Calcium-dependent signaling protein family member S100A11 promotes liver and adipose lipid accumulation through fibroblast growth factor 21

Abstract

BACKGROUND

S100A11, a calcium-binding protein implicated in cholesterol metabolism, and fibroblast growth factor (FGF) 21, a key liver-fat signaling factor elevated in metabolic disorders, were investigated for their roles in obesity and fatty liver.

AIM

To determine if S100A11 promotes hepatic lipid accumulation via FGF21.

METHODS

Transgenic mice overexpressing S100A11 in the liver and normal control wild-type mice were fed with normal or high-fat diet. S100A11 was overexpressed or knocked down in liver cell lines. Expression of triglycerides (TGs) and genes related to liver lipid metabolism was measured by real-time reverse transcription polymerase chain reaction. The liver tissues of mice fed with a high-fat diet were used for proteomic analysis. FGF21 neutralizing antibody was administered in both in vivo and in vitro models to validate its role in S100A11-induced hepatic and adipose lipid accumulation.

RESULTS

S100A11 overexpression elevated hepatic TGs and FGF21 in male mice. In hepatocytes, S100A11 overexpression upregulated TGs, lipogenic genes, and FGF21, which were reversed by S100A11 silencing. Coculture of S100A11-overexpressing hepatocytes with adipocytes suppressed brown-fat-related genes and promoted white-fat-related genes upon oleic acid stimulation. Proteomic analysis highlighted enrichment of the peroxisome proliferator-activated receptor pathway, where FGF21 is pivotal. Hepatic and adipose FGF21 was upregulated by S100A11 but downregulated upon its silencing. FGF21-neutralizing antibody partially reversed S100A11-induced hepatic and adipose lipid accumulation.

CONCLUSION

S100A11 promotes hepatic and adipose lipid accumulation in male mice, a process mediated by FGF21. This identifies S100A11 as a potential therapeutic target for obesity-related metabolic diseases.

Key Words: Non-alcoholic fatty liver disease; S100A11; Fibroblast growth factor 21; Type 2 diabetes mellitus; Obesity

Core Tip: This study reveals that the calcium-binding protein S100A11 promotes hepatic and adipose lipid accumulation by upregulating fibroblast growth factor 21, identifying S100A11 as a novel potential therapeutic target for obesity-related metabolic disorders.

INTRODUCTION

Obesity has become a major threat to world health and is closely related to metabolic syndromes such as diabetes, hypertension, and hyperlipidemia. Obesity is caused by long-term energy intake exceeding energy consumption, and it is continuously aggravated by various environmental factors[1]. With the acceleration of life rhythm, the intake of high-fat food is increasing, and the incidence of fatty liver is increasing annually[2]. Fatty liver is defined histologically by the presence of fat in more than one third of hepatocytes or when hepatic fat content exceeds 5% of liver wet weight. Patients with fatty liver exhibit dysregulated hepatic lipid, leading to excessive triglyceride (TG) accumulation in the liver[3,4]. Hepatic lipid metabolism involves multiple processes, including uptake, storage, synthesis, and breakdown.

Dietary fat is digested and incorporated into chylomicrons, which are absorbed by the small intestine. A portion of these lipids is transported to the liver, while the remainder is delivered to peripheral tissues for storage as adipose fat. Following adipose tissue dysfunction or consumption of a high-fat diet (HFD), excess lipids accumulate in the liver, promoting inflammation[5]. Fatty acid transport proteins (FATPs) on hepatocyte membranes play a key role in hepatic lipid uptake, particularly FATP2, FATP4, and FATP5[6]. Excess lipids are converted to peripheral fat and stored under the skin and internal organs. There are three main sources of hepatic lipids: Dietary free fatty acids, which are absorbed via the small intestine and transported to the liver in the form of chylomicrons; de novo lipogenesis within the liver; and free fatty acids released from the breakdown of peripheral adipose tissue and subsequently delivered to the liver[7]. Free fatty acids are first activated by fatty acyl-CoA synthetase to form fatty acyl-CoA. Together with acetyl CoA generated in the liver during aerobic metabolism, fatty acyl-CoA is converted into lysophosphatidic acid by glycerol-3-phosphate O-acyltransferase in the mitochondria. Through a series of enzymatic reactions, lysophosphatidic acid is subsequently converted to phosphatic acid. Phosphatic acid is converted to diacylglycerol through removal of its phosphate group from the glycerol backbone. Under the catalysis of diacylglycerol acyltransferase (DGAT), diacylglycerol is subsequently converted into TGs[8]. Hepatic lipid metabolism can be broadly divided into two main pathways: Β-oxidation of free fatty acids and export of TGs from the liver following their incorporation into very low-density lipoproteins (VLDL).

S100 protein is a group of small acidic calcium binding proteins expressed in a variety of tissues. S100 protein regulate cell proliferation, differentiation, survival and migration, inflammation and tissue repair, and/or exerts antibacterial activity under normal and pathological conditions. S100A11, also known as S100C or calmodulin, is an EF hand Ca²⁺ binding protein belonging to the S100 protein family. It was first identified in chicken smooth muscle and has also been found in several mammalian species and tissues[9-11]. S100A11 protein has a molecular weight of 13 kDa and is acidic. Existing research has demonstrated that S100A11 is associated with the occurrence of many cancers, including papillary thyroid cancer, colon cancer, pancreatic cancer and breast cancer[12-14]. Genome-wide association studies (GWASs) shows that S100A11 is closely related to cholesterol, peroxisome proliferator-activated receptor (PPAR), and diabetes, and its expression is affected by the hypoglycemic drugs rosiglitazone, pioglitazone, and metformin.

The fibroblast growth factor (FGF) family contains many members with various roles in cell growth, cell differentiation, and embryonic development[15]. Most FGFs are small, with molecular weights ranging from 17 kDa to 34 kDa. The FGF peptide binds to one of several tyrosine kinase FGF receptors (FGFR) expressed on the plasma membrane to trigger signal propagation[15]. FGF21 is a member of the endocrine FGF subfamily. The functional FGF21 receptor complex consists of FGFR and its co-receptor β-klotho; both of which are essential for FGF21 signal transduction[16]. FGF21 is an important regulator of heat generation in brown adipose tissue (BAT)[17,18]. In neonatal mice, FGF21 treatment increases the expression of heat-related genes in BAT, leading to an increase in body temperature[17]. FGF21 also inhibits lipolysis by reducing lipase activity and downregulating the activity of adipose tissue[19]. FGF21 expression can lead to more severe obesity[20]. The absence of cardiomyocyte-specific KLF5 aggravates diet-induced obesity through FGF21, and the ultimate positive or negative role of FGF21 in obesity depends on a multifactorial signaling network that has not been fully elucidated[21].

Whether S100A11 affects glucose and lipid metabolism and whether FGF21 is involved in this process are still unknown.

MATERIALS AND METHODS

Animals and treatments

C57BL/6J mice and transgenic mice expressing human S100A11 specifically in the liver (S100A11-hTg) were used in the present study. The protocol was designed to minimize pain or discomfort to the animals. Five-week-old male mice were housed in specific pathogen-free microisolators and maintained in a regulated environment (24 °C, 12-hour light-dark cycle, with lights on at 7:00 am). Regular chow and water were available ad libitum. Mice were assigned to receive standard laboratory normal-chow diet (NCD) or a HFD (45% fat, D12451; Research Diets, New Brunswick, NJ, United States) for 12 weeks. Animals used in this study were handled in accordance with the Guide for the Care and Use of Laboratory Animals published by the United States National Institutes of Health (No. 85-23, revised 1996). All experimental procedures involving animals were conducted in strict compliance with the guidelines of the Animal Research Ethics Committee of Peking University, approval No. LA2018061.

Administration of FGF21-neutralizing antibody in vivo

To investigate the functional role of endogenous FGF21 in our model, FGF21-neutralizing antibody (Biogradetech, BGT-ANT-37337) was administered to S100A11-hTg mice via intraperitoneal injection. The antibody was diluted in sterile phosphate-buffered saline (PBS) and injected at a dose of 200 μg per mouse in a final volume of 100 μL. Control mice received an equal volume (100 μL) of species and isotype-matched control IgG diluted in PBS at the same concentration. Injections were performed every 3 days over a period of 12 weeks, ensuring consistent timing throughout the study. All solutions were freshly prepared before each injection and stored on ice until use.

Cell culture

AML12, LO2, and HepG2 cells were cultured in growth medium: High-glucose Dulbecco’s modified Eagle’s medium (Invitrogen, Grand Island, NY, United States) supplemented with 10% heat-inactivated fetal bovine serum (FBS; United States Biotechnologies), 100 U/mL penicillin, and 100 U/mL streptomycin (Invitrogen) and incubated at 37 °C with 50 mL/L CO₂. Cells were passaged weekly after trypsin-ethylenediaminetetraacetic acid detachment. All studies were performed on cells at passages 20-25.

In vitro administration of FGF21-neutralizing antibody

To assess the direct effects of FGF21 inhibition in cultured cells, an FGF21-neutralizing antibody (Biogradetech, BGT-ANT-37337) was added to the culture medium of AML12 hepatocytes in both monoculture and co-culture systems with 3T3-L1 adipocytes. The antibody was diluted in complete culture medium to a final concentration of 10 μg/mL.

Plasmid construction and amplification

The S100A11 gene was synthesized according to the designed sequence and amplified by polymerase chain reaction (PCR) with the following primers: Forward: 5’-GATCCCAGCTAGATTTCTCAGAATTTCTCTGTGCTTGAGAAATTCTGAGAAATCTAGCTGTTTTTA-3’; reverse: 5’-AGCTTAAAAACAGCTAGATTTCTCAGAATTTCTCAAGCACAGAGAAATTCTGAGAAATCTAGCTGG-3’. PCR was performed under the following conditions: Initial denaturation at 95 °C for 10 minutes; 40 cycles of denaturation at 95 °C for 30 seconds, annealing at 50 °C for 30 seconds, and extension at 72 °C for 30 seconds. The reaction mixture consisted of 2 μL dNTPs, 2 μL 10 × buffer, 0.4 μL Taq DNA polymerase, 1 μL of each primer, and 13.6 μL nuclease-free water.

The PCR product was ligated into the target vector using T4 DNA ligase (1 μL ligase, 1 μL 10 × buffer, 1 μL T-DNA, 7 μL DNA-PolyA) at 4 °C overnight. Ampicillin (100 mg/mL stock) was prepared in sterile saline and stored at -20 °C. Luria-Bertani (LB) medium was prepared with 20 g tryptone, 10 g yeast extract, and 20 g NaCl per liter; for solid medium, 20 g agar was added. After autoclaving, ampicillin was added to a final concentration of 50 μg/mL for plates or 100 μg/mL for liquid culture.

The ligation product was transformed into chemically competent E. coli DH5α cells. Briefly, 1 ng DNA was mixed with 100 μL competent cells on ice for 30 minutes, heat-shocked at 42 °C for 90 seconds, and returned to ice for 2-3 minutes. Cells recovered in 900 μL antibiotic-free LB medium at 37 °C with shaking (150 rpm) for 45 minutes, then plated on ampicillin-containing LB agar. After overnight incubation at 37 °C, single colonies were selected and cultured in LB liquid medium with 100 μg/mL ampicillin at 37 °C with shaking (180 rpm) for 12 hours. Plasmid DNA was extracted using a commercial column-based kit according to the manufacturer's instructions. The extracted plasmid was verified by Sanger sequencing and stored at -80 °C in 50% glycerol.

For transfection, cells were seeded in 6-well plates at 2-7 × 10⁵ cells per well and cultured until 70%-80% confluency. Prior to transfection, the medium was replaced with 2 mL fresh complete medium. For each well, 2.5 μg plasmid DNA and 4 μL Lipo8000™ transfection reagent was diluted in 125 μL serum and antibiotic-free Dulbecco’s modified Eagle’s medium, mixed gently, and incubated for 5 minutes at room temperature. The mixture was added dropwise to the cells, and the plate was gently rocked to ensure even distribution. Transfected cells were cultured for 24-48 hours before further analysis.

Metabolic cage experiments

Prior to the experiment, mice were acclimatized to the metabolic cages for a minimum of 48 hours to minimize stress-induced variability in metabolic parameters, with ad libitum access to food and water provided throughout this period. Metabolic cages (CLAMS, Columbus Instruments, United States) were calibrated according to the manufacturer’s instructions to ensure accurate measurement of oxygen consumption, carbon dioxide production, food intake, water consumption, and locomotor activity, with the cages maintained in a controlled environment featuring a 12-hour light/dark cycle, constant temperature (22 ± 1 °C), and humidity (50% ± 10%). Following acclimatization, baseline metabolic parameters were recorded over a 24-hour period to establish individual reference values for each mouse, including oxygen consumption, carbon dioxide production, respiratory quotient, energy expenditure, food and water intake, and spontaneous activity.

Magnetic resonance imaging

For body composition analysis using nuclear magnetic resonance (NMR), mice were sequentially numbered and weighed, and then individually placed into the NMR machine (TRIO, Siemens, Munich, Germany). The corresponding body weight reading of each mouse was entered into the system prior to measurement. The NMR analysis provided data on fat mass and lean mass, which were statistically analyzed to calculate the ratios of fat mass to body weight and lean mass to body weight, respectively. This method allowed for precise and non-invasive quantification of body composition in treated mice.

RNA extraction and quantitative real-time PCR

Tissue samples, either freshly harvested or flash-frozen in liquid nitrogen and stored at -80 °C, were homogenized in liquid nitrogen using a pre-chilled mortar and pestle. The resulting powder was immediately lysed in TRIzol reagent with thorough vortexing. Following a brief incubation at room temperature, chloroform was added, and the mixture was vigorously shaken before centrifugation at 12000 g for 15 minutes at 4 °C. The upper aqueous phase was carefully transferred to a fresh tube, and RNA was precipitated by adding an equal volume of isopropanol. After incubation at room temperature for 10 minutes, the sample was centrifuged at 12000 g for 10 minutes at 4 °C. The RNA pellet was washed twice with 75% ethanol, air-dried, and finally dissolved in RNase-free water. RNA concentration and purity were determined spectrophotometrically, ensuring an A260/A280 ratio between 1.8 and 2.0. RNA integrity was assessed via 1% agarose gel electrophoresis to visualize distinct 28S and 18S ribosomal RNA bands. For cDNA synthesis, 1 μg total RNA was reverse-transcribed using a commercial reverse transcription kit with oligo primers according to the manufacturer’s instructions. The reaction was performed at 25 °C for 10 minutes, 42 °C for 50 minutes, and terminated at 70 °C for 15 minutes. Quantitative PCR was performed using a SYBR Green master mix on a real-time PCR system. Each 20 μL reaction contained cDNA template, specific forward and reverse primers for target and reference genes, and the master mix. The thermal cycling conditions were as follows: Initial denaturation at 95 °C for 30 seconds, followed by 40 cycles of 95 °C for 5 seconds and 60 °C for 30 seconds. A melt curve analysis was conducted from 60 °C to 95 °C to verify amplification specificity. All reactions were performed in triplicate, and no-template controls were included. Relative gene expression levels were calculated using the 2^(-ΔΔCt) method with normalization to a stable housekeeping gene. Primer sequences used in this study are listed in detail in Supplementary Table 1.

Western blotting

Freshly isolated or snapfrozen liver tissues were homogenized on ice in RIPA lysis buffer supplemented with protease and phosphatase inhibitors. The homogenate was incubated on ice for 30 minutes and then centrifuged at 12000 g for 15 minutes at 4 °C. The supernatant was collected, and protein concentration was determined using a bicinchoninic acid protein assay kit. Protein samples were denatured by boiling in sodium dodecyl sulfate loading buffer at 95 °C for 10 minutes. Equal amounts of protein (40 µg per lane) were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis using 8%-12% gels. After electrophoresis, proteins were transferred onto PVDF membranes using a wet or semi-dry transfer system. The membranes were blocked with 5% nonfat milk or bovine serum albumin in tris-buffered saline with tween 20 (TBST) for 1 hour at room temperature and then incubated overnight at 4 °C with primary antibodies diluted in blocking solution. Following three washes with (TBST), membranes were incubated with horseradish peroxidaseconjugated secondary antibodies for 1 hour at room temperature. Protein bands were visualized with an enhanced chemiluminescence substrate, and images were acquired using a chemiluminescence imaging system. Band intensities were quantified with ImageJ software.

Oil Red O staining

Freshly collected mouse liver tissue was fixed in 4% paraformaldehyde for 24 hours, followed by routine paraffin embedding and sectioning at a thickness of 5-8 μm. Alternatively, frozen tissues were embedded in optimal cutting temperature compound and sectioned at 8-10 μm thickness using a cryostat. Prior to staining, paraffin sections were deparaffinized in xylene and rehydrated through a graded ethanol series, while frozen sections were taken directly from -80 °C, equilibrated to room temperature, post-fixed in 4% paraformaldehyde for 10 minutes, and then rinsed with distilled water. The Oil Red O working solution was prepared by dissolving 0.5 g Oil Red O powder in 100 mL isopropyl alcohol to make a stock solution. Immediately before use, 6 mL of the stock solution was mixed with 4 mL of distilled water, allowed to stand for 10 minutes, and filtered. Sections were immersed in the freshly prepared Oil Red O working solution and stained at room temperature, protected from light, for 10-15 minutes. They were then differentiated in 60% isopropyl alcohol for a few seconds to remove nonspecific background staining and gently rinsed with distilled water. After dehydration through a graded ethanol series and clearing in xylene, the sections were mounted with neutral mounting medium. Stained sections were observed and photographed under a light microscope, with lipid droplets appearing bright red. Quantitative analysis of the stained lipid droplet area was performed using image analysis software ImageJ.

Hepatic lipid extraction

Lipids were extracted from liver tissues using a modified chloroform/methanol extraction method. Approximately 20 mg of liver tissue was accurately weighed and homogenized in 1 mL of chloroform/methanol (2:1, volume/volume) mixture using a high-speed homogenizer on ice (three cycles of 30 seconds each). The homogenates were incubated in a chromatography cabinet at 4 °C for at least 18 hours. Subsequently, 200 μL of deionized water was added to each sample, followed by vigorous vortex mixing. The mixture was centrifuged at 3000 rpm for 10 minutes at 4 °C to separate phases. The lower lipid-containing organic phase was carefully collected using a syringe and transferred into a 1.5 mL microcentrifuge tube. The organic solvent was evaporated under a gentle stream of nitrogen gas. The dried lipid residue was reconstituted in PBS containing 5% Triton X-100 and incubated at 55 °C for approximately 1.5 hours, with intermittent vortex mixing every 30 seconds. The lipid content was subsequently quantified using commercial cholesterol (Applygen, Cat. No. E1005) and TG assay kits (Beyotime, Cat. No. S0219FT) according to the manufacturer’s instructions.

Proteomics and analysis

Fresh liver tissue (n = 4) was obtained and frozen in liquid nitrogen, followed by grinding with a pestle. Subsequently, 120 μL protein lysis buffer and protease inhibitor cocktail (1:100 dilution) were added. The mixture was placed on ice and ground with a pestle. After centrifugation at 13000 rpm, 4 °C for 30 minutes, the supernatant was collected to extract total liver tissue protein. The protein lysis buffer was prepared by dissolving 10% trichloroacetic acid and 0.07% β-mercaptoethanol in acetone, with 2.7 g urea and 0.2 g 3-[(3-cholamidopropyl) dimethylammonio]-1-propanesulfonate dissolved in 3 mL sterile, deionized water (final volume: 5 mL). Before use, 65 μL/mL of 1 M dithiothreitol was added. Protein quantification was performed using the Bradford method, and protein concentration was calculated by subtracting the absorbance at 488 nm from that at 595 nm. Proteomics analysis was conducted at the Proteomics Core Facility of Peking University Health Science Center. Mass spectrometry raw data were processed and analyzed using the Peaks-Online software for database searching. The heatmap depicts genes differentially expressed according to the threshold of |fold change| ≥ 1.5 with a false discovery rate-adjusted P value < 0.05 (Benjamini-Hochberg method).

Quantification of hepatic extracellular signal-regulated kinase 1/2 phosphorylation by enzyme-linked immunosorbent assay

The levels of phosphorylated and total extracellular signal-regulated kinase (ERK) 1/2 in mouse liver tissues were quantified using the ab126445 enzyme-linked immunosorbent assay kit. Liver tissues were rinsed with ice-cold PBS to remove residual blood and homogenized in 1 × cell lysis buffer containing protease and phosphatase inhibitors at a ratio of 1 mL buffer per 100 mg tissue. The homogenates were lysed with shaking at 2-8 °C for 30 minutes, followed by centrifugation at 13000 rpm for 10 minutes at 4 °C; the resulting supernatants were collected as tissue protein lysates. Prior to analysis, samples were appropriately diluted with 1 × assay diluent (a 50-fold dilution was determined optimal based on preliminary experiments). According to the manufacturer’s protocol, microwells pre-coated with anti-total ERK1/2 antibody were loaded with 100 μL of sample or positive control per well and incubated at room temperature for 2.5 hours. After washing, 100 μL of diluted rabbit anti-phospho-ERK1 (T202/Y204)/ERK2 (T185/Y187) antibody or rabbit anti-total ERK1/2 antibody was added to each well, followed by incubation with shaking at room temperature for 1 hour. Following another wash step, 100 μL of 1 × horseradish peroxidase-conjugated goat anti-rabbit IgG secondary antibody was added and incubated for 1 hour at room temperature. After a final wash, 100 μL of TMB One-Step Substrate Reagent was added per well and incubated in the dark with shaking for 30 minutes. The reaction was terminated by adding 50 μL of Stop Solution, and the absorbance was measured immediately at 450 nm. Phosphorylated ERK1/2 levels were expressed as the ratio of phosphorylated ERK (p-ERK) 1/2 to total ERK1/2. All samples were assayed in duplicate.

Statistical analysis

Data were expressed as means ± SE and were analyzed using Graph Pad Prism software. One-way analysis of variance, Student-Newman-Keul’s test (comparison between multiple groups), or unpaired Student’s t test (between two groups) was used as appropriate. P < 0.05 denotes statistical significance.

RESULTS

General indexes of S100A11-hTg mice

Five-week-old S100A11-hTg male mice and littermate control mice were randomly divided into NCD and 60% HFD groups for 12 weeks. To validate the S100A11 transgenic mouse model, genotyping of tail DNA confirmed the presence of the transgene (Supplementary Figure 1A), and real-time PCR and western blot analyses demonstrated significant upregulation of S100A11 mRNA and protein levels in the liver, respectively (Supplementary Figure 1B and C). After HFD, the body weight of mice increased significantly, fat mass increased, and lean mass decreased, suggesting that HFD induced obesity in mice (Figure 1). Under NCD condition, there was no difference between control and S100A11-hTg mice, but under HFD, the body weight of S100A11-hTg mice significantly increased (Figure 1A and B), fat mass increased (Figure 1C), and lean mass decreased (Figure 1D) compared with control mice. Between NCD and HFD conditions, the energy consumption of S100A11-hTg mice was reduced (Figure 1E and F), but there was no significant change in activity level or food intake (Figure 1G and H, Supplementary Figure 1D and E). Between NCD and HFD, there was no significant change in body weight of female S100A11-hTg mice (Supplementary Figure 2A and B). The above results indicate that S100A11 gene overexpression in the liver can increase body weight and body fat content in male mice.

Figure 1 General indexes of S100A11-hTg mice. All data are presented as mean ± SE. n = 6 mice/group. Statistical analysis was performed using one-way analysis of variance. P values were adjusted for multiple testing using the Bonferroni correction: ^aP < 0.05 vs normal-chow diet (NCD) wild type, ^bP < 0.05 vs high-fat (HFD) diet wild type. Five-week-old male S100A11-hTg mice and littermate control mice were fed NCD or HFD for 12 weeks. A: General picture of mice at 17 weeks of age, scale bar: 1 cm; B: Body weight of mice at 17 weeks of age; C: Fat mass as a percentage of body weight; D: Lean mass as a percentage of body weight; E: Energy consumption of mice under NCD; F: Energy consumption of mice under HFD; G: Activity of mice under HFD; H: Food intake of mice under HFD. NCD: Normal-chow diet; HFD: High-fat diet; WT: Wild type.

HFD increases hepatic TG concentration in male S100A11-hTg mice

Five-week-old male S100A11-hTg mice and littermate control mice were randomly divided into NCD and 60% HFD for 12 weeks (Figure 2). The liver TG concentration of male S100A11-hTg mice increased on HFD (Figure 2A, B, and G), while S100A11 overexpression had the opposite effect on female mice (Supplementary Figure 3). HFD upregulated mRNA expression levels of liver lipid-synthesis-related genes in male S100A11-hTg mice, including Acc, Dgat1, Dgat2, and Fasn (Figure 2C), and lipid-uptake-related genes, such as Cd36 and Fabp3 (Figure 2D). In contrast, lipid-oxidation-related genes, such as Cpt1α, were downregulated (Figure 2E). Cholesterol-metabolism-related gene expression did not change (Figure 2F).

Figure 2 High-fat diet increased liver triglyceride levels in male S100A11-hTg mice. All data are presented as mean ± SE. n = 5-6 mice/group. Statistical analysis was performed using one-way analysis of variance, P values were adjusted for multiple testing using the Bonferroni correction: ^aP < 0.05 vs normal-chow diet wild type, ^bP < 0.05 vs high-fat diet wild type. Five-week-old male S100A11-hTg mice and littermate control mice were fed normal-chow diet or high-fat diet for 12 weeks. A: General liver image of mice at 17 weeks of age, scale bar: 1 cm; B: Triglyceride concentration in the liver of mice at 17 weeks of age; C: Real-time polymerase chain reaction (PCR) was used to detect expression of hepatic lipid synthesis-related genes; D: Real-time PCR was used to detect expression of hepatic lipid uptake-related genes; E: Real-time PCR was used to detect expression of hepatic lipid oxidation-related genes; F: Real-time PCR was used to detect expression of hepatic cholesterol synthesis-related genes; G: Oil Red O staining of mouse liver and quantification of positive areas, scale bar: 100 μm. NCD: Normal-chow diet; HFD: High-fat diet; WT: Wild type; ACC: Acetyl-CoA carboxylase; DGAT: Diacylglycerol acyltransferase; FASN: Fatty acid synthase; PPAR: Peroxisome proliferator-activated receptor; Cpt1α: Carnitine palmitoyltransferase 1α; MCADH: Medium-chain acyl-CoA dehydrogenase; HMGCR: 3-hydroxy-3-methyl-glutaryl-CoA reductase; SREBP2: Sterol regulatory element-binding protein 2; HMGCS1: 3-hydroxy-3-methylglutaryl-CoA synthase 1.

Intervention of S100A11 gene expression in isolated hepatocytes affects lipid accumulation

To confirm the effect of S100A11 on hepatic lipid metabolism, we overexpressed or knocked down the S100A11 gene in the mouse-derived normal liver AML12 cell line to observe changes in cell lipid metabolism. AML12 cells were cultured for 3 days and infected with recombinant adenoviruses encoding (AD) green fluorescent protein (GFP) or AD-S100A11. After 48 hours, green fluorescence was observed under a fluorescent microscope, indicating virus infection (Figure 3A). Compared with the control group, S100a11 mRNA expression significantly increased (Figure 3B). S100a11-overexpressing AML12 cells were treated with oleic acid at 0 nmol/L, 62.5 nmol/L or 125 nmol/L for 36 hours. The TG concentration (Figure 3C) and cholesterol levels significantly increased (Figure 3D). Expression of fatty acid synthesis-related genes Srebp1, Acc, and Dgat2 significantly increased, and Fasn also showed an upward trend (Figure 3E). Expression of genes related to fatty acid uptake, transport, and activation, such as Fabp3 and Cd36, significantly decreased (Figure 3F). Genes related to fatty acid β-oxidation such as Cpt1α were significantly downregulated (Figure 3G). There was no significant change in genes related to cholesterol synthesis (Figure 3H). The above results suggest that S100A11 overexpression inhibits fatty acid β-oxidation in AML12 cells and promotes fatty acid synthesis, thereby promoting lipid deposition.

Figure 3 S100A11 gene overexpression of in vitro promotes lipid accumulation in AML12 hepatocyte cell line. All data are presented as mean ± SE. n = 3-5 biological repeats. Statistical analysis was performed using unpaired two-tailed Student’s t test: ^aP < 0.05 vs recombinant adenoviruses encoding (AD) green fluorescent protein (GFP). A: AML12 cells infected with AD-S100A11 or AD-GFP as a control and observed under a fluorescent microscope after 48 hours, scale bar: 100 μm; B: Real-time polymerase chain reaction was used to detect expression of S100A11 mRNA; C: AML12 cells were infected with AD-S100A11 or AD-GFP as a control and treated with 0 nmol/L, 62.5 nmol/L, or 125 nmol/L oleic acid for 36 hours to detect triglyceride concentration; D: AML12 cells were infected with AD-S100A11 or AD-GFP as a control and treated with 0 nmol/L, 62.5 nmol/L, or 125 nmol/L oleic acid for 36 hours to detect cholesterol concentration; E-H: Real-time polymerase chain reaction was used to detect expression of liver lipid metabolism. AD-GFP: Recombinant adenoviruses encoding green fluorescent protein; AD-SA/AD-S100A11: Recombinant adenoviruses encoding S100A11; OA: Oleic acid; ACC: Acetyl-CoA carboxylase; DGAT: Diacylglycerol acyltransferase; FASN: Fatty acid synthase; PPAR: Peroxisome proliferator-activated receptor; Cpt1α: Carnitine palmitoyltransferase 1α; MCADH: Medium-chain acyl-CoA dehydrogenase; HMGCR: 3-hydroxy-3-methyl-glutaryl-CoA reductase; SREBP2: Sterol regulatory element-binding protein 2; HMGCS1: 3-hydroxy-3-methylglutaryl-CoA synthase 1.

To confirm the effect of S100A11 on hepatic lipid metabolism, we knocked down S100A11 expression of in liver cells and observed changes in lipid metabolism. After 48 hours, green fluorescence was observed under a fluorescent microscope, indicating efficient shRNA delivery into the cells (Supplementary Figure 4A). Compared with the control group, S100a11 mRNA expression significantly decreased (Supplementary Figure 4B). S100A11-knockdown AML12 cells were treated with oleic acid at 0 nmol/L, 62.5 nmol/L and 125 nmol/L for 36 hours. The concentration of TGs and cholesterol significantly decreased (Supplementary Figure 4C and D). We next measured the expression of markers related to hepatic lipid metabolism. Expression of the fatty acid synthesis-related gene Fasn was significantly downregulated, and Srebp1, Acc, Dgat1, and Dgat2 also showed a decreasing trend (Supplementary Figure 4E). Genes related to fatty acid uptake, transport, and activation such as Fabp3 were significantly downregulated (Supplementary Figure 4F-H). The above results suggest that S100A11 knockdown inhibits fatty acid synthesis and uptake by liver cells and inhibits lipid deposition.

S100A11 overexpression in the liver promotes lipid accumulation in adipose tissue

After HFD, the adipose tissue volume (Figure 4A-C) and weight (Figure 4D) increased in S100A11-hTg mice. Expression of BAT-related genes, such as Ucp1 and Pgc1α, was significantly downregulated in the white adipose tissue (WAT) of the epididymis of S100A11-hTg mice, while that of WAT-related genes, such as Ednra and Psat, was significantly upregulated (Figure 4E). In BAT, expression of related genes, such as Ucp1 and Pgc1α, was significantly downregulated, while expression of WAT-related genes, such as Ednra and Adiponectin, was significantly upregulated (Figure 4F).

Figure 4 Adipose tissue and expression of fat-related gene mRNA in different tissues of male S100A11-hTg mice. All data are presented as mean ± SE. n = 6 mice/group. Statistical analysis was performed using one-way analysis of variance, P values were adjusted for multiple testing using the Bonferroni correction: ^aP < 0.05 vs normal-chow diet wild type, ^bP < 0.05 vs high-fat diet wild type. Five-week-old male S100A11-hTg mice and control mice were fed with normal-chow diet or high-fat diet for 12 weeks. A: General picture of subcutaneous fat at 17 weeks of age; B: General picture of perirenal fat at 17 weeks of age; C: General picture of epididymal fat at 17 weeks of age; D: At 17 weeks, the weight of adipose tissue in each region of the mice; E: Real-time polymerase chain reaction was used to detect expression of brown fat-related genes (left) and white fat-related genes (right) in white adipose tissue of the epididymis. F: Real time polymerase chain reaction was used to detect the expression level of brown fat-related genes (left) and white fat-related genes (right) in brown adipose tissue. Scale bar: 1 cm. NCD: Normal-chow diet; HFD: High-fat diet; WT: Wild type; sWAT: Subcutaneous white adipose tissue; eWAT: Epididymal white adipose tissue; rWAT: Retroperitoneal white adipose tissue; BAT: Brown adipose tissue; UCP1: Uncoupling protein 1; PGC1α: Peroxisome proliferator-activated receptor gamma coactivator 1-α; PRDM16: PR domain containing 16; PAI-1: Plasminogen activator inhibitor-1.

To verify the role of S100A11, we infected brown adipocytes with AD-S100A11 and examined lipid accumulation. After 48 hours of virus infection, the overexpression and control groups both showed green fluorescence, confirming successful infection. Oil Red O staining showed that several fat droplets of different sizes were dispersed in brown fat cells, while the control group had a low staining positive rate (Supplementary Figure 5A). Six days after differentiation, S100a11 expression was significantly upregulated, as shown by real-time PCR (Supplementary Figure 5B). We also found that BAT- and WAT-related gene expression was increased upon S100a11 overexpression (Supplementary Figure 5C and D). These results suggest that S100A11 overexpression can promote lipid accumulation in brown adipocytes.

FGF21 plays a crucial role in promoting S100A11-mediated lipogenesis

To investigate the mechanisms underlying S100A11-induced lipid accumulation, we conducted proteomic profiling of liver tissues from HFD-fed mice, followed by Kyoto Encyclopedia of Genes and Genomes analysis. We identified 159 differentially expressed proteins (fold change > 1.2, P < 0.05) (Figure 5A). Subsequent analysis focused on the top 10 enriched processes, with significant P values highlighted (0.01 in red, 0.05 in blue) (Figure 5B). Enrichment analysis revealed significant associations with metabolic pathways, particularly fatty acid degradation. Additionally, the PPAR signaling pathway exhibited the most significant enrichment (Figure 5C, verification details in Supplementary Figure 6). We analyzed the genes enriched in the PPAR signaling pathway (Figure 5D), among which, FGF21 exhibited the most pronounced increase in the S100A11-hTg group (details provided in Supplementary Table 2). Previous studies have confirmed that FGF21 is a key factor in liver–fat signal communication. FGF21 acts upstream of PPAR signaling by enhancing PPARγ transcriptional activity through β-klotho-dependent receptor activation, thereby amplifying metabolic responses[22]. We propose that FGF21 may serve as a key mediator in S100A11-induced lipid accumulation.

Figure 5 Fibroblast growth factor 21 may play a crucial role in promoting S100A11-mediated lipogenesis. Proteomic analysis of liver tissues from high-fat diet-fed male mice. A: Heatmap of differentially expressed genes according to the threshold of |fold change| ≥ 1.5 with a false discovery rate-adjusted P < 0.05 (Benjamini-Hochberg method); B: Distribution of enriched Kyoto Encyclopedia of Genes And Genomes pathway; C: Class of enriched Kyoto Encyclopedia of Genes And Genomes pathway; D: Gene heatmap enriched in the peroxisome proliferator-activated receptor signaling pathway. n = 4 mice/group. WT: Wild type.

Role of FGF21 in S100A11-mediated lipid accumulation

To clarify whether FGF21 participates in S100A11-mediated promotion of lipid metabolism and lipid accumulation, we conducted the following experiments. Five-week-old male S100A11-hTg mice and littermate control mice were randomly divided into NCD and 60% HFD groups and fed for 12 weeks. We examined the plasma FGF21 levels and Fgf21 expression in various tissues. Under NCD and HFD conditions (Figure 6), FGF21 plasma levels of S100A11-hTg mice significantly increased (Figure 6A). Under HFD, Fgf21 mRNA expression in the liver and epididymis WAT of S100A11-hTg mice was significantly increased (Figure 6B and D), and expression of Fgf21 mRNA in BAT also increased (Figure 6C). Despite significantly increased plasma FGF21 levels in S100A11-hTg mice, hepatic activation of its key downstream effector, p-ERK1/2, was markedly attenuated. Quantitative enzyme-linked immunosorbent assay further confirmed a significant reduction in the p-ERK/total ERK ratio (Supplementary Figure 7). Together, these results demonstrate impaired activation of the FGF21 signaling cascade in the liver, indicating the presence of pronounced FGF21 resistance in this model.

Figure 6 Fibroblast growth factor 21 levels in tissues and plasma of male S100A11-hTg mice. All data are presented as means ± SE. n = 5 or 6 mice/group. Statistical analysis was performed using one-way analysis of variance, P values were adjusted for multiple testing using the Bonferroni correction: ^aP < 0.05 vs normal-chow diet wild type, ^bP < 0.05 vs high-fat diet wild type. Five-week-old male S100A11-hTg mice and littermate control mice were fed normal-chow diet or high-fat diet for 12 weeks. A: Plasma fibroblast growth factor 21 (FGF21) level at 17 weeks of age; B: Liver FGF21 mRNA expression level at 17 weeks of age; C: Fgf21 mRNA expression level in brown adipose tissue at 17 weeks of age; D: Fgf21 mRNA expression level in epididymal white adipose tissue of mice at 17 weeks of age; E: Protein interaction network of human S100A11. Protein-protein interaction network of human S100A11 generated by the STRING database (Homo sapiens). High-confidence interactors are shown (minimum required interaction score: 0.700). Key interaction partners are labeled, including members of the S100 protein family, annexin family proteins, the receptor for advanced glycation end products, and the multifunctional nucleolar protein nucleolin. FGF21: Fibroblast growth factor 21; NCD: Normal-chow diet; HFD: High-fat diet; WT: Wild type.

To investigate the human-relevant mechanisms by which S100A11 upregulates FGF21, we systematically analyzed the Homo sapiens S100A11 protein interactome using the STRING database. This analysis revealed evolutionarily conserved interactions with key partners that suggest two plausible regulatory pathways (Figure 6E). Human S100A11 shows a strong association with nucleolin, a multifunctional nuclear protein involved in transcription regulation, ribosome biogenesis, and RNA metabolism[23]. This interaction suggests that S100A11 may modulate the transcriptional activity of target genes, including FGF21[24], through functional interplay with nuclear regulatory complexes. The analysis also suggests a well-established interaction between S100A11 and the receptor for advanced glycation end products (RAGE, encoded by AGER)[25]. In human cells, S100 protein binding to RAGE classically activates downstream signaling cascades, including nuclear factor kappa B and mitogen-activated protein kinase pathways[25]. This supports a model in which S100A11 acts as a signaling ligand, activating intracellular kinase cascades via membrane-bound RAGE to ultimately regulate FGF21 transcription. In summary, S100A11 may upregulate FGF21 through dual, potentially convergent mechanisms: Receptor-mediated signal transduction and nuclear transcriptional co-regulation. These findings provide specific, human-relevant targets for future mechanistic validation.

FGF21 neutralization partially reverses S100A11-induced lipid accumulation in vivo

To further elucidate the role of FGF21 in S100A11-induced hepatic and systemic fat accumulation, we administered an FGF21-neutralizing antibody to S100A11-hTg mice (Figure 7A). Neutralization of FGF21 partially reversed hepatic steatosis and WAT lipid accumulation induced by S100A11 overexpression. Specifically, we observed a significant reduction in body weight and fat mass (Figure 7B and C), along with decreased TG content in the liver (Figure 7D).

Figure 7 Fibroblast growth factor 21 neutralization partially reverses S100A11-induced lipid accumulation in vivo. All data are presented as mean ± SE. n = 6 biological replicates. Statistical analysis was performed using unpaired two-tailed Student’s t-test: ^aP < 0.05 vs wild type_IgG, ^bP < 0.05 vs transgenic _IgG. Five-week-old male wild type and transgenic mice were fed a high-fat diet for 12 weeks, during which they were injected with either a fibroblast growth factor 21-neutralizing antibody or its isotype control every 3 days. A: Schematic diagram of the experimental design; B: Body weight; C: Body fat percentage; D: Hepatic triglyceride content. FGF21: Fibroblast growth factor 21; WT: Wild type; BW: Body weight.

S100A11 gene overexpression in hepatocytes promotes lipid accumulation in adipocytes

To clarify whether FGF21 is a key factor in the network regulation of liver and adipose tissue, we cocultured AML12 cells with 3T3-L1 cells. When 3T3-L1 cells were approximately 70% confluent, we induced adipogenic differentiation, and AML12 cells were treated with a 106 AD-S100A11 titer or a AD-GFP titer (control). Compared with the control group, S100a11 mRNA expression significantly increased in AML12 cells infected with AD-S100A11 (Figure 8A), and Fgf21 mRNA expression also increased (Figure 8B). The level of FGF21 in the culture supernatant was significantly higher than that of the control group (Figure 8C). At the same time, compared with the control group, 3T3-L1 cells cocultured with AD-S100A11-transfected AML12 cells had significantly higher Fgf21 mRNA expression (Figure 8D). Expression of the WAT-related gene Prdm16 was significantly downregulated, while expression of WAT-related genes such as Ednra and Resistin was significantly upregulated (Figure 8E and F). The above results indicate that FGF21 is involved in the network regulation process of liver S100A11 expression in adipose tissue.

Figure 8 S100A11 overexpression in hepatocytes promotes lipid accumulation in adipocytes. All data are presented as mean ± SE. n = 5 or 6 biological repeats. Statistical analysis was performed using unpaired two-tailed Student’s t test: ^aP < 0.05 vs recombinant adenoviruses encoding green fluorescent protein. Mouse-derived normal liver AML12 cells were cocultured with preadipocyte 3T3-L1 cells in vitro. When 3T3-L1 cells were approximately 70% confluent, adipogenic differentiation was performed, and recombinant adenoviruses encoding S100A11 were transfected into hepatocytes. Real-time polymerase chain reaction was used to detect related mRNA levels of AML12 cells and 3T3-L1 cells after 72 hours. A: MRNA level of S100A11 in AML12 cells; B: MRNA level of fibroblast growth factor 21 (FGF21) in AML12 cells; C: FGF21 level in AML12 cell culture supernatant; D: MRNA level of Fgf21 in 3T3-L1 cells; E and F: Expression levels of genes related to brown adipose tissue and white adipose tissue in 3T3-L1 cells. FGF21: Fibroblast growth factor 21; AD-GFP: Recombinant adenoviruses encoding green fluorescent protein; AD-SA/AD-S100A11: Recombinant adenoviruses encoding S100A11; UCP1: Uncoupling protein 1; PGC1α: Peroxisome proliferator-activated receptor gamma coactivator 1-α; PRDM16: PR domain containing 16.

FGF21 neutralization partially reverses S100A11-induced lipid accumulation in vitro

To further validate the role of FGF21 in S100A11-induced lipid accumulation in vitro, we employed an FGF21-neutralizing antibody in both monocultures of AML12 cells and in co-cultures with 3T3-L1 cells. In AML12 hepatocytes overexpressing S100A11, co-treatment with the FGF21-neutralizing antibody markedly attenuated intracellular (Figure 9A and B) and 3T3-L1 (Figure 9C and D) lipid accumulation. This further supports that S100A11-promoted lipid deposition requires the presence of FGF21.

Figure 9 Fibroblast growth factor 21 neutralization partially reverses S100A11-induced lipid accumulation in vitro. All data are presented as mean ± SE. n = 6 biological replicates. Statistical analysis was performed using unpaired two-tailed Student’s t-test: ^aP < 0.05 vs recombinant adenoviruses encoding green fluorescent protein, ^bP < 0.05 vs recombinant adenoviruses encoding S100A11_IgG. A: Triglyceride content in AML12; B: Cholesterol content in AML12; C and D: Expression levels of brown adipose tissue and white adipose tissue related genes in 3T3-L1 cells. FGF21: Fibroblast growth factor 21; AD-GFP: Recombinant adenoviruses encoding green fluorescent protein; AD-SA/AD-S100A11: Recombinant adenoviruses encoding S100A11; UCP1: Uncoupling protein 1; PGC1α: Peroxisome proliferator-activated receptor gamma coactivator 1-α; PRDM16: PR domain containing 16.

DISCUSSION

S100A11 is an EF-hand calcium-binding protein and a member of the S100 family. The gene encoding S100A11 is located on chromosomal region 1q21, an area prone to genomic rearrangements in cancer, which may lead to its dysregulated expression[26]. The S100 protein family is implicated in a wide range of cellular functions, including contact inhibition, epidermal differentiation, cellular senescence, apoptosis, inflammatory responses, and oncogenesis[27-29]. Accumulating studies have linked S100A11 to the pathogenesis of several malignancies, such as papillary thyroid carcinoma, colorectal cancer, pancreatic ductal adenocarcinoma, and breast cancer[30-32]. Beyond its role in cancer, GWASs suggest that S100A11 is associated with cholesterol metabolism, PPAR signaling, and diabetes, with its expression being modulated by therapeutic agents including rosiglitazone, pioglitazone, and metformin. Despite these connections, the specific impact of S100A11 on systemic metabolic homeostasis, particularly in obesity, glucose regulation, and lipid metabolism, remains poorly understood. Therefore, this study investigated the role of S100A11 in key metabolic tissues such as the liver and adipose tissue, and elucidated its mechanisms of action in lipid metabolism. By doing so, we sought to provide insights that may help prevent disorders of lipid metabolism and related metabolic diseases.

The liver plays a central role in systemic metabolism. Hepatic steatosis is formally defined when histological examination demonstrates fatty infiltration in more than one-third of hepatocytes or when hepatic fat content exceeds 5% of liver wet weight[33]. This condition is characterized by disrupted hepatic lipid homeostasis, leading to excessive TG deposition and diffuse fatty degeneration[34]. Hepatic lipid metabolism involves a tightly regulated balance between synthesis, uptake, oxidation, and export.

While previous studies have established the involvement of S100 proteins in metabolic pathways and identified S100A11 as a factor in cancer progression[27] and diabetes-related pathways[35] via GWASs, its direct functional role in hepatic lipid metabolism under obesogenic conditions remains largely undefined[10]. Our findings address this gap by demonstrating that S100A11 actively regulates hepatic lipid balance in vitro and in vivo. In contrast to earlier work that primarily associated S100A11 with tumorigenesis or systemic metabolic traits, our study mechanistically linked S100A11 overexpression to the transcriptional reprogramming of lipid metabolism genes, thereby promoting steatosis.

We found that although overall liver weight and volume remained similar to controls, significant alterations occurred in the expression of key lipid-metabolism-related genes in the liver. This suggests that while HFD and altered S100A11 expression did not markedly impair overall energy metabolism in male mice, the liver exhibited an early and pronounced metabolic adaptation. Specifically, S100A11 exacerbated HFD-induced obesity and promoted hepatic lipid accumulation in male mice, a finding that extended beyond the previously reported correlations with metabolic disease. Supporting this, we found that HFD significantly upregulated hepatic S100A11 expression, indicating that S100A11 acted as a diet-responsive regulator. Furthermore, liver-specific overexpression of S100A11 directly induced lipid deposition. Gene expression analysis via real-time PCR revealed upregulation of genes involved in lipid synthesis and uptake, alongside downregulation of those governing lipid oxidation. These transcriptional shifts provide a plausible molecular explanation for the lipid-accumulating phenotype and distinguish our work from earlier descriptive or associative studies.

Notably, this effect was not observed in female mice, a discrepancy potentially attributable to the modulating influence of sex hormones on metabolic pathways and experimental readouts. This sex-specific dimension adds a novel layer to the understanding of S100A11 in metabolism, suggesting that its role is context-dependent and hormonally regulated, an aspect not previously emphasized in the literature.

Existing literature offers several plausible explanations for this observation. First, differences in sex hormones represent the most direct mechanistic hypothesis. Estrogen, particularly signaling through estrogen receptor α, has been extensively demonstrated to exert potent protective and regulatory effects on hepatic metabolism and inflammatory responses[36]. We speculate that the basal levels of endogenous estrogen in female mice may already activate protective networks that overlap with or lie downstream of the S100A11/FGF21 pathway, thereby creating a “ceiling effect” that masks the additional impact of our experimental intervention within the observation window. In contrast, in male mice, the absence of comparably strong basal estrogen signaling allows S100A11 upregulation to more clearly drive an FGF21-dependent phenotypic outcome. Second, inherent sexual dimorphism in gene expression may also contribute. Multiple studies have reported that the expression of numerous metabolic genes in the liver, including Fgf21 and its upstream regulators such as PPARA, exhibits intrinsic sex differences, often programmed by differential growth hormone secretion patterns[37].

Although the present study did not replicate the male phenotype in female models, this finding carries significant biological and translational implications. It strongly suggests that therapeutic strategies targeting the S100A11/FGF21 pathway may exhibit differential efficacy between males and females. Future studies are essential to directly validate the causal role of sex hormones through experimental approaches such as castration in males and ovariectomy combined with hormone replacement in females. Furthermore, investigating whether S100A11 expression or function is directly transcriptionally regulated at the cellular level by estrogen or androgen receptors will help elucidate the molecular origins of this sex-specific disparity.

To further delineate the cell-autonomous role of S100A11 in hepatocytes, we utilized the AML12 mouse hepatocyte line. Consistent with our in vivo observations, S100A11 overexpression in AML12 cells promoted cellular lipid synthesis and TG accumulation, whereas S100A11 knockdown yielded the opposite effect, inhibiting lipogenesis and reducing TG content.

WAT functions as the body’s primary energy reservoir, storing excess energy in the form of TGs. In response to metabolic stimuli, some WAT depots, particularly visceral fat, exhibit plasticity, possessing the capacity to undergo “beiging”, wherein white adipocytes transdifferentiate toward a thermogenic phenotype resembling brown adipocytes, a process marked by the upregulation of BAT-related genes[38]. Conversely, BAT can undergo “whitening” under obesogenic conditions, losing its thermogenic properties and accumulating lipids, as demonstrated in studies of ob/ob mice[39]. This bidirectional adipocyte plasticity plays a critical role in systemic energy balance and metabolic health.

While existing literature has established involvement of S100A11 in cancer[40] and metabolic diseases[41], its role in adipose tissue plasticity and obesity progression remains largely unexplored. Our study advances the field by identifying S100A11 as a novel regulator of adipocyte phenotype switching in the context of HFD-induced obesity. We found that liver-specific overexpression of S100A11 in male mice exacerbated HFD-induced obesity and significantly increased the mass and volume of key WAT depots, including subcutaneous, epididymal, and perirenal fat. Importantly, gene expression analysis revealed consistent downregulation of BAT-related genes alongside upregulation of WAT-related genes in these adipose tissues, suggesting that S100A11 promotes a shift from a thermogenic, energy-dissipating phenotype toward a lipid-storing, white-like phenotype, i.e., the whitening of adipose tissue. To further investigate the mechanistic basis of this observation, we established a coculture system using mouse hepatocyte (AML12) and adipocyte (3T3-L1) lines. When 3T3-L1 adipocytes were cocultured with AML12 cells overexpressing S100A11, they exhibited a gene expression profile consistent with reduced browning and enhanced lipid storage, mirroring our in vivo findings. This result not only supports the notion that S100A11 promotes adipose whitening but also suggests that its effect may be mediated, at least in part, through hepatocyte-adipocyte crosstalk.

A dynamic, multilevel dialog exists between the liver and adipose tissue, positioning this axis as a central coordinator of systemic energy metabolism. This communication is largely mediated through the secretion of hepatokines and adipokines, which target peripheral and central organs to regulate metabolic homeostasis[42]. Among these signaling molecules, FGF21, a liver-derived endocrine factor, has emerged as a key metabolic regulator. Although primarily produced by the liver[43], FGF21 exerts significant effects on energy expenditure and glucose metabolism in adipose tissue[44]. Our findings reveal a novel layer within this hepatic–adipose axis by implicating S100A11 in the regulation of FGF21. In male mice with liver-specific S100A11 overexpression under conditions of energy excess, FGF21 levels significantly increased in the liver, BAT, WAT, and plasma. This systemic elevation of FGF21 suggests a potential link between S100A11 function and the promotion of lipid accumulation in liver and fat. To investigate this mechanistically, we used a coculture system of mouse AML12 hepatocytes and 3T3-L1 adipocytes. Consistent with the in vivo data, S100A11 overexpression in AML12 cells led to a marked increase in both Fgf21 mRNA expression and FGF21 protein secretion. Importantly, 3T3-L1 adipocytes cocultured with S100A11-overexpressing hepatocytes also exhibited elevated Fgf21 mRNA levels. Concomitantly, these adipocytes showed a gene expression signature indicative of suppressed thermogenesis and enhanced lipid storage.

While our study provides mechanistic insights into the role of the S100A11/FGF21 axis in hepatic lipid metabolism, several limitations should be acknowledged. First, the findings were primarily derived from mouse models. Although murine systems offer powerful genetic and experimental tools, intrinsic differences in metabolism, immune response, and disease progression between mice and humans necessitate caution when extrapolating these results directly to human pathophysiology. Second, our study lacked validation in human clinical samples or cell lines. Future studies correlating S100A11 and FGF21 levels with metabolic parameters in human cohorts will be crucial to establish the translational relevance of our findings. Finally, while we identified potential upstream signaling pathways, the precise molecular cascade linking S100A11 to FGF21 transcription warrants further analysis, including the identification of direct binding partners and downstream effectors in metabolically relevant cell types.

Future perspectives

To establish the clinical relevance of the S100A11/FGF21 axis, future studies will analyze human NAFLD/NASH cohorts by correlating hepatic and circulating S100A11 and FGF21 levels with histopathological and metabolic parameters, and by evaluating the prognostic value of baseline S100A11 in disease progression or therapeutic response. In parallel, preclinical studies will assess the therapeutic potential of this axis, including testing RAGE antagonists in S100A11-transgenic and diet-induced obese models to determine whether blocking upstream signaling mitigates FGF21-driven lipid accumulation, as well as exploring combination strategies targeting both S100A11/RAGE and FGF21 pathways. Furthermore, the mechanistic hypotheses generated here, particularly the S100A11-AGER-nuclear factor kappa B/mitogen-activated protein kinase and S100A11-nucleolin regulatory nodes, should be functionally validated in human hepatic cell systems through genetic and pharmacological interventions, aiming to clarify whether S100A11 operates primarily through secretory or intracellular modes in metabolic stress. Together, these efforts will help translate our mechanistic insights into clinically actionable strategies for metabolic diseases.

CONCLUSION

Our study shows that S100A11 promotes lipid synthesis in the liver and lipid accumulation in adipose tissue through FGF21 in male mice (Supplementary Figure 8). These results provide us with a new entry point to treat obesity and liver steatosis. S100A11 may become a new target for clinical treatment of obesity-related metabolic diseases.

ACKNOWLEDGEMENTS

We appreciate Professor Eugene Y Chen, Dr Ji Zhang and Dr Ji-Feng Zhang of University of Michigan for kindly providing the S100A11-hTg mice.