with the GHSR‐1a inverse agonist PF‐5190457 (10 mg/kg, twice daily) in 10% DMSO for 28 days. Mice treated with 10% DMSO in saline served as the control group. Baseline measurements of food intake, body composition, and muscle function were taken before treatment and at weekly intervals until day 28. Weekly measurements were normalized to baseline levels and expressed as percentages. Decreases in cumulative food intake due to PF‐5190457 administration were observed at week 1 (p = 0.051) and reached significance in weeks 2 and 3 (Figure 6A). Body weight and fat mass were lower in PF‐5190457‐treated mice at weeks 3 and 4, while lean body mass remained unchanged throughout the treatment (Figure 6B–D). Improved treadmill running time in PF‐5190457‐treated mice started to show at week 3 (p = 0.059) and reached significance at week 4 (Figure 6E). After normalizing running time to body weight, the PF‐5190457 effect persisted, although it did not reach statistical significance (p = 0.081 at week 4, Figure 6E). Grip strength was not significantly different between groups (Figure 6F). At the end of the study, hindlimb skeletal muscles were collected for mass measurements. The normalized soleus muscles tended to be larger in PF‐5190457‐treated mice compared to DMSO‐treated mice (p = 0.087), whereas the normalized Quad tended to be larger in DMSO‐treated mice (p = 0.071). No significant differences were found in other muscles between groups (Figure 6G). Protein levels of the mitochondrial biogenesis marker PGC‐1α were elevated in the GAS muscles of PF‐5190457‐treated mice. Similarly, the transcript level of nuclear respiratory factor 1 (Nrf1) showed an increasing pattern with PF‐5190457 treatment (p = 0.056, Figure 6H). Markers of mitophagy, including LC3II protein and BCL2 Interacting Protein 3 (Bnip3) mRNA levels, were also increased following administration of PF‐5190457. Although no significant differences were detected in p62, Park2, and PINK1, their expression patterns were consistent with the observation in LC3II (Figure 6I). Mitochondrial respiration in isolated mitochondria in PL muscles from each group at the end of the study showed no significant difference across groups (Figure 6K).

Additionally, we examined the effects of PF‐5190457 on muscle mass and function in older mice (25–27 months of age). As in the younger cohort, these mice received either PF‐5190457 or DMSO for 4 weeks. Food intake, body composition, treadmill running time, and grip strength were assessed at baseline and weekly after the first dose (Figure S5A–F). Although no significant differences were detected across these parameters, PF‐5190457–treated mice showed lower cumulative food intake by week 3 (p = 0.081, Figure S5A) and exhibited numerically higher treadmill running times (Figure S5E). At the end of the study, hindlimb muscle mass was evaluated, and no significant difference was observed between treatments (Figure S5G). Notably, PF‐5190457 treatment resulted in significant increases in both basal and ADP‐stimulated maximal respiration in isolated mitochondria in PL muscles (Figure S5H).

The study used male growth hormone secretagogue receptor (GHSR)‐1a wild type (WT) and knockout (KO) mice on a C57BL/6J background, categorized as 6‐months (5–7 months), 24‐months (23–26 months), and 28‐months (27–30 months). The original Ghsr−/− mice, obtained from Dr. Roy Smith's laboratory, were backcrossed with the C57BL/6J strain for at least ten generations to minimize potential background genetic variations (Liu et al. 2021). The mice used in this study were descendants of these congenic lines and were bred in the Animal Research Facilities at the Veterans Affairs Puget Sound Health Care System. In this model, GHSR‐1a has been deleted from all tissues. It is worth noting that this receptor is not expressed in skeletal muscles (Sun et al. 2007). A total of 67 WT and 99 KO mice were evaluated in this study, and the experimental unit was each single animal. To minimize potential confounders, all mice were individually housed in the same room, maintained on a 12/12 light/dark cycle (lights on at 6 AM), and acclimated to their environment and human handling for 1 week before experiments (day 0–6). Daily food intake was then evaluated for 5 days after acclimation (day 7–12). In the following week, mice underwent a treadmill test (day 14), grip test (day 16), and body composition (day 18) analysis via nuclear magnetic resonance (NMR) as described previously (Kerr et al. 2024; Liu et al. 2021). We used CO₂ euthanasia for direct tissue harvest, whereas mice undergoing procedures requiring immediate functional assessment (e.g., Aurora muscle physiology or mitochondrial isolation from fresh tissue) were anesthetized with isoflurane followed by cervical dislocation. Hindlimb muscles were collected for mass estimation and biochemical analysis. Muscles for Western blotting, proteomics or mtDNA analysis were flash‐frozen in liquid nitrogen and stored at −80°C. Tissues for RNA extraction were preserved in RNAlater and stored at −80°C. Muscles designated for histology were embedded in OCT and flash‐frozen in liquid nitrogen–prechilled isopentane before storage at −80°C. Samples for mitochondrial isolation were processed immediately after collection. The specific muscles used for each assay are summarized in Table S4. Survival analysis was conducted on WT and KO mice that were born within the same 2‐year period. All procedures were approved by the Institutional Animal Care and Use Committee (IACUC) at the VA Puget Sound Health Care System and adhered to NIH Guidelines for the Use and Care of Laboratory Animals (IACUC protocol #0950).

Treadmill tests were conducted using the Exer‐6 M treadmill (Columbus Instruments, Columbus, OH). Mice were allowed to acclimate for 5 min on the stationary treadmill before testing. The protocol began at a speed of 5 meters (m) per minute for 5 min, with the speed increased by 1 m per minute thereafter. Mice were encouraged to run by gently tapping their rear with a cotton swab until exhaustion. A mouse was considered exhausted, and the test was concluded if, after three attempts to encourage it to run by tapping, it still refused and remained near the tester at the end of the treadmill. The protocol and volitional fatigue criterion were adapted from Kwak et al. (2020). The total time (seconds) the mouse remained on the treadmill was recorded.

Grip strength is the most widely accepted method to estimate muscle strength, allowing for more generalizable results. It has also been used as one of the criteria for sarcopenia diagnosis by our group (Kerr et al. 2024) and others (Borsch et al. 2021; Xie et al. 2021). A grip strength meter with a digital force gauge was used to measure forelimb grip strength (Columbus Instruments, Columbus, OH). Mice were acclimated to the environment before grip strength testing by placing them on the gripping grid without applying any pulling force. The mouse was allowed to use only its forelimbs to grasp a pull bar connected to a force gauge to assess forelimb grip strength. The maximum grip strength was recorded on the force gauge in kilograms. The test was performed three times at 1‐min intervals. Maximum grip strength was recorded. Absolute grip strength (g), and grip strength normalized to body weight (g/g) or hindlimb muscle mass (g/g) were reported.

Sarcopenia status was evaluated using three clinically relevant criteria adapted for mice: muscle strength, muscle mass, and physical performance. Muscle strength was assessed by forelimb grip strength (g), muscle mass by total hindlimb muscle weight (mg), and physical performance by treadmill running time (seconds to exhaustion). Cutoff values for each parameter were defined as 2 standard deviations below the mean of a young adult male C57BL/6J reference cohort, adapted from our previous study (Kerr et al. 2024). Specifically, low grip strength was defined as < 131 g, low muscle mass as < 708 mg, and impaired physical performance as < 514 s. Mice were then classified as nonsarcopenic (non‐S, no deficits), probable sarcopenic (PS, 1 deficit), sarcopenic (S, 2 deficits), or severe sarcopenic (SS, 3 deficits), depending on the number of criteria met.

Male C57BL/6J mice aged 9–11 months (middle‐aged) and aged 25–27 months (old, age at week 0) were included to test the effects of the GHSR‐1a inverse agonist PF‐5190457. Baseline measurements were conducted at week 0 after the mice had been individually housed for 5 days in the same room for acclimation. These measurements included food intake, body weight, body composition, treadmill running time, and grip strength. Animals were nonsystematically assigned to one of the following groups: the treatment group (PF‐5190457) received 10 mg/kg PF‐5190457 subcutaneously twice daily for 28 days (ghrelin receptor inverse agonist, MedChemExpress) in 10% DMSO in saline, while the control group received 10% DMSO in saline at a volume adjusted to match the body weight‐to‐liquid ratio used in the PF‐5190457 group. Throughout the experiment, food intake and body weight were measured daily, while body composition, treadmill tests, and grip tests were conducted weekly. Group allocation and data analysis were performed by one researcher, while treatment administration and behavioral testing were conducted by two researchers in a blinded manner using coded identifiers. At the conclusion of the 28‐day dosing, the mice were euthanized for tissue collection as previously described. Gastrocnemius muscles from middle‐aged mice were used to identify molecular markers. Plantaris muscles from both middle‐aged and old mice were used to determine mitochondrial respiration. No animals were excluded at the end of the experiment.

An in situ muscle physiology test was performed on the TA muscle using the Aurora 1300A 3‐in‐1 Whole Animal System for Mice (Aurora Scientific, Ontario, Canada), as previously described (Kerr et al. 2024). A pair of needle electrodes was used to stimulate the TA muscle, with the electrodes placed directly on the deep side of the muscle's proximal end while the mouse was under isoflurane‐induced anesthesia. Optimal length (Lo, mm) was determined through single twitch tests, followed by tetanus (150 Hz) and force‐frequency tests (10, 30, 50, 75, 100, 125, 150, 200 Hz) at one‐minute intervals. Peak force (Po) was measured as the highest force production during the force‐frequency test. The physiological cross‐sectional area (pCSA) was calculated using the equation: pCSA (mm²) = muscle mass (mg) / [1.06 × Lo (mm) × 0.6], where 1.06 represents skeletal muscle density and 0.6 is the TA length‐to‐fiber ratio (Brooks and Faulkner 1988; Burkholder et al. 1994). Specific force (SPo) was then calculated by dividing peak force by the pCSA. Fatigue tests were conducted by stimulating the muscle at 150 Hz every 2 s for 4 min.

The cross‐sectional area (CSA) of individual fibers from plantaris (PL) muscle was determined as previously described (Kerr et al. 2024, 2023). Plantaris was selected for fiber typing due to its mixed fiber composition and established use in previous studies (Bloemberg and Quadrilatero 2012). The muscle was mounted in OCT and sectioned into 10 μm slices at −22°C using a Cryostat (Leica CM3050S, Nußloch, Germany). After being dehydrated for 30 min, the muscle sections were blocked with 10% normal goat serum (NGS) in Phosphate Buffered Saline (PBS) for 1 h. The sections were then incubated with the following primary antibodies at room temperature for 1 h: SC‐71 (1:300) and BF‐F3 (1:25) to detect myosin heavy chain (MHC) isoforms IIA and IIB respectively (Developmental Studies Hybridoma Bank, Iowa City, IA), along with anti‐dystrophin to detect membranes (1:100, ab15277, Abcam). After three washes in PBS, sections were incubated for 1 h with the corresponding secondary antibodies (Thermo Fisher Scientific, Waltham, MA): Goat anti‐Mouse IgG1, Alexa Fluor 488 (A21121, 1:500) for MHC‐IIA; Goat anti‐Mouse IgM (Heavy chain), Alexa Fluor 555 (A21426, 1:500) for MHC‐IIB; and Goat anti‐Rabbit IgG (H + L), Alexa Fluor Plus 647 (A32733, 1:200) for membranes. After another three washes in PBS, the sections were mounted with Prolong Gold AntiFade reagent (P36934↗, Thermo Fisher Scientific). A Nikon Ni‐E microscope equipped with NIS‐Elements software (Nikon, Tokyo, Japan) was used to capture 10× images of the PL muscle cross‐sections. Approximately 100 type IIA, 200 type IIB, and 60–80 type nonstained fibers from the entire PL area were analyzed to determine the muscle fiber CSA, as previously described (Kerr et al. 2024). The percentage of each fiber type in the PL muscles was determined using the Cell Counter plugin in ImageJ analysis software (National Institutes of Health, http://rsb.info.nih.gov/ij/↗). To identify colocalization of Tomm20 and LC3 in tibialis anterior (TA) muscles, primary antibodies (Anti‐TOMM20 antibody, ab56783, Abcam, and LC3B (E7X4S) XP Rabbit mAb, 43566 T, Cell Signaling) were used, followed by corresponding secondary antibodies. The entire TA area was imaged at 40× using the Nikon Ni‐E's stitching function. A colocalization plugin in ImageJ was used to quantify Tomm20/LC3‐positive cells.

The SDH staining protocol was performed as previously described (Kerr et al. 2024). Plantaris (PL) muscle sections were incubated for 30 min at 37°C in a solution containing 10 mL of 0.2 M phosphate buffer (pH 7.6), 270 mg of sodium succinate (S2378, Sigma), and 10 mg of nitro blue tetrazolium (NBT, N6876, Sigma). After incubation, sections were washed three times with dH2O to remove unbound NBT, followed by three washes in acetone solutions of increasing concentration (30%, 60%, and 90%) to ensure thorough removal. The sections were then rinsed with dH2O and mounted with an aqueous mounting medium. Three trained and blinded researchers quantified the whole PL section using ImageJ (Fiji) software.

Masson's Trichrome stain was performed to assess collagen levels in TA muscles. A Trichrome Stain (Connective Tissue Stain) Kit (ab150686) from Abcam was used, and the procedure followed the manufacturer's instructions. Whole TA sections were imaged at 10× using the Nikon Ni‐E's stitching function. The “Color Deconvolution2” plugin was installed in Image J, and the “Masson Trichrome” vector was selected to quantify collagen‐positive area in blue. The percentage of collagen‐positive area in the whole TA muscle was calculated as the ratio of collagen‐positive area to total TA area in ImageJ (Fiji).

Quad (aging cohorts) or GAS muscles (PF‐5190457 interventions) were preserved in RNAlater (Qiagen) after collection. RNA was isolated using the Qiagen RNeasy mini kit, and transcription levels were quantified with the BioTek Cytation 5. cDNA synthesis was performed using the QuantiTect Reverse Transcription Kit (Qiagen). RT‐PCR was conducted on an ABI 7500 instrument using TaqMan Expression Assays (Thermo Fisher). Bnip3, Sirt1, PINK1, Nrf1, and Park2 gene expression were normalized to Hprt and expressed as relative fold‐change using the 2^−ΔCT method.

Approximately 15 mg of the PL muscle was used to isolate DNA with the QIAamp DNA Mini Kit (Qiagen). Transcription levels were measured using the BioTek Cytation 5. Quantitative real‐time PCR was performed with 100 ng of DNA, predesigned primers, SYBR Green master mix, and an ABI 7500 instrument to detect mtDNA depletions. mtDNA gene 16S was normalized to the nuclear gene Hexokinase 2 (HK2) and expressed as the relative mtDNA to nDNA copy number ratio using the 2‐ΔCT method. The following primers were purchased from Thermo Fisher Scientific: 16S rRNA: Forward: 5′‐CCGCAAGGGAAAGATGAAAGAC‐3′; Reverse: 5′‐TCGTTTGGTTTCGGGGTTTC‐3′; HK2: Forward: 5′‐GCCAGCCTCTCCTGATTTTAGTGT‐3′; Reverse: 5′‐GGGAACACAAAAGACCTCTTCTGG‐3′.

Protein extraction from whole muscle, protein quantification, and electrophoresis were described previously (Kerr et al. 2024). OXPHOS complex, Parkin, VDAC1, and cytochrome C protein levels were evaluated in isolated mitochondria as described in the following session (Mitochondria isolation). In brief, the BIO‐RAD Criterion Cell system was used to separate 50 μg of whole muscle homogenate or 10 μg of isolated mitochondria per lane on NuPAGE Bis‐Tris Midi Protein Gels, 4%–12% (WG1402A, Thermo Fisher Scientific). Kaleidoscope or Precision Plus Protein Standards (1610375 and 1610374, Bio‐Rad) were loaded at both ends of the membrane for reference. Proteins were transferred to nitrocellulose or PVDF membranes using a BIO‐RAD Criterion Blotter (170–4070, BIO‐RAD) or XCell4 SureLock Midi‐Cell (WR0100, Thermo Fisher Scientific). Nitrocellular membranes were stained with Ponceau S (59803, Cell Signaling), imaged, and cut based on molecular weight. After blocking in 5% nonfat dry milk, membranes were incubated overnight at 4°C with the following primary antibodies: anti‐SQSTM1/p62 (1:1000, Ab56416, Abcam), anti‐PGC1 alpha (1:1000, ab191838, Abcam), LC3B (1:500, NB100‐2220 Novus), total OXPHOS Rodent WB Antibody Cocktail (1:1000, ab110413, Abcam), anti‐Parkin antibody (1:500, ab77924, Abcam), anti‐VDAC1 (1:1000, ab14734, Abcam), and Cytochrome c Antibody (1:1000, 4272, Cell Signaling). After washing, membranes were incubated with HRP‐conjugated secondary antibodies for 1 h at room temperature. Detection was performed using SuperSignal West Dura Substrate (34075, Thermo Fisher Scientific) and imaged with ImageQuant LAS 4000 (GE Health Care). Densitometry was analyzed using ImageJ and normalized to the Ponceau red signal.

Mitochondria were isolated from freshly harvested PL muscles using a modified protocol (Boutagy et al. 2015; Rogers et al. 2011) with mitochondria isolation buffer (MIB: 210 mM sucrose, 2 mM EGTA, 40 mM NaCl, 30 mM HEPES, pH 7.4) on ice. Protein concentration was quantified using the BCA method, and the mitochondria pellet was resuspended in mitochondrial assay solution (MAS: 70 mM sucrose, 220 mM d‐mannitol, 10 mM KH2PO4, 5 mM MgCl2, 2 mM HEPES, 1 mM EGTA, 0.2% fatty‐acid free BSA, pH 7.4) with substrates (5 mM malate, 5 mM pyruvate). The sample (7.5 μg/well) was loaded in triplicate, and the oxygen consumption rate (OCR) was measured in real time using an XFe24 Seahorse Analyzer (Agilent Technologies). Sequential injections included ADP (2 mM), oligomycin (2 μM), FCCP (4 μM), and antimycin A (2 μM).

Citrate synthase (CS) activity in isolated mitochondria from plantaris (PL) muscle was measured using the Citrate Synthase Activity Kit (CS0720, Sigma) according to the manufacturer's instructions. For each assay, 3 μg of isolated mitochondrial protein was analyzed in duplicate. CS activity was normalized to protein load and expressed as μmol/mL/min/mg. To estimate total CS activity in the whole PL muscle, the normalized CS activity (μmol/mL/min/mg) was multiplied by the total mitochondrial protein content determined by BCA assay and expressed as μmol/mL/min.

Gastrocnemius muscles from male WT and KO mice aged 6–7 months and 27–29 months were used for proteomic analysis (N = 6/group). Batch design was randomly balanced based on condition ratios. The samples were divided into 2 batches of 12 individual samples and 2 pooled references for a total of 14 samples per batch. A pool of all 24 samples was used to create a reference pool to be used as a common reference, which was homogenized, aliquoted, frozen, and used to compare between the two batches. The processes of sample lysis and digestion, liquid chromatography and data‐independent acquisition (DIA) mass spectrometry, DIA signal processing were described previously (Kerr et al. 2024). Briefly, peptide identification and quantification were performed using chromatogram libraries generated from pooled reference samples, and peptide fragment ion intensities were integrated and summed to derive relative protein abundance across samples. Linear regression analyses were then carried out for all mice to explore potential correlations between individual proteins and any of the sarcopenia measures, including grip strength, muscle mass, or treadmill running time (p < 0.05). To further identify pathways that may be related to sarcopenia and evaluate how GHSR‐1a deletion affects these relationships differently in 6‐ and 28‐month‐old mice, we performed a competitive gene set test based on KEGG pathways using the ‘camera’ function from the Bioconductor limma package in R. This method can perform competitive gene set testing using any linear model and adjusts for interprotein correlations to reduce bias (Wu and Smyth 2012). We fit a model that included an interaction term between individual sarcopenia measurement (treadmill, muscle mass or grip strength), age group, and genotype and performed the gene set test based on the t‐statistic for the interaction term (False discovery rate, FDR < 0.05). For the top five proteins from each KEGG pathway, gene names were submitted to the UniProt ID mapping tool (https://www.uniprot.org↗) to retrieve Gene Ontology Cellular Component (GO‐CC) annotations and assign cellular locations. Mitochondrial proteins were further analyzed by GO Biological Process. The chord diagram demonstrating the cellular localization of proteins was generated using the R packages circlize and ComplexHeatmap (Gu 2022; Gu et al. 2014). The Skyline documents, raw files for quality control and DIA data are available at Panorama Public. ProteomeXchange ID: PXD048723. Access URL: https://panoramaweb.org/mouse‐gastroc‐sarcopenia‐proteomics.url↗.

The sample size was determined based on our previous study investigating the effects of aging on WT mice, where the sample size provided sufficient statistical power to detect significant differences in the same outcome measurements as those described in the current study (Kerr et al. 2024). Two‐way analysis of variance (ANOVA) was performed to identify differences between genotypes (WT vs. KO) across age groups (6, 28 and/or 24 months) followed by Tukey post hoc test if there were three age groups (p ≤ 0.05) or LSD post hoc test if there were two age groups (p ≤ 0.05). For data not normally distributed based on the Shapiro–Wilk normality test (p < 0.05), nonparametric tests (Mann–Whitney) were performed for pairwise comparisons. Pearson Chi‐square tests were used to detect genotype differences across the rates of sarcopenic status in aged mice. An independent t‐test (two‐tailed) was used to compare the treatment effects of PF‐5190457 (vs. DMSO). Log‐rank (Mantel‐Cox) test was used to detect genotype difference in survival (p < 0.05). p‐values were reported and discussed if greater than 0.05 and less than 0.1 for abovementioned tests. Values are presented in mean ± SE. No data points were excluded from analysis. All statistical testing was performed using IBM SPSS version 18 software and GraphPad Prism 10. Statistics for proteomics are described in the Proteomic analysis. Data is available upon request to the corresponding author and first author.

Values for all data points associated with the manuscript and supplemental results are provided in the Supporting Data Values document. Data availability for proteomics is provided under “Proteomic analysis”. Additional information and detailed methods in this paper will be available upon direct request from the authors.