Age-related adiponectin resistance in human skeletal muscle dysfunction: in vivo and in vitro evidence

Abstract

Background

Sarcopenia is an age-related condition characterized by the progressive decline of skeletal muscle mass and function. Although adiponectin is known for its anti-inflammatory and insulin-sensitizing effects that support muscle regeneration, paradoxically, elevated levels in older adults are linked to decreased muscle mass, strength, and performance. This study aimed to investigate the relationship between adiponectin levels, age, body composition, and functional status in elderly individuals, as well as to perform in vitro analyses of adiponectin resistance.

Methods

A cohort of 393 elderly subjects underwent anthropometric, bioimpedance, and functional assessments. Plasma adiponectin levels were measured by ELISA, and AdipoR1/AdipoR2 expression in peripheral blood mononuclear cells (PBMCs) was evaluated. In vitro, human skeletal muscle cells (SkMCs) were exposed to high concentrations (50 µM) of AdipoRon, a dual AdipoR1/AdipoR2 agonist, for 24 and 72 h. Analyses include cell viability, oxidative stress, protein homeostasis, autophagy, proteasome activity, and lipid metabolism.

Results

In elderly subjects, plasma adiponectin levels negatively correlated with BMI (r =  -0.129; p = 0.03), lean mass (r =  -0.252; p = 0.001), muscle mass (r =  -0.296; p = 0.001), and physical performance (SPPB score; r =  -0.163; p = 0.007). After adjusting for BMI and fat mass, adiponectin levels positively correlated with age (r = 0.281; p = 0.001). AdipoR2 expression in peripheral blood mononuclear cells was inversely associated with both age and adiponectin levels, suggesting adiponectin resistance in aging. In vitro, high dose of AdipoR agonist -AdipoRon exposure leads to oxidative stress, impaired proteostasis, dysregulated lipid metabolism, AdipoR2 receptor downregulation, and reduced cell viability. Together, these findings support a model in which elevated adiponectin in aging reflects adiponectin resistance and cellular stress rather than beneficial adiponectin signaling, contributing to muscle dysfunction.

Conclusions

These findings highlight a shift in adiponectin signaling during aging, with the downregulation of AdipoR2 promoting systemic adiponectin resistance. Excessive AMPK activity, in the context of impaired AdipoR2 function, contributes to redox imbalance and metabolic dysfunction in the skeletal muscle, favoring a “senescent-like” phenotype.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-026-07959-9.

Background

Sarcopenia, an aging-related syndrome, is a progressive and generalized skeletal muscle disorder characterized by the loss of muscle mass and function [1–4]. It is associated with adverse outcomes such as falls, functional decline, frailty, and mortality [5, 6]. Although several factors, such as neuromuscular aging, malnutrition, alteration of inflammation status, and endocrine function, have been identified as pathophysiological mechanisms of sarcopenia, its etiology is not entirely understood [7].

Adiponectin, the most abundantly expressed human adipokine, is primarily produced by adipose tissue (AT) and, to a lesser extent, by skeletal muscle (SM) [8]. It is well demonstrated that adiponectin mediates inflammatory processes by inhibiting the expression of inflammatory cytokines. Moreover, it plays a protective role in cardiovascular health by reducing foam cell formation and increasing systemic insulin sensitivity [9–11].

More intriguingly, recent studies have revealed that adiponectin, by binding to its receptors Adiponectin Receptor AdipoR1 and AdipoR2, acts as a myogenic factor by participating in muscle differentiation and regeneration, thereby demonstrating a protective role against muscle atrophy [12–16]. Indeed, models designed to mimic the aging process, such as adiponectin-knockout mice and angiotensin II-infused mice, overexpression of adiponectin improved muscle regeneration [17]. This suggests that its presence has a significant influence on the regenerative process. Furthermore, the restoration of adiponectin levels in adiponectin knockout mice resulted in: (i) reduced markers of inflammation (TNFα, IL-1β, and CD68); (ii) increased expression of regeneration markers (Mrf4, myogenin, Myh3, Myh7); and (iii) morphological improvements, including larger muscle fibers and less inflammation and extracellular matrix between fibers [18, 19].

In addition, recent epidemiological studies have also demonstrated a negative relationship between circulating adiponectin and skeletal muscle mass and function. A meta-analysis of 557 individuals with sarcopenia and 832 controls showed that sarcopenic participants had significantly higher levels of adiponectin than controls [20]. Moreover, older Japanese women exhibited cross-sectional associations between low grip strength, body composition, and elevated serum adiponectin levels (≥ 20 mg/L) [21]. In older adults, further prospective analyses showed that high circulating adiponectin concentrations inversely correlated with muscle strength, a functional marker of physical performance [22, 23]. These conflicting data gave rise to the “adiponectin paradox”. Despite different explanations encompassing adiponectin resistance and compensatory effects, the underlying mechanisms explaining the negative association between age, adiponectin levels, and physical performance in elderly subjects remain elusive.

Therefore, this study aims to investigate the relationship between age, adiponectin levels, body composition, and physiological function in a cohort of elderly subjects.

Additionally, it aims to evaluate the cellular consequences of adiponectin receptor overactivation by treating human skeletal muscle cells with a high concentration of AdipoRon (50µM), with a particular focus on cell viability, oxidative stress, protein homeostasis, and metabolic function.

Methods

Subjects’ enrolment

Three hundred ninety-three eligible participants were older adults consecutively admitted to the Internal Medicine and Geriatrics Division of the University of Campania “Luigi Vanvitelli” between October 2017 and October 2019 and were clinically stable and free of major metabolic, inflammatory, neoplastic, or neuropsychiatric conditions. Patients with diabetes, obesity, NYHA class III-IV, corticosteroid therapy, previous or ongoing anti-tumor therapy, severe respiratory failure, continuous anti-inflammatory therapy, kidney or liver failure, infectious pathologies, autoimmune disorders, mental disorders, terminal diseases (SPV < 6 months), and dementia were excluded from the study. Diabetes was diagnosed according to the criteria of the American Association of Clinical Endocrinologists and the American Diabetes Association [24] completed a specific questionnaire regarding the medications used before the initiation of the study, including the dates of treatment initiation and termination, the route of administration, and the duration of use. The study was approved by the Ethics Committee of the University of Campania Luigi Vanvitelli, protocol no. 372 dated 29/05/2017, and informed written consent was obtained from each patient.

Anthropometric and bioelectrical impedance analysis measures

Weight and height were measured using standard techniques. BMI was calculated as weight (kg) divided by height (m²). Waist circumference was measured midway between the lower rib and iliac crest, and hip circumference was measured at the level of the greater trochanter. Both measurements were taken to the nearest 0.5 cm using a a non-stretchable plastic tape, and the waist-to-hip ratio (WHR) was calculated. Baseline blood pressure was recorded using a mercury sphygmomanometer; the disappearance of Korotkoff sound (phase V) defined the diastolic value. Body composition was assessed using bioelectrical impedance analysis (BIA 101 BIVA, Akern/RJL, Italy), which measures fat mass, fat-free mass, lean mass, muscle mass, and phase angle. Body cell mass (BCM), reflecting metabolically active lean tissue, was also calculated. Participants lay supine with limbs slightly apart. Electrodes were placed on the right wrist and ankle for voltage sensing, and on the hand and foot for current injection.

Handgrip test

Handgrip strength was measured using an electronic hand dynamometer (CAMRY MODEL EH101, HANDCREW, Guangdong, China). Two consecutive measurements of grip strength in both hands were recorded to the nearest kilogram, with the participant placed in an upright position and the arm of the measured hand parallel to the body. Maximum grip strength was calculated by averaging the highest measurement from both hands.

Physical performance

The Short Physical Performance Battery (SPPB) assesses lower limb function through balance, gait speed, and chair-stand tests. Balance was evaluated in three progressively challenging stances (side-by-side, semi-tandem, tandem), each held unaided for 10 s. Gait speed was measured using the GAITRite system. For the five-times sit-to-stand test, participants began seated, performed one practice trial, then repeated the action five times as quickly as possible, keeping feet flat. Scores were assigned based on performance, with a maximum total score of 12.

Adiponectin ELISA assay

Whole blood was centrifuged to obtain plasma. According to the manufacturer’s protocol, the Adiponectin assay was performed using the Human ADP/Acrp30 ELISA Kit (Elabscience, Cat. # E-EL-H6122). Optical density was determined using an Infinite M Nano⁺ plate reader (Tecan), set at 450 nm.

Cell culture

Human skeletal muscle cells (SkMC; Lonza, cat #2561, lot. #0000418985), were cultured in SkGM™-2 basal medium (CC-3246), supplemented with SkGM™-2 SingleQuots™ (CC-3244), at 37 °C in 5% CO₂. AdipoRon (ab141867, Abcam), an adiponectin receptor agonist, was dissolved in DMSO to prepare a 50 mM stock solution. Cells were treated with 10, 25, or 50 µM AdipoRon for 24–72 h.

Cell viability assay

Cell viability was assessed using the Cell Counting Kit-8 (CCK-8, cat. #CK04, Dojindo) according to the manufacturer’s instructions. The samples were measured by reading the absorbance at 450 nm using a Sunrise absorbance reader (TECAN).

Western blotting

After treatment, cells were lysed in ice-cold RIPA buffer for 10 min, vortexed three times, and centrifuged at 15,000 rpm for 30 min. Supernatants were collected, and protein concentration was determined using the Bradford assay (Bio-Rad, cat. #5000006).

Proteins were separated by SDS-PAGE and transferred to nitrocellulose membranes (Bio-Rad, cat. #1704158) using the Bio-Rad Trans-Blot Turbo system. Membranes were blocked in 5% milk/TBS-T for 1 h, then incubated overnight at 4 °C with primary antibodies, FAS (Cell signaling cat. #3180); PPAR-γ (Cell signaling cat. #2443); oxidative stress cocktail (Abcam, ab179843); SOD2 (Abcam, ab13533), AMPK (Abcam, ab32047), phospho-AMPK (Elabscience, cat. #E-AB-21121), Bcl-2 (Elabscience, cat. #E-AB- 15522), PPAR-α (Abcam, ab227074), LC3B (Abcam, ab192890), ADIPOR1 (Abcam, ab70362), ADIPOR2 (Thermo Fisher Scientific, PA5-114166), BECN1 (Elabscience, cat. #E-AB-53242), p62 (Elabscience, cat. #E-AB-70387), Ubiquitin (Abcam, ab134953), Caspase-3 (Abcam, ab184787), Vinculin (Abcam, ab129002). After three 5-min washes in TBS-T, membranes were incubated for 1 h at room temperature with HRP-conjugated secondary antibodies: goat anti-rabbit (Bethyl, cat. #A120-101P) and donkey anti-mouse (Bethyl,, cat. #A90-137P), both at 1:5000. Each Western blot experiment was performed in three independent replicates (n = 3). Detection was performed using Clarity Max ECL substrate (Bio-Rad,, cat. #1705062) and visualized with the ChemiDOC Imaging System and Image Lab Software (v6.1). Protein size was determined with Opti-Protein Marker (ABM,, cat. #G623), and densitometric analysis was done using ImageJ.

RNA extraction and real-time PCR

RNA was extracted using the TRIzol reagent (QIAzol Lysis Reagent, cat. no. 79306) and reverse-transcribed using the SuperScript III Reverse Transcriptase Kit (Invitrogen Corp). qPCR with SsoAdvanced Universal SYBR Green Supermix (Bio-Rad, cat. #1725270) was used to determine mRNA levels using the Rotor-Gene Q (Qiagen). A threshold cycle (Ct) value was obtained for each amplification cycle, and ΔC_t was calculated as the C_t difference between target mRNA and housekeeping mRNA (β-actin). The fold increase of mRNA expression compared with the control sample was calculated using 2^−ΔΔCt method.

Primer’s list: (hADIPOR1 fw: 5’-GGGCGCTGA TCG GG-3’, rv: 5’-CCCGTTTGCCCTTCTCTTCT-3’; hADIPOR2 fw: 5’-GGCTCCCAAGAAGTCCGAG-3’, rv: 5’-TCGGTTTTCTGTTGGCTCGT-3’; MURF1: Hs.PT.58.39092203; Atrogin1: Hs.PT.58.19947148; β-actin: fw: 5′-CATCCGCAAAGACCTGTACG-3′, rv: 5′-CCTGCTTGC TGATCCACATC-3′).

ROS/superoxide detection assay

Intracellular ROS levels were measured using an ROS detection kit (Abcam, ab139476). The staining protocol followed the manufacturer’s instructions for the ROS/Superoxide Detection Assay Kit. The measurements were performed using a BD Accuri C6 Plus Personal Flow Cytometer (BD Biosciences), and data processing was carried out using FlowJo BD Accuri C6 Plus Software for Windows.

Autophagy detection assay

The autophagy detection assay kit (Abcam, ab139484) was used following the manufacturer’s instructions. Samples were analyzed for green fluorescence (FL1) using flow cytometry (BD Accuri C6 Plus). Measurements were performed using a BD Accuri C6 Plus Personal Flow Cytometer (BD Biosciences), and data processing was performed using FlowJo BD Accuri C6 Plus Software for Windows.

Protein synthesis assay

The protein synthesis assay was conducted according to the manufacturer’s instructions for the protein synthesis assay kit (Abcam, ab239725), and the measurements were carried out using a flow cytometer.

Lipid droplet detection assay

Cells were washed three times with PBS for the lipid droplet assay (Dojindo Lipid Droplets Assay Kit Deep Red, cat. #LD06). Image analysis software was used to quantify the number, size, and intensity of lipid droplets. Thresholding methods were applied to distinguish droplets from the background. Data generated from image analysis were analyzed to determine the lipid droplet content in the samples.

Proteasome activity assay

The Proteasome Activity Assay Kit (Abcam, ab107921) was performed according to the assay protocol instructions. Samples were mixed and incubated at 37 °C for 25 and 30 min while protected from light. The output was measured using a fluorometric microplate reader at an excitation/emission wavelength of 350/440 nm, and fluorescent intensity at T1 and T2 was recorded. Proteasome activity was calculated using the equations provided in the kit instructions.

Hydrogen peroxide assay

Cell culture and treatment were carried out as described in Sect.  1.6. The cell lysate was obtained with RIPA lysis buffer. Hydrogen peroxide generation was detected by a hydrogen peroxide colorimetric assay kit (Elabscience cat. #E-BC-K102-M).

Cell staining: autophagic vacuole detection

SkMC cells were seeded in a Nunc glass-bottom dish, 12 mm (Thermo Fisher Scientific, cat. no. 150680), and were treated with AdipoRon. Cells were washed three times with PBS and stained with an autophagy detection assay kit (Abcam, ab139484) for 30 min in a cell incubator. Then, the cells were washed with PBS and fixed with 4% paraformaldehyde (PFA).

Microscope imaging

SkMC cells were analyzed using confocal microscopy (Zeiss LSM700). A standard FITC filter set was used to image the autophagic signal, and a DAPI filter set was used to detect the nuclear signal. Fluorescence emitted by Hoechst 33,342 was detected using the DAPI light cube (Invitrogen, AMEP4950), and Alexa Fluor 488 was detected using the GFP light cube (Invitrogen, AMEP4951).

Statistical analysis

The sample size was estimated using G*Power software on an IBM computer to investigate the difference between groups. The experimental data were presented as mean ± standard error from at least three independent experiments. Differences between groups were compared using a one-way analysis of variance (ANOVA). Differences were considered statistically significant at a p-value < 0.05. Statistical analyses were performed using SPSS v23 software (IBM SPSS, Chicago). Quantitative image analysis was performed using ImageJ.

Results

Clinical study

Table 1 presents the clinical characteristics of the enrolled patients. The mean age was 66.9 ± 13.4 years. Age demonstrated a negative correlation with Short Physical Performance Battery (SPPB) scores (r =  -0.360; p < 0.001) and muscle strength (r =  -0.248; p = 0.004). After adjustment for body mass index (BMI) and fat mass (kg), age positively correlated with adiponectin plasma levels (r = 0.281; p = 0.001). Plasma adiponectin levels negatively correlated with BMI (r =  -0.129; p = 0.03), waist-to-hip ratio (WHR) (r =  -0.137; p = 0.02), lean body mass (kg) (r =  -0.252; p = 0.001), muscle mass (kg) (r =  -0.296; p = 0.001), and SPPB scores (r =  -0.163; p = 0.007). The association between SPPB score and adiponectin persisted even after adjustment for BMI and fat mass (kg) (r =  -0.169; p = 0.04). The expression levels of adiponectin receptors ADIPOR1 and ADIPOR2 were assessed in peripheral blood mononuclear cells (PBMCs) of a subgroup of patients (n = 100). A negative correlation between ADIPOR2 expression with age (r =  -0.273; p = 0.007) was found, even after adjustment for BMI and fat mass (Table 2). While no correlation was observed with age (r =  -0.154; p = 0.17), the expression levels of the adiponectin receptor ADIPOR1 negatively correlated with lean body mass (kg) (r =  -0.221; p = 0.04) and with muscle mass (kg) (r = -0.281; p = 0.01). Both Adiponectin levels and PBMCA DIPOR2 expression did not differ significantly between males and females in our study cohort (p = > 0.05). Furthermore, the associations of age with adiponectin levels (r=-0.201; p = 0.001) and with PBMCs ADIPOR2 (r=-0.276; p = 0.007) expression remained significant after adjustment for sex, indicating that these relationships were independent of sex.

Table 1.

Clinical characteristic of the study population (n = 393)

Clinical parameters
Age (yrs)	66.98 ± 13.46
Sex (F/M %)	45.5/54.5
BMI (kg/m2)	26.57 ± 4.5
Systolic arterial pressure (mmHg)	128.35 ± 14.53
Diastolic arterial pressure (mmHg)	79.40 ± 10.50
Glucose (mmol/L)	4.68 ± 0.50
HbA1c %	5.83 ± 0.39
Cholesterol (mmol/L)	182.37 ± 45.53
HDL (mg/dL)	50.80 ± 16.31
LDL (mg/dL)	107.29 ± 39.64
Triglycerides (mg/dL)	126.37 ± 88.21
SPPB	10.18 ± 3.02
Fat Mass (Kg)	23.33 ± 12.41
Lean Mass (Kg)	55.20 ± 12.85
Muscle mass (Kg)	26.56 ± 7.69
Adiponectin (ng/mL)	10.22 ± 7.87

BMI: body mass index; HDL: high-density lipoprotein; LDL: high-density lipoprotein;

SPPB: short physical performance battery

Table 2.

Correlations of adiponectin, and adiponectin receptors expression with physical performance and body composition

Variable	Adiponectin r (p)	ADIPOR1 r (p)	ADIPOR2 r (p)
Age (years)	0.205 (0.001)*	-0.134 (0.17)	-0.273 (0.007)*
SPPB score	-0.163 (0.007)*	0.135 (0.22)	0.143 (0.15)
Muscle strength	-0.145 (0.140)	0.154 (0.095)	0.138 (0.18)
Lean body mass (kg)	-0.252 (0.001)*	-0.221 (0.04)*	-0.047 (0.68)
Muscle mass (kg)	-0.296 (0.001)*	-0.281 (0.012)*	-0.101 (0.33)

Values are Pearson’s correlation coefficients (r) with p-values in parentheses. ADIPOR1 and ADIPOR2 expression were assessed in PBMCs in a subgroup of patients (n = 100). SPPB: Short Physical Performance Battery. *p < 0.05, even after adjustament for BMI and fat mass

In vitro study

Dose-response curve of AdipoRon in human skeletal muscle cells (SKMC)

A dose-response curve was performed to evaluate cell viability and granularity and determine the optimal concentration of AdipoRon for further experiments.

SKMCs exposed to AdipoRon at concentrations of 10 µM and 25 µM for 24 h did not show a difference in cell viability compared to untreated SKMCs (CTRL). In contrast, skeletal muscle cells treated with AdipoRon at 50 µM concentration for 24 h exhibited a statistically significant reduction in cell viability compared to CTRL (p < 0.01) (Fig. 1a).

Fig. 1.

Dose–response curve of AdipoRon in human skeletal muscle cells (SkMCs) evaluating cell viability and granularity. (a, b) Cell viability, cell size, and cell granularity after AdipoRon treatment at 10, 25, and 50 µM for 24 and 72 h. The relative cell viability was normalized to the CTRL using optical density values. Cell size and granularity were determined by flow cytometry. Data are presented as mean ± SEM from three independent experiments. *p < 0.05; **p < 0.001

Extension treatment with AdipoRon to 72 h did not provide any statistically significant difference between AdipoRon 10 µM and CTRL in cell viability (p > 0.05) (Fig. 1a). Conversely, SKMCs treated with AdipoRon at 25 µM and 50 µM concentrations for 72 h displayed a statistically significant reduction in cell viability compared to CTRL (p < 0.01) (Fig. 1a).

The flow cytometry SSC/FSC plots reveal a progressive increase in the proportion of cells exhibiting larger size and higher granularity, consistent with senescent-like morphological changes. In the upper panel, the percentage of senescent-like cells increases gradually from approximately 9.6% to 19.1%. Similarly, in the lower panel, senescent-like cells rise from 12.3% to 24.1%, followed by a pronounced escalation to 44.9% in the final condition (Fig. 1B).

Overall, these data indicate a dose- or time-dependent induction of senescent-like morphology, with prolonged exposure to 50 µM AdipoRon (72 h) eliciting a markedly enhanced senescent-like phenotype.

AdipoRon effects on adiponectin receptors and downstream signaling

The impact of AdipoRon treatment on the expression of AdipoR1 and AdipoR2 was also evaluated.

Specifically, AdipoRon treatment did not affect AdipoR1 expression at either time point, although an up-regulation of its downstream target p-AMPK was observed after 24 h (p < 0.05), returning to control levels after 72 h. Moreover, both 24 h and 72 h AdipoRon treatments resulted in a significant decrease in AdipoR2 expression, as well as in its downstream target PPAR-α (p < 0.05) (Fig. 2).

Fig. 2.

AdipoRon effects on adiponectin receptors. Western blotting analysis of AdipoR1, p-AMPK, AMPK, AdipoR2, and PPAR-α. The histograms show the densitometric analysis of three independent experiments, representing the relative expression, with the CTRL value set as 1. Data are presented as mean ± SEM. *p < 0.05; **p < 0.001

AdipoRon effects on apoptosis and oxidative stress in SKMC

AdipoRon induced a statistically significant reduction in BCL-2 protein levels at both 24 h (p < 0.05) and 72 h (p < 0.01) (Fig. 3). In contrast, 24 h treatment with 50 µM AdipoRon did not show any difference in pro-Caspase-3 protein levels compared to CTRL. This result was reversed after 72 h of treatment, demonstrating a significant decline in pro-Caspase-3 compared to CTRL (p < 0.05) (Fig. 3).

Fig. 3.

AdipoRon effects on apoptosis. Western blotting analysis of Bcl-2 and pro-Casp3. The histograms show the densitometric analysis of three independent experiments, representing the relative expression, with the CTRL value set as 1. Data are presented as mean ± SEM. *p < 0.05; **p < 0.001

As far as changes in oxidative stress are concerned, AdipoRon treatment for 24 h and 72 h induced a statistically significant increase in reactive oxygen species (ROS) production compared to CTRL (p < 0.01) (Fig. 4a), while a significant up-regulation of superoxide levels was only observed after 72 h of treatment with 50 µM AdipoRon (p < 0.05) (Fig. 4a). Consistently, the analysis of the main proteins involved in oxidative stress defense showed a substantial reduction in catalase and thioredoxin protein levels, together with an up-regulation of SOD2 protein levels, following treatment with 50 µM AdipoRon for 24 h and 72 h compared to CTRL (p < 0.01) (Fig. 4b). Additionally, only 50 µM AdipoRon for 72 h increased H₂O₂ and carbonyl group levels compared to CTRL (p < 0.05) (Fig. 4c and d).

Fig. 4.

AdipoRon effects on oxidative stress. (a) Intracellular ROS and superoxide levels were measured using a ROS/Superoxide detection assay kit. The histograms show the analysis of three independent experiments, and the figures below are representative flow cytometer images. (b) Western blotting analysis of catalase, SOD2, and thioredoxin. (c) H₂O₂ levels detected using a hydrogen peroxide colorimetric assay kit. (d) Western blotting analysis of carbonyl groups. The histograms show the densitometric analysis of three independent experiments, representing the relative expression, with the CTRL value set as 1. Data are presented as mean ± SEM. *p < 0.05; **p < 0.001

Effects of AdipoRon on protein homeostasis in SKMC

No changes in protein synthesis and ubiquitinated protein levels were observed with 50 µM AdipoRon at 24 h and 72 h (Fig. 5a and b). Similar results were observed in atrogin mRNA levels and in the gene expression of the skeletal muscle–specific E3 ligase MuRF1 at 50 µM AdipoRon for 24 h. In contrast, 50 µM AdipoRon for 72 h showed an increase in MuRF1 gene expression (Fig. 5c), a reduction in proteasome activity (Fig. 5d), and increased autophagic vacuole formation compared to CTRL (p < 0.05) (Fig. 5e).

Fig. 5.

AdipoRon effects on protein homeostasis. (a) Protein synthesis measured by flow cytometry. (b) Western blotting analysis of ubiquitinated proteins. The histograms show the densitometric analysis of three independent experiments, representing the relative expression, with the CTRL value set as 1. (c) qRT-PCR for Atrogin-1 and MuRF1. The fold increase in mRNA expression compared with CTRL was calculated using the 2^−ΔΔCt method. (d) Proteasome activity was measured using a microplate reader at Ex/Em = 350/330 nm. (e) Representative images of autophagic vacuole detection (green) and nuclear staining with Hoechst 33,342 (blue). Scale bar: 100 μm. Histogram graph and fluorescence median of autophagy vacuoles estimated by flow cytometry; (f) Western blotting analysis of LC3-II, BECN1, and p62. The histograms show the densitometric analysis of three independent experiments, representing the relative expression, with the CTRL value set as 1. Data are presented as mean ± SEM. *p < 0.05; **p < 0.001

Finally, a significant increase in LC3-II and p62 protein levels, but not in BECN1, was detected as autophagic markers in response to 50 µM AdipoRon (p < 0.01) at the different time points (Fig. 5g).

AdipoRon effects on metabolic pathways in SKMC

The effect of high-dose AdipoRon on metabolic pathways was also investigated by analyzing the expression levels of FAS, a key regulator of fatty acid synthesis, and PPAR-α and PPAR-γ, proteins involved in lipid oxidation and insulin sensitivity. Treatment with 50 µM AdipoRon for 72 h significantly upregulated FAS, downregulated PPAR-γ (Fig. 6a), and induced lipid droplet accumulation compared to CTRL (Fig. 6b).

Fig. 6.

AdipoRon effects on metabolic pathways. (a) Western blotting analysis of FAS and PPAR-γ. The histograms show the densitometric analysis of three independent experiments, representing the relative expression, with the CTRL value set as 1. (b) Lipid droplet detection by flow cytometry. Data are presented as mean ± SEM. *p < 0.05; **p < 0.001

Discussion

Our study demonstrates that in a cohort of elderly subjects: (i) age is associated with increased plasma adiponectin levels and a decrease in AdipoR2 expression in PBMCs independent of BMI and fat mass (ii) increased adiponectin levels are associated with reduced physical performance and muscle mass and strength. In vitro studies in SKMC treated with high concentrations of AdipoRon, a synthetic small-molecule agonist of AdipoR1 and AdipoR2, showed: (i) activation of AMPK, and inhibition of AdipoR2/PPAR−α signaling pathways; (ii) a decrease in cell viability (iii) the acquisition of a senescent-like morphology characterized by an increase in cell granularity and the development of a senescent-like phenotype, marked by oxidative stress, disturbances in protein homeostasis, and metabolic deterioration, including alterations in lipid metabolism.

In skeletal muscle, adiponectin, a peptide hormone secreted by adipocytes, plays a multifaceted role in skeletal muscle function, exhibiting anti-inflammatory properties, improving insulin sensitivity, and contributing to muscle differentiation and regeneration [25]. By activating the IRS-1/Akt signaling pathway, adiponectin prevents muscle protein degradation and supports myogenesis [26]. Restoration of adiponectin in knockout mouse models has been shown to reduce pro-inflammatory markers (TNFα, IL-1β, CD68), enhance expression of regeneration markers (Mrf4, myogenin, Myh3, Myh7), and improve muscle fiber morphology [27].

Paradoxically, despite these established protective roles, clinical studies have consistently reported elevated adiponectin levels in older adults, which correlate with adverse outcomes, such as reduced skeletal muscle mass, impaired physical function, and an increased risk of disability and mortality [28, 29]. Moreover, cross-sectional studies have associated higher adiponectin concentrations with sarcopenia [30].

This contradictory evidence has led to the “adiponectin paradox,” an issue whose underlying mechanisms remain poorly understood. A plausible hypothesis is that the age-related increase in adiponectin levels may represent a compensatory response to development of adiponectin resistance, characterized by impaired receptor expression and signaling efficiency [31]. Consistent with this idea, we found that age is associated not only with elevated plasma adiponectin levels but also with reduced expression of AdipoR2 in PBMCs in a cohort of elderly individuals. Elevated adiponectin could reflect a compensatory response that becomes ineffective because target tissues and circulating cells express less AdipoR2, limiting appropriate downstream responses. Such a finding mirrors alterations previously reported in skeletal muscle tissue [32, 33] and suggests that receptor downregulation may be a systemic response to aging, potentially undermining the efficacy of adiponectin signaling despite elevated circulating levels.

To further dissect the impact of impaired adiponectin signaling in skeletal muscle, we conducted in vitro experiments treating human skeletal muscle cells with AdipoRon. The in vitro experiments were not intended to replicate physiological adiponectin signaling, but rather to model the cellular consequences of excessive or unbalanced adiponectin receptors activation. Consistent with previous evidence [34], exposure to high concentrations of AdipoRon (50 µM) induced activation of the AdipoR1/AMPK pathway, coupled with inhibition of the AdipoR2/PPAR-α signaling axis. This imbalance between pathways appears to have detrimental consequences, as treated muscle cells displayed reduced viability and acquired a senescent-like phenotype within 72 h. Cellular senescence is a hallmark of tissue aging, characterized by increased cell size, enhanced granularity, and profound disturbances in cellular homeostasis [35–37].

In our model, AdipoRon-treated skeletal muscle cells showed significant increases in both size and granularity, suggesting that activation of adiponectin signaling under certain pathological conditions may paradoxically promote a senescent-associated phenotype in muscle tissue. In line with this, AdipoRon-treated cells exhibited increased ROS and superoxide production and an upregulation of SOD2 as a mitochondrial antioxidant response. A concomitant downregulation of key cytosolic antioxidant enzymes, catalase and thioredoxin, led to an imbalance in redox homeostasis and to the accumulation of hydrogen peroxide. This oxidative environment led to protein oxidative modifications (carbonylation) and activation of autophagy — a process responsible for the clearance of long-lived insoluble protein aggregates and entire organelles [38, 39].

Numerous studies have consistently shown increased autophagy in senescent cells, emphasizing its crucial role in preserving replicatively senescent cells [40–42]. At the same time, the ubiquitin-proteasome system (UPS) is responsible for the degradation of short-lived proteins and soluble misfolded proteins [43]. Interestingly, it has been reported that senescent cells prefer to utilize autophagy for the turnover of polyubiquitinated proteins, rather than relying on the ubiquitin-proteasome system [34, 44]. Notably, in vitro experiment, an increase in LC3 levels and the formation of lysosomal/autophagic vacuoles were observed. Furthermore, p62 levels rose dose- and time-dependently, consistent with autophagy induction following proteasomal impairment [45]. Protein homeostasis is a highly dynamic and finely tuned process. Previous research has underscored the role of adiponectin in maintaining this balance [46]. Our in vitro data demonstrate that AdipoRon does not affect de novo protein synthesis but significantly promotes protein degradation through autophagy. Notably, MuRF1 expression increased in a dose- and time-dependent manner, while Atrogin-1 remained unchanged. Despite MuRF1 upregulation and reduced proteasome activity, we did not observe an accumulation of ubiquitinated proteins, suggesting efficient degradation via the autophagic pathway rather than the UPS. Moreover, disturbances in lipid metabolism were also evident. AdipoRon treatment significantly increased the expression of FAS in a time- and dose-dependent manner, leading to lipid droplet accumulation within skeletal muscle cells. Simultaneously, the expression of critical regulators of lipid oxidation and insulin sensitivity, such as PPAR-α and PPAR-γ, was suppressed. This shift favors lipid storage over utilization; a phenomenon typically observed in aged or metabolically impaired skeletal muscle. These findings suggest that reduced AdipoR2 expression, despite elevated circulating adiponectin, may drive maladaptive cellular responses that impair muscle homeostasis during the ageing process. Specifically, the preferential activation of the AMPK pathway in the setting of reduced AdipoR2 signaling appears to promote oxidative stress, disrupt proteostasis, impair lipid metabolism, and ultimately induce a senescent-like phenotype in muscle cells.

To strengthen the systemic relevance of our observations, we additionally assessed AdipoR1 and AdipoR2 expression in PBMCs of a subgroup of participants. Notably, AdipoR2 expression showed a significant inverse correlation with age and circulating adiponectin levels, even after adjusting for BMI and fat mass. These results further support the notion that AdipoR2 downregulation is a systemic and age-related feature, contributing to impaired adiponectin signaling. This mirrors the molecular alterations observed in skeletal muscle, reinforcing the concept of widespread adiponectin resistance. In contrast, AdipoR1 expression did not correlate with age or adiponectin levels but was inversely associated with lean body mass and muscle mass. This suggests a distinct regulatory mechanism for AdipoR1, possibly reflecting compensatory adaptations to progressive muscle loss. These findings highlight the divergent roles of adiponectin receptors and underscore the complexity of adiponectin signaling in aging, with AdipoR2 impairment playing a dominant role in systemic resistance. The observed decrease in AdipoR2 expression in peripheral blood mononuclear cells and skeletal muscle models suggests that adiponectin resistance may be a critical mechanism underpinning the adiponectin paradox during advancing age. Our in vitro data further reveal that excessive activation of the AMPK pathway, coupled with impaired AdipoR2/PPAR-α signaling, promotes oxidative stress, disrupts protein and lipid homeostasis, and induces cellular senescence. Thus, while adiponectin maintains critical protective roles in young and metabolically healthy individuals, alterations in receptor expression and downstream signaling pathways in the aging muscle may shift its role from beneficial to maladaptive. These findings offer novel insights into the molecular basis of the adiponectin paradox and identify potential targets for therapeutic strategies aimed at preserving skeletal muscle health during aging. Several limitations of our study must be acknowledged. First, the cross-sectional nature of the study limits the ability to infer causality; thus, longitudinal studies are necessary to confirm whether changes in adiponectin signaling precede and contribute to muscle function decline over time. Furthermore, the potential selection bias due to the study setting and recruitment procedures, and the possibility that concomitant medications could influence adiponectin signaling and related endpoints. Indeed, despite these unavoidable constraints, the use of stringent exclusion criteria was specifically designed to minimize clinically meaningful confounders and strengthen the internal validity of our findings. Second, assessing AdipoR2 expression in peripheral blood mononuclear cells is a proxy for systemic receptor alterations; however, this may not fully reflect receptor expression and functionality within skeletal muscle tissue. PBMCs are circulating immune cells increasingly used in gene expression studies because they can be easily collected multiple times in sufficient quantities. In contrast, adipose, muscle, and liver tissue samples require more invasive sampling. Thus, PBMCs have been used as a surrogate for tissues that are not easily accessible in gene expression studies investigating the molecular mechanisms underlying several human diseases [16]. Importantly, the gene expression profiles in PBMCs and other cell types are remarkably similar (approximately) 80% concordant [47]. In particular, similarities between myocytes and lymphocytes in AdipoR expression have been previously demonstrated, suggesting that the genes encoding the two AdipoRs are regulated similarly in these tissues [48, 49]. Future studies should incorporate direct assessments of adiponectin receptors in muscle biopsies to strengthen the translational relevance of these findings. Third, although informative, the in vitro experiments utilizing human skeletal muscle cells treated with AdipoRon may not entirely recapitulate the complexity of the in vivo environment. Finally, our in vitro system represents stress-induced senescence, rather than replicative aging. The relatively rapid onset of senescence-associated features is consistent with acute metabolic and oxidative stress–induced senescence. In particular, the observed effects may represent exaggerated responses due to high agonist exposure. Future studies using lower doses, selective receptor agonists, or genetic modulation of AdipoR2 will be essential to validate these mechanisms under more physiological conditions.

Overall, while these limitations should be considered, they do not detract from the significance of our findings, which provide critical new insights into the complex role of adiponectin signaling in skeletal muscle aging. Rather than serving solely as a protective factor, its dysregulated signaling may actively contribute to the decline of skeletal muscle health. Understanding the balance between adiponectin receptor signaling pathways and their downstream effects may offer novel therapeutic opportunities to prevent or attenuate age-related muscle deterioration and sarcopenia.

Conclusions

In conclusion, our findings reveal that in elderly individuals, elevated plasma adiponectin levels are paradoxically associated with reduced physical performance, loss of muscle mass, and molecular features of cellular senescence. The observed age-related decrease in AdipoR2 expression both in mononuclear cells and skeletal muscle supports the hypothesis of systemic adiponectin resistance as a key contributor to these adverse outcomes. Moreover, in vitro exposure of skeletal muscle cells to AdipoRon, a dual AdipoR1/AdipoR2 agonist, showed that excessive and imbalanced adiponectin signaling characterized by the preferential activation of the AdipoR1/AMPK axis and suppression of AdipoR2/PPAR-α pathways induces oxidative stress, disrupts proteostasis, impairs lipid metabolism, and promotes a senescent-like phenotype. These data suggest that dysregulation of adiponectin receptor expression and signaling may play a direct and deleterious role in age-related skeletal muscle deterioration. Future research should aim to clarify the molecular mechanisms underlying adiponectin resistance in aging, including the distinct regulatory roles of AdipoR1 and AdipoR2. Future studies will include the analysis of established senescence markers, such as SA-β-gal activity, p16^INK4a, γ-H2AX, and SASP factors, to confirm the senescent phenotype induced by AdipoRon.

Furthermore, identifying therapeutic strategies capable of restoring balanced adiponectin signaling either by enhancing AdipoR2 expression or selectively modulating downstream pathways may offer novel opportunities to preserve skeletal muscle integrity and function in older adults.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

ADIPOQ

Adiponectin

AdipoR1/AdipoR2

Adiponectin Receptor 1/Adiponectin Receptor 2

AMPK

AMP-activated Protein Kinase

AT

Adipose Tissue

BCM

Body Cell Mass

BIA

Bioelectrical Impedance Analysis

BMI

Body Mass Index

BSA

Bovine Serum Albumin (se presente nei reagenti; se non serve, lo rimuovo)

DMSO

Dimethyl Sulfoxide

ECM

Extracellular Matrix

ELISA

Enzyme-Linked Immunosorbent Assay

FAS

Fatty Acid Synthase

GAITRite

Gait Analysis System

H₂O₂

Hydrogen Peroxide

HRP

Horseradish Peroxidase

IL-1β

Interleukin-1 beta

IRS-1

Insulin Receptor Substrate-1

LC3

Microtubule-associated Protein 1 A/1B-light Chain 3

MuRF1

Muscle RING-finger Protein-1

PBMCs

Peripheral Blood Mononuclear Cells

PBS

Phosphate-Buffered Saline

PCR

Polymerase Chain Reaction

PFA

Paraformaldehyde

PPAR-α / PPAR-γ

Peroxisome Proliferator-Activated Receptor alpha / gamma

qPCR

Quantitative Polymerase Chain Reaction

ROS

Reactive Oxygen Species

RT

Room Temperature

SkMC / SkMCs

Skeletal Muscle Cells

SM

Skeletal Muscle

SOD2

Superoxide Dismutase 2

SPPB

Short Physical Performance Battery

TNFα

Tumor Necrosis Factor alpha

UPS

Ubiquitin–Proteasome System

WHR

Waist-to-Hip Ratio

Authors’ contribution

SS: acquisition of data, investigation, validation, writing - original draft; LS: investigation, validation, writing - original draft; MGB: data curation; AP: data curation; RAF: Data curation; NB: acquisition of data; AMMP: acquisition of data; AZ: acquisition of data; GT: acquisition of data; ZU: acquisition of data; MA: acquisition of data; RJ: acquisition of data MTV: writing - review & editing; AC: writing - review & editing; GP: supervision, conceptualization, writing - review & editing, funding acquisition; MB: supervision, conceptualization, writing - review & editing, funding acquisition. All the authors read and approved the final manuscript.

Funding

This work was supported by: the European Union – Next Generation EU, under the National Recovery and Resilience Plan (PNRR), Mission 4 – Component 1 CUP: B53D23030780001 Project Code: P2022RHFSS; PNRR Project ANTHEM (AdvaNced Technologies for Human-cEntred Medicine) Project Code PNC0000003, CUP: B53C22006540001; PRIN2022 - CUP: B53D23020210006; ARtificial Intelligence for Early RisK PrEdicTIon of Heart Failure by Combining Circulating EPi Signature tO Clinical Features Project Code F/310107/05/X56, CUP: B29J23000310005.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by the ethics committee of the University of Campania, L. Vanvitelli, prot. N. 372 of 29/05/2017. All methods were performed following the relevant guidelines and regulations (Declaration of Helsinki), and all participants provided their informed consent.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.