The human normal hepatocyte cell line THLE-2 was purchased from American Type Culture Collection (ATCC; cat. no. CRL-2706), and human hepatoma cells HepG2 were purchased from ATCC (cat. no. HB-8065) and preserved in the laboratory. All cell lines used in the present study were authenticated using short tandem repeat (STR) profiling (Genewiz, Inc.) within 6 months prior to experimentation, and the STR profiles of THLE-2 and HepG2 matched the reference databases (ATCC), confirming no cross-contamination. The THLE-2 cells were cultured in BEGM^™ Bronchial Epithelial Cell Growth Medium (cat. no. CC-3170; Lonza Group, Ltd.), which included 5 ng/ml EGF, 0.5 µg/ml hydrocortisone, 0.5 µg/ml epinephrine, 10 µg/ml transferrin, 5 µg/ml insulin, 0.1 ng/ml retinoic acid, 6.5 ng/ml triiodothyronine and 50 µg/ml gentamicin/amphotericin B. HepG2 cells were cultured in DMEM (cat. no. SH30243.01; HyClone^™; Cytiva) containing 10% fetal bovine serum (FBS; Gibco; cat. no. 10099-141; Thermo Fisher Scientific, Inc.) and 1% penicillin-streptomycin (cat. no. P1400; Beijing Solarbio Science & Technology Co., Ltd.), and both cell lines were routinely cultured in an incubator at 37˚C and 5% CO₂.

When the confluence of THLE-2 and HepG2 cells reached 70-80%, the medium was replaced with serum-free DMEM containing oleic acid (OA; cat. no. P1145; Sigma-Adrich; Merck KGaA) and palmitic acid (PA; cat. no. P5585; Sigma-Adrich; Merck KGaA). Various OA/PA combinations were tested (0.25/0.125, 0.5/0.25 and 1.0/0.5 mM) and the cells were induced for 24 h. The concentrations (0.5 mM OA + 0.25 mM PA) were selected because this combination induced optimal lipid droplet accumulation (as detected by BODIPY staining) without excessive cytotoxicity [viability >80% via Cell Counting Kit-8 (CCK-8) assay (Beyotime Institute of Biotechnology); data not shown]. For the CCK-8 assay, 10 µl of CCK-8 solution was added to the cells after 48 h of culture. The cells were then incubated for an additional 2 h, and the absorbance at 450 nm (OD value) was measured using a microplate reader. Cells were divided into the following groups for treatment: The control group, normal cultured THLE-2 and HepG2 cells; the HF group, cells induced with OA/PA without additional treatment; the CDCA group, during HF induction, 10 and 50 µM CDCA (cat. no. HY-76847; MedChemExpress) were added for intervention; and the CDCA + Shikonin (cat. no. HY-N0822; MedChemExpress) group, 1 h before adding 50 µM CDCA for intervention, 10 µM Shikonin was added for pretreatment. Each group had 6 replicate wells and subsequent detection was performed after 24 h of treatment.

THLE-2 and HepG2 cells were fixed in 4% ice-cold paraformaldehyde (cat. no. P0099; Beyotime Institute of Biotechnology) at 4˚C for 20 min, then treated with 0.1% Triton X-100 (Beyotime Institute of Biotechnology) at room temperature for 5 min, and washed twice with phosphate-buffered saline (PBS) to remove the residual 4% paraformaldehyde. A working fluid solution containing 10 µM BODIPY 493/503 (cat. no. GC42959; GLPBIO Technology LLC) was added to the cells, and the mixture was incubated at 37˚C for 30 min in the dark. Subsequently, the cell nuclei were stained with DAPI at room temperature for 5 min after two washes with PBS. After three more washes with PBS, the cells were immediately imaged using a fluorescence microscope (IX73; Olympus Corporation) at a magnification of x200. The area and fluorescence intensity of the lipid droplets were quantified using ImageJ software (version 1.8.0; National Institutes of Health) by analyzing five random fields per well (6).

All animal experiments were performed in compliance with all relevant ethical regulations for animal testing and research and in accordance with the animal protocols approved (approval no. SW-2020R0126-F03) by Pingxiang People's Hospital (Pingxiang, China). A total of 24 male, 6-week-old C57BL/6J mice were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. The mice were housed in SPF-grade animal rooms with controlled temperature (22±2˚C), humidity (50±10%), and a 12-h light/dark cycle. Each cage contained 6 mice (to ensure social interaction). Sterile food and filtered water were provided ad libitum. After 1 week of adaptive feeding in an SPF-grade animal room, the experiment was performed. The NAFLD model was established by feeding the mice a high-fat diet (HFD; cat. no. D12492; 60% fat content Research Diets, Inc.) for 12 weeks. Successful NAFLD modeling was determined based on body weight, with mice in the HF group exhibiting a body weight ≥20% higher than the normal diet group (average HF group weight ~38 g vs. normal group ~31 g).

After successful NAFLD modeling, 18 mice were randomly divided into three groups (n=6): i) A high-fat diet group (HF); ii) HF + CDCA (50 mg/kg/d); and iii) HF + CDCA (100 mg/kg/d). A normal diet group (fed with normal feed) was used as a control (n=6). During the entire study period, all the mice included at the start were monitored closely. There were no exclusions or losses; all mice completed the experimental procedures as scheduled, which ensured the integrity and reliability of the experimental data. The normal feed (cat. no. D12450B; Research Diets, Inc.) contained 10% fat, 70% carbohydrate and 20% protein, which was used as a direct contrast to the HFD (60% fat). CDCA (purity ≥98%; MedChemExpress) was dissolved in 0.5% carboxymethylcellulose sodium solution (cat. no. HY-Y0703; MedChemExpress) and administered by oral gavage at the corresponding dose once a day for 8 weeks. The total experiment duration (21 weeks) included 1 week of adaptation, 12 weeks of HFD feeding, and 8 weeks of CDCA intervention.

After the intervention, the mice were weighed and fasted for 12 h without water deprivation. For blood collection, 50-60 mg/kg sodium pentobarbital was administered intraperitoneally (anesthetic route, intraperitoneal injection) to induce temporary anesthesia, and 0.2 ml of blood was collected by eyeball enucleation. The animals were euthanized using an intraperitoneal overdose of sodium pentobarbital (100 mg/kg body weight; euthanasia method, intraperitoneal injection of anesthetic overdose). Death was verified by: i) No spontaneous breathing (chest movement) or heartbeat (chest palpation) for ≥5 min; and ii) no hind paw withdrawal reflex when pinched with forceps.

During the experiment, mice were monitored daily for health (coat condition, wounds, jaundice) and behavior (activity, feeding/drinking); body weight was measured weekly. Humane endpoints for euthanasia included: Severe weight loss (>20% of initial weight in 1 week), persistent lethargy/refusal to eat/drink for >24 h, visible organ damage (abdominal distension, and jaundice), or abnormal behaviors (trembling, convulsions). No mice met these endpoints or died unexpectedly; all 24 mice were euthanized at the end of the intervention.

The serum was separated, and the liver was quickly removed and fixed in 4% paraformaldehyde at 4˚C for 24 h. After fixation, the tissues were dehydrated and then cut into 8-10 µm-thick frozen sections using a cryostat. The sections were air-dried at room temperature for 10 min, followed by staining with freshly prepared Oil Red O (Thermo Fisher Scientific, Inc.) working solution at room temperature for 20 min in the dark. Finally, the stained sections were mounted with glycerol gelatin, and lipid droplet morphology was observed under a bright-field light microscope (Nikon E100) at a magnification of x400 (scale bar, 10 µm) and the remaining tissue was frozen at -80˚C for biochemical and molecular biology detection.

After cell collection, the cell lysate [2% Triton X-100 (Beyotime Institute of Biotechnology) lysate] was added to the cell pellet and incubated for 30 min to completely lyse the cells. Experiments were performed according to the instructions of the following kits: TC and TG detection kits (cat. nos. A111-1-1 and A110-1-1; Nanjing Jiancheng Bioengineering Institute), and SOD (cat. no. BC0170) and MDA (cat. no. BC0020) detection kits (Beijing Solarbio Science & Technology Co., Ltd.).

Total RNA from HepG2 cells or mouse liver tissues was extracted using TRIzol^® reagent (cat. no. 15596026; Inivtrogen; Thermo Fisher Scientific, Inc.), reverse-transcribed into cDNA (cat. no. RR037A; Takara Bio, Inc.), and qPCR reactions of lipid synthesis [fatty acid synthase (FASN) and sterol regulatory element-binding transcription factor 1 (SREBP-1c)] and β-oxidation [CTP synthase 1 (CPT1α) and peroxisome proliferator-activated receptor α (PPARα)]-related genes were performed with SYBR Green Master Mix (cat. no. 04707516001; Roche Diagnostics), according to the manufacturer's instructions. The thermocycling conditions were as follows: An initial denaturation step was performed at 95˚C for 10 min. This was followed by 40-45 cycles consisting of three sequential steps of denaturation at 95˚C for 10 sec, annealing at 60˚C for 20 sec, and elongation at 72˚C for 20 sec. After the cycling stage, a final extension step was conducted at 72˚C for 3 min. GAPDH was used as an internal reference gene, and the relative gene expression was calculated using the 2^-ΔΔC method (11). The primer sequences for the target and reference genes are listed in Table I.

Experiments were performed according to the instructions of the PK activity detection kit and pyruvate detection kit (cat. nos. BC0540 and BC2200; Beijing Solarbio Science & Technology Co., Ltd.).

To determine the binding site of CDCA and PK, after treating cells with CDCA, the cells were lysed in Pierce^™ RIPA Buffer (cat. no. 89900; Thermo Fisher Scientific, Inc.) to extract the total protein, and a PK antibody was added for IP. For each IP reaction, 500 µl of normalized cell lysate (containing 1,000 µg total protein) was incubated with 2 µg of anti-PK polyclonal antibody (cat. no. ab137852; Abcam), while normal rabbit IgG (cat. no. ab37415; Abcam) at the same amount was used as the negative control. Prior to use, 20 µl of settled Protein A agarose beads (cat. no. 20334; Thermo Fisher Scientific, Inc.) were pre-washed three times with cold PBS to remove the storage buffer, and then added to the antibody-lysate mixture for binding. After eluting the bound protein, the CDCA content was determined using HPLC. A Waters-C18 chromatographic column (250x4.6 mm; 5 µm) was used, with a volume flow rate of 1 ml/min, an injection volume of 20 µl, a column temperature of 30˚C and a mobile phase consisting of 0.1% formic acid aqueous solution (A phase) and acetonitrile (B phase) at a ratio of 40:60. Under these chromatographic conditions, the retention time of CDCA was 8.5±0.2 min, which was used to confirm the specificity of CDCA detection.

Discovery Studio software (version 2023; BIOVIA; Dassault Systèmes) was used to dock the small-molecule structure of CDCA with the crystal structure of PK, and to analyze the key amino acid residues and binding modes of their interaction.

Experimental data are expressed as the mean ± standard deviation (SD) with the number of independent replicates indicated in the figure legends or corresponding table notes. The experimental data were analyzed using one-way analysis of variance using SPSS Statistics software (Version 21.0; IBM Corp.), followed by Tukey's honestly significant difference (HSD) test for post hoc multiple comparisons. P<0.05 was considered to indicate a statistically significant difference. Origin 2024b software (OriginLab Corporation) was used to create the graphs.

The effect of CDCA on hepatic lipid deposition was evaluated with both in vitro and in vivo experiments. Screening of OA/PA combinations (0.25/0.125, 0.5/0.25 and 1.0/0.5 mM) revealed that 0.5 mM OA combined with 0.25 mM PA induced optimal lipid droplet accumulation without significant cytotoxicity (cell viability >80% as determined using a Cell Counting Kit-8 assay; data not shown). Following the establishment of the cell model using 0.5 mM OA combined with 0.25 mM PA, 10 and 50 µM CDCA were applied for 24-h intervention. The results demonstrated that compared with the HF group, intracellular TG content was significantly reduced in the CDCA-treated groups (P<0.05), with a comparable trend observed in HepG2 cells (Fig. 1A). TC content also exhibited a significant reduction (Fig. 1A), indicating that CDCA effectively mitigated intracellular lipid accumulation.

BODIPY staining visually illustrated intracellular lipid deposition. Abundant lipid droplets were observed in the HF group cells, whereas both the number and volume of lipid droplets were markedly decreased in the CDCA-treated groups, further confirming the inhibitory effect of CDCA on cellular lipid deposition (Fig. 1B). Additionally, cellular oxidative stress indices were assessed. Results indicated that, compared with the HF group, intracellular MDA content was significantly decreased in the CDCA-treated groups (P<0.05), while SOD activity was significantly elevated (Fig. 1C). To further investigate the molecular mechanisms underlying CDCA-mediated regulation of hepatic lipid metabolism, RT-qPCR analysis was performed. Results revealed that, in cell-based experiments, mRNA expression levels of lipid synthesis-related genes (FASN and SREBP-1c) were significantly downregulated following CDCA treatment compared with the HF group, whereas expression of lipid metabolism-related genes (CTP1α and PPARα) exhibited an opposite trend (Fig. 1D).

In the in vivo experiments, for the NAFLD mouse model established by 12-week HFD feeding, 50 and 100 mg/kg/d of CDCA were administered via gavage for 8-week intervention. The biochemical detection results showed that, compared with the HF group, the TC in the high-dose CDCA-treated group and TG in the high- and low-dose CDCA-treated groups were significantly reduced (P<0.01; Fig. 2A). Oxidative stress analysis in the liver showed that, compared with the HF group, the MDA content in the liver tissues of the CDCA-treated groups was significantly reduced (P<0.05), and the SOD activity was significantly increased (Fig. 2B). Oil Red O staining of liver tissue sections showed that the livers of mice in the HF group were filled with numerous red lipid droplets, whereas the lipid droplets in the CDCA-treated groups were notably reduced, especially in the high-dose group. Hepatic lipid deposition was markedly improved, which was consistent with the biochemical results, indicating that CDCA also reduced hepatic lipid deposition in vivo (Fig. 2C). In the liver tissues of mice, the mRNA expression levels in the high-dose CDCA-treated group of lipid synthesis genes, such as FASN and SREBP-1c, were significantly decreased compared with those in the HF group, and the expression of fatty acid β-oxidation-related genes (CPT1α and PPARα) were significantly upregulated (Fig. 2D), indicating that CDCA improved hepatic lipid deposition by inhibiting fatty acid synthesis and promoting fatty acid β-oxidation.

To explore whether there was a direct interaction between CDCA and PK, a co-IP experiment was performed. After treating HepG2 cells with CDCA, the cells were lysed to extract the total protein, and a PK antibody was added for co-IP. The results showed that in the CDCA-treated group, CDCA bound to PK was successfully detected, while no obvious binding signal was detected in the control group, indicating specific binding between CDCA and PK (Fig. 3A). Furthermore, the results of molecular docking showed that CDCA specifically binds to the PK-M2 subtype. The binding site was mainly located in the catalytic domain of PK-M2, and the key amino acid sites were Arg246, Lys247, Glu275, Arg278 and Arg279 in PKM2. The binding free energy score between CDCA and PK-M2 was -10.33 kcal/mol, indicating a strong binding stability between the two molecules (Fig. 3B). To study the regulatory effect of CDCA on PK activity, intracellular PK activity was measured. The results showed that CDCA enhanced PK activity in a concentration-dependent manner. When the CDCA concentration was in the range of 10-50 µM, the PK activity gradually increased with increasing CDCA concentration, and the difference was statistically significant (P<0.05; Fig. 3C). The enhancement of PK activity promoted the production of pyruvate, a product of glycolysis process (Fig. 3D), indicating that CDCA affects glycolysis by regulating PK activity.

To verify the role of PK in the CDCA-mediated improvement of hepatic lipid metabolism, a functional rescue experiment was performed. HepG2 cells were pretreated with 10 µM of the PK inhibitor (Shikonin) 1 h before intervention with 50 µM CDCA. These results showed that Shikonin pretreatment completely reversed the inhibitory effect of CDCA on lipid synthesis and enhanced β-oxidation. Compared with the CDCA-alone treatment group, the intracellular TG and TC levels in the CDCA + Shikonin group were significantly increased (P<0.01; Fig. 4A).

Immunofluorescence results showed that the fluorescence intensity of carbohydrate-responsive element-binding protein (ChREBP) in the nucleus was markedly reduced in the CDCA-treated group, whereas the nuclear translocation of ChREBP was restored in the CDCA + Shikonin group, and the fluorescence intensity in the nucleus was enhanced (Fig. 4B). Meanwhile, the CDCA-PK interaction promoted the nuclear localization of PPARα (Fig. 4B). Concurrently, 50 µM CDCA treatment reduced the mRNA expression levels of FASN and SREBP-1c, and these reduced levels returned to a state similar to the baseline after Shikonin intervention. By contrast, the mRNA expression levels of CPT1α and PPARα were increased following 50 µM CDCA treatment, and this upregulatory effect was reversed by Shikonin (Fig. 4C), indicating that PK serves a key mediating role in the regulation of hepatic lipid metabolism by CDCA.