Introduction
Cholelithiasis is a highly prevalent disorder of the digestive system, afflicting 6% of adults worldwide.1 Its incidence is increasing, particularly among females and South Americans. Cholesterol gallstones are the leading subtype.
Trimethylamine-N-oxide (TMAO) is synthesized by the hepatic enzyme flavin-containing monooxygenase 3 (FMO3) from the substrate trimethylamine, which is produced by the gut microbiota through the conversion of dietary phosphatidylcholine, choline, L-carnitine, and betaine.2 TMAO has been implicated in the pathogenesis of cardiovascular diseases, including hypertension,3 atherosclerosis,4 coronary heart disease,5 stroke,6 and abdominal aortic aneurysm7; it is also established as a risk factor for metabolic disorders such as obesity,8 type 2 diabetes,9 osteoporosis,10 graft-versus-host disease,11 and chronic kidney disease.12 Additionally, proline/serine-rich coiled-coil protein 1 deficiency may increase FMO3 expression, thereby upregulating TMAO production.13 TMAO binds and activates endoplasmic reticulum stress kinase, inducing the transcription factor FOXO1, which exacerbates metabolic dysfunction and atherosclerosis.14
The relationship between gut microbiota and lithogenesis has garnered increasing attention. Certain bacterial species not only affect bile microenvironment but also actively participate in lithogenesis by producing enzymes such as β-glucuronidase and mucins that influence bile acid and cholesterol metabolism.15,16 However, the underlying mechanisms of TMAO’s role in promoting lithogenesis remain unclear.
We utilized RNA sequencing, bioinformatic approaches, and both in vivo and in vitro experiments to confirm that TMAO promotes cholelithiasis by dysregulating hepatic cholesterol metabolism through its effects on the key regulatory genes apolipoprotein A4 (APOA4) and proprotein convertase subtilisin/kexin type 9 (PCSK9).
Methods
Clinical sample collection
Serum samples were collected from ten patients with asymptomatic cholelithiasis undergoing cholecystectomy (gallstone disease group, GSD group). Exclusion criteria included histories of: antibiotic or probiotic treatment within three months prior to sample collection; other digestive tract-related diseases; allergies or autoimmune diseases or ongoing immunomodulatory therapy; neurological or neurodevelopmental disorders or chronic pain syndromes; obesity based on a body mass index ≥30 kg/m2; metabolic syndrome or severe malnutrition; malignancy or ongoing cancer treatment; vegetarian diet; and fecal microbiota transplantation. For comparison, serum samples from 10 healthy liver transplant donors were included as a negative control (NC) group, using the same inclusion and exclusion criteria. After extraction, samples were promptly transported in liquid nitrogen and stored long-term at −80°C.
Animal experiments
Male C57BL/6J mice (six to eight weeks old, non-siblings), weighing 20–22 grams, were purchased from Liaoning Changsheng Biotechnology Co., Ltd. (Liaoning Province, China). All mice were maintained on a 12 h light/dark cycle at 22°C, with unrestricted access to standard rodent chow and water, and were housed under pathogen-free conditions. Mice were allowed to acclimatize for one week before experiments. All animal experiments were approved by the Animal Ethics Committee of The First Affiliated Hospital of Harbin Medical University. We employed a murine model of cholelithiasis induced by a lithogenic diet. Mice were fed either normal diet (ND, containing 0.02% cholesterol) or a lithogenic diet (LD, containing 1.25% cholesterol, 15% total fat, and 0.5% cholic acid) for four weeks.17 To investigate the effects of TMAO administration, 30 mice were randomly divided into three groups (NC, LD, and LD+TMAO), with each group housed separately. Mice in the NC group received an intraperitoneal injection of 200 µL saline as a control after 1 week. Mice in the LD group received an intraperitoneal injection of 200 µL saline for four consecutive weeks after 1 week. Mice in the LD+TMAO group received intraperitoneal injections of 200 µL TMAO (20 mg/kg/day, Sigma-Aldrich, St. Louis, MO, USA) for four consecutive weeks after 1 week.18 To investigate the effects of TMAO inhibition on lithogenesis, 20 mice were randomly divided into two groups, with each group housed separately. Mice in both groups received the LD, while the intervention group received LD+1% 3,3-dimethyl-1-butanol (DMB, Sigma-Aldrich, St. Louis, MO, USA), an inhibitor of TMAO synthesis, added to their drinking water.19
Measurement of gallstones and bile components
Mice were fasted for 6 h but allowed free access to water. Following euthanasia, the presence or absence of gallstones was determined. Gallbladders were removed intact, and all contents were extracted and placed on a glass dish to allow complete evaporation of liquid content. The residual solid mass was weighed.
We conducted a separate experiment following the methodology described by Matsumoto et al.20 Mice were anesthetized with isoflurane and maintained at 37°C on a heating pad. The common bile duct was ligated and cannulated with a PE-10 catheter to collect bile for 120 min, during which time the flow rate was measured. Total cholesterol (catalog number A111-1-1) and total bile acids (catalog number E003-2-1) were assayed according to the manufacturer’s instructions (Nanjing JIANCHENG Bioengineering, Nanjing, China). Phospholipid concentrations were quantified using a Wako kit (catalog number 296-63801, Osaka, Japan), also following the manufacturer’s instructions. Activity levels were measured using a multi-mode microplate reader (BioTek Synergy NEO).
Calculation of cholesterol saturation index (CSI)
The CSI of bile was calculated using the method proposed by Carey et al.21 The steps were as follows: (1) Calculation of the molar percentages (mol%) of cholesterol, bile acids, and phospholipids based on their molecular weights. (2) Calculation of the ratio of phospholipids to the sum of phospholipids and bile acids. (3) Using the ratio obtained in the previous step, reference the Carey table to obtain the theoretical molar percentage concentration of cholesterol. (4) CSI = actual molar percentage concentration of cholesterol/theoretical molar percentage concentration of cholesterol.
RNA sequencing
Three mice were randomly selected from each group and underwent partial hepatectomy to harvest liver tissue for RNA sequencing. cDNA libraries were sequenced on the Illumina sequencing platform by Wuhan Metware Biotechnology Co., Ltd. RNA was extracted using the Trizol method, dissolved in 50 µL of diethyl pyrocarbonate-treated water,22 and qualified and quantified using a Qubit Fluorometer and a Qsep400 high-throughput biological fragment analyzer. Raw data were filtered using fastp, and gene alignment was calculated using featureCounts.23 The fragments per kilobase of transcript per million fragments mapped (hereinafter referred to as FPKM) value of each gene was calculated based on its length.
Enzyme-linked immunosorbent assay (ELISA)
Serum samples from 10 GSD and 10 NC patients, as well as from mice in the LD and LD+DMB groups, were subjected to ELISA testing. ELISA kits for measuring human TMAO, PCSK9, APOA4, and murine TMAO were obtained from Lptech (China) and used according to the manufacturer’s instructions. The absorbance of the samples was read at 450 nm using a microplate reader (Multiskan-FC, Thermo Scientific).
Immunohistological analysis
Immunohistological analysis, tissue preparation, mounting, and blocking were conducted as previously described.24 Paraffin-embedded tissue sections were incubated overnight at 4°C with antibodies against APOA4 (Proteintech, China), 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR) (CUSABIO, China), ATP-binding cassette sub-family G member 5 (ABCG5) (Abmart, China), PCSK9, and ATP-binding cassette sub-family G member 8 (ABCG8) (Abclonal, China). After PBS washing, the sections were incubated with secondary antibodies for 1 h at room temperature, followed by counterstaining with hematoxylin and eosin. Histological sections were incubated with primary antibodies overnight at 4°C. After thorough washing with PBS, CoraLite 594 Goat Anti-Rabbit IgG (Proteintech, China) was applied and incubated for 1 h at room temperature. The slides were then stained with 4′,6-diamidino-2-phenylindole (SEVEN, China) and observed under a fluorescence microscope.
Cell culture and treatment
The murine normal hepatocyte line AML12 was acquired from Cyagen Biosciences Inc. (Shanghai, China) and cultured using AML12 cell-specific culture medium (Procell, Wuhan). AML12 cells were categorized into the following groups, analogous to those used in previous studies4: (1) Control (treated with complete culture medium only); (2) TMAO (50 µM of TMAO); (3) TMAO+APOA4OE (50 µM of TMAO and APOA4 overexpression); (4) TMAO+PCSK9si (50 µM of TMAO and PCSK9 inhibition); (5) APOA4OE (APOA4 overexpression). (6) PCSK9si (PCSK9 inhibition).
Plasmid and small interfering RNA (siRNA) transfection
Targeting siRNAs and overexpression plasmids were designed and synthesized by Hanbio Biotechnology (Shanghai, China). Cells were seeded in 6-well plates at a density of 2 × 105 cells per well and cultured overnight until they reached 50–70% confluence. Lipofectamine 2000 (Invitrogen, USA) was employed to transfect the siRNAs and plasmids into AML12 cells in a serum-free medium following the manufacturer’s instructions.25 Eight hours post-transfection, the cells were transferred to a complete culture medium and further incubated for 24 h. The siRNA sequences are summarized in Supplementary Table 1.
Real-time fluorescence quantitative PCR (qPCR)
Total RNA was extracted from tissues or cells using the RNA extraction kit from Axygen Scientific Inc. (Silicon Valley, USA) and then reverse-transcribed into cDNA using the Toyobo Reverse Transcription Kit. SYBR GREEN reagent was employed to detect the expression of target genes. Real-time-qPCR samples were preheated at 95°C for 10 min, followed by 40 cycles of denaturation at 95°C for 15 s and annealing/extension at 60°C for 1 min.26 GAPDH served as the internal reference gene. Data were analyzed using the 2−ΔΔCt method. Supplementary Table 2 provides detailed information regarding the primers.
Western blot (WB) analysis
Equal amounts of protein samples were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred onto a polyvinylidene difluoride membrane. The polyvinylidene difluoride membrane was incubated with the primary antibody overnight at 4°C, washed repeatedly with tris-buffered saline containing Tween 20, and then incubated with the secondary antibody for 1 h at room temperature.27 Proteins were visualized using an Enhanced Chemiluminescence Kit (Millipore Corporation), and quantitative analysis of proteins was performed using ImageJ software.28
Statistical analysis
Bioinformatic analysis was conducted using R software (version 4.0.2). GraphPad Prism 8 was employed for statistical analysis. Continuous data with normal distributions were compared between two groups using the t-test or Mann-Whitney U test. The Kruskal-Wallis test was applied for comparisons between three groups of samples. Spearman correlation analysis was utilized for all correlation analyses. In all statistical analyses, p < 0.05 was considered significant (*p < 0.05, **p < 0.01, ****p < 0.0001).
Results
TMAO promotes lithogenesis by dysregulating hepatic metabolism
Serum TMAO levels were significantly higher in GSD patients compared to the control group (Fig. 1A). Our murine model demonstrated that lithogenesis was more prevalent in LD mice than in NC mice and further elevated in the LD+TMAO group (Fig. 1B, C). DMB treatment reduced serum TMAO levels in LD+TMAO-treated mice to the physiological range observed in human serum (Supplementary Fig. 1). The weight of intracystic solid materials increased significantly after TMAO treatment (Fig. 1E). Hematoxylin and eosin staining revealed that TMAO disrupted hepatic lipid metabolism (Fig. 1D). Furthermore, LD significantly increased cholesterol content and decreased bile acid and phospholipid concentrations in bile, resulting in an elevated CSI. TMAO further increased bile cholesterol content, leading to a further elevation of the CSI (Fig. 1F–I).
Altered cholesterol metabolism in TMAO-treated mice revealed by RNA sequencing
RNA sequencing of hepatic tissue from three randomly selected mice in each group yielded FPKM distribution box plots, showing no significant intergroup differences in overall gene expression levels (Fig. 2A). 3D principal component analysis disclosed significant intragroup similarities and intergroup differences in gene expression (Fig. 2B). Differential gene analysis revealed differentially expressed genes among the NC, LD, and LD+TMAO groups, with 100 genes exhibiting significant differential expression between the NC and LD+TMAO groups, as well as between the LD and LD+TMAO groups, but no significant difference between the NC and LD groups (Fig. 2C, D). We therefore hypothesized that these 100 genes were altered by TMAO. Furthermore, KEGG enrichment analysis of these 100 differentially expressed genes indicated significant enrichment of cholesterol metabolic pathways (Fig. 2E). By selecting the top 10 pathways with the smallest q-values and constructing enrichment analysis and chord diagrams, we identified PCSK9 and APOA4 as key regulatory genes in the cholesterol metabolism pathway (Fig. 2F).
TMAO promotes hepatic cholesterol synthesis and efflux
Murine bile flow rate and cholesterol output were significantly higher in the LD group compared to the NC group and further increased in the LD+TMAO group (Fig. 3A, B). qPCR analysis of murine hepatic tissue disclosed significantly increased mRNA expression levels of the cholesterol biosynthesis-related gene HMGCR and the transporter-related genes ABCG5 and ABCG8 in the LD group compared to the NC group; this trend was even more pronounced in the LD+TMAO group (Fig. 3C–E). WB analysis of HMGCR, ABCG5, and ABCG8 revealed significantly increased expression of these proteins related to cholesterol biosynthesis and transport in the LD group compared to the NC group, and even more significant elevations in the LD+TMAO group (Fig. 3G–J). Immunohistochemistry revealed that the differential expression of HMGCR, ABCG5, and ABCG8 was consistent with results obtained from qPCR and WB assays (Fig. 3F).
APOA4 and PCSK9 are key factors in promoting TMAO-mediated lithogenesis
A volcano plot of gene expression revealed decreased APOA4 and increased PCSK9 expressions in the LD+TMAO group compared to the LD group (Fig. 4A). qPCR of murine hepatic tissue disclosed no significant differences in APOA4 and PCSK9 mRNA levels between the LD and NC groups; however, the LD+TMAO group exhibited significantly decreased APOA4 and increased PCSK9 mRNA levels compared to the other two groups (Fig. 4B, C). WB results were consistent with those of the qPCR assays (Fig. 4D–F). GSD patients exhibited significantly lower serum APOA4 protein levels and higher serum PCSK9 levels compared to the NC group (Fig. 4G, H).
Inhibited TMAO synthesis suppresses gallstone formation
DMB treatment significantly reduced gallstone formation in the LD group and markedly decreased the mass of intracystic solid materials (Fig. 5A, B). Additionally, DMB treatment reduced serum TMAO levels (Fig. 5C). qPCR and immunofluorescence staining revealed that DMB treatment attenuated the LD-induced APOA4 hypoexpression and PCSK9 hyperexpression in murine hepatic tissue (Fig. 5D–F). Further rescue experiments confirmed our conclusion that, in a lithogenic background, TMAO promotes gallstone formation by dysregulating the expression of APOA4 and PCSK9 proteins.
TMAO regulates APOA4, PCSK9 expressions, and cholesterol metabolism in vitro
PCSK9, HMGCR, ABCG5, and ABCG8 mRNA levels were significantly elevated (Fig. 6A–C, E), while APOA4 mRNA expression was markedly decreased (Fig. 6D) in TMAO-treated compared to control AML12 cells. Increasing TMAO concentrations downregulated APOA4 expression and upregulated PCSK9 expression, demonstrating a concentration-dependent effect of TMAO on both APOA4 and PCSK9 (Supplementary Fig. 2A, B).
Immunofluorescence staining showed that the fluorescence intensities of PCSK9, HMGCR, ABCG5, and ABCG8 were significantly higher, while that of APOA4 was notably lower in TMAO-treated AML12 cells compared to the control group (Fig. 6F). These findings further demonstrated that TMAO can upregulate the protein expression levels of PCSK9, HMGCR, ABCG5, and ABCG8, while downregulating that of APOA4, thereby modulating cholesterol metabolism.
TMAO dysregulates cholesterol metabolism by reducing APOA4 and increasing PCSK9 expression, initiating a feedback loop that promotes lithogenesis
To elucidate the specific mechanisms underlying the dysregulation of cholesterol metabolism by TMAO’s effects on APOA4 and PCSK9 expressions, we conducted an APOA4 gene functional overexpression study using specific plasmids. The overexpression efficiency of APOA4 was verified through qPCR (Supplementary Fig. 3A). Protein expression levels of PCSK9, HMGCR, ABCG5, and ABCG8 were significantly reduced in AML12 cells after APOA4 overexpression, partially rescuing their TMAO-induced overexpression (Fig. 7A). qPCR analysis of AML12 cells from both NC and APOA4-overexpressed groups revealed decreased PCSK9, HMGCR, ABCG5, and ABCG8 mRNA levels in the APOA4-overexpressed group compared to the NC group (Fig. 7B–E). Furthermore, immunofluorescence staining confirmed these findings, demonstrating markedly decreased immunofluorescence intensities of PCSK9, HMGCR, ABCG5, and ABCG8 in the APOA4-overexpressed group, consistent with the qPCR results (Fig. 7F).
Our PCSK9 gene knockdown study using specific siRNAs verified knockdown efficiencies in AML12 cells via qPCR (Supplementary Fig. 3B). PCSK9 knockdown significantly reduced protein expression levels of HMGCR, ABCG5, and ABCG8, while APOA4 expression increased to some extent. Additionally, PCSK9 knockdown partially rescued TMAO-induced hyperexpression of HMGCR, ABCG5, ABCG8, and APOA4 hypoexpression (Fig. 7G). qPCR analysis revealed decreased HMGCR, ABCG5, and ABCG8, and increased APOA4 mRNA levels in the PCSK9-knockdown group compared to the NC group (Fig. 7H–K). Immunofluorescence staining further confirmed these findings, demonstrating markedly decreased immunofluorescence intensities of HMGCR, ABCG5, and ABCG8 and a significant increase in APOA4 immunofluorescence in the PCSK9-knockdown group, consistent with the qPCR results (Fig. 7F). Furthermore, we observed a reciprocal feedback loop between PCSK9 and APOA4 expression, in which PCSK9 loss of function promoted APOA4 expression, and APOA4 overexpression inhibited PCSK9 expression.
Discussion
TMAO promotes atherogenesis by upregulating scavenger receptors in macrophages, which promotes cholesterol accumulation and foam cell formation. Furthermore, TMAO enhances inflammation via the MAPK and NF-κB pathways, which have been widely implicated in the progression of atherosclerosis.29 TMAO may also perturb the bile acid profile within the gallbladder.30 However, studies investigating the relationship between TMAO and lithogenesis are scarce. An animal study demonstrated that FMO3 knockout can prevent obesity and metabolic syndrome.31 Additionally, FMO3 plays a pivotal role in regulating glucose and lipid homeostasis. FMO3 knockout mice have exhibited significant reductions in hepatic and plasma lipid levels, ketone bodies, glucose, and insulin levels.32FMO3 transcription also mediates the activation of the TMA/Fmo3/TMAO pathway, upregulating ABCG5/8, thereby promoting the formation of cholesterol gallstones.33 These findings are consistent with the conclusions of the current study, which investigated the role of TMAO in lithogenesis. Cholesterol supersaturation in bile is a prerequisite for gallstone formation.34 Located on the membrane of hepatic canaliculi, the ABCG5/8 cholesterol transporter facilitates cholesterol excretion into bile, leading to bile cholesterol supersaturation and gallstone formation.35
Serum TMAO levels were significantly elevated in patients with cholelithiasis compared to the control group. In vivo, we observed that TMAO increased the cholesterol content and decreased the bile acid and phospholipid contents of bile, thereby exacerbating gallstone formation in mice. Murine studies by Luo33 and Chen36et al. also associated plasma TMAO levels with cholelithiasis. This association may be caused by TMAO-induced hypersecretion of bile cholesterol through increased expression of hepatic ABCG5/8, lipoprotein receptor SRB1, and HMGCR, which is consistent with our findings.
Differential and enrichment analyses of our RNA sequencing results from murine hepatic tissue revealed significant enrichment of cholesterol metabolism signaling pathways, with APOA4 and PCSK9 identified as key regulatory genes. Notably, the relationship between apolipoproteins and PCSK9 inhibition in cholelithiasis remains controversial and warrants further investigation. As a constituent of high-density lipoprotein, APOA4 may enhance insulin secretion and exert anti-inflammatory, antioxidant, and anti-atherosclerotic effects.37 To our knowledge, no previous studies have demonstrated a direct link between TMAO and APOA4. PCSK9 blocks the recycling of low-density lipoprotein receptors, leading to excessive accumulation of LDL-C,38 which is associated with hyperlipidemia39 and an increased risk of coronary heart disease.40 Cross-sectional studies have shown a significant association between TMAO and PCSK9,41 possibly mediated by IL-8. Furthermore, PCSK9 can promote lithogenesis by inhibiting PPARα-mediated CYP7A1 expression and preventing the conversion of cholesterol into bile acids.42 These findings align with our in vitro results. A slight association between PCSK9 and an increased risk of gallstones has been suggested but has not been confirmed by co-localization analysis.43 Inhibition of PCSK9 promotes hepatic cholesterol biosynthesis, leading to cholesterol oversaturation and gallstone formation.44 In contrast, Sanderson et al. suggested an association between APOA4 downregulation and decreased plasma cholesterol,45 while Mendelian randomization analyses have speculated that LDL-C reductions caused by PCSK9 or apolipoprotein variants should not affect the risk of symptomatic cholelithiasis.46 These observations are incongruent with our conclusions. Specifically, PCSK9 has been associated with apolipoprotein levels,47 and specific inhibition of PCSK9 reduces plasma lipoprotein levels.48 Interestingly, the relationship between PCSK9 and apolipoprotein levels may vary between ethnic groups.
To further validate the relationship between PCSK9 and APOA4, we conducted gene silencing experiments using siRNA. We knocked down PCSK9 and observed increased APOA4 expression. Interestingly, the insertion of an APOA4 overexpression plasmid decreased PCSK9 expression. We propose that in TMAO-treated murine hepatic tissue, elevated PCSK9 expression inhibits APOA4 expression, while low APOA4 expression further promotes PCSK9 expression, initiating a feedback loop that dysregulates cholesterol metabolism and thereby promoting lithogenesis by increasing cholesterol synthesis and efflux via HMGCR and ABCG5/8, respectively. Consequently, biliary concentrations of cholesterol and bile acids increase and decrease, respectively, thereby promoting the formation of cholesterol gallstones (Fig. 8).
Nevertheless, our understanding of the genetic and molecular basis of lithogenesis is still limited. First, the direct molecular interactions between TMAO and PCSK9/APOA4, as well as the reciprocal expressions of APOA4 and PCSK9, require further clarification. Furthermore, the collection of liver tissue samples from patients with cholelithiasis to validate our hypothesis would be challenging. Finally, the effectiveness and feasibility of targeting APOA4 and PCSK9 in clinical treatment remain to be demonstrated.
Conclusions
We have discovered that TMAO dysregulates cholesterol metabolism by initiating a feedback loop, upregulating PCSK9 expression, and downregulating APOA4 expression in murine hepatic tissue, thereby promoting gallstone formation. In clinical practice, hyperlipidemia, a risk factor for cardiovascular diseases, also promotes gallstone formation, which should warrant attention. Regulation of blood lipid levels is necessary for patients with cholelithiasis. Targeting PCSK9 and APOA4 may represent a promising approach for the prevention and treatment of cholelithiasis.
Supporting information
Supplementary Table 1
siRNA sequences used in this study.
(DOCX)
Supplementary Table 2
Primers used for qPCR.
(DOCX)
Supplementary Fig. 1
Serum TMAO concentrations in the NC, LD, and LD+TMAO groups.
(TIF)
Supplementary Fig. 2
Changes in APOA4 and PCSK9 mRNA expression in AML12 cells with increasing concentrations of TMAO.
(TIF)
Supplementary Fig. 3
Efficiency of APOA4 overexpression and PCSK9 knockdown in AML12 cells was verified through qPCR.
(TIF)
Declarations
Acknowledgement
We thank Medjaden Inc. for their assistance in the preparation of this manuscript. The authors sincerely thank all the institutions and individuals who contributed to this research.
Ethical statement
The study protocol was approved by the Ethics Committee of The First Affiliated Hospital of Harbin Medical University (No.202322), in strict accordance with the Declaration of Helsinki. This study was conducted in strict accordance with the Guide for the Care and Use of Laboratory Animals formulated by the Ministry of Science and Technology of China in 2006, and all procedures involving animal experiments were conducted according to the regulations of the Institutional Animal Care and Use Committee of The First Affiliated Hospital of Harbin Medical University (No. 2024006).
Data sharing statement
Data supporting the findings of this study are available from the corresponding author upon reasonable request.
Funding
This study was supported by the National Natural Science Foundation of China (No. 82270598), the National Natural Science Foundation of China (No.82100675), and the Open Fund of the Key Laboratory of Hepatosplenic Surgery, Ministry of Education, Harbin, China (No. GPKF202404).
Conflict of interest
The authors have no conflicts of interest related to this publication.
Authors’ contributions
Conceptualization (CS, JY), investigation and formal analysis (CS, JY, ZM), resources (DL, HD, GS), writing – original draft (CS, ZM), writing –review & editing (XM, DX), supervision (XM) and funding acquisition (XM, DX, HS). All authors have read and agreed to the published version of the manuscript.