Prior to study participation, all subjects provided written informed consent in accordance with the ethical guidelines established by the Ethics Review Committee of the First Affiliated Hospital of Dalian Medical University (Approval No. PJ-KS-KY-2021–229).

This study investigated the potential mechanisms underlying aging-related AF by analyzing both blood and tissue samples from patients and controls. Between July 2022 and July 2023, 100 AF patients aged 60 or older, diagnosed according to the European Society of Cardiology 2020 guidelines [11], and with complete clinical data were recruited at the First Affiliated Hospital of Dalian Medical University. The exclusion criteria comprised valvular heart disease, acute coronary syndrome, dilated or hypertrophic cardiomyopathy, congenital heart disease, hyperthyroidism, inflammatory diseases, systemic diseases, moderate to severe renal impairment (estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m2), and any malignancy. Additionally, to avoid confounding with advanced cardiac structural remodeling, patients with low ejection fraction (< 50%), elevated B-type natriuretic peptide (BNP) (> 105 ng/mL), or elevated N-terminal pro-B-type natriuretic peptide (NT-ProBNP (> 365 ng/mL) were excluded [12]. Additionally, two control groups were selected, matched by gender: 100 young individuals without atrial fibrillation aged under 60 years and 100 older individuals without atrial fibrillation aged over 60 years. Blood samples were collected from all participants after an overnight fast between 6:00 and 7:00 AM, anticoagulated, stored at 4 °C for < 1 h, centrifuged, and frozen at -80 °C. Additionally, left atrial tissues were collected from each group of 8 elderly patients with and without atrial fibrillation who underwent valve replacement surgery between December 2023 and August 2024, excluding patients with mitral stenosis. These tissues were fixed with 4% paraformaldehyde, 2.5% glutaraldehyde, or frozen at -80℃ depending on downstream analysis.

Patient data collection entailed a comprehensive review of medical records, encompassing demographic attributes, lifestyle details, medication usage, and laboratory test results such as fasting blood glucose (FBG), uric acid (UA), lipid profiles, blood urea nitrogen (BUN), and creatinine. Age was recorded at admission, smoking defined as daily cigarette use exceeding one cigarette for over a year, and BMI calculated. Blood pressure was measured in the right arm supine position using an automated monitor in the morning. All subjects underwent transthoracic echocardiography performed by a skilled echocardiographer to assess left atrial diameter (LAD) and left ventricular ejection fraction (LVEF).

β-OHB, Citrate Synthase activity and Acetyl-CoA assay in plasma and tissues.

β-hydroxybutyrate (β-OHB) and CS activity were quantified in plasma (n = 100 per group) and tissue samples (5 mg for β-OHB, 15 mg for CS; n = 6 per group). β-OHB concentrations were determined using a commercially available colorimetric assay kit (E-BC-K785-M, Elabscience Biotechnology) according to the manufacturer's protocol. For CS activity, Sigma's assay kit (MAK193, Merck) was employed as per the provided instructions. Briefly, samples, standards, and reduced glutathione (GSH) positive controls were loaded into a 96-well plate followed by a reaction mixture containing CS developer and substrate [13]. Absorbance at 412 nm was measured at 5-min intervals over 20–40 min to quantify CS activity. Acetyl-CoA levels were similarly assessed in plasma and left atrial tissues (n = 100 and n = 6 per group, respectively) using an ELISA kit (E-EL-0125c, Elabscience Biotechnology) [14].

Three-Dimensional Electroanatomic Mapping and Delineation of Low-Voltage Areas.

For comprehensive localization of the entire left atrium (LA), including its posterior, anterior, inferior, parietal, and pulmonary vein (PV) orifices, a 10-pole circular catheter was employed in conjunction with a 3D electroanatomical positioning system (CARTO, Biosense Webstar, USA). This approach yielded an average of approximately 1700 mapping points per patient across the LA. To distinguish normal and low-voltage tissue, reference voltage values were utilized: regions exceeding 0.5 mV were classified as normal, areas below 0.15 mV were categorized as low-voltage, and values between 0.15 and 0.5 mV defined a border zone [15, 16]. Using CARTO 3 software (Biosense Webstar, USA), the percentage of low-pressure area acquisition points relative to the total number of LA points was assessed. Subsequently, in both anterior–posterior, posterior-anterior, left anterior oblique and right anterior oblique four orientations, 3D electroanatomical maps were acquired for each patient. Quantification of the percentage of low-voltage areas within these maps was then performed using ImageJ software, with independent measurements conducted by two technicians.

Briefly, total protein samples were extracted from left atrial tissues with lysis buffer containing protease/phosphatase inhibitors (Thermo Fisher Scientific, Carlsbad, California, USA). Proteins (40–50 μg) were subjected to separation via 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis, followed by their transfer onto a polyvinylidene difluoride membrane (Millipore, Billerica, Massachusetts, USA). The membrane was subsequently blocked using 5% skimmed milk powder in TBST for 1 h at room temperature, followed by overnight incubation at 4 °C for 12 h with the respective primary antibody. All blots were generated utilizing the Shenhua Technology Chemiluminescence System (SH-523, Hangzhou, China). Quantification of protein levels involved assessing band intensity through ImageJ (NIH, Bethesda, Maryland, USA) and normalization relative to glyceraldehyde-3-phosphate dehydrogenase protein levels, as described previously [17]. The antibodies used in this study are shown in Supplementary Table 1.

Tissue from each group underwent fixation in 4% paraformaldehyde, paraffin embedding, and sectioning at 5 μm intervals for analysis. Sections were subjected to histopathological analysis utilizing Masson's trichrome staining (Sigma-Aldrich, St Louis, MO), following established procedures, as described previously [18]. Immunofluorescence analysis was conducted on samples of atrial tissue to assess the expression of CS protein. In brief, sections fixed with 4% paraformaldehyde underwent a 15-min PBS wash at room temperature. Subsequently, the sections were blocked for 30 min in PBS containing 1% BSA and 0.5% Triton X-100. Following this, the monoclonal anti-CS antibody was applied at a dilution of 1:50 and left to incubate overnight at 4 °C. Following this, the samples were subjected to a 1-h incubation at room temperature with the secondary antibodies tetramethylrhodamine isothiocyanate (TRITC) or fluorescein fisothiocyanate (FITC). Nuclear staining was achieved using 4',6-diamidino-2-phenylindole (DAPI) for 5 min [19]. CS protein content was quantified using ImageJ software. Apoptosis was identified utilizing an in situ cell death assay kit (Roche, Basel, Switzerland) following the provided protocol [20]. Cardiac sections were captured in 5–8 randomly selected fields of view, and the quantification of TUNEL-positive cells and DAPI-stained total nuclei was performed using ImageJ software. The apoptosis rate was determined by computing the ratio between the number of TUNEL-positive cells and the total count of nuclei. For dihydroethidium (DHE) staining, the atrial samples were cut into 5-μm-thick sections and incubated with 10 μmol/l DHE in a light-protected humidified chamber at 25 °C for 60 min [21]. Images were captured using a fluorescence microscope (CX40-RFL, SOPTOP, Ningbo, China).

Transmission electron microscopy (TEM) was employed for the visualization of mitochondria in atrial myocytes. Specimens were initially fixed in 2.5% glutaraldehyde and subsequently post-fixed in 1% osmium tetroxide. Following fixation, the tissues underwent dehydration through a graded series of ethanol before being embedded in an 812 (90,529–77-4, SPI, USA) embedding medium. Ultrathin sections were sliced using a microtome (UC7, Leica, German), stained with a 2% uranyl acetate-saturated alcoholic solution and lead citrate, and subsequently visualized using a transmission electron microscope (JEM1400PLUS, Japanese electronics, Japan). For mitochondrial analysis, ImageJ software was utilized to quantify mitochondria in five randomly selected fields of view within each tissue. This process involved quantifying the mitochondrial region area, measuring the length of the longest axis, and calculating the aspect ratio using the multi-measurement ROI tool in ImageJ software [22].

Statistical analyses were performed using SPSS 20.0 or GraphPad Prism 9. The continuous variables were exhibited as mean ± SD, and categorical variables were expressed as percentages. Categorical variables were tested with chi-squared tests. Data normality was assessed for continuous variables using the Kolmogorov–Smirnov or Shapiro–Wilk tests. For normally distributed data within each group, group differences were evaluated using the t-test. The Mann–Whitney test was employed for non-normal distributions. The potential association between citrate synthase-related molecules and aging-related AF was subsequently examined via multivariable logistic regression analysis. Receiver operating characteristic (ROC) curves were constructed to evaluate the specificity and sensitivity of predicting aging-related AF using Plasma β-OHB as well as CS activity and Combined diagnosis of plasma β-OHB and CS activity, and the area under curve (AUC) was calculated. Evaluate the relationship between plasma β-OHB levels, CS activity, and left atrial low-voltage regions utilizing a multivariate linear regression model. Statistical significance was established at P < 0.05, Fig. 1.

Overview of the workflow for the present study. AF indicates atrial fibrillation;NAF indicates non-atrial fibrillation;β-OHB indicates β-Hydroxybutyrate; CS indicates citrate synthase