TUDCA-3S is Associated with Acute Pancreatitis Severity and Pro-Inflammatory Macrophage Activation Involving S100A8/IL-17 Signaling

Abstract

Objective

Early identification of patients at risk of severe acute pancreatitis (SAP) remains challenging, and the contribution of sulfated bile-acid metabolites to AP-associated inflammatory responses remains incompletely understood. This study investigated the association between serum TUDCA-3S and AP severity and explored its effects on inflammatory macrophage activation in vitro.

Methods

The study cohort comprised 100 patients with acute pancreatitis (AP) and 50 healthy controls. Serum TUDCA-3S levels were quantified using targeted metabolomics, and associations with disease severity and receiver operating characteristic (ROC) performance were evaluated. An LPS-stimulated RAW264.7 inflammatory macrophage model was used to examine cellular responses associated with TUDCA-3S exposure. Flow cytometry, RT-qPCR, Western blotting, transcriptomics, untargeted metabolomics, lipidomics, and S100A8 knockdown were used for further characterization.

Results

Serum TUDCA-3S levels differed significantly among the study groups, with the highest median level observed in patients with SAP. TUDCA-3S distinguished AP patients from healthy controls with an AUC of 0.873. However, its performance for severity discrimination was substantially lower, with an AUC of 0.534 for MAP versus MSAP and 0.714 for MAP/MSAP versus SAP. In LPS-stimulated RAW264.7 macrophages, TUDCA-3S exposure was associated with enhanced M1-related inflammatory features and increased S100A8 and IL-17A expression. S100A8 knockdown attenuated M1-associated markers and IL-17A expression. Transcriptomic, untargeted metabolomic, and lipidomic analyses further identified inflammatory and lipid-metabolic changes associated with TUDCA-3S exposure.

Conclusion

Serum TUDCA-3S is associated with AP and disease severity, although its discriminatory performance across AP severity categories is limited to moderate. The in vitro findings support an association between TUDCA-3S exposure and pro-inflammatory macrophage features involving S100A8-associated IL-17A signaling and metabolic remodeling. These findings are exploratory and require independent clinical and experimental validation.

Introduction

Acute pancreatitis (AP) is a critical gastrointestinal disorder characterised by localised pancreatic inflammation that can rapidly progresses to the systemic inflammatory response syndrome (SIRS).¹ The aetiology of AP is diverse, including biliary, alcoholic, and drug-induced subtypes, with biliary acute pancreatitis (BAP) accounting for approximately 50% of cases. The current management of AP primarily relies on supportive care, including intravenous fluid resuscitation and intensive monitoring.^1,2 However, the absence of specific drugs targeting key pathological pathways means that approximately 20% of patients still develop severe acute pancreatitis (SAP), leading to multiple organ dysfunction syndrome (MODS) and mortality rates ranging from 20% to 40%. Survivors often experience severe complications.^3–5 Therefore, early detection and control of SAP progression are crucial, highlighting the urgent need to elucidate the pathological mechanisms of AP and identify specific potential therapeutic targets.

The pathogenesis of AP is initiated by premature intracellular activation of pancreatic enzymes within acinar cells, triggering local inflammation. Subsequently, inflammasome activation and the release of inflammatory cytokines exacerbate pancreatic injury, potentially leading to SIRS and MODS.^6,7 Among immune cells, macrophages and neutrophils are the first to infiltrate the pancreas and contribute to tissue injury.⁸ Recent studies have highlighted the central role of macrophage polarisation in regulating inflammation. Proinflammatory M1 macrophages exacerbate inflammation by releasing cytokines such as interleukin-6 (IL-6) and tumour necrosis factor-α (TNF-α), whereas anti-inflammatory M2 macrophages may limit tissue damage through reparative functions.⁹ However, the key molecular mechanisms driving macrophage polarisation and their dynamic changes during AP progression remain to be elucidated. Because pro-inflammatory macrophage activation contributes to inflammatory amplification, we focused on M1-associated inflammatory features in the subsequent cell-based experiments. CD11c and CD86 were used as complementary M1-associated markers, whereas CD206 was used as an M2-associated marker in the flow-cytometry analyses.

Dysregulation of bile acid (BA) metabolism in AP has attracted increasing attention. In patients with BAP, bile reflux can induce acinar cell injury through calcium overload and mitochondrial dysfunction, thereby accelerating pancreatic damage and triggering SIRS.^10,11 Furthermore, studies have shown that BAs can directly activate inflammatory signalling pathways in macrophages, suggesting a potential role in modulating systemic inflammation through macrophage polarisation.^12–14 However, the mechanism by which specific bile acid metabolites influence macrophage polarisation and AP progression remains unclear. Circulating bile-acid metabolites in AP may be influenced by biliary obstruction, hepatic dysfunction, altered enterohepatic circulation, disease etiology, systemic inflammation, and metabolic status. Therefore, associations between a specific bile-acid metabolite and AP severity should be interpreted cautiously.

Tauroursodeoxycholic acid (TUDCA) is a secondary bile acid formed by conjugation of ursodeoxycholic acid with taurine.¹⁵ Studies have indicated that TUDCA alleviates endoplasmic reticulum (ER) stress in AP by suppressing the unfolded protein response (UPR), thereby reducing inflammation.¹⁶ Tauroursodeoxycholic acid-3-sulphate (TUDCA-3S) is a sulphated derivative of TUDCA formed by conjugation of a sulphate group at the 3-position of the steroid nucleus. Research in intestinal inflammation models has demonstrated that, while secondary bile acids generally exhibit anti-inflammatory properties, sulfation abolishes these protective effects.^17,18 Sulfation can alter the physicochemical properties and inflammatory effects of bile acids. Therefore, the biological behavior of TUDCA-3S cannot necessarily be inferred directly from that of its unsulfated parent compound, TUDCA. This provides a rationale for investigating TUDCA-3S independently rather than assuming that it shares the biological effects reported for TUDCA.

Clinical assessment of AP severity currently relies on combinations of laboratory biomarkers and scoring systems, including C-reactive protein (CRP), procalcitonin, the Bedside Index for Severity in Acute Pancreatitis (BISAP), the Acute Physiology and Chronic Health Evaluation II (APACHE II) score, and CT-based severity indices. Although these approaches provide clinically useful information, they differ in timing, complexity, availability, and predictive performance. Therefore, additional molecular markers may complement, rather than replace, established clinical assessment. Our previous study investigated NorCA-associated macrophage polarization and Wnt/β-catenin-related signaling in AP.^19,20 The present study investigates a distinct sulfated bile-acid metabolite, TUDCA-3S, and integrates clinical targeted metabolomics with independent transcriptomic, untargeted metabolomic, lipidomic, and cell-based datasets. None of the clinical samples, transcriptomic datasets, metabolomic datasets, lipidomic datasets, or cell-based experimental data analyzed in the present study overlap with those used in our previous NorCA study. The conceptual connection between the two studies is limited to the broader research theme of bile-acid metabolites and macrophage-associated inflammation in AP. The novelty of the present work therefore lies in identifying a TUDCA-3S-associated clinical, inflammatory, and metabolic signature in AP.

Because pro-inflammatory macrophage activation contributes to inflammatory amplification in AP, we first examined whether TUDCA-3S exposure was associated with an M1-related macrophage phenotype. Transcriptomic profiling was subsequently used as an unbiased approach to identify inflammatory pathways associated with TUDCA-3S exposure. Enrichment of IL-17-related signaling together with differential S100A8 expression then provided the rationale for focused analysis of S100A8 and IL-17A. In parallel, untargeted metabolomic and lipidomic analyses were used to characterize metabolic changes associated with TUDCA-3S exposure. Accordingly, this study aimed to evaluate the association between circulating TUDCA-3S and AP severity and to explore the inflammatory macrophage responses associated with TUDCA-3S exposure using an LPS-stimulated macrophage model.

Materials and Methods

Study Population and Sample Collection

Fasting serum samples were collected from 100 patients with acute pancreatitis (AP group) and 50 healthy controls (control group). The AP group comprised individuals with biliary AP (BAP) (n=36), hyperlipidaemic AP (HLAP) (n=27), and alcoholic AP (AAP) (n=37). AP severity was classified according to the 2012 Revised Atlanta Classification as mild acute pancreatitis (MAP), moderately severe acute pancreatitis (MSAP), or severe acute pancreatitis (SAP). MAP was defined by the absence of organ failure and local or systemic complications. MSAP was defined by transient organ failure resolving within 48 h and/or local or systemic complications. SAP was defined by persistent organ failure lasting >48 h. The present analysis evaluates the association of TUDCA-3S with these established severity categories and should not be interpreted as prospective validation of TUDCA-3S for predicting future persistent organ failure.

The exclusion criteria for both patients and controls were as follows: age <18 or >85 years; severe cardiac, pulmonary, hepatic, or renal comorbidities; malignancy; pregnancy; traumatic pancreatitis; or presentation more than 48 h after symptom onset. Patients presenting more than 48 h after symptom onset were excluded. However, serum sampling was not standardized to an identical post-onset time point for every patient. The control group consisted of healthy individuals with no history of metabolic, gastrointestinal, or cardiovascular disease and was matched to the AP group by age, sex, and body mass index (BMI). This study was conducted in accordance with the Declaration of Helsinki (1975) and was approved by the Ethics Committee of Affiliated Hospital of North Sichuan Medical College. Informed consent was obtained from all patients (2021ER133-1). This was a single-center exploratory observational cohort rather than a prospectively powered biomarker-validation study. Patients presenting more than 48 h after symptom onset were excluded. Serum sampling was not standardized to an identical post-onset time point for every patient; therefore, temporal variation in circulating bile-acid metabolites cannot be completely excluded.

Cell Culture and LPS-Induced Inflammatory Macrophage Model

All cell lines used in this study were authenticated by short tandem repeat (STR) profiling and confirmed to be free of mycoplasma contamination. RAW264.7 murine macrophages were purchased from SUNNCELL (Wuhan, China) in 2022 and immediately expanded for experimental use. Cells were maintained in high-glucose Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% foetal bovine serum (FBS) and 1% penicillin-streptomycin at 37°C in a humidified atmosphere containing 5% CO2.

For subculturing, cells were passaged when they reached approximately 90% confluence in 6-cm dishes. Briefly, the old medium was aspirated, and the cells were washed with pre-warmed phosphate-buffered saline (PBS). After dissociation, the cell suspension was collected and centrifuged at 800 × g for 3 min. The supernatant was discarded, and the cell pellet was resuspended in 2 mL of fresh, pre-warmed complete medium. The cells were then split into two new dishes, each containing 3 mL of fresh medium. All procedures were performed gently to avoid unintended activation or polarisation of macrophages.

Cells were divided into three experimental groups: (1) Ctrl group, cultured in standard medium for 24 hours; (2) LPS group, treated with 100 ng/mL LPS for 24 h to induce a pro-inflammatory macrophage phenotype; and (3) TUDCA-3S group, co-treated with 100 ng/mL LPS and 2.0 μg/mL TUDCA-3S for 24 h. LPS and TUDCA-3S were dissolved in dimethyl sulfoxide (DMSO; cell-culture grade) before application.

Other Experimental Procedures

Targeted metabolomics was used to quantify serum TUDCA-3S. Transcriptomic sequencing was performed using three biological samples per group. Differentially expressed genes were identified using |log2FC| ≥1 and FDR ≤0.05. Multiple-testing correction was incorporated into GO, KEGG, and metabolomic pathway analyses according to the respective analytical pipelines. Untargeted metabolomics and lipidomics were used to characterize metabolite and lipid alterations associated with TUDCA-3S exposure. S100A8 knockdown was performed using siRNA followed by RT-qPCR, Western blotting, and flow cytometry. Flow cytometry was performed using two parallel staining combinations, CD11c-PE/CD206-APC and CD86-PE/CD206-APC. Flow-cytometry data were analyzed using FlowJo software. The principal cellular population was first identified according to forward- and side-scatter characteristics to exclude events with aberrant scatter characteristics and obvious debris. Singlet events were subsequently selected using FSC-A versus FSC-H to minimize the contribution of cell aggregates and doublets. The resulting cell population was then analyzed using PE-A versus APC-A fluorescence plots. For the CD86/CD206 panel, CD86 was detected in the PE channel and CD206 in the APC channel; the CD86⁺CD206⁻ population was used as the CD86-defined M1-associated population. For the parallel CD11c/CD206 analysis, CD11c was detected in the PE channel and CD206 in the APC channel, and the CD11c⁺CD206⁻ population was used as the corresponding M1-associated population. The same analysis strategy and quadrant settings were applied consistently across experimental groups within each experiment.

All cell-based experiments were independently repeated three times (n=3 biological replicates), and RT-qPCR reactions were additionally performed in technical triplicate. The M1-associated inflammatory macrophage phenotype was used as the principal cellular endpoint. S100A8 and IL-17A were subsequently prioritized for focused validation based on the transcriptomic differential-expression and pathway-enrichment results. Detailed experimental procedures are provided in Supplementary Material 1.

Statistical Analysis

Cell-based quantitative data are presented as mean ± standard deviation (SD) from three independent biological experiments (n=3). Normality was assessed using the Shapiro–Wilk test. Comparisons between two groups were performed using an unpaired two-tailed Student’s t-test when appropriate, whereas comparisons among more than two groups were performed using one-way ANOVA followed by the post-hoc procedure used in the original analysis. Clinical continuous variables are presented as median (interquartile range) where appropriate. Non-normally distributed continuous variables were compared among multiple independent groups using the Kruskal–Wallis test. Categorical variables are presented as n (%) and were compared using the chi-square test or Fisher’s exact test, as appropriate. A two-sided P<0.05 was considered statistically significant. For high-dimensional omics analyses, multiple-testing correction was performed according to the respective analytical pipelines. Transcriptomic differential-expression analysis used |log2FC| ≥1 and FDR ≤0.05. GO, KEGG, and untargeted-metabolomic pathway analyses incorporated FDR/Benjamini–Hochberg correction as described in Supplementary Material 1. All cell-based experiments were independently repeated three times (n=3 biological replicates), and RT-qPCR measurements were additionally performed in technical triplicate. Statistical significance was denoted as ns, not significant (P > 0.05); *P < 0.05; **P < 0.01; ***P < 0.001; and ****P < 0.0001.

Results

Baseline Characteristics and Spearman Correlation Analysis

Table 1 summarizes the baseline demographic and clinical laboratory profiles for the healthy control group and AP patients, stratified by disease severity (MAP, MSAP, SAP). Significant intergroup differences were observed in key clinical laboratory parameters, such as white blood cell and neutrophil counts, creatinine, blood glucose, liver function tests, and lipid profiles (p < 0.05). Spearman correlation analysis (Table 2) identified positive correlations of pancreatitis severity with multiple hematological parameters, liver function markers, and lipid profiles. Notably, TUDCA-3S levels also exhibited a significant positive correlation with disease severity.

Table 1.

Baseline Characteristics of Patients with Acute Pancreatitis of Different Severity

Variables	Ctrl (n=50)	MAP (n=59)	MSAP (n=27)	SAP (n=14)	P-value
Age, years	56 (39, 65)	55 (39, 61)	50 (34, 71)	45 (38, 61)	P=0.849
Sexuality, n (%)					P=0.828
Male	33 (66)	39 (66.1)	18 (66.7)	11 (78.6)
Female	17 (34)	20 (33.9)	9 (33.3)	3 (21.4)
White blood cell count, ×109/L	5.33 (4.72, 7.22)	9.14 (7.59, 13.52)	14.65 (11.22, 16.94)	14.86 (11.35, 18.35)	P<0.0001
Neutrophil percentage, %	58.35 (51.78, 63.93)	77.9 (67.60, 84)	85.60 (78.5, 87.5)	86.67 (83.6, 90.83)	P<0.0001
Creatinine, umol/L	66 (57.75, 75.25)	63 (59, 70)	67 (61, 90)	83 (67.58, 118.75)	P=0.005
Blood sugar, mmol/L	5.59 (5.03, 6.07)	7.35 (6.21, 9.65)	8.13 (6.51, 12.98)	10.06 (6.35, 15.51)	P<0.0001
ALT, IU/L	25 (16, 33)	34 (22, 154)	45 (24, 104)	42.5 (29.25, 87.75)	P<0.0001
AST, IU/L	25.5 (21, 32.25)	33 (23, 97)	36 (24, 62)	42.5 (36, 142.25)	P<0.0001
Total bilirubin, umol/L	12.65 (11.58, 14.63)	17.8 (13.5, 27.1)	20.6 (15.6, 37.5)	31.35 (14.84, 58.88)	P<0.0001
Direct bilirubin, umol/L	4.45 (3.5, 5.5)	11.2 (6.7, 20.9)	10.8 (5.3, 15.7)	14.65 (6.49, 22.4)	P<0.0001
Total cholesterol, mmol/L	5.25 (4.9, 5.83)	5.54 (5.05, 7.12)	5.76 (4.42, 8.09)	5.28 (4.47, 8.14)	P=0.275
Triglyceride, mmol/L	1.95 (1.52, 2.73)	2.27 (1.64, 7.68)	2.15 (0.96, 12.17)	3.41 (2.16, 8.24)	P=0.05
High density lipoprotein, mmol/L	1.43 (1.22, 1.57)	1.24 (1.14, 1.35)	1.26 (1.11, 1.41)	1.25 (1.2, 1.35)	P=0.003
Low density lipoprotein, mmol/L	3.29 (2.92, 3.54)	3.26 (2.79, 3.81)	3.6 (2.26, 3.93)	2.97 (2.31, 3.47)	P=0.427
TUDCA-3S, nMol/L	3.77 (2.13, 5.47)	19.03 (7.3, 83.98)	18.43 (5.56, 73.65)	209.08 (18.84. 291.51)	P<0.0001
Enzymology negative, n (%)		19 (32.2)	10 (37.0)	3 (21.4)	P=0.599
Type of pancreatitis, n (%)					P<0.0001
BAP		21 (35.6)	10 (37)	5 (35.7)
HLAP		16 (27.1)	7 (26)	4 (28.6)
AAP		22 (37.3)	10 (37)	5 (35.7)

Abbreviations: Ctrl, healthy controls; MAP, mild acute pancreatitis; MSAP, moderately severe acute pancreatitis; SAP, severe acute pancreatitis; ALT, alanine aminotransferase; AST, aspartate aminotransferase; TUDCA-3S, tauroursodeoxycholic acid-3-sulphate; BAP, biliary acute pancreatitis; HLAP, hyperlipidemia acute pancreatitis; AAP, Alcoholic acute pancreatitis.

Table 2.

Spearman Correlation Matrix Between Each Research Variable and the Severity of Pancreatitis

Variables	Correlation Coefficient	P-value
White blood cell count, ×109/L	0.721	P<0.0001
Neutrophil percentage, %	0.739	P<0.0001
Creatinine, umol/L	0.210	P=0.01
Blood sugar, mmol/L	0.627	P<0.0001
ALT, IU/L	0.368	P<0.0001
AST, IU/L	0.324	P<0.0001
Total bilirubin, umol/L	0.475	P<0.0001
Direct bilirubin, umol/L	0.515	P<0.0001
Total cholesterol, mmol/L	0.105	P=0.201
Triglyceride, mmol/L	0.191	P=0.019
High density lipoprotein, mmol/L	−0.217	P=0.008
Low density lipoprotein, mmol/L	−0.072	P=0.382
TUDCA-3S, nMol/L	0.568	P<0.0001

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; TUDCA-3S tauroursodeoxycholic acid-3-sulphate.

Association of Serum TUDCA-3S with AP and Disease Severity

Therefore, to investigate metabolic alterations in AP, we performed targeted metabolomics on serum samples from patients with AP and healthy controls. The results revealed a significant elevation in serum TUDCA-3S levels in patients with AP compared with healthy controls (Figure 1A). Furthermore, serum TUDCA-3S levels were markedly higher in patients with SAP than in those with MSAP (Figure 1C). In contrast, no significant difference was observed between patients with MAP and those with MSAP (Figure 1B). Significantly elevated TUDCA-3S levels were also observed in the two major aetiological subtypes of AP: BAP (Figure 1D) and AAP (Figure 1F). However, no significant increase was observed in HLAP (Figure 1E). The levels of TUDCA-3S exhibit distinct patterns across different etiologies of AP: significant elevations are observed in both BAP and AAP, whereas no significant elevation is noted in HLAP. These observations suggest the potential presence of etiological heterogeneity.

Figure 1.

The expression level and clinical diagnostic value of TUDCA-3S in AP and its subtypes: (A) To compare the expression level of TUDCA-3S in Ctrl group and AP group. (B) Comparison of TUDCA-3S expression levels between MAP group and MSAP group. (C) Comparison of TUDCA-3S expression levels between MSAP group and SAP group. (D) Comparison of TUDCA-3S expression levels between Ctrl group and BAP group. (E) Comparison of the expression level of TUDCA-3S between Ctrl group and HLAP group. (F) Comparison of the expression level of TUDCA-3S between Ctrl group and AAP group. (G) ROC curve, AUC, sensitivity and specificity of TUDCA-3S in the diagnosis of AP. (H) ROC curve, AUC, sensitivity and specificity of TUDCA-3S in the diagnosis of amylase-negative AP. (I) ROC curve, AUC, sensitivity and specificity of TUDCA-3S in the discrimination between MAP and MSAP. (J) ROC curve, AUC, sensitivity and specificity of TUDCA-3S in the discrimination between MAP/MSAP and SAP.

Next, the diagnostic potential of TUDCA-3S was evaluated. The area under the receiver operating characteristic curve (AUC) for distinguishing patients with AP from healthy controls was 0.873, with a sensitivity of 74% and specificity of 90% (Figure 1G). The area under the curve (AUC) for distinguishing amylase-negative acute pancreatitis (AP) was 0.816, with a sensitivity of 62.5% and a specificity of 90% (Figure 1H). The AUC for distinguishing MAP from MSAP was 0.534, with 92.6% sensitivity and 20.3% specificity (Figure 1I). The AUC for distinguishing MAP/MSAP from SAP was 0.714, with 64.3% sensitivity and 86.1% specificity (Figure 1J).

Although TUDCA-3S showed good discrimination between AP and healthy controls, this comparison has limited incremental clinical relevance after AP has been diagnosed. Its performance for severity discrimination was poor for MAP versus MSAP and moderate for MAP/MSAP versus SAP. Therefore, the present data support further evaluation of TUDCA-3S as a severity-associated candidate but do not establish it as a stand-alone diagnostic or predictive biomarker for SAP.

Establishment of an LPS-Induced Inflammatory Macrophage Model and Selection of the TUDCA-3S Experimental Concentration

RAW264.7 macrophages were stimulated with 100 ng/mL LPS to induce a pro-inflammatory macrophage phenotype.^21,22 Western blotting and RT-qPCR showed increased expression of the M1-associated markers CD86 and CD11c in LPS-treated cells compared with Ctrl cells (Figure 2A and B), supporting successful induction of an LPS-stimulated inflammatory macrophage phenotype.

Figure 2.

Validation of AP model establishment and construction of TUDCA-3S model: (A) Western blot was used to verify the expression of M1 macrophage protein on AP model induced by LPS. (B) RT-qPCR was used to verify the mRNA expression of M1 macrophages on the AP model induced by LPS. (C) CCK8 was used to verify the effect of TUDCA-3S on the survival rate of macrophages. Data are presented as mean ± SD from three independent biological experiments (n=3).

To characterize the concentration-dependent effect of TUDCA-3S on RAW264.7-cell viability, cells were exposed to 0, 2.0, 4.0, 6.0, 8.0, or 10.0 μg/mL TUDCA-3S. Cell viability decreased at 2.0 μg/mL and showed a further reduction at 4.0 μg/mL, whereas relatively limited additional changes were observed across the higher concentration range tested (Figure 2C). Based on this concentration-response pattern, 2.0 μg/mL was selected as the lowest tested treatment concentration producing a clear cellular response while avoiding exposure to the higher concentration range associated with lower cell viability. Importantly, 2.0 μg/mL should not be considered a non-cytotoxic concentration, and a contribution of nonspecific cellular stress to the downstream findings cannot be completely excluded.²³

TUDCA-3S Exposure is Associated with Enhanced Pro-Inflammatory Macrophage Features

To investigate the role of TUDCA-3S in AP progression, we assessed the proportion of M1 macrophages by flow cytometry in the control (Ctrl), AP model, and TUDCA-3S intervention groups. The results showed a significant increase in the proportion of M1 macrophages in the AP group compared with the control group, and a further increase after TUDCA-3S treatment (Figure 3A). Consistent with this, Western blot analysis revealed that TUDCA-3S treatment significantly increased M1 marker protein levels compared with those in the LPS group (Figure 3B). RT-qPCR results further supported these findings, showing that M1 marker mRNA expression was highest in the TUDCA-3S group (Figure 3C).

Figure 3.

Detection of M1 macrophage markers: (A) Representative flow-cytometry plots and quantitative analysis of M1-associated macrophage populations using the CD11c-PE/CD206-APC and CD86-PE/CD206-APC staining combinations. PE fluorescence is shown on the x-axis and APC fluorescence on the y-axis. The lower-right CD11c⁺CD206⁻ or CD86⁺CD206⁻ population was used as the corresponding M1-associated population. The same gating and quadrant settings were applied consistently across the experimental groups within each analysis. (B) Western Blot was used to detect the expression of CD11c and CD86 on the surface of macrophages. (C) The expression of CD11c and CD86 mRNA in macrophages was detected by RT-qPCR. Data are presented as mean ± SD from three independent biological experiments (n=3).

Together, these findings indicate that TUDCA-3S exposure is associated with enhanced pro-inflammatory macrophage features under the present LPS-stimulated conditions.

Transcriptomic Analysis of TUDCA-3S-Associated Inflammatory Changes

To identify pathways associated with TUDCA-3S exposure, transcriptomic sequencing was performed in the LPS and TUDCA-3S groups. GO enrichment analysis showed that differentially expressed genes were enriched in immune-response, cytokine-activity, and neutrophil-chemotaxis categories (Figure 4A). KEGG pathway analysis identified enrichment of inflammation-related pathways, including IL-17 signaling, viral protein-cytokine/cytokine-receptor interactions, and TNF signaling (Figure 4B).

Figure 4.

Transcriptome analysis: (A and B) GO, KEGG gene enrichment pathway. (C) Differential genes in AP group and TUDCA-3S group were screened out. (D) The expression of S100a8 mRNA in Ctrl group, AP group and TUDCA-3S group. (E) The mRNA expression of siS100a8 in AP group and TUDCA-3S group. Cell-based validation experiments were independently repeated three times (n=3).

A heatmap showed the top differentially expressed genes between the LPS and TUDCA-3S groups (Figure 4C). S100A8 was prioritized for subsequent validation because it showed prominent differential expression, has established relevance to myeloid inflammatory responses, and was linked to the concurrently enriched IL-17 signaling pathway. Other altered genes and pathways remain candidates for future investigation.

These transcriptomic findings provided the rationale for focused evaluation of S100A8 and IL-17A in the subsequent experiments.

S100A8-Associated IL-17A Signaling is Involved in the TUDCA-3S-Associated Macrophage Phenotype

To validate the role of S100A8 in macrophage polarisation and its relationship with TUDCA-3S, we first confirmed by RT-qPCR that TUDCA-3S treatment significantly increased S100A8 mRNA expression (Figure 4D). An S100A8 knockdown model was established using small interfering RNA (siRNA), and the most efficient siRNA was selected for subsequent experiments (Figure 4E). Using the AP model as a reference control, siS100A8 treatment was applied to both the AP and TUDCA-3S groups.

Flow cytometry showed that the proportion of M1 macrophages was significantly reduced in both the siS100A8+AP and siS100A8+TUDCA-3S groups compared with the AP group (Figure 5A). Western blot analysis further confirmed a significant decrease in M1 marker protein expression following S100A8 knockdown (Figure 5B). Similarly, RT-qPCR results showed that siS100A8 significantly reduced the mRNA expression of M1-related markers (Figure 5C).

Figure 5.

Detection of M1 macrophage markers: (A) Representative flow-cytometry plots and quantitative analysis of M1-associated macrophage populations after S100A8 knockdown using the CD11c-PE/CD206-APC and CD86-PE/CD206-APC staining combinations. PE fluorescence is shown on the x-axis and APC fluorescence on the y-axis. The lower-right CD11c⁺CD206⁻ or CD86⁺CD206⁻ population was used as the corresponding M1-associated population. Identical analysis and quadrant settings were applied across the Ctrl, AP, siS100A8+AP, and siS100A8+TUDCA-3S groups within each analysis. (B) Western Blot was used to detect the expression of CD11c and CD86 proteins on macrophages treated with siS100a8. (C) RT-qPCR was used to detect the expression of CD11c and CD86 mRNA on macrophages treated with siS100a8. (D) The mRNA expression of IL-17A was detected by RT-qPCR. (E) Western Blot was used to detect the expression of IL-17A protein on macrophages treated with siS100a8. (F) RT-qPCR was used to detect the mRNA expression of IL-17A in macrophages treated with S100a8. Data are presented as mean ± SD from three independent biological experiments (n=3).

To investigate the effect of S100A8 knockdown on the IL-17 pathway, we measured the mRNA expression of IL-17A and found that it progressively increased from the control to the AP and TUDCA-3S groups (Figure 5D). Furthermore, both the protein and mRNA levels of IL-17A were significantly lower in the siS100A8+TUDCA-3S group compared with the AP group (Figure 5E and F).

Flow cytometry showed attenuation of the M1-associated macrophage phenotype after S100A8 knockdown. Consistently, Western blotting and RT-qPCR demonstrated reduced CD11c and CD86 expression after S100A8 knockdown. These findings support involvement of S100A8 in the pro-inflammatory macrophage phenotype observed under the present experimental conditions. IL-17A expression was higher under LPS-stimulated and TUDCA-3S-exposed conditions than in Ctrl cells. S100A8 knockdown reduced IL-17A protein and mRNA expression under TUDCA-3S-exposed conditions, supporting an association between S100A8 and IL-17A-related inflammatory signaling in this model. However, because IL-17-specific inhibition, rescue experiments, and downstream pathway analyses were not performed, these findings do not establish a complete causal S100A8/IL-17 signaling pathway.

Metabolic Changes Associated with TUDCA-3S Exposure

Untargeted metabolomic analysis was performed to characterize metabolic differences associated with TUDCA-3S exposure in the LPS-stimulated macrophage model. ClassyFire annotation showed that lipids and lipid-like molecules constituted the largest category of differential metabolites, followed by organic heterocyclic compounds, organic acids, and their derivatives (Figure 6A). Hierarchical clustering of lipid and lipid-like metabolites, including fatty acids, glycerophospholipids, glycerolipids, and prenol lipids (Figure 6B), showed distinct profiles between the LPS and TUDCA-3S groups, indicating metabolic differences associated with TUDCA-3S exposure.

Figure 6.

Non-targeted metabolomics and lipid metabolomics analysis: (A) The proportion of differential metabolites in AP group and TUDCA-3S group. (B) Cluster heat maps of differential metabolites in lipids and lipids. (C) Pathways enriched by lipid differential metabolites in AP group and TUDCA-3S group. (D) Cluster heat map of lipid differential metabolites.

TUDCA-3S Exposure is Associated with Altered Glycerophospholipid and Triglyceride Profiles

To delineate the link between TUDCA-3S and lipid metabolism in AP progression, lipidomic analysis was conducted on the AP and TUDCA-3S intervention groups. KEGG pathway enrichment of differential lipid metabolites highlighted pathways such as adipocytokine signalling, regulation of lipolysis in adipocytes, and sphingolipid signalling (Figure 6C). Further analysis confirmed that TUDCA-3S induced a lipid-metabolic imbalance characterised by disruption of triglyceride (TG) and phosphatidylcholine (PC) metabolism (Figure 6D).

These lipidomic findings identify metabolic differences associated with TUDCA-3S exposure but do not establish whether lipid remodeling is a cause, mediator, or consequence of the inflammatory macrophage phenotype.

Discussion

The current management of AP primarily involves early supportive care and close monitoring.²⁴ Although serum amylase and lipase are commonly used for AP diagnosis, they do not reliably predict disease severity.²⁵ Close monitoring within the first 48–72 h is crucial to prevent progression to SAP and associated complications.² Therefore, there is an urgent need to identify reliable early biomarkers for SAP and elucidate the pathological mechanisms driving AP progression, which could provide novel potential therapeutic targets. In the present study, we investigated the role of TUDCA-3S in AP progression and its association with SAP. Our targeted bile-acid metabolomic analysis of patient serum samples revealed that TUDCA-3S levels were significantly elevated across different AP subtypes, demonstrating its potential as a biomarker for AP diagnosis and severity prediction. These ROC findings should be regarded as exploratory because the study was conducted in a single-center cohort without external validation. In addition, the small number of SAP cases limited the feasibility of adequately powered multivariable adjustment. The different TUDCA-3S patterns observed in BAP, AAP, and HLAP suggest possible etiological heterogeneity. Importantly, re-analysis of the baseline dataset showed that the overall etiological composition was comparable among MAP, MSAP, and SAP (P=0.9998), indicating that the severity-group differences were not explained by a major imbalance in etiological composition. Nevertheless, the present sample size is insufficient for robust severity-by-etiology interaction analyses. Therefore, larger etiology-stratified cohorts will be required to determine whether the association between circulating TUDCA-3S and AP severity is consistent within individual etiological categories.

Bile acids (BAs) are steroids molecules synthesized from cholesterol in the liver. They facilitate lipid absorption, maintain cholesterol homeostasis, regulate lipid synthesis and enterohepatic circulation, and influence hepatic lipid, glucose, and energy homeostasis.²⁶ Furthermore, BAs modulate inflammatory responses via receptors such as the farnesoid X receptor (FXR) and G protein-coupled bile acid receptor TGR5.^27,28 SAP, characterised by a systemic inflammatory response and persistent organ failure, is closely associated with BA dysregulation. Conjugated BAs alleviate pancreatic and systemic inflammation by activating the TGR5 pathway and suppressing NLRP3 inflammasome activation.²⁹ Conversely, BAs can induce abnormal Ca²⁺ release into acinar cells, triggering AP initiation and exacerbation.³⁰

Tauroursodeoxycholic acid (TUDCA) is an endogenous bile acid produced in the liver by conjugation of taurine with ursodeoxycholic acid.³¹ Recently, TUDCA has attracted significant attention owing to its role in alleviating endoplasmic reticulum (ER) stress.¹⁵ Studies indicate that TUDCA mitigates AP by suppressing ER-stress-associated pro-apoptotic signalling and reducing the activation of the unfolded protein response (UPR).¹⁶ It also exerts cytoprotective effects by enhancing enzyme secretion in acinar cells, alleviating ER stress, and inhibiting subsequent apoptosis.³² Tauroursodeoxycholic acid-3-sulphate (TUDCA-3S) is a sulphated derivative of TUDCA. Using an LPS-induced inflammatory macrophage model, we found that TUDCA-3S promoted macrophage polarisation from M0 to M1, as shown by flow cytometry, RT-qPCR, and Western blot analyses. Given that BA-induced M1 polarisation worsens AP outcomes,¹² these findings indicate that TUDCA-3S exacerbates AP progression, leading to a more severe inflammatory response. The opposing effects of TUDCA and TUDCA-3S in AP are likely due to sulfation, which abolishes the anti-inflammatory properties of secondary bile acids.^17,18 To the best of our knowledge, this is the first study to report the proinflammatory role of TUDCA-3S in AP through M1 macrophage activation. Because TUDCA-3S is a bile-acid-derived metabolite, its circulating concentration may be influenced by biliary obstruction, hepatic dysfunction, altered enterohepatic circulation, disease etiology, systemic inflammation, and metabolic disturbance. Therefore, the observed clinical association should not be interpreted as evidence of an independent pathogenic effect. Larger prospective studies with standardized sampling and adequate adjustment for these potential confounders are required.

To elucidate the mechanism underlying TUDCA-3S-induced M1 macrophage activation, we performed transcriptome sequencing, which revealed that TUDCA-3S targeted S100A8 to regulate M1 polarisation. S100A8, a Ca²⁺-binding protein of the S100 family, is actively released during inflammation and plays a key role in regulating inflammatory responses by promoting leukocyte recruitment and inducing cytokine secretion.^33,34 For instance, in a TRPA1-knockout dilated cardiomyopathy model, elevated S100A8 levels promoted M1 polarisation, which was attenuated by S100A8 inhibition.³⁵ S100A8 is a pathogenic molecule with proinflammatory and immunosuppressive properties that are crucial in the pathogenesis of conditions such as sepsis.³⁶ Our experimental data corroborated this finding. S100A8 knockdown significantly reduced the proportion of M1 macrophages in the TUDCA-3S treatment group, indicating that silencing S100A8 attenuated inflammatory progression in SAP. KEGG pathway analysis linked S100A8 to the IL-17 signalling pathway. IL-17 is a versatile proinflammatory cytokine essential for host defence, tissue repair, and the pathogenesis of inflammatory diseases.³⁷ The IL-17 receptor family, comprising five subunits, transduces signals through specific intracellular domains,³⁸ allowing different IL-17 members to exert distinct functions in various tissues.³⁹ IL-17A is involved in autoimmunity, inflammation, and tumour development.^40,41 Previous studies support the S100A8/IL-17 axis. For instance, S100A8/A9 promotes Mycoplasma pneumoniae-induced epithelial cell apoptosis via the IL-17 pathway,⁴² whereas S100A8/A9 deficiency exacerbates psoriasiform dermatitis by enhancing IL-17A and IL-17F production, which is ameliorated by anti-IL-17 treatment.⁴³ Transcriptomic analysis identified S100A8 and IL-17-related signaling as candidates associated with TUDCA-3S exposure. In the present experiments, S100A8 knockdown attenuated M1-associated markers and reduced IL-17A expression under TUDCA-3S-exposed conditions, supporting involvement of S100A8-associated IL-17A signaling. To our knowledge, the present study provides initial evidence linking TUDCA-3S exposure with S100A8- and IL-17A-related inflammatory macrophage changes. However, the molecular events linking TUDCA-3S exposure to increased S100A8 expression remain unresolved. Potential mechanisms may involve receptor-dependent signaling, cellular stress responses, or metabolic remodeling. Moreover, because IL-17-specific blockade, rescue experiments, and downstream pathway analyses were not performed, a complete causal S100A8/IL-17 pathway has not been established. These mechanisms should be regarded as hypotheses for future investigation rather than established upstream events in the current study.

Given the growing interest in BA metabolism, we employed LC-MS analysis to investigate the relationship between TUDCA-3S and AP progression. Untargeted metabolomic profiling revealed that lipids and lipid-like molecules were the most significantly altered metabolites between the AP and TUDCA-3S groups, including fatty acids, glycerophospholipids, glycerolipids, and prenol lipids. Fatty acyls are closely associated with inflammation, playing direct roles in regulating immune and inflammatory processes.^44,45 Subsequent lipidomic analysis identified disrupted phosphatidylcholine (PC) and triglyceride (TG) metabolism as key features. PC plays a vital role in modulating systemic inflammation, potentially alleviating LPS-induced systemic inflammation via the gut–brain axis.⁴⁶ PC can also promote adipose-tissue lipolysis by upregulating hormone-sensitive lipases and downregulating other lipases and adipogenesis markers,⁴⁷ it can alleviate synovial inflammation by downregulating the TLR/NF-κB and MAPK signalling pathways.⁴⁸ Hypertriglyceridaemia is a major risk factor for AP. High TG levels are associated with increased serum free fatty acids (FFAs), which can inhibit mitochondrial complex function in pancreatic acinar cells, induce mitochondrial calcium overload and cytokine release, cause tissue damage, and impair pancreatic ductal function, thereby contributing to AP initiation and exacerbation.^49–51 Interestingly, cholesterol-lowering therapy reduces both serum TG and S100A8/A9 levels, thereby attenuating inflammation.⁵² Another study showed that reducing TG and LDL levels also decreased hepatic expression of pro-inflammatory factors such as S100A8/A9.⁵³ In our study, we found through untargeted metabolomics and lipidomics identified prominent changes in lipid and lipid-like metabolites associated with TUDCA-3S exposure, including TG- and PC-related profiles. Lipid metabolism and inflammatory macrophage activation are closely interconnected, providing biological plausibility for an association between the metabolic and inflammatory changes observed in this study. However, the present data do not establish directionality between lipid remodeling and inflammatory activation. The observed TG- and PC-related changes should therefore be regarded as exploratory metabolic signatures that warrant further mechanistic investigation rather than as validated targets for risk stratification or treatment.

Our study has several important limitations. First, this was a single-center exploratory cohort without external validation. Although several inflammatory, hepatic, metabolic, and renal variables differed among severity groups, the SAP subgroup included only 14 patients. Given the limited number of SAP events relative to the number of potentially relevant and correlated covariates, a conventional multivariable logistic-regression model would carry a substantial risk of overfitting and unstable effect estimates. Accordingly, the present clinical analyses should be considered exploratory and unadjusted, and TUDCA-3S should not be interpreted as an independent predictor of SAP. Second, serum sampling was not standardized to an identical time point after symptom onset, and circulating TUDCA-3S may be influenced by biliary, hepatic, etiological, metabolic, and inflammatory factors. Third, the cellular findings were obtained using a single LPS-stimulated RAW264.7 macrophage model, which represents inflammatory macrophage activation rather than the complete pathophysiology of AP. No independent macrophage model or in vivo validation was performed. Fourth, the TUDCA-3S concentration used in the cellular experiments was associated with reduced cell viability. Consequently, a contribution of nonspecific cellular stress cannot be completely excluded. Fifth, the original experimental design did not include additional lower-dose experiments, TUDCA-3S-alone treatment, parent-TUDCA comparison, or separate matched vehicle-control experiments. These omissions limit the ability to distinguish TUDCA-3S-specific effects from concentration- or vehicle-related effects. Sixth, although S100A8 knockdown supports involvement of S100A8, IL-17-specific blockade, rescue experiments, and downstream pathway analyses were not performed. Therefore, a complete causal S100A8/IL-17 signaling pathway has not been established. Seventh, the lipidomic findings are exploratory and do not establish the direction of the relationship between inflammatory activation and lipid remodeling. These limitations necessitate further independent clinical validation, additional macrophage models, pathway-specific intervention studies, and in vivo investigation.

Conclusion

In conclusion, serum TUDCA-3S was elevated in AP and was associated with disease severity in this single-center cohort, although its discriminatory performance across AP severity categories was limited to moderate. In an LPS-stimulated RAW264.7 macrophage model, TUDCA-3S exposure was associated with enhanced M1-related inflammatory features, increased S100A8 and IL-17A expression, and altered lipid-metabolic profiles. S100A8 knockdown attenuated macrophage inflammatory markers and IL-17A responses, supporting involvement of S100A8-associated signaling in this in vitro phenotype. However, the present study does not establish that TUDCA-3S causally drives SAP progression in vivo or that a complete causal S100A8/IL-17 signaling pathway has been demonstrated. Further independent clinical validation, pathway-specific intervention studies, additional macrophage models, and in vivo studies are required.

Funding Statement

This study was supported by National Natural Science Foundation of China (Grant no. 82300737). This study was supported by the 2024 Scientific Research Development Plan Project of the Clinical Medical College-Affiliated Hospital of North Sichuan Medical College (Grant no. 2024MPZK020). The Doctoral Start-up Fund of The Affiliated Hospital of North Sichuan Medical College (Grant no. 2023GC010). The Nanchong City Science and Technology Bureau and School Strategic Cooperation Project (Grant no. 22SXQT0110).

Data Sharing Statement

Complete data of this study are available upon reasonable request from the corresponding author.

Author Contributions

Linfeng Yang: Conceptualization, Methodology, Visualization, Writing – review and editing. Duan Yan: Investigation, Data curation, Writing – review and editing. Jun Yu: Investigation, Data curation, Writing – review and editing. Yi Liu: Methodology, Formal analysis, Writing – review and editing. Yiming Zhang: Methodology, Formal analysis, Writing – review and editing. Pengsheng Yi: Conceptualization, Methodology, Visualization, Writing – review and editing. Dawei Deng: Conceptualization, Methodology, Visualization, Writing – review and editing. All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.