ST3Gal1 modulates intestinal barrier function and impacts human ulcerative colitis

Introduction

Inflammatory bowel diseases (IBDs) comprise chronic as well as recurrent gastrointestinal conditions. A recent study has reported on the notable increase in the global prevalence of IBD, predominantly characterized by Crohn's disease (CD) and ulcerative colitis (UC) (1). Impaired intestinal barrier integrity is a contributing factor in the pathogenesis of IBD and is closely associated with its pathophysiology (2,3). Despite this, there is a characteristic lack of therapeutic options that specifically enhance mucosal barrier function and promote its regeneration in IBD, which is likely owing to a limited understanding of the underlying biological pathways and regulatory molecules involved (4). Although various potential therapeutic approaches have been developed for improving mucosal barrier function (4,5), to the best of our knowledge, no regenerative strategies have yet been developed specifically for IBD.

Glycosylation is a post-translational modification affecting proteins, lipids and RNA, which can profoundly influence the structural and functional characteristics of the modified molecules (6–8). This modification is important for regulating a variety of biological mechanisms, including protein maturation, transport and immune modulation (9). Glycosylation is also implicated in the pathophysiology of severe diseases, such as during carcinogenesis, and inflammation; a previous study also demonstrated its association with IBD (10). In healthy individuals, protein glycoforms exhibit substantial stability over time; however, they tend to undergo notable alterations in response to pathological conditions, particularly during inflammatory states. For example, patients diagnosed with IBD exhibit increased expression levels of short-chain O-glycans in the intestine compared with healthy individuals, accompanied by alterations in the expression of terminal glycan structures (for example, increased sialylation, enhanced fucosylation, reduced sulfation and dysregulated Lewis antigens). These changes adversely affect the integrity of the intestinal mucus layer, impede glycan-lectin interactions, disrupt the dynamic balance of the intestinal microbiota and impair the functionality of the intestinal mucosal immune system (11).

Different sialylation modifications are catalyzed by various sialyltransferases. However, any alterations in the expression levels of sialyltransferases change their physiological functions. For example, a decrease in sialyltransferase ST6GalNAc1 (ST6)-catalyzed α2,6-sialylation in mice results in impaired mucus barrier function, gut microbiota dysbiosis and increased susceptibility to intestinal inflammation (12). The α2,6-sialylation catalyzed by ST6Gal1 promotes the activation of CD4+ T cells and induces the development of UC (13), whereas α2,8-sialylation catalyzed by ST8Sia4 facilitates the metastasis of breast cancer cells (14). However, the role of ST3Gal1-catalyzed α2,3-sialylation in intestinal inflammation remains to be fully elucidated.

The upregulation of ST3Gal1 augments the sialylation of O-glycan Tn and catalyzes its conversion into sialyl-Tn. This modification introduces an α2,3-linkage between Gal and β1,3-GalNAc, which is frequently observed in various cancer types, including breast cancer and hematological malignancies, where the upregulation of sialyltransferases leads to premature termination of aberrant mucin-type O-glycan chains (15,16). Notably, mucin-type O-linked glycosylation is not restricted to mucins; this modification also occurs in an array of cell-surface glycoproteins (17). Emerging evidence has highlighted that glycosylation in intestinal epithelial cells (IECs) is associated with specific glycosyltransferases; however, the functional role of ST3Gal1-catalyzed α2,3-sialylation, a key glycosylation event mediated by this glycosyltransferase, in the regulation of intestinal barrier function has often been neglected in relevant research. Dysregulation of glycosylation exacerbates intestinal inflammation by damaging the intestinal barrier integrity, interfering with glycan-lectin interactions, disrupting intestinal microbiota dynamics and impairing mucosal immunity (18–20).

Given that ST3Gal1-catalyzed α2,3-linkage formation at the end of N-glycan or O-glycan is associated with carcinogenesis, as well as a host of other diseases, it is necessary to evaluate prognostic factors, identify predictive biomarkers and explore possible therapeutic targets associated with ST3Gal1 activity, particularly in IBD (15,21). However, there is limited research on ST3Gal1 and its impact on the function of the intestinal barrier.

The present study aimed to explore the potential role of ST3Gal1-catalyzed α2,3-linkage in IBD, along with its underlying mechanisms. The present experimental study employed an in vitro triple-cell co-culture system to replicate both the healthy and inflamed states of the human intestine to investigate the impact of ST3Gal1-mediated α2,3-sialylation on the integrity of intestinal mucosal barrier function. Notably, in the present model, each cell type embodied distinct roles: i) Caco-2 cells formed a monolayer that mimicked IECs; ii) HT29-MTX-E12 cells secreted mucus and acted as a surrogate for intestinal goblet cells; and iii) THP-1 cells, upon induction with phorbol 12-myristate 13-acetate (PMA), differentiated into macrophage-like cells, simulating intestinal inflammatory cells (22–24). The present model adhered to the principles of the 3R framework, which are replacement, reduction and refinement of animal model usage, thus underscoring the increasing importance of in vitro models in intestinal health research (25).

Materials and methods

Cell culture

The colorectal cancer cell line HT-29-MTX was sourced from Beijing Bohui Innovation Biotechnology Co., Ltd., whereas the THP-1 human monocytic leukemia and Caco-2 human colorectal adenocarcinoma cell lines were obtained from The Cell Bank of Type Culture Collection of The Chinese Academy of Sciences. All cells were authenticated by short tandem repeat profiling and were cultured in RPMI-1640 medium (Merck KGaA) supplemented with 10% heat-inactivated fetal bovine serum (HyClone™; Cytiva). Cell cultures were maintained in a humidified incubator at 37°C with 5% CO2.

Cell culture model

To establish an IEC monolayer, Caco-2 and HT29-MTX-E12 cells were co-seeded at a total density of 8×104 cells/well into the apical compartment of 24-well Transwell inserts (pore size, 0.4 µm; Wuxi NEST Biotechnology Co., Ltd.) at a ratio of 9:1, based on this ratio and total cell density, the seeding number of Caco-2 cells was 7.2×104 cells/well, and that of HT29-MTX-E12 cells was 8×103 cells/well, followed by a differentiation period of 19 days. Concurrently, the culture medium was replaced three times a week to maintain optimal growth conditions. Additionally, THP-1 cells were stimulated to differentiate into macrophage-like cells by applying PMA (100 ng/ml; Beyotime Biotechnology) at 37°C with 5% CO2 for 48 h in 24-well plates. Subsequently, the differentiated macrophage-like THP-1 cells were cultured in the basolateral compartment of the same Transwell plate. The IEC monolayer and macrophage-like THP-1 cells were co-cultured for 24 h under standard conditions (37°C, 5% CO2) to simulate the intestinal mucosa in vitro.

ST3Gal1 overexpression (ST3Gal1-OE) assay

Lentiviral vectors designed for the overexpression of ST3Gal1 were procured from Shanghai GenePharma Co., Ltd. The ST3Gal1-OE lentiviral vector was constructed using the LV5(EF-1α/CopGFP&Puro) backbone, based on the 3rd-generation lentiviral system, with a working concentration of 1×1010 plaque-forming units (PFU)/ml. The lentiviral particles were produced by transfecting 293T cells (American Type Culture Collection). For transfection, a total of 10 µg plasmids was used, and the molar ratio of lentiviral expression plasmid (ST3Gal1-OE)/packaging plasmid/envelope plasmid was 4:3:1. The transfection was performed using Lipofectamine® 3000 Transfection Reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions, under the conditions of 37°C and 5% CO2 for a duration of 48 h to facilitate lentivirus production. The negative control for overexpression experiments was an empty LV5 vector (Shanghai GenePharma Co., Ltd.), used at the same concentration (1×1010 PFU/ml).

Following the precise guidelines set forth by the manufacturer, infection of the cells in the triple co-culture model was performed within the Transwell system, with lentiviral vectors added to both the upper and lower chambers. For lentiviral transduction, polybrene (Shanghai GenePharma Co., Ltd.) was added to the culture medium at a final concentration of 5 µg/ml to enhance infection efficiency, and the lentiviral particles were used at a multiplicity of infection of 10. The infected cells were then incubated at 37°C for 24 h. Subsequently, 4 µg/ml puromycin (MilliporeSigma; Merck KGaA) was used for the initial selection of stably infected cell lines for 7 days, followed by maintenance with 2 µg/ml puromycin thereafter. These cell lines were subsequently classified as OE-Ctr/IEC (negative control group) and ST3Gal1-OE/IEC groups. Cells were collected on day 9 post-infection (after the completion of the 1-day infection plus a 7-day selection process), and the successful overexpression of ST3Gal1 was confirmed through reverse transcription-quantitative PCR (RT-qPCR) and western blot analysis.

ST3Gal1 interference assay

Adenoviral particles designed for interfering with the ST3Gal1 enzyme were obtained from Shanghai GenePharma Co., Ltd. The adenoviral constructs were based on the ADV1(U6/CMV-GFP) vector (2nd-generation adenoviral expression system), which separately harbored three distinct ST3Gal1-targeting short hairpin RNA (shRNA; ST3Gal1-sh) inserts or a non-targeting scrambled negative control shRNA (sh-Ctr). The specific interference target sequences and full-length sense/antisense shRNA fragments were as follows: ST3Gal1 sh1, target sequence 5′-GCACCATTTCCCACACCTACA-3′, sense strand′-AATTCGCACCATTTCCCACACCTACATTCAAGAGATGTAGGTGTGGGAAATGGTGCTTTTTTG-3′, antisense strand 5′-GATCCAAAAAAGCACCATTTCCCACACCTACATCTCTTGAATGTAGGTGTGGGAAATGGTGCG-3′; ST3Gal1 sh2, target sequence 5′-GCATCCTCTCGGTCATCTTCT-3′, sense strand 5′-AATTCGCATCCTCTCGGTCATCTTCTTTCAAGAGAAGAAGATGACCGAGAGGATGCTTTTTTG-3′, antisense strand 5′-GATCCAAAAAAGCATCCTCTCGGTCATCTTCTTCTCTTGAAAGAAGATGACCGAGAGGATGCG-3′; ST3Gal1 sh3, target sequence 5′-GGTTCGATGAGAGGTTCAACC-3′, sense strand 5′-AATTCGGTTCGATGAGAGGTTCAACCTTCAAGAGAGGTTGAACCTCTCATCGAACCTTTTTTG-3′, antisense strand 5′-GATCCAAAAAAGGTTCGATGAGAGGTTCAACCCTCTCTTGAAGGTTGAACCTCTCATCGAACCG-3′; and sh-Ctr, target sequence 5′-GTTCTCCGAACGTGTCACGT-3′, sense strand 5′-GATCCGTTCTCCGAACGTGTCACGTTTCAAGAGAACGTGACACGTTCGGAGAACTTTTTTG-3′, antisense strand 5′-AATTCAAAAAAGTTCTCCGAACGTGTCACGTTCTCTTGAAACGTGACACGTTCGGAGAACG-3′. The adenoviral particles were produced by transfecting 293T cells. For transfection, 10 µg plasmids were used per 10-cm culture dish, and the molar ratio of adenoviral expression plasmid (ST3Gal1-sh or sh-Ctr)/packaging plasmid/envelope plasmid was 1:1:1. The transfection was performed using Lipofectamine 3000 Transfection Reagent, according to the manufacturer's instructions, at 37°C in a humidified atmosphere containing 5% CO2 for 48 h before collecting the adenoviral supernatant.

The triple co-culture cells cultured within the Transwell system were infected with these adenoviral particles in accordance with the manufacturer's guidelines, with adenoviral particles added to both the upper and lower chambers of the Transwell system. The infection procedure was consistent with that used for the infection of overexpression lentiviral vectors aforementioned. These cell lines were subsequently classified as sh-Ctr/IEC and ST3Gal1-sh/IEC groups. The working concentration of adenovirus particles carrying interfering fragments was 1×1010 PFU/ml, and that of negative control adenovirus particles was 1×1010 PFU/ml. For adenoviral infection, the MOI was set at 40. Specifically, cells were collected at 48 h post-infection, and the efficiency of the interference was assessed through RT-qPCR and western blot analysis.

Induction of inflammation in the IEC monolayer

To establish an in vitro UC cell model, ST3Gal1-sh/IEC or ST3Gal1-OE/IEC and their controls were treated with 2% DSS (MilliporeSigma) and incubated in a cell culture incubator at 37°C for 4 days (26,27). The supernatants were collected by centrifuging the cell cultures at 3,000 × g and 4°C for 10 min, and were employed for subsequent experiments.

RT-qPCR

RNA was isolated and PCR was carried out following the methodologies detailed in our earlier study (28). Briefly, total RNA was extracted from triple co-culture cells, including ST3Gal1-sh, ST3Gal1-OE and their respective negative control cells, using the RNA Isolation Kit (MilliporeSigma; Merck KGaA). RNA purity was assessed by evaluating the A260/A280 ratio (1.8–2.0) with a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Inc.). RNA integrity was verified through 1% agarose gel electrophoresis and by using an Agilent 2100 Bioanalyzer (Agilent Technologies, Inc.), obtaining an RNA integrity number >8.0. cDNA synthesis was performed using the PrimeScript™ RT Reagent Kit with gDNA Eraser (Takara Bio Inc., Japan), according to the manufacturer's protocol. For qPCR, the SYBR® Green Premix Pro Taq HS qPCR Kit (Hunan Accurate Biotechnology Co., Ltd.) was used, with SYBR Green as the PCR fluorophore. Thermocycling conditions were as follows: Pre-denaturation at 95°C for 3 min, followed by 40 cycles at 95°C for 30 sec and 60°C for 30 sec. β-actin was used as the internal reference gene for normalization. Quantification of the qPCR results was performed using the 2−ΔΔCq method (29), and result analysis was performed using Bio-Rad CFX Manager software (version 3.1; Bio-Rad Laboratories, Inc.). Table I displays the sequences of PCR primers utilized in the present experimental investigation.

Table I. Primer sequences used in reverse transcription-quantitative PCR.

Table I.

Primer sequences used in reverse transcription-quantitative PCR.

Primer	Sequence	Product length, bp
ST3Gal1		74
Forward	5′-TTCCGGGAGCTGGGAGATAA-3′
Reverse	5′-CTCACCACCCACTCCAAGTC-3′
MUC2		165
Forward	5′-GGGGAGTGCTGTAAGAAGTGTGA-3′
Reverse	5′-GTTGGAGACGGACGAGATGAG-3′
TFF3		143
Forward	5′-CTCCAGCTCTGCTGAGGAGT-3′
Reverse	5′-CAGGGATCCTGGAGTCAAAG-3′
CDX2		85
Forward	5′-CAGGACGAAAGACAAATATC-3′
Reverse	5′-GATGTAGCGACTGTAGTG-3′
β-actin		132
Forward	5′-GTGATCTCCTTCTGCATCCTGT-3′
Reverse	5′-CCACGAAACTACCTTCAACTCC-3′
IL-1β		153
Forward	5′-CAGAAGTACCTGAGCTCGCC-3′
Reverse	5′-AGATTCGTAGCTGGATGCCG-3′
IL-6		166
Forward	5′-TACCCCCAGGAGAAGATTCC-3′
Reverse	5′-AGTGCCTCTTTGCTGCTTTC-3′
IL-8		210
Forward	5′-GTGCAGTTTTGCCAAGGAGT-3′
Reverse	5′-ACTTCTCCACAACCCTCTGC-3′
STAT3		237
Forward	5′-CATCCTGAAGCTGACCCAGG-3′
Reverse	5′-TCCTCACATGGGGGAGGTAG-3′

[i] MUC2, mucin 2; TFF3, trefoil factor 3; CDX2, homeobox protein CDX-2.

Western blotting

The procedures for protein extraction and detection were carried out in accordance with the methodologies outlined in our previous publication (26). Briefly, total protein was isolated from the triple co-culture cells, including ST3Gal1-sh cells, ST3Gal1-OE cells and their corresponding negative control cells. Table II presents the information regarding the manufacturer, cat. no. and dilution ratios for the primary and secondary antibodies utilized in the present study.

Table II. Antibodies used for western blotting.

Table II.

Antibodies used for western blotting.

A, Primary antibodies
Name	Company	Cat. no.	Dilution
Rabbit polyclonal IgG to ST3Gal1	Invitrogen; Thermo Fisher	PA5-42726	1:1,000
	Scientific, Inc.
Rabbit monoclonal IgG to CDX2	Abcam	ab76541	1:1,000
Rabbit monoclonal IgG to p-STAT3	Abcam	ab76315	1:1,000
Mouse monoclonal IgG to MUC2	Abcam	ab11197	1:1,000
Rabbit Polyclonal IgG to TFF3	Proteintech Group, Inc.	23277-1-AP	1:1,000
Rabbit polyclonal IgG to STAT3	Proteintech Group, Inc.	10253-2-AP	1:500
Rabbit polyclonal IgG to α-tubulin	Proteintech Group, Inc.	11224-1-AP	1:1,000
B, Secondary antibodies
HRP-conjugated Affinipure Goat	Proteintech Group, Inc.	SA00001-2	1:10,000
Anti-Rabbit IgG (H+L)
HRP-conjugated Affinipure Goat	Proteintech Group, Inc.	SA00001-1	1:10,000
Anti-Mouse IgG (H+L)

[i] CDX2, homeobox protein CDX-2; p-, phosphorylated.

ELISA

Cell culture supernatants from each group were collected by centrifuging the cell cultures at 3,000 × g and 4°C for 10 min. The levels of multiple inflammatory mediators, such as IL-1β, IL-6 and IL-8, present in the harvested cell supernatants were measured in accordance with the instructions outlined in the ELISA kits specific to these inflammatory factors (Beijing Solarbio Science & Technology Co., Ltd.): IL-1β (cat. no. SEKH-0002), IL-6 (cat. no. SEKH-0013), and IL-8 (cat. no. SEKH-0016).

Trans-epithelial electrical resistance (TEER) assay

Cells from each group were plated at a density of 1×105 cells/well into the upper chambers of a 24-well Transwell plate with a polyester membrane (pore size, 0.4 µm; Costar; Corning, Inc.). For the Transwell co-culture assay, both the upper and lower chambers were initially supplemented with DMEM containing 10% FBS. Cells were co-cultured at 37°C with 5% CO2 for 14 days, and 2 days before TEER measurement, the media in both chambers were replaced with serum-free DMEM. The resistance values of different cell models, including inflammatory and non-inflammatory cell models, as well as their ST3Gal1 knockdown or overexpression cell models, were measured using a resistance meter. The relative TEER values were calculated for each group by comparison with the control.

FITC-dextran permeability assay

Subsequently, the permeability of different cell models was studied. Notably, the setup of Transwell inserts used in this permeability assay (other than the addition of FITC-dextran) was consistent with that of the aforementioned protocol for TEER measurement. FITC-dextran (4 kDa; MilliporeSigma) was added to the apical compartment of Transwell inserts, with a final concentration of 1 mg/ml. These cell models were incubated at 37°C for 2 h. Subsequently, a 100-µl aliquot was aspirated from the lower chamber, and the fluorescence value was measured using a microplate reader (Bio-Rad Laboratories, Inc.) at an excitation wavelength of 492 nm and an emission wavelength of 520 nm. The relative fluorescence values were calculated for each group by comparison with the control.

Bioinformatics analysis

For ST3Gal1 expression analysis, datasets including human normal samples and UC samples (GDS3119) (30), as well as UC lesional and non-lesional samples (GSE107499) were obtained from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geoprofiles/?term=) and analyzed. To investigate the relationship between ST3Gal1 and the severity score of UC, relevant clinical data were extracted from the GEO dataset GSE92415 (31) for analysis. The GEO dataset GDS3859 (32) was used to analyze changes in the mRNA levels of ST3Gal1 in DSS-treated mice. Additionally, the GEO dataset GSE107499 was used to analyze the expression changes of ST3Gal1, as well as the expression profiles of inflammatory mediators, intestine-associated secretory proteins and signaling pathway-related proteins in lesional and non-lesional tissues of UC. Furthermore, the relationships among these factors and proteins in the GSE107499 dataset, specifically within the lesional tissues of UC, were investigated. Dataset processing and statistical analysis were performed using GEO2R (https://www.ncbi.nlm.nih.gov/geo/geo2r/), an official GEO tool. In the present study, this tool was specifically utilized to extract and compare the expression levels of pre-specified target genes (including ST3Gal1, MUC2, TFF3, CDX2, IL-1β, IL-6, IL-8 and STAT3) between experimental groups (for example, UC lesional mucosa vs. UC non-lesional mucosa, and healthy human colon mucosa vs. UC colon mucosa).

Statistical analysis

GraphPad Prism 6.0 software (Dotmatics) was utilized for the graphical representation and statistical evaluation of diverse data obtained during the experiment. Data are presented as box and whisker plots (displaying median and interquartile range), scatter plots or as the mean ± SD from at least three independent replicates, with unpaired two-tailed t-test used for comparisons between two groups and Pearson's correlation test used for assessing linear correlations. In addition, one-way ANOVA with Tukey's HSD post hoc multiple comparisons test was applied for multi-group analyses. P<0.05 was considered to indicate a statistically significant difference.

Results

ST3Gal1 expression in intestinal mucosa from patients with UC and the mouse DSS-induced colitis model is related to the inflammatory status of colitis

The present study compared the mRNA levels of ST3Gal1 in normal human colon mucosa and the mucosa from patients with UC obtained from the GDS3119 dataset in the GEO. The mRNA expression levels of ST3Gal1 in the mucosal tissue of individuals diagnosed with UC were significantly higher than those in healthy colon mucosa (derived from healthy control individuals) (P<0.05; Fig. 1A). Furthermore, mRNA levels of ST3Gal1 in the inflamed mucosal regions of patients with UC were significantly higher than those observed in normal colon mucosa. Additionally, ST3Gal1 expression levels were significantly higher in inflamed UC intestinal mucosa than in non-inflamed UC intestinal mucosa (P<0.01; Fig. 1B). Furthermore, the mRNA expression levels of ST3Gal1 in non-lesional mucosa were significantly lower than those in lesional mucosa (GSE107499; P<0.01; Fig. 1C). Additionally, in patients with UC, ST3Gal1 mRNA within the inflamed mucosa exhibited progressive, mostly significant upregulation associated with an increasing modified Mayo score (GSE92415; P<0.01; Fig. 1D).

Figure 1. ST3Gal1 mRNA levels in intestinal mucosa from patients with UC and a mouse DSS-induced colitis model are associated with inflammatory states. (A) Comparison of ST3Gal1 mRNA levels between human normal colon mucosa from healthy controls (n=5) and colonic mucosa from patients diagnosed with UC (n=21) in the GDS3119 dataset from the GEO database. (B) ST3Gal1 mRNA levels in human normal colonic tissues from healthy controls (n=5), inflamed colonic tissues from patients with UC and visible macroscopic inflammation (n=8), and non-inflamed colonic mucosal tissues from patients with UC without macroscopic signs of inflammation (n=13) in the GEO dataset GDS3119. (C) ST3Gal1 mRNA levels in non-lesional mucosa (n=44) and lesional mucosa (n=75) from different patients with UC, obtained from the GEO dataset GSE107499. (D) ST3Gal1 mRNA levels in the normal mucosa (n=21) from healthy controls and the inflamed mucosa (n=162) from patients with UC with varying modified Mayo scores in the GEO dataset GSE92415. (E) ST3Gal1 mRNA levels in colon tissues from mice treated with DSS to induce colitis (independent, non-overlapping mice) divided into four groups: Before DSS administration (n=5), and at 2, 4 and 6 days post-DSS treatment (n=6/group). Data were obtained from the GEO dataset GDS3859. Data in A, C and D are presented as box and whisker plots, whereas data in B and E are presented as the mean ± SD. Statistical significance was determined using specific tests for each subpart based on group comparisons: (A and C) Unpaired t-test and (B, D and E) one-way analysis of variance followed by Tukey's post-hoc test. *P<0.05 and **P<0.01. DSS, dextran sulfate sodium; GEO, Gene Expression Omnibus; UC, ulcerative colitis.

The present study further analyzed the mRNA levels of ST3Gal1 in the mucosa of the DSS-induced mouse colitis model. The obtained ST3Gal1 mRNA levels were classified into four groups based on DSS usage: The normal group, classified as mRNA levels obtained prior to DSS administration, and those obtained 2, 4 and 6 days after DSS feeding. ST3Gal1 mRNA expression in the mucosa of mice 6 days after DSS feeding was significantly higher than that in all other groups analyzed (GDS3859; P<0.01; Fig. 1E).

Establishment of an IEC monolayer and ST3Gal1-knockdown or ST3Gal1-OE models

The present study first prepared an IEC monolayer model through a co-culture of three distinct cell types, and then obtained ST3Gal1-knockdown and ST3Gal1-OE IEC monolayer cell models using either an ADV1(U6/CMV-GFP) vector and an ST3Gal1 interference sequence insert or a ST3Gal1-OE lentiviral vector. The third interference sequence (ST3Gal1-sh3) had the most efficient inhibitory effect on ST3Gal1 expression in the IEC monolayer model, named ST3Gal1-sh3/IEC for further experiments (P<0.01; Fig. 2A and C). A ST3Gal1-OE assay showed that the ST3Gal1-OE group exhibited a significant increase in ST3Gal1 expression at both mRNA and protein levels in the IEC monolayer model compared with those in the untreated and overexpression control groups (P<0.01; Fig. 2B and D). This model was designated ST3Gal-OE/IEC for further experiments.

Figure 2. Establishment of ST3Gal1-interfered IEC and ST3Gal1-overexpressed IEC models, and the schematic representation of IEC monolayer grouping. (A) ST3Gal1 mRNA levels in IEC, sh-Ctr/IEC and ST3Gal1-interfered IEC groups 1–3. (B) ST3Gal1 mRNA levels in IEC, OE-Ctr/IEC and ST3Gal1-OE/IEC groups. (C) ST3Gal1 protein levels in IEC, sh-Ctr/IEC and ST3Gal1-interfered IEC groups. (D) ST3Gal1 protein levels in IEC, OE-Ctr/IEC and ST3Gal1-OE/IEC groups. (E) Establishment of IEC monolayers containing sh-Ctr, ST3Gal1-sh3, OE-Ctr and ST3Gal1-OE. The stable triple culture models were treated with fresh complete medium supplemented with 2% DSS for 4 days to obtain inflamed models. Data are presented as the mean ± SD. Statistical significance was assessed using one-way analysis of variance followed by Tukey's post-hoc test. **P<0.01. IEC, intestinal epithelial cell; sh, short hairpin; Ctr, control; OE, overexpression; PMA, phorbol 12-myristate 13-acetate; DSS, dextran sulfate sodium.

The present study used IEC monolayer models to replicate both healthy and inflamed states of the human intestine. To establish an inflamed IEC monolayer model, the aforementioned infection models were administered fresh complete medium enriched with 2% DSS for 4 days. The resulting inflamed IEC monolayer model was employed for investigating intestinal barrier integrity and inflammatory mediator secretion (Fig. 2E).

ST3Gal1 alters the barrier function of the IEC monolayer

The IEC monolayer barrier was assessed by measuring TEER. The TEER levels in the ST3Gal1-sh3/IEC group were significantly higher than those in the IEC and sh-Ctr/IEC groups after stimulation with 2% DSS (P<0.01; Fig. 3A). By contrast, no notable differences in the TEER levels were observed among the non-inflamed IEC monolayers of the stable triple culture models without 2% DSS treatment (P>0.05; Fig. 3A). The TEER levels of the inflamed ST3Gal1-OE/IEC group were significantly lower compared with those in the IEC or Ctr/IEC groups (P<0.01; Fig. 3B). By contrast, no notable differences were observed among the non-inflamed IEC monolayers of the stable triple culture models (P>0.05; Fig. 3B). The impact of modifications in ST3Gal1 expression on the permeability of IEC monolayers was measured by performing a trans-epithelial permeability assay using FITC-dextran. The assays demonstrated a notable decrease in the FITC-dextran permeability of the inflamed ST3Gal1-sh3/IEC population compared with that in the IEC and sh-Ctr/IEC groups (P<0.01; Fig. 3C). However, no significant difference was observed among the non-inflamed stable triple culture models (P>0.05; Fig. 3C). Notably, a significant increase in the permeability of the inflamed ST3Gal1-OE/IEC group was noted compared with that in the IEC and Ctr/IEC groups (P<0.01; Fig. 3D). By contrast, the non-inflamed triple culture models did not show any significant differences in permeability (P>0.05; Fig. 3D). Therefore, it may be suggested that ST3Gal1 interference in IECs enhanced the barrier function of the IEC monolayer. However, enforced expression of the ST3Gal1 gene in IECs damaged the barrier function of the IEC monolayer.

Figure 3. ST3Gal1 expression in triple culture models modulates the epithelial barrier of the IEC monolayers. (A) TEER levels in the IEC monolayers containing sh-Ctr/IEC or ST3Gal1-sh3/IEC with 2% DSS stimulation (inflamed triple culture) or without 2% DSS treatment (non-inflamed triple culture). (B) TEER levels in the inflamed (DSS-treated) and non-inflamed IEC monolayers containing OE-Ctr/IEC or ST3Gal1-OE/IEC. (C) Results of the FITC-dextran permeability assay indicated that there was a significant reduction in the inflamed IEC monolayer containing ST3Gal1-sh3/IEC. (D) There was a significant increase in the permeability of the inflamed IEC monolayer containing ST3Gal1-OE/IEC. Data are presented as the mean ± SD. Statistical significance was assessed using one-way analysis of variance followed by Tukey's post-hoc test. **P<0.01. TEER, trans-epithelial electrical resistance; IEC, intestinal epithelial cell; sh, short hairpin; Ctr, control; OE, overexpression; DSS, dextran sulfate sodium.

ST3Gal1 expression in the IEC monolayer alters MUC2, TFF3 and CDX2 expression, as well as STAT3 phosphorylation

Owing to alterations in the intestinal epithelial barrier caused by altered ST3Gal1 expression in the IEC monolayer, the present study detected intestinal mucus barrier-associated MUC2 and TFF3, goblet cell differentiation-associated CDX2 expression and inflammation-associated STAT3 phosphorylation in IEC monolayers. The present study found that ST3Gal1 knockdown in IECs significantly increased the mRNA levels of MUC2, TFF3 and CDX2, but significantly decreased the mRNA levels of STAT3 in the inflamed stable triple culture model compared with those in the control groups (P<0.01; Fig. 4A-D). No significant differences in MUC2, TFF3, CDX2 and STAT3 mRNA levels were observed among the non-inflamed stable triple culture models (P>0.05; Fig. 4A-D). Overexpression of ST3Gal1 in IECs significantly reduced the mRNA levels of MUC2, TFF3 and CDX2 in the inflamed stable triple culture model compared with those in the control groups (P<0.05 or P<0.01; Fig. 4E-H), and the mRNA expression levels of STAT3 were decreased in the inflamed stable triple culture model compared with those in the control groups (P<0.05; Fig. 4H); however, no significant difference was observed when compared with the OE-Ctr group (P>0.05; Fig. 4H). No significant differences were observed in the mRNA levels of MUC2, TFF3, CDX2 and STAT3 among the non-inflamed stable triple-culture models (P>0.05; Fig. 4E-H).

Figure 4. ST3Gal1 expression in inflamed IEC monolayers alters MUC2, TFF3, CDX2, STAT3 and p-STAT3 expression. ST3Gal1 knockdown in ST3Gal1-sh3/IEC significantly increased (A) MUC2, (B) TFF3 and (C) CDX2, but significantly reduced (D) STAT3 mRNA levels in the inflamed IEC monolayer. ST3Gal1 OE in ST3Gal1-OE/IEC significantly reduced (E) MUC2, (F) TFF3 and (G) CDX2 mRNA levels in the inflamed IEC monolayer. (H) ST3Gal1 OE in ST3Gal1-OE/IEC slightly decreased STAT3 mRNA levels in the inflamed IEC monolayer. (I) Western blotting showed no changes in protein expression in non-inflamed knockdown samples. (J) Western blotting showed that MUC2, TFF3 and CDX2 protein levels were increased, and p-STAT3 expression was decreased in the inflamed IEC monolayer comprising ST3Gal1-sh3/IEC. (K) Western blotting showed no notable changes in protein expression in non-inflamed OE samples. (L) Western blotting showed that MUC2, TFF3 and CDX2 protein levels were decreased, and p-STAT3 expression was increased in the inflamed IEC monolayer comprising ST3Gal1-OE/IEC. Data are presented as the mean ± SD. Statistical significance was assessed using one-way analysis of variance followed by Tukey's post-hoc test. *P<0.05 and **P<0.01. IEC, intestinal epithelial cell; sh, short hairpin; Ctr, control; OE, overexpression; MUC2, mucin 2; TFF3, trefoil factor 3; CDX2, homeobox protein CDX-2; p-, phosphorylated.

The present study also detected the protein expression levels of MUC2, TFF3, CDX2, phosphorylated (p)-STAT3 and STAT3 in both the inflamed and non-inflamed stable triple culture models. In the inflamed stable triple culture model comprising ST3Gal1-sh3/IEC, the protein levels of MUC2, TFF3 and CDX2 were increased, whereas p-STAT3 levels were notably decreased compared with those in the control groups (Fig. 4J). However, no notable difference was noted in the MUC2, TFF3, CDX2, STAT3 and p-STAT3 protein levels of the non-inflamed stable triple culture models (Fig. 4I). Additionally, the protein levels of MUC2, TFF3 and CDX2 were notably decreased in the inflamed stable triple culture model comprising ST3Gal1-OE/IEC, whereas p-STAT3 levels were increased compared with those in the control groups (Fig. 4L). However, there were no marked differences observed in the MUC2, TFF3, CDX2, STAT3 and p-STAT3 protein levels of the non-inflamed stable triple culture overexpression model (Fig. 4K).

ST3Gal1 expression is correlated with TFF3 and CDX2 mRNA levels in lesional mucosa obtained from patients with UC

To support the relationship between ST3Gal1 expression and MUC2, TFF3 and CDX2 mRNA levels in patients with UC, the expression dataset GSE107499 was downloaded, and the mRNA levels of TFF3 and CDX2 in the non-lesional and lesional mucosa from patients with UC, as well as their correlations with ST3Gal1 mRNA expression, were subsequently compared. As illustrated in Fig. 5A, the mRNA expression levels of MUC2 in the affected lesional mucosal tissue were significantly elevated compared with those in the non-lesional mucosa (P<0.05). By contrast, the TFF3 mRNA levels in the lesional mucosa showed no significant difference when compared with those in the non-lesional mucosa of patients with UC (P>0.05; Fig. 5B). Notably, the mRNA levels of CDX2 in the lesional mucosa were significantly lower than those in non-lesional mucosa (P<0.01; Fig. 5C). The correlation analysis revealed that the mRNA expression levels of ST3Gal1 in the lesional mucosal tissue exhibited a negative correlation with the mRNA levels of TFF3 (R=−0.4883; P<0.0001; Fig. 5E) and CDX2 (R=−0.4797; P<0.0001; Fig. 5F) in patients diagnosed with UC. However, no notable correlation was observed between ST3Gal1 and of MUC2 expression in lesional tissues of patients with UC (R=−0.2792; P<0.05; Fig. 5D).

Figure 5. Comparison of (A) MUC2, (B) TFF3 and (C) CDX2 mRNA levels between non-lesional mucosa (n=44) and lesional mucosa (n=75) from distinct patients with ulcerative colitis. Correlations between (D) MUC2, (E) TFF3 and (F) CDX2 mRNA levels and ST3Gal1 expression. Data in A-C are presented as box and whisker plots, and those in D-F as scatter plots. Statistical significance for group comparisons in A-C was assessed using unpaired t-test. Correlation analyses in D-F were performed using Pearson's correlation coefficient. *P<0.05 and **P<0.01. Data were obtained from the GSE107499 dataset from the Gene Expression Omnibus database. *P<0.05 and **P<0.01. NS, not significant; MUC2, mucin 2; TFF3, trefoil factor 3; CDX2, homeobox protein CDX-2.

ST3Gal1 expression in the IEC monolayer alters the expression of inflammatory mediators

Supernatants from the inflamed and non-inflamed stable triple culture models comprising ST3Gal1-sh3/IEC and ST3Gal1-OE/IEC were collected to detect the levels of inflammatory mediators. The protein concentrations of inflammatory mediators were quantified via ELISA, whereas the mRNA levels of inflammatory mediators in cell lysates were assessed via RT-qPCR. Notably, knockdown of ST3Gal1 significantly reduced the mRNA expression levels of IL-1β, IL-6 and IL-8 in the inflamed intestinal IEC monolayer, compared with that in the control groups (P<0.01; Fig. 6A-C). This analysis, however, revealed no significant differences in the mRNA levels of IL-1β, IL-6 and IL-8 among the non-inflamed IEC monolayers (P>0.05; Fig. 6A-C). However, the overexpression of ST3Gal1 in ST3Gal1-OE/IEC cells significantly elevated the mRNA levels of IL-1β, IL-6 and IL-8 within the inflamed IEC monolayer compared with the control groups (P<0.01; Fig. 6D-F). By contrast, the non-inflamed IEC monolayer exhibited no significant differences in the expression of these cytokines (P>0.05; Fig. 6D-F). Additionally, a significant decrease in the protein concentrations of IL-1β (720.51±61.01 pg/ml), IL-6 (461.20±98.06 pg/ml) and IL-8 (4,325.37±579.25 pg/ml) was observed in supernatants derived from the inflamed IEC monolayer containing ST3Gal1-sh3/IEC when compared with the control groups (P<0.01; Fig. 6G-I). No significant differences were noted in the protein concentrations of IL-1β (318.43±24.54 pg/ml), IL-6 (404.52±20.55 pg/ml) and IL-8 (3,151.23±156.78 pg/ml) in ST3Gal1-sh3/IEC compared with in the control groups in the supernatants obtained from the non-inflamed IEC monolayer (P>0.05; Fig. 6G-I). In addition, the protein concentrations of IL-1β (1,863.41±62.25 pg/ml), IL-6 (1,797.57±94.93 pg/ml) and IL-8 (18,319.25±1,893.01 pg/ml) were significantly higher in the supernatants derived from the inflamed IEC monolayer containing ST3Gal1-OE/IEC compared with those from the control groups (P<0.01; Fig. 6J-L). Consistent with previous observations in the present study, there were no significant differences in the concentrations of IL-1β (282.01±48.33 pg/ml), IL-6 (408.14±28.22 pg/ml) and IL-8 (4,203.54±433.62 pg/ml) the aforementioned cytokines in the supernatants obtained from the non-inflamed IEC monolayer (P>0.05; Fig. 6J-L). Therefore, it may be inferred that the expression of ST3Gal1 in IECs markedly modified the expression, and subsequent release of IL-1β, IL-6 and IL-8 in inflamed triple culture models.

Figure 6. ST3Gal1 expression in the IEC monolayer alters the levels of inflammatory mediators. ST3Gal1 knockdown in ST3Gal1-sh3/IEC significantly reduced the mRNA levels of the inflammatory mediators (A) IL-1β, (B) IL-6 and (C) IL-8 from the inflamed IEC monolayer supernatant. ST3Gal1 OE in ST3Gal1-OE/IEC significantly increased the mRNA levels of the inflammatory mediators (D) IL-1β, (E) IL-6 and (F) IL-8 in the supernatant of the inflamed IEC monolayer. Protein levels of the inflammatory mediators (G) IL-1β, (H) IL-6 and (I) IL-8 were decreased in the supernatant of inflamed ST3Gal1-sh3/IEC. Protein levels of the inflammatory mediators (J) IL-1β, (K) IL-6 and (L) IL-8 were increased in the supernatant of inflamed ST3Gal1-OE/IEC. Data are presented as the mean ± SD, and statistical significance was assessed using one-way ANOVA followed by Tukey's post-hoc test. **P<0.01. IEC, intestinal epithelial cell; sh, short hairpin; Ctr, control; OE, overexpression.

ST3Gal1 levels are correlated with IL-1β, IL-6, IL-8 and STAT3 mRNA levels in lesional mucosa obtained from patients with UC

The present study conducted a comparative analysis of the mRNA expression levels of IL-1β, IL-6, IL-8 and STAT3 between the non-lesional and lesional mucosal tissues of patients diagnosed with UC, which was obtained from the GSE107499 dataset, alongside the relationships of cytokine expression with ST3Gal1 mRNA expression levels. The present results demonstrated that the mRNA levels of IL-1β (P<0.01; Fig. 7A), IL-6 (P<0.01; Fig. 7B), IL-8 (P<0.01; Fig. 7C) and STAT3 (P<0.01; Fig. 7D) in the lesional mucosa were significantly elevated compared with those in the non-lesional mucosa. Furthermore, correlation analyses revealed positive correlations between ST3Gal1 mRNA expression and the expression levels of the cytokines IL-1β (R=0.5675; P<0.0001; Fig. 7D), IL-6 (R=0.4117; P=0.0002; Fig. 7E), IL-8 (R=0.5765; P<0.0001; Fig. 7F) and STAT3 (R=0.4079; P=0.0003; Fig. 7H) in the lesional mucosa of patients with UC.

Figure 7. Comparison of the expression of inflammatory mediators (A) IL-1β, (B) IL-6 and (C) IL-8, as well as (D) STAT3 between non-lesional mucosa (n=44) and lesional mucosa (n=75) from distinct patients with ulcerative colitis. Correlations between (E) IL-1β, (F) IL-6, (G) IL-8 and (H) STAT3 mRNA levels and ST3Gal1 expression. Data were obtained from the GSE107499 dataset from the Gene Expression Omnibus database. Data in A-D are presented as box and whisker plots, and those in E-H as scatter plots. Statistical significance for group comparisons in A-D was assessed using the unpaired t-test; correlation analyses in E-H were performed via Pearson's correlation coefficient. **P<0.01.

Discussion

The results of the present study indicated a strong relationship between ST3Gal1 expression in the intestinal epithelium and the barrier function of IECs, as well as with the secretion of inflammatory mediators within the context of intestinal inflammation. ST3Gal1 knockdown in the IEC monolayer increased the barrier function, whereas its overexpression resulted in intestinal barrier function deterioration. Overexpression of ST3Gal1 in the IEC monolayer reduced the expression of intestinal mucus barrier-associated MUC2, TFF3 and CDX2, as well as the expression of inflammation-associated p-STAT3. By contrast, overexpression of ST3Gal1 elevated the levels of inflammatory mediators (IL-1β, IL-6, and IL-8), which are important for maintaining intestinal mucosa homeostasis (33). The DSS-induced colitis model is linked to the inflammatory status of the colonic tissue, as evidenced by the dynamic association between disease progression and the expression profiles of key molecules: Specifically, inflammatory factors (IL-1β, IL-6, IL-8), goblet cell secretory proteins (MUC2, TFF3) and signaling pathway-related proteins (STAT3, p-STAT3). This is supported by the bioinformatics analysis of the GEO dataset GDS3859, where ST3Gal1 mRNA expression in mouse colonic mucosa was significantly upregulated with prolonged DSS treatment (peaking at 6 days post-administration), concurrent with the exacerbation of colonic inflammation. Additionally, in the in vitro inflamed triple-culture model induced by 2% DSS, the expression patterns of the aforementioned molecules were consistently altered in response to ST3Gal1 knockdown or overexpression, further confirming that the DSS-induced model recapitulates the intrinsic association between colonic inflammatory status and the expression of these functional proteins in UC. ST3Gal1 facilitates the formation of α2,3-sialic acid residues and contributes to the modification of glycan ends on proteins that possess α2,3-sialic acid residues (34). Dysregulation of ST3Gal1 within IECs can result in the abnormal expression of sialic acid, which is a unique carbohydrate that anchors glycoproteins and glycolipids to the cell membrane through α2,3-sialic acid residue modifications (35). The present study elucidated the role of ST3Gal1 in various IEC monolayers, including inflamed and non-inflamed states, to identify knowledge gaps, and explore its potential application as a novel diagnostic and therapeutic target.

Sialyltransferases are key enzymes in the regulatory mechanisms of various life processes, including cell signaling, cellular recognition, interactions between cells and pathogens, and cancer metastasis (36–38). Among these enzymes, the ST3Gal family stands out as one of the most important enzyme families, comprising six distinct members in both mice and humans: ST3Gal1-6 (39). The ST3Gal family catalyzes the synthesis and attachment of α2,3-linked sialic acid residues to the terminal galactose residues of glycans in proteins through a process known as sialylation. The evolution of distinct human ST3Gal family members, which establish specific α2,3-linkages, suggests a degree of functional specialization; however, the roles of α2,3-linkage in health and disease remain insufficiently investigated (39). Prior research has documented increased sialylation in IBD, as well as in other autoimmune conditions and instances of acute inflammation (40). Notably, α2,6- and α2,3-sialylation of glycans with extensively branched structures appear to escalate during inflammatory responses in CD and UC, respectively, which are mediated by α2,6-sialyltransferases and α2,3-sialyltransferases (41,42). Similarly, ST6 mutations in animal models have been shown to cause impairment of the intestinal mucus barrier, dysbiosis of the intestinal microbiota and increased susceptibility to intestinal inflammation (11,43). Therefore, it can be suggested that ST6 is important for inhibiting bacterial translocation and mitigating inflammation within the intestinal mucus barrier (44). Consequently, the role of α2,3-linkage, catalyzed by ST3Gal1 sialyltransferases, in the pathogenesis of patients with IBD and its underlying mechanisms remain largely unexplored.

IBD refers to a persistent inflammatory disorder affecting the intestines, which is influenced by a combination of genetic predispositions, immune responses and environmental elements (45). To facilitate research on IBD, 66 distinct animal models have been developed at present, which can be classified into chemical treatment models, cell migration models, gene mutation models and gene transfer models (46). However, in some mouse models with few IBD-related genes, the specific pathogenic mechanisms of IBD are more complex than previously predicted (46). To investigate the pathogenic mechanisms and therapeutic approaches of IBD under healthy conditions and inflammatory microenvironments, respectively, researchers have established tri-culture cell models of healthy and inflamed states in vitro. Based on these models, various complex in vitro culture systems and derivative systems have been developed, including primary cell organoid culture systems, multicellular co-culture systems and chip device systems (24,47,48). The present study established an in vitro intestinal barrier simulation model by co-culturing three cell lines: i) HT29-MTX-E12 cells, which exhibit goblet cell characteristics and have a mucus-secreting ability, can be used to simulate the intestinal mucus layer in vitro; ii) Caco-2 cells with epithelial cell properties, which were seeded in the upper layer; and iii) differentiable THP-1 inflammatory cells, which were seeded in the basal side. Subsequently, this model was treated with 2% DSS to induce an inflammatory state. The effect of ST3Gal on the integrity of the IEC monolayer barrier was then evaluated by measuring TEER and FITC-dextran permeability. Additionally, as reported in previous studies, the secretion levels of relevant inflammatory factors and the expression of intestinal barrier-related proteins were compared between inflammatory and non-inflammatory states (49–51). The tri-culture system designed to simulate intestinal conditions streamlined the present model, enhancing its robustness and reproducibility (24). The epithelial Transwell cultures incorporating Caco-2 and HT29-MTX-E12 cell lines exhibited traits that closely resemble those of human intestinal mucosal epithelium. Furthermore, the inclusion of immune-competent THP-1 cells contributed to the immune mechanisms associated with the development of IBD (21). Notably, both the presence of these immune effector cells and the persistence of inflammatory processes substantially influenced the sensitivity of the IEC monolayer.

However, there remains a lack of α2,3-sialylation-linked target molecules bridging ST3Gal1 and intestinal mucus barrier-associated MUC2, TFF3 and CDX2, as well as inflammation-associated molecules, such as IL-1β, IL-6, IL-8, STAT3 and p-STAT3. Further study is required to identify ST3Gal1-mediated molecules with post-translation α2,3-sialylation at their glycan termini. This may help identify novel target molecules to modulate the intestinal epithelial barrier function important for the pathogenesis of human IBD.

The present study provided a new direction for the treatment of UC and other IBDs and showed that ST3Gal1 can act as a potential therapeutic target. Overexpression of ST3Gal1 was directly associated with impaired intestinal barrier function, elevated expression of the pro-inflammatory cytokines IL-1β, IL-6 and IL-8, and increased disease activity in patients with UC, supported by in vitro and clinical data: i) In vitro, ST3Gal1 overexpression in the inflamed triple-culture model showed significantly reduced TEER, increased FITC-dextran permeability, downregulated expression of barrier-protective proteins MUC2, TFF3 and CDX2, and upregulated p-STAT3, as well as elevated mRNA and protein levels of IL-1β, IL-6 and IL-8. ii) Clinically, bioinformatics analysis of GEO datasets (GSE107499 and GSE92415) revealed that ST3Gal1 mRNA expression was significantly higher in lesional vs. non-lesional mucosa inpatients with UC, it was positively associated with modified Mayo scores (a measure of UC disease activity), and it was positively correlated with the mRNA levels of IL-1β, IL-6, IL-8 and STAT3 in UC lesional tissues. By contrast, ST3Gal1 knockdown upregulated the barrier-protective molecules (MUC2, TFF3 and CDX2) and reduced p-STAT3 expression, thereby repairing the mucosal barrier and alleviating inflammation. Compared with traditional broad-spectrum immunosuppressive therapies, ST3Gal1-targeted interventions may act precisely on the glycosylation pathway, thereby potentially minimizing off-target effects. Additionally, the characteristic high expression of ST3Gal1 in inflamed mucosal tissues enables it to serve as a potential biomarker for assessing disease activity or treatment response. Future studies should validate the efficacy of ST3Gal1 inhibitors in animal models or explore their synergistic effects with existing drugs, such as mesalamine, to provide more precise therapeutic options for IBD.

While the Caco-2/HT29-MTX-E12/THP-1 triple-culture model recapitulated key intestinal mucosal features and followed the 3R principles, it failed to fully replicate the in vivo intestinal microenvironment, lacking fibroblasts, endothelial cells and diverse gut microbiota, which are all regulators of mucosal barrier function and IBD inflammation (52). Furthermore, immortalized cell lines such as those used in the present study may not reflect the phenotypes or functions of primary cells of patients with IBD, thus limiting translational relevance.

The clinical analyses of the present study, which relied on retrospective GEO data, also had inherent constraints: i) The lack of access to detailed patient metadata, such as age and treatment history, or long-term paired samples, which would preclude ST3Gal1-prognosis/treatment response analyses; and ii) limited sample sizes of subgroups, such as the UC non-inflamed mucosa group, reduced the power of the correlation analysis for ST3Gal1 with MUC2, which as a key intestinal mucosal barrier component and goblet cell marker, is closely implicated in UC pathogenesis (53). To address these limitations, future studies should: i) Validate the findings of the present study by using primary IEC-organoid models with microbiota or stromal cells for improved physiological relevance; ii) use prospective, well-characterized clinical cohorts to clarify the clinical utility of ST3Gal1; and iii) apply glycoproteomic methods, such as lectin affinity chromatography-mass spectrometry, to identify ST3Gal1-mediated α-2,3-sialylation target molecules to strengthen the mechanistic and translational value of the present results.

In conclusion, the present study provided evidence that ST3Gal1-catalyzed α2,3-linkage formation in IEC may be closely associated with intestinal barrier function. This relationship could be mediated through the distinct influence of ST3Gal1 on the expression of intestinal barrier-associated proteins, including MUC2, TFF3 and CDX2, which are important for preserving the integrity of the intestinal barrier. Additionally, ST3Gal1 modulated the expression of associated inflammatory mediators and transcription factors, including the expression of IL-1β, IL-6, IL-8 and STAT3, which serve important roles in regulating intestinal mucosal inflammation, thereby further linking ST3Gal1 to barrier function.

Acknowledgements

Not applicable.

Funding

The present work was supported by the Natural Science Foundation of Chongqing (grant no. cstc2020jcyj-msxmX1094).

Availability of data and materials

The data generated in the present study may be requested from the corresponding author.

Authors' contributions

YT, YL, YS, LR and LL were responsible for performing the experiments, and collecting and analyzing the data; specifically, they performed in vitro cell culture (triple-culture model establishment, ST3Gal1 overexpression/knockdown), functional assays (TEER, FITC-dextran permeability), molecular experiments (RT-qPCR, western blotting, ELISA), and bioinformatics data processing (GEO dataset analysis). JY substantially contributed to study conception and design, provided critical project oversight (experimental direction, data validation), and led manuscript drafting and critical revision for key intellectual content. RW critically revised the manuscript by interpreting the biological significance of core findings (for example, linking ST3Gal1-mediated α2,3-sialylation to intestinal barrier dysfunction in UC) and standardizing the academic expression of results (refining data description and mechanistic discussion logic). The manuscript was authored by JY and RW, with JY providing project oversight. JY and YT confirm the authenticity of all the raw data. All authors have read, critically revised, and approved the final manuscript.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References