Melatonin attenuates glandular cystitis via modulation of THBS1 and TLR4/p38 MAPK/STAT3 signaling Authors: Binghai Chen, Sun Gong, Jialong Zhang, Yang Zhang, Haomin Wang Published: 2025 ABSTRACT Melatonin (MEL) is a natural hormone widely found in animals and humans. A multitude of studies have illustrated MEL exerts anti-inflammatory impacts on diverse organs. However, it remains unclear how MEL plays a role in glandular cystitis (GC) at present. We investigated how MEL affected GC through thrombospondin-1 (THBS1). MEL was employed to treat human bladder epithelial cells (SV-HUC-1) subsequent to their stimulation by lipopolysaccharide (LPS). Subsequently, the expression levels of inflammatory factors in human bladder epithelial cells under different treatment methods were measured. We perfused LPS into the bladders of rats once every three days for eight consecutive weeks. The rats belonging to the treatment group underwent a four-week treatment regimen with MEL (10 mg/kg). Ultimately, we meticulously detected the expression levels of Toll-like receptor 4 (TLR4), Nuclear Factor κB p65 Subunit (NF-κB p65), p38 Mitogen-Activated Protein Kinase (p38 MAPK) as well as Signal Transducer and Activator of Transcription 3 (STAT3). Additionally, to gain deeper insights, further proteomic analysis was conducted, which was expected to disclose more detailed molecular mechanisms. In vitro, we found that LPS had a remarkable effect on human bladder epithelial cells, significantly augmenting the expression levels of several key inflammatory factors. Specifically, the expressions of interleukin 1β (IL-1β), interleukin 6 (IL-6), interleukin 10 (IL-10), and tumor necrosis factor alpha (TNF-α) were noticeably enhanced after LPS stimulation. Meanwhile, the expression of key molecules in related signaling pathways—including TLR4, NF-κB p65, p38 MAPK, p-p38, STAT3 and p-STAT3—also increase significantly. Similar results were observed in a rat model of GC as well. After treatment with MEL, the expressions of the inflammatory factors decreased, and the expressions of various signaling pathways also decreased accordingly. According to our proteomics analysis, THBS1 might be an important protein involved in the progression of GC. MEL is able to reduce the expression of THBS1, leading to the alleviation of the inflammation of GC. MEL reduces the inflammatory response in GC by regulating the TLR4/p38 MAPK/STAT3 axis. THBS1 plays a critical role in promoting the progression of GC, and this finding provides a new perspective for understanding the molecular mechanisms of GC or identifying potential therapeutic targets for GC.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-025-28128-w. Introduction Glandular cystitis (GC) is a proliferative disease of the bladder, and the etiology has not been fully clarified1. Upon chronic inflammation or bacterial infection2, the urothelium underwent chemical changes, thus forming GC. The diagnosis of GC necessitates the utilization of cystoscopy and pathological examination3.The pathological changes of GC were as high as 71% in the continuous sections of the bladder, especially in the triangular area of the bladder4. The proliferating urothelium invaginates into the shallow lamina propria, which is called the brunn’s nests. Most researchers believe that GC develops from brunn’s nests. The characteristic structure of GC is the central glandular differentiation within brunn’s nests. The surface of the glandular lumen is lined with cuboidal or columnar epithelial cells, which constitute the glandular cavity, while several layers of urothelium surround it. This represents typical GC and is also the most common type of this condition. GC has two manifestations: typical and enteric, the latter having a similar glandular structure in the lamina propria but containing a profusion of mucus-secreting goblet cells in the intimal epithelium5,6. In 1950, a study by Immergut and Cottler linked GC to the development of bladder adenocarcinoma7. Since then, various studies have suggested that GC is a precancerous lesion of adenocarcinoma8,9. GC is regarded as a precancerous lesion10–12. In subsequent studies, it was found that enteric GC has a greater possibility of causing bladder cancer. However, some scholars also believe that GC is a chronic, non-progressive histological lesion and has no association with various adenocarcinomas13,14. In a large number of subsequent clinical studies, it is also believed that intestinal metaplasia is not the main risk factor for bladder cancer15. Our study focuses specifically on infection-associated GC, aiming to explore the regulatory mechanisms underlying its progression. Our experimental model employs LPS stimulation to recapitulate the pathological features of infection-driven GC, including bladder mucosal hyperplasia, inflammatory cell infiltration, and upregulation of pro-inflammatory mediators—phenotypes that align with the histological and molecular hallmarks of human infection-associated GC observed in clinical specimens. While we acknowledge that this LPS-based approach does not encompass non-infectious GC subtypes, such as those mediated by autoimmune responses or neurogenic inflammation, its relevance lies in addressing a critical knowledge gap: the lack of targeted mechanistic insights into infection-triggered GC. Toll-like receptor 4 (TLR4) is one of the cell transmembrane receptors and pathogen pattern recognition receptors discovered in recent years16. TLR4 is a bridge connecting innate immunity and adaptive immunity and is referred to as the “gate” of the inflammatory response. TLR4 plays a vital and essential part in triggering innate immunity17. TLR4 binds to pathogen-associated molecular patterns, and this interaction triggers the activation of downstream signaling cascades. These cascades then act through specific pathways to ultimately drive the overexpression of various inflammatory mediators. Among these pathways, the canonical Nuclear Factor κB (NF-κB) pathway—with NF-κB p65 as its central component—functions as a core regulatory hub that amplifies the inflammatory response18. TLR4 is also closely related to the p38 Mitogen-Activated Protein Kinase (p38 MAPK) pathway and the Signal Transducer and Activator of Transcription 3 (STAT3) pathway19,20. TLR4 activation can activate NF-κB p65, while also triggering the activation and phosphorylation of p38 MAPK and STAT3. Notably, TLR4 functions as a potent initiator in signal transduction. It can set off complex signaling cascades, and these cascades intricately engage a multitude of intracytoplasmic and nuclear transcription factors. Among these transcription factors, NF-κB p65 is a crucial one. Upon activation, NF-κB p65 translocates into the nucleus. It then binds to the promoters of pro-inflammatory cytokine genes and directly drives the transcription of these genes. TLR4 activation initiates a series of intricate molecular events. TLR4 binds to specific family members, which serve as crucial bridges connecting TLR4 to serine/threonine kinases. These kinases play a central role in the signaling cascade. Through enzymatic reactions, they can precisely regulate the expression levels of various immunomodulatory cytokines. This regulatory mechanism is of fundamental significance, as it dictates the dynamics of the immune response and finely tunes the balance between immune activation and immune suppression. Melatonin (MEL) exhibits a broad distribution pattern, being ubiquitously present in several key regions of the central nervous system, including the pituitary gland, hypothalamus, cerebral cortex, and hippocampus21,22. MEL is involved in many regulatory processes, including biorhythms, prevention of inflammation, anti-oxidation, metabolism, immunity and reproduction23. Combining findings from previous studies and the present research, it can be observed that MEL exerts an inhibitory effect on the expression or activity of inflammatory factors, thereby interfering with the progression of various bladder injuries and GC24–2624–26. Meanwhile, MEL can also play a certain anti-inflammatory role by inhibiting the TLR4/p38 MAPK/STAT3 axis. THBS1 is a large extracellular matrix glycoprotein that can mediate the interactions between cells or between the extracellular matrix and a variety of proteins, playing a crucial role in controlling extracellular matrix remodeling27. In the course of our in-depth study, we made a notable discovery that the expression level of THBS1 was conspicuously boosted in GC. Moreover, MEL demonstrates a remarkable therapeutic effect by precisely targeting THBS1. Through this targeted action, it successfully curtails the TLR4/p38 MAPK/STAT3 axis, which are pivotal in the inflammatory cascade. As a result, MEL effectively reduces the severity of GC-related inflammation. Materials and methods Regents LPS (ST1470-50 mg) was from Beyotime Biotechnology; MEL (A600605) was from Sangon Biotech (Shanghai) Co., Ltd; Human bladder epithelial cells were from Najing Wanmuchun Biotechnology Co., Ltd; ELISA reagents (IL-1β:KE00021, IL-6: KE00139, IL:10 KE00170, TNF-α: KE00154) were from Proteintech Group, Inc; Trizol lysis buffer (R0016) was from Beyotime Biotechnology; Real-time PCR reagent (Q321-02/03) was from Vazyme Biotech Co., Ltd; TLR4 inhibitor (CAS No.: 243984-11−4) was from Sigma-Aldrich Corporation; Transfection reagent (TL201-01) was from Vazyme Biotech Co., Ltd; THBS1-specific siRNA and THBS1-specific lentivirus were from Sangon Biotech (Shanghai) Co., Ltd. The details of all the antibodies utilized in the study are presented as follows: TLR4 (Proteintech, 66350-1-lg, RRID: AB_2881730), NF-κB p65 (Proteintech, 66535-1-Ig, RRID: AB_2881898), STAT3 (Proteintech,10253-2-AP, RRID: AB_2302876), Phospho-STAT3 (Affinity, AF3293, RRID: AB_2810278), MAPK p38 (Proteintech, 14064-1-AP, RRID: AB_2878007), Phospho-p38 (Affinity, AF4001, RRID: AB_2835330), interleukin 1β (IL-1β) (Affinity, AF5103, RRID: AB_2837589), interleukin 6 (IL-6) (Affinity, DF6087, RRID: AB_2838055), interleukin 10 (IL-10) (Affinity, DF6894, RRID: AB_2838853), tumor necrosis factor alpha (TNF-α) (Affinity, AF7014, RRID: AB_2835319), THBS1 (Affinity, DF6848, RRID: AB_2838807), HRP-conjugated Goat Anti-Mouse IgG(H + L) (Proteintech, SA00001-1, RRID: AB_2722565), HRP-conjugated Goat Anti-Rabbit IgG(H + L) (Proteintech, SA00001-2, RRID: AB_2722564). We strictly followed the requirements in the antibody manufacturers’ instructions and used the same primary antibodies for both human and rat tissue samples. Cell culture and treatments Human bladder epithelial cells were cultured in Ham’s F12 nutrient medium supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 µg/mL streptomycin, under a humidified atmosphere of 5% CO₂ at 37 °C. To explore the impact of MEL on the progression of GC, the cells underwent stimulation by LPS (0.01 g/L). After determining the optimal stimulation time, we added different concentrations of MEL and used it together with LPS to stimulate the cells. Enzyme-linked immunosorbent assays (ELISA) When cells reached 70–80% confluency with good morphology, they were stimulated with LPS (0.01 g/L) at 37 °C with 5% CO₂ for 2 h. After stimulation, cell supernatants were collected by centrifugation (1500 rpm, 4 °C, 5 min) and stored at −80 °C. For ELISA, standard cytokine standards were serially diluted to gradient concentrations; biotinylated antibodies and enzyme conjugates were prepared as working solutions (1:100 dilution). 100µL standards or samples (in triplicate) were added to pre-coated plates, incubated at 37 °C for 90 min, then washed 5 times. Biotinylated antibodies (100µL/well) were added and incubated for 60 min, followed by washing and addition of enzyme conjugates (100µL/well) for 30 min incubation and 6 washes. TMB substrate (100µL/well) was added for 15 min dark incubation, then reaction was stopped with 50µL stop solution. A microplate reader measured OD values at 450 nm, a standard curve was generated using OD values of gradient standards, and sample concentrations were calculated accordingly28. Real-time PCR Total RNA was isolated from cells/tissues using Trizol reagent, and its concentration and purity were quantified via a spectrophotometer (A260/A280 ratio 1.8–2.0.8.0). First-strand cDNA was synthesized from 1 µg of qualified total RNA using a reverse transcription kit following the manufacturer’s protocol. Gene-specific primers (sequences retrieved from NCBI Gene Bank) were used for real-time PCR, which was performed in a 20µL reaction system containing cDNA template, forward/reverse primers, and SYBR Green Master Mix. The reaction conditions were: 95 °C for 30 s, followed by 40 cycles of 95 °C for 5 s and 60 °C for 30 s, with a final melting curve analysis to confirm amplicon specificity. Relative gene expression levels were calculated using the 2−(∆∆Ct) method, with GAPDH as the internal reference gene29,30. The sequences of primers used were as follows: Human: GAPDH: F:5′-AACTTTGGTATCGTGGAAGGACTC-3′, R:5′-CAGTAGAGGCAG. GGATGATGTTC-3′; IL-1β: F: 5′-AATGATGGCTTATTACAGTGGCAATG-3′, R:5′-TAGTGGTG. GTCGGAGATTCGTAG-3′; IL-6: F: 5′-CTTCGGTCCAGTTGCCTTCTC − 3′, R:5′-CAATTCGTT. CTGAAGAGGTGAGTG-3′; IL-10: F:5′-CCGTGGAGCAGGTGAAGAATG-3′, R:5′-CATTGTC. ATGTAGGCTTCTATGTAGTTG-3′; TNF-α: F: 5′-GAGGCCAAGCCCTGGTATG − 3′, R:5′-CG. GGCCGATTGATCTCAGC-3′. Rat: GAPDH: F: 5′-AGACAGCCGCATCTTCTTGT-3′, R: 5′-CTTGCCGTGGGTAGAGTC AT −3′; IL-1β: F: 5′-AGGCTTCCTTGTGCAAGTGT − 3′, R: 5′-TGAGTGACACTGCCTTCCTG − 3′; IL-6: F: 5′-CCGGAGAGGAGACTTCACAG-3′, R: 5′-ACAGTGCATCATCGCTGTTC − 3′; IL-10: F: 5′- CCTGCTCTTACTGGCTGGAG − 3′, R: 5′-TGTCCAGCTGGTCCTTCTTT-3′; TNF-α༚F: 5′- GCCTCTTCTCATTCCTGCTT-3′, R: 5′-TGGGAACTTCTCATCCCTTTG-3′. Western blot analysis Cell and animal tissue lysates were prepared in RIPA lysis buffer (supplemented with 1% PMSF and protease inhibitor cocktail) and protein concentration was quantified using BCA assay. Equal amounts of proteins (30 µg per lane) were loaded onto SDS-PAGE gels for electrophoretic separation. After electrophoresis, proteins were transferred to PVDF membranes at 300 mA for 90 min. Membranes were blocked with 5% non-fat milk in TBST for 1 h at room temperature, then incubated with primary antibodies overnight at 4 °C, followed by HRP-conjugated secondary antibodies for 1 h at room temperature with TBST washes (3 × 10 min) between steps. Western blot signals were detected using ECL substrate and quantified via a calibrated exposure system31,32. Cell transfection Log-phase target cells in the logarithmic growth phase were seeded in 6-well plates, and upon reaching 70–80% confluency, they were stimulated with LPS at a final concentration of 0.01 g/L for 2 h. After the LPS stimulation period, the cell culture medium was replaced with normal F12 medium. Two sterile EP tubes were pre-labeled as Tube A (for siRNA dilution) and Tube B (for transfection reagent dilution). Each tube was filled with 250 µL of serum- and antibiotic-free basal medium. Then, 6 µL of siRNA stock solution (20µM) was added to Tube A and 6 µL of transfection reagent to Tube B. After gentle pipetting to mix, the contents of both tubes were slowly combined and incubated at room temperature for 15–20 min to allow formation of siRNA–transfection reagent complexes. Meanwhile, the old medium in the 6-well plates was aspirated, and the cell monolayer was gently washed twice with serum-free medium. Then, 1.5 mL of complete medium containing 10% fetal bovine serum was added to each well. The transfection complexes were added dropwise to the corresponding wells, and the plates were rocked gently to ensure even distribution. The cells were cultured at 37 °C under 5% CO₂ for 6 h. Subsequently, the medium containing the transfection complexes was replaced with 2 mL of fresh complete medium. Culture was continued until 24 h post-transfection, after which cells were harvested and transfection efficiency was assessed by measuring the expression level of the target protein via Western blot33. The sequences of the siRNA are as follows: siRNA1: F: 5′-GCGUUGGUGAUGUAACAGAAA-3′, R: 5′-UUUCUGUUACAUCACCAA CGC-3′; siRNA2: F: 5′-GAAGCAAGUCACCCAGUCCUA-3′, R: 5′- UAGGACUGGGUGACU UGCUUC-3′; siRNA3: F: 5′-CCUUGACAACAACGUGGUGAA-3′, R: 5′-UUCACCACGUUG UUGUCAAGG-3′. Knock down of the THBS1 gene Cells were stimulated with 0.01 g/L LPS for 2 h in a 37 °C, 5% CO₂ incubator. After the LPS stimulation period, the cell culture medium was replaced with normal F12 medium. Subsequently, 8µL of THBS1-interfering lentivirus (along with a negative control lentivirus) was added at a multiplicity of infection (MOI) of 10. To enhance transduction efficiency, 5 µg/mL polybrene was supplemented, and the cells were incubated for 24 h. After 48 h of lentiviral transduction, when cell density reached approximately 80%, the medium was replaced with fresh complete medium containing 2µL of 10 mg/mL puromycin (final concentration 2 µg/mL). Puromycin selection was performed daily for two consecutive days. The surviving cells were then passaged and expanded, and THBS1 stable knockdown cell lines were validated by Western blot analysis34,35. The sequences of the shRNA are as follows: F:5′-CCGGGACCAAAGCCTGCAAGAAAGATCTCTTGAATCTTTCTTGCAGGCTTTG. GTCTTTTTTG-3′, R:5′-AATTCAAAAAAGACCAAAGCCTGCAAGAAAGATTCAAGAGAT. CTTTCTTGCAGGCTTTGGTC-3′. Animals The animal experiments conducted within this study were granted official approval by the Ethics Committee as well as the Animal Experiment Committee of Jiangsu University. We used female Sprague-Dawley (SD) rats, precisely aged 8 weeks and weighing within the range of 150 to 200 g. Rats were anesthetized by an intraperitoneal injection of 2% sodium pentobarbital (5 mg/100 g body weight). The bladder was catheterized with a 1 mm epidural catheter to drain the urine, followed by instillation with LPS. Eight weeks later, the bladder tissues were collected, and the rats were euthanized by an overdose of 2% sodium pentobarbital (15 mg/100 g body weight) administered intraperitoneally. We adopted the method of bladder perfusion with LPS to induce the GC model in rats and used hematoxylin and eosin (HE) staining for histological and morphological analyses. The first step: Twenty rats were randomly and evenly partitioned into 4 distinct groups (5 rats/group). Group 1: bladder perfusion with 300µL of PBS; Group 2: bladder perfusion with 300µL of LPS (0. 5 mg/ml); Group 3: bladder perfusion with 300µL of LPS (0.75 mg/ml); Group 4: bladder perfusion with 300µL of LPS (1 mg/ml). After maintaining the LPS in the bladder for a 30-minute contact period, the bladder was gently emptied via transurethral catheterization to terminate the stimulation. The perfusion process was meticulously executed once every three days, maintaining a consistent frequency throughout a consecutive 8-week period. The second step: Fifteen rats were randomly and evenly partitioned into 3 distinct groups (5 rats/group). The control group: Bladder perfusion with 300µL of PBS; the LPS-treated group: Bladder perfusion with 300µL of LPS for 8 weeks; the LPS + MEL-treated group: Bladder perfusion was carried out using 300µL of LPS. After a continuous 4-week period of LPS perfusion, MEL at a dosage of 10 mg/kg/d, which was dissolved in 1% ethanol, was administered via intraperitoneal injection on a daily basis for another successive 4 weeks. Similarly, after maintaining the LPS in the bladder for a 30-minute contact period, the bladder was gently emptied via transurethral catheterization to terminate the stimulation. Mass spectrometry analysis Sample preparation When the adherent cell confluency reaches 80% or higher, rinse the cell culture dish with PBS. After that, add trypsin digestion solution and place the cell culture dish in an incubator for digestion. Subsequently, aspirate the cell suspension into a sterile centrifuge tube. Set the centrifuge speed to 2000 rpm and the centrifugation time to 5 min. After centrifugation, remove the supernatant, leaving only the cell pellet at the bottom of the centrifuge tube. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) Using a pipette, carefully load an appropriate amount of protein sample into the wells of the gel, avoiding introduction of bubbles. Simultaneously, load protein molecular weight standards into one well as a reference. Set the initial voltage to 80 V. Once the samples have entered the stacking gel, increase the voltage to 120 V and continue electrophoresis until the bromophenol blue tracking dye reaches the bottom edge of the gel. Then, terminate the electrophoresis process. The entire procedure takes approximately 2 h. Subsequent steps: Following electrophoresis, carefully remove the gel and immerse it in Coomassie Brilliant Blue staining solution. Shake gently on an orbital shaker for 2 h to ensure complete staining of the protein bands. After staining, discard the staining solution and add decolorizing solution. Decolorize with continuous gentle shaking, changing the decolorizing solution periodically until the background becomes clear and protein bands are distinctly visible. Finally, capture an image of the gel using a gel imaging system and analyze the position and intensity of the protein bands to obtain information such as molecular weight and expression levels of the target proteins. Filter aided sample preparation (FASP) Transfer an appropriate amount of protein sample (50–200 µg) to an ultrafiltration centrifuge tube. Centrifuge at 15,000 × g for 20 min at 4 °C to retain the protein on the filter membrane and remove low-molecular-weight impurities. Add a suitable volume of buffer, mix gently by pipetting or swirling, and centrifuge again at 15,000 × g for 20 min at 4 °C. Repeat this washing step twice to ensure complete denaturation of the proteins. Next, add an appropriate amount of iodoacetamide solution and incubate the tube at room temperature in the dark for 30 min to facilitate alkylation of cysteine residues. Following alkylation, centrifuge at 15,000 × g for 20 min at 4 °C to remove excess iodoacetamide. Add buffer and centrifuge again under the same conditions; repeat this step twice to eliminate residual SDS and iodoacetamide. Then, add ammonium bicarbonate solution and centrifuge at 15,000 × g for 20 min at 4 °C. Repeat this buffer exchange step three times to ensure the solution surrounding the filter membrane is replaced with 50 mM ammonium bicarbonate. Add trypsin solution at a protein-to-enzyme ratio of 100:1, and incubate at 37 °C for 18 h to allow complete digestion of the protein into peptides. After digestion, centrifuge at 15,000 × g for 20 min at 4 °C and collect the filtrate containing the digested peptides into a new centrifuge tube. To recover any residual peptides, add 50 µL of 50 mM ammonium bicarbonate solution to the ultrafiltration unit, centrifuge again under the same conditions, and combine this eluate with the previously collected peptide solution. Statistical evaluation The task of statistical evaluation was executed with great meticulousness using GraphPad Prism 9.0. All experiments were replicated a minimum of three independent times. Prior to statistical analysis, the normality of the data was verified using the Shapiro-Wilk test. For comparisons involving two groups: if the data were normally distributed, results were presented as mean ± standard deviation (mean ± SD) and analyzed using an unpaired Student’s t-test; if non-normally distributed, the Mann-Whitney U test (non-parametric alternative) was applied. For comparisons involving three or more groups: normally distributed data were systematically presented as mean ± SD, analyzed by one-way analysis of variance (ANOVA), followed by Tukey’s post hoc test for multiple comparisons; non-normally distributed data were analyzed using the Kruskal-Wallis test, with Dunn’s post hoc test for pairwise comparisons. The p values were distinctly labeled following a standardized notation: *p < 0.05, denoting a moderate level of significance; **p < 0.01, highlighting a more pronounced significance; and ***p < 0.001, signifying an extremely high level of significance. Ethics declaration The animal experiment reports in this study adhere to the ARRIVE guidelines to ensure the transparency, integrity, and reproducibility of the experimental procedures and results. We confirm that all experimental methods were carried out in accordance with the specific relevant guidelines and regulations. Throughout the entire experimental process, we strictly abided by the ethical norms of animal experiments and were committed to safeguarding animal welfare. Results LPS induced increased GC-related cytokines and signaling pathways in human bladder epithelial cells To commence, human bladder epithelial cells were exposed to stimulation using LPS (0.01 g/L) for a series of durations, and measured the expressions of twelve common cytokines upon ELISA. We screened out four cytokines, namely IL-1β, IL-6, IL-10, and TNF-α with the highest increase in expression levels (Fig. 1A). Next, the cells were subjected to stimulation using an identical concentration of LPS. It was demonstrated that, two hours subsequent to the stimulation with LPS, the levels of the above four inflammatory factors, were all markedly elevated (Fig. 1B). This result was consistent with the trend shown by the previous ELISA, and was also confirmed in the subsequent western blot analysis (Fig. 1C). To investigate the expression of signaling pathways in bladder epithelial cells under chronic LPS stimulation, the cells were treated with LPS for 2, 24, and 48 h. Real-time PCR results showed that the expression of TLR4, p65, STAT3, and p38 remained active throughout this period (Figure S1). Fig. 1(A) Human bladder epithelial cells were stimulated with lipopolysaccharide (LPS) (0.01 g/L). The levels of twelve common cytokines were then quantified by ELISA. The results showed that the increases in IL-1β, IL-6, IL-10, and TNF-α were the most pronounced. (B, C) Human bladder epithelial cells were subjected to stimulation with LPS (0.01 g/L) for durations of 2 h, 4 h, and 6 h, respectively. Immediately upon the cells reaching the designated time points, the signal intensities of IL-1β, IL-6, IL-10, and TNF-α within the cells, were measured. We examined the expression of these four inflammatory factors in relation to β-Tubulin. We ascertained that the optimal stimulation time for LPS (0.01 g/L) was 2 h. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (A) Human bladder epithelial cells were stimulated with lipopolysaccharide (LPS) (0.01 g/L). The levels of twelve common cytokines were then quantified by ELISA. The results showed that the increases in IL-1β, IL-6, IL-10, and TNF-α were the most pronounced. (B, C) Human bladder epithelial cells were subjected to stimulation with LPS (0.01 g/L) for durations of 2 h, 4 h, and 6 h, respectively. Immediately upon the cells reaching the designated time points, the signal intensities of IL-1β, IL-6, IL-10, and TNF-α within the cells, were measured. We examined the expression of these four inflammatory factors in relation to β-Tubulin. We ascertained that the optimal stimulation time for LPS (0.01 g/L) was 2 h. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. LPS contributed to GC in vivo We constructed rat models of GC by the method of bladder perfusion of LPS. (Figure 2A and B). Upon detailed examination, no conspicuous pathological alterations were detected in the bladders of the rats belonging to Group 1, while chronic non-specific inflammatory changes were observed in the bladders of the rats in Groups 2, 3 and 4. In addition, brunn’s nests and cystic cavities appeared in the lamina propria mucosa of the rats in Group 4. The glandular epithelium was columnar epithelium covered with transitional epithelium, but no intestinal metaplasia was observed, indicating that the constructed model was a typical GC (Fig. 2C). Therefore, we came to the conclusion that bladder perfusion with 300µL of LPS (1 mg/ml) could effectively induce rats to form typical GC. Fig. 2(A) The process of bladder perfusion with LPS (1 mg/ml), the 1 mm epidural catheter is indicated by the red arrow. (B) The procedure for establishing the GC model in SD rats. A total of 20 SD rats were randomly and equally divided into four experimental groups (5 rats/group), with group allocation performed in a blinded manner. (C) Representative results of HE staining on rat bladder tissues under varying experimental circumstances. (Magnification: 40×; Scale bar: 500 μm). (A) The process of bladder perfusion with LPS (1 mg/ml), the 1 mm epidural catheter is indicated by the red arrow. (B) The procedure for establishing the GC model in SD rats. A total of 20 SD rats were randomly and equally divided into four experimental groups (5 rats/group), with group allocation performed in a blinded manner. (C) Representative results of HE staining on rat bladder tissues under varying experimental circumstances. (Magnification: 40×; Scale bar: 500 μm). MEL alleviated LPS-induced GC in vitro and vivo In order to comprehensively and meticulously explore the impact of MEL on GC, we co-treated human bladder epithelial cells with different concentrations of MEL (25µM, 50µM, 100µM) and LPS for 2 h respectively. It was demonstrated that different concentrations of MEL reduced inflammation in GC. When the concentration of MEL was precisely set at 100µM, the most pronounced and statistically significant decreases in IL-1β, IL-6, IL-10, and TNF-α were clearly demonstrated (Fig. 3A and B). The above data suggested that MEL indeed has a certain role in the inhibition of GC. Fig. 3(A, B) After human bladder epithelial cells were treated with LPS (0.01 g/L) and Melatonin (MEL) (25µM, 50µM, 100µM) for 2 h, the expression of IL-1β, IL-6, IL-10, and TNF-α in human bladder epithelial cells were detected. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (C) Representative HE staining results of rat bladder tissues. A total of 15 SD rats were randomly and equally divided into three experimental groups (5 rats/group), with group allocation performed in a blinded manner. (The control group: bladder perfusion with PBS for 8 weeks; the LPS-treated group: bladder tissue exposed to LPS (1 mg/ml) for 8 weeks; LPS + MEL-treated group: after a continuous 4-week period of LPS perfusion, MEL (10 mg/kg) was administered via intraperitoneal injection on a daily basis for another successive 4 weeks.) (Magnification: 40×; Scale bar: 500 μm) (D) Schematic diagram illustrating the experimental design for MEL intervention in the glandular cystitis (GC)model. (E, F) After extracting RNA and proteins from the bladder tissues of SD rats in the control group, LPS-treated, and LPS + MEL-treated group, the four inflammatory factors were detected through real-time PCR and Western blot, and the corresponding statistical analyses were conducted. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (G) Representative IHC results of the above four inflammatory factors in bladder tissues. (Magnification: 100×; Scale bar: 500 μm). (A, B) After human bladder epithelial cells were treated with LPS (0.01 g/L) and Melatonin (MEL) (25µM, 50µM, 100µM) for 2 h, the expression of IL-1β, IL-6, IL-10, and TNF-α in human bladder epithelial cells were detected. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (C) Representative HE staining results of rat bladder tissues. A total of 15 SD rats were randomly and equally divided into three experimental groups (5 rats/group), with group allocation performed in a blinded manner. (The control group: bladder perfusion with PBS for 8 weeks; the LPS-treated group: bladder tissue exposed to LPS (1 mg/ml) for 8 weeks; LPS + MEL-treated group: after a continuous 4-week period of LPS perfusion, MEL (10 mg/kg) was administered via intraperitoneal injection on a daily basis for another successive 4 weeks.) (Magnification: 40×; Scale bar: 500 μm) (D) Schematic diagram illustrating the experimental design for MEL intervention in the glandular cystitis (GC)model. (E, F) After extracting RNA and proteins from the bladder tissues of SD rats in the control group, LPS-treated, and LPS + MEL-treated group, the four inflammatory factors were detected through real-time PCR and Western blot, and the corresponding statistical analyses were conducted. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (G) Representative IHC results of the above four inflammatory factors in bladder tissues. (Magnification: 100×; Scale bar: 500 μm). Fifteen rats were randomly divided into 3 groups in equal numbers. The Control group: Bladder perfusion with PBS; the LPS-treated group: Bladder perfusion with 300µL of LPS; the LPS + MEL-treated group: Bladder perfusion was carried out using 300µL of LPS. After a 4-week period of LPS perfusion, MEL was administered via intraperitoneal injection on a daily basis for another successive 4 weeks. The results obtained from HE staining vividly manifested that, when contrasted with the LPS-treated group, the quantity of Brunn’s nests within the LPS + MEL-treated group was conspicuously diminished, and the degree of non-specific inflammation also showed a decreasing trend (Fig. 3C and D). We detected the expressions of IL-1β, IL-6, IL-10 and TNF-α. Following the treatment with LPS, the expression abundances of the aforesaid inflammatory factors were markedly elevated as compared to those in the control group. However, these inflammatory factors were inhibited upon MEL (Fig. 3E and F). The subsequent immunohistochemical (IHC) staining results further confirmed that MEL indeed alleviated LPS-induced GC (Fig. 3G). These obtained data provided compelling evidence suggesting that MEL is capable of effectively attenuating the severity of inflammation within the rat GC model. MEL inhibited the p38 MAPK/STAT3 axis induced by LPS-stimulated TLR4 activation in GC To carry out a more profound and exhaustive exploration of the influence and specific traits of MEL on the signaling pathways within the GC model of human bladder epithelial cells that are mediated by LPS, we measured the expressions of the TLR4, p65, p38, p-p38, STAT3, and p-STAT3. Compared with normal cells, the expressions of the pathways increased after the cells were treated with LPS, while MEL could significantly reduce their expressions (Fig. 4A). Following pretreatment with a TLR4 inhibitor prior to LPS stimulation, real-time PCR results demonstrated a significant decrease in the expression levels of p38 and STAT3 (Fig. 4B). These experimental results strongly indicate that LPS acts as a potent inducer: it triggers the activation of TLR4 as well as the p38 MAPK/STAT3 axis, thereby promoting the phosphorylation of p38 and STAT3. Notably, MEL can effectively suppress this entire cascade of events. Fig. 4(A) After human bladder epithelial cells were treated with LPS (0.01 g/L) and MEL (25µM, 50µM, 100µM) for 2 h, the expression of TLR4, p65, p38, p-p38, STAT3, and p-STAT3 in human bladder epithelial cells were detected by Western blot. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (B) The cells were treated with a TLR4 inhibitor prior to LPS stimulation. The mRNA expression levels of molecules associated with the p38 MAPK and STAT3 signaling pathways were determined by real-time PCR. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (C) Protein expression levels of the above signaling pathways in the bladder tissues of SD rats. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (D) Representative IHC outcomes of TLR4, p65, STAT3 and p38 in bladder tissues. (Magnification: 100×; Scale bar: 500 μm). (A) After human bladder epithelial cells were treated with LPS (0.01 g/L) and MEL (25µM, 50µM, 100µM) for 2 h, the expression of TLR4, p65, p38, p-p38, STAT3, and p-STAT3 in human bladder epithelial cells were detected by Western blot. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (B) The cells were treated with a TLR4 inhibitor prior to LPS stimulation. The mRNA expression levels of molecules associated with the p38 MAPK and STAT3 signaling pathways were determined by real-time PCR. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (C) Protein expression levels of the above signaling pathways in the bladder tissues of SD rats. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (D) Representative IHC outcomes of TLR4, p65, STAT3 and p38 in bladder tissues. (Magnification: 100×; Scale bar: 500 μm). With the aim of delving deeper into the influence and unique characteristics of MEL on the TLR4 signaling pathway within the context of rat GC models induced by LPS, we carried out in vivo investigations to precisely measure the expression of TLR4, p65, p38, p-p38, STAT3, and p-STAT3. Following the treatment with LPS, the expression of the above signaling pathways was markedly elevated when compared with those in the control group. Similarly, MEL exerted an inhibitory effect on the LPS-induced p38 MAPK/STAT3 axis in vivo (Fig. 4C). These experimental findings cogently demonstrated MEL effectively reduced the p38 MAPK/STAT3 axis activated by TLR4, which is consistent with the results in vitro. It could be further identified by subsequent IHC staining analysis (Fig. 4D). THBS1 was critical in the progression of GC To gain a more profound understanding of the underlying mechanism, a proteomic analysis was carried out on the specimens of the control group, LPS-treated (exposed to 0.01 g/L LPS for 2 h), and LPS + MEL-treated group (with a concentration of 100µM) derived from human bladder epithelial cells (n = 3) (Fig. 5A). A total of more than 6,000 non-redundant proteins were subjected to analysis (Table S1). We meticulously screened and identified the top 9 proteins that manifested the most pronounced up-regulated expression alterations when comparing the LPS-treated group with the control group. The main functions of these proteins are as follows: COX6B1: A subunit of mitochondrial cytochrome c oxidase, critical for cellular respiration and ATP production. SLC25A3: A mitochondrial carrier protein that transports ADP/ATP across the inner mitochondrial membrane, regulating energy metabolism. MRPL41/RPL30: Ribosomal proteins (mitochondrial/cytoplasmic) involved in protein synthesis; they ensure accurate translation of genetic information. UQCRH: Part of the mitochondrial respiratory chain complex III, mediating electron transfer and energy generation. PTBP1: A splicing factor that regulates alternative splicing of pre - mRNAs, influencing gene expression diversity. USP10: A deubiquitinase that modulates protein stability and signaling by removing ubiquitin tags. PCYOX1: Catalyzes post-translational modifications of polyamines, participating in cellular redox homeostasis. Concurrently, we also singled out the 9 proteins that exhibited the most conspicuous down-regulated expression shifts between the LPS + MEL-treated group and the LPS-treated group. The main functions of these proteins are as follows: EPB41L5: Involved in cytoskeleton organization and cell adhesion, regulating cell shape and migration. TWSG1: A secreted protein that modulates BMP (bone morphogenetic protein) signaling, influencing embryonic development and tissue repair. TMUB2: May regulate protein degradation pathways (e.g., ubiquitin-proteasome system) and cell cycle progression. PAX8: A transcription factor essential for thyroid development and function; it controls gene expression in epithelial cells. ISOC2: Associated with calcium-mediated signaling and immune responses, potentially regulating T-cell activation. JUNB: A component of the AP-1 transcription complex, controlling cell proliferation, differentiation, and stress responses. TNS4: Promotes cell migration and invasion by modulating integrin - based adhesion complexes, often linked to cancer progression. TENM3: Functions in axon guidance and neural circuit formation during development, influencing neuronal connectivity. By integrating the data from these two sets of groups, it was discovered that THBS1 is highly likely to assume a pivotal role in the dynamic process of melatonin’s intervention in the progression of GC (Fig. 5B, C and D). The protein expression abundances of THBS1 were validated via Western blot analysis. Through close observation, it was found that the expression of THBS1 was significantly elevated in the LPS-treated rat bladder tissues, while in the LPS + MEL-treated group, its expression showed a distinct downward trend (Fig. 5E). Fig. 5(A) The workflow of proteomics analysis. (B) Proteomics analysis was conducted on the control group, LPS-treated, and LPS + MEL-treated group derived from human bladder epithelial cells. Bioinformatics analysis targeting differentially expressed proteins was carried out by means of a particular protein database. The outcomes of differential protein screening were exhibited in the format of volcano plots. (C, D) The top nine proteins exhibiting the highest upregulation when comparing the LPS-treated group with the control group, along with the top nine proteins showing the most pronounced downregulation between the LPS + MEL-treated group and the LPS-treated group were identified. According to the analysis outcomes, we tentatively concluded that THBS1 might have a certain function in the progression of GC. All experiments were repeated three times, data processing was performed using Student’s t-test (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (E) The protein expression of THBS1 in rat bladder tissues under different conditions. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (F) The top 20 pathways that were highly enriched in the outcomes of KEGG enrichment analysis. (G) These two GO enrichment analysis plots respectively present the differential enrichment patterns in terms of cellular components, molecular functions. (A) The workflow of proteomics analysis. (B) Proteomics analysis was conducted on the control group, LPS-treated, and LPS + MEL-treated group derived from human bladder epithelial cells. Bioinformatics analysis targeting differentially expressed proteins was carried out by means of a particular protein database. The outcomes of differential protein screening were exhibited in the format of volcano plots. (C, D) The top nine proteins exhibiting the highest upregulation when comparing the LPS-treated group with the control group, along with the top nine proteins showing the most pronounced downregulation between the LPS + MEL-treated group and the LPS-treated group were identified. According to the analysis outcomes, we tentatively concluded that THBS1 might have a certain function in the progression of GC. All experiments were repeated three times, data processing was performed using Student’s t-test (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (E) The protein expression of THBS1 in rat bladder tissues under different conditions. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (F) The top 20 pathways that were highly enriched in the outcomes of KEGG enrichment analysis. (G) These two GO enrichment analysis plots respectively present the differential enrichment patterns in terms of cellular components, molecular functions. These two KEGG pathway enrichment analysis charts display the enrichment results of differentially expressed proteins (DEPs) for different treatment group. Left chart (LPS-treated group vs. control group): DEPs are enriched in multiple KEGG pathways. The most enriched pathway is “Metabolic pathways,” which contains the highest number of DEPs. Other significantly enriched pathways include those related to neurodegenerative diseases (e.g., Alzheimer disease), viral infection response (e.g., Human immunodeficiency virus 1 infection), and signaling pathways (e.g., PI3K-Akt signaling pathway). These results indicate that LPS treatment markedly affects cellular metabolism, neuroinflammatory processes, and immune-related pathways. Right chart (LPS + MEL- treated group vs. LPS-treated group): The “Metabolic pathways” and “PI3K-Akt signaling pathway” remain among the most enriched, but additional pathways such as “Focal adhesion,” “Human papillomavirus infection,” and “ECM-receptor interaction” are also prominent. This pattern suggests that MEL intervention partially counteracts LPS-induced pathway alterations, particularly in metabolic and inflammatory signaling processes. (Fig. 5F). The left chart reflects widespread pathway disturbances induced by LPS, potentially driven by THBS1-related signaling changes that affect metabolic and neurodegeneration-associated pathways. The right chart indicates that MEL treatment restores certain pathways, possibly through THBS1-mediated modulation of PI3K-Akt signaling and metabolic functions. Together, these results provide insights into how THBS1 may contribute to LPS-induced pathway disruptions and MEL-mediated protection, highlighting its role in regulating inflammatory and repair mechanisms. Next, we analyzed the results of Gene Ontology (GO) enrichment analysis. Comparing the LPS-treated group with the control group: For cellular components: Significant enrichment of “extracellular space” and “cell junction” indicates obvious LPS-induced protein changes in these structures. For molecular functions: Enrichment of “G-protein beta/gamma-subunit complex” suggests potential effects on G-protein-related functions. For biological processes: Enrichment of “regulation of biological quality” implies LPS impacts multiple biological processes. Comparing the LPS + MEL-treated group with the LPS-treated group: For cellular components: Enrichment of “sarcoplasmic reticulum membrane” and “perinuclear region of cytoplasm” demonstrates MEL-induced significant protein changes in these structures. For molecular functions: Enrichment of “signaling receptor activity” suggests MEL may affect related signal transduction. For biological processes: Enrichment of “positive regulation of membrane potential” indicates MEL regulates biological processes differently from LPS (Fig. 5G). Thus, LPS and MEL exert specific effects on distinct cellular functions: LPS primarily targets extracellular/cell junction-related structures/functions (potentially involving THBS1-mediated matrix regulation) and biological quality regulation, while MEL acts more on intracellular regional structures and membrane potential regulation (possibly via THBS1-associated intracellular signaling). These differences provide clues for dissecting THBS1-centered mechanisms of LPS injury and MEL protection. MEL inhibited the progression of GC by targeting THBS1 It could be determined that siRNA2 and siRNA3 exhibited remarkable efficacy in knocking down the expression of THBS1 within the LPS-treated group (Fig. 6A). Next, we determined the expression profiles of IL-1β, IL-6, IL-10 and TNF-α after knockdown of THBS1 via siRNA. When contrasted with the control group, a notably significant decreasing trend was distinctly observed in the levels of the aforesaid four inflammatory factors within both the siRNA2 group and the siRNA3 group (Fig. 6A). The expression profiles of TLR4, p65, p38, p-p38, STAT3, and p-STAT3 are also found to be in perfect alignment with the aforementioned data (Fig. 6B). Subsequently, to further validate the specificity of the THBS1 knockdown, we conducted a rescue experiment through co-transfection of si-Thbs1 with an siRNA-resistant expression vector (Res-THBS1). Real-time PCR analysis demonstrated that the suppressed expression of THBS1 was effectively restored, confirming the target specificity of si-THBS1 (Figure S2). Fig. 6(A, B) The interference of THBS1 in human bladder epithelial cells was carried out using siRNA2 and siRNA3, as well as the alterations in the expression of the four inflammatory factors and the signaling pathways were analyzed (Normal control: Untreated normal human bladder epithelial cells; Control: Human bladder epithelial cells stimulated with LPS and then transfected with negative control siRNA). All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (C, D) The THBS1 interference lentivirus (sh-THBS1) was transfected into the cells. Western blot was conducted on the alterations in the expression of the four inflammatory factors, as well as on the signaling pathways. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (E) Following transfection with shRNA in normal cells, the cells were stimulated with LPS (0.01 g/L) for 2 h. The changes in expression levels of IL-1β, IL-6, IL-10, and TNF-α were then measured using real-time PCR. (F) Schematic diagram of the process of MEL interfering with LPS-induced GC. (A, B) The interference of THBS1 in human bladder epithelial cells was carried out using siRNA2 and siRNA3, as well as the alterations in the expression of the four inflammatory factors and the signaling pathways were analyzed (Normal control: Untreated normal human bladder epithelial cells; Control: Human bladder epithelial cells stimulated with LPS and then transfected with negative control siRNA). All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (C, D) The THBS1 interference lentivirus (sh-THBS1) was transfected into the cells. Western blot was conducted on the alterations in the expression of the four inflammatory factors, as well as on the signaling pathways. All experiments were repeated three times, data processing was performed using one-way ANOVA (mean ± SD), *p < 0.05, **p < 0.01, ***p < 0.001. (E) Following transfection with shRNA in normal cells, the cells were stimulated with LPS (0.01 g/L) for 2 h. The changes in expression levels of IL-1β, IL-6, IL-10, and TNF-α were then measured using real-time PCR. (F) Schematic diagram of the process of MEL interfering with LPS-induced GC. After human bladder epithelial cells were stimulated with LPS (0.01 g/L, 2 h), the THBS1 interference lentivirus (sh-THBS1) was transfected into the cells (Fig. 6C). The expression of the aforesaid four inflammatory factors, along with the signaling pathways, were precisely quantified and determined through the application of Western blot analysis. When compared directly with the LPS-treated group, a strikingly prominent downward trend was distinctly manifested in the expression of the aforesaid four inflammatory factors, within the sh-THBS1 group, these signaling pathways also exhibited a consistent changing trend, that is, their expression levels were also decreased (Fig. 6C and D). To further validate the role of THBS1 in the inflammation associated with GC, we first knocked down THBS1 expression using shRNA in normal human bladder epithelial cells, followed by stimulation with LPS (0.01 g/L). Real-time PCR results showed that, after THBS1 knockdown combined with LPS stimulation, the expression levels of IL-1β, IL-6, IL-10, and TNF-α were significantly reduced compared to those in normal cells. This further demonstrates that THBS1 plays a non-negligible role in the pathogenesis of GC (Fig. 6E). Based on the above data, it has been demonstrated that within the context of GC, LPS is capable of mediating GC-induced inflammation and the activation of downstream signaling via THBS1. Concurrently, MEL exhibits the capacity to attenuate the expression level of THBS1, and thus have an obvious effect on the inflammatory progression of GC (Fig. 6F). Discussion The etiological factors underlying GC remain incompletely elucidated. The majority of researchers hold the view that it was the consequence of metaplasia in the transitional epithelium of the bladder under the long-term stimulatory effect of chronic inflammation. This phenotypic alteration emerges as a consequence of the synergistic and concerted actions of multiple elements. Bacterial endotoxin LPS, a common component of Gram-negative bacteria and an endotoxin36,37, is inextricably linked to the pathogenicity of bacteria. LPS has already been applied in numerous disease models of organ-specific inflammations38, It can be utilized to establish animal models of chronic bronchitis39, liver injury40, lung injury41 and gingivitis, among others. However, there haven’t been many studies on LPS in GC. In this study, we established in vivo and in vitro models of GC via two distinct approaches: rapid stimulation of human epithelial cells with LPS for the in vitro model, and long-term intravesical perfusion of LPS in rats for the in vivo model. LPS, stands out as a highly potent agonist of TLR42. Upon binding with its ligand, TLR4 is capable of setting off a series of signaling pathways. It can activate the cascade reactions of the p38 MAPK/STAT3 axis and upregulate the secretion of IL-1β, IL-6, IL-10, TNF-α, as well as chemokines and adhesion molecules43. Upon exposure to infectious agents and innate stimuli, p38 MAPK gets activated and engages in a variety of cellular activities. These encompass cell multiplication, the specialization process of cells, their movement capabilities, as well as apoptosis, all of which are essential aspects within the complex cellular functioning landscape44,45. The STAT3 pathway is also considered to be of great significance in diverse cellular functions, like the multiplication and viability of different cell types46,47. After being phosphorylated, p38 MAPK can also activate STAT348. Once activated, STAT3 has the capacity to boost the expression levels of several downstream target genes that are tied to the cell cycle regulation, cell growth, and apoptosis processes46. What’s more, there are sequences specific to STAT proteins, which are subject to recognition and phosphorylation by p38 MAPK. When the action of p38 MAPK is blocked, the activity of STAT3 in promoting gene transcription decreases, which demonstrates that there exists a bypass or regulatory mode of p38 MAPK/STAT3 signal transduction in cells49. The outcomes of our research demonstrate that the p38 MAPK/STAT3 axis plays a pivotal role in the pathological progression of GC50, and it furnishes a theoretical groundwork for deeper exploration of the pathological mechanism of GC. LPS initially sets off the TLR4 signaling cascade, thereafter instigating the activation and phosphorylation procedures within the p38 MAPK/STAT3 axis. The interconnection between the signal transduction mediated by p38 MAPK and STAT3 forms a key axis that is successively activated by LPS, which ultimately leads to the increase of the aforesaid four inflammatory factors in vivo and the occurrence of GC. LPS effectively recapitulates the core pathological manifestations characteristic of infection-associated GC. In preclinical modeling, LPS challenge elicits robust bladder mucosal inflammation, upregulation of inflammatory factors such as IL-1β, IL-6, IL-10, TNF-α, and activation of the TLR4 signaling pathway. These induced responses closely mirror the histopathological and molecular signatures observed in clinical specimens of human infection-associated GC51[,52, thereby substantiating the validity of this approach for investigating the pathogenic mechanisms underlying infection-driven disease processes. Nevertheless, the LPS-based model fails to emulate the pathophysiological features of non-infectious GC subtypes, including autoimmune or neurogenic variants. Autoimmune GC is characterized by dysregulated adaptive immune responses, such as autoantibody deposition and T-cell–mediated inflammation, whereas neurogenic GC arises from persistent neural dysregulation. These mechanisms operate through pathways fundamentally distinct from LPS-induced innate immune activation. Consequently, the applicability of the LPS model remains circumscribed to infection-associated GC, and it offers limited insight into the etiology or pathogenesis of GC subtypes with non-infectious origins. Future studies should focus on developing novel experimental models that can simulate non-infectious etiologies, so as to fully elucidate the complex disease spectrum of GC. The anti-inflammatory capabilities of MEL are essential for the preservation of physiological functions and the maintenance of homeostasis within the body. This is attributed to its capacity to synergistically amalgamate anti-inflammatory efficacies, thereby exerting a comprehensive regulatory impact on various biological processes53. With its unique endogenous characteristics, MEL shines brightly in the fields of diet and therapy, opening up new avenues for human health and well-being and demonstrating broad application prospects and potential. MEL’s potency in alleviating inflammation was confirmed by its therapeutic effect on LPS-induced GC. Both under in vitro conditions and under in vivo conditions, the inflammatory factors were significantly reduced upon MEL. These results can be obtained from the corresponding real-time PCR, Western blot and IHC staining results. Furthermore, after the addition of MEL, the expression levels of TLR4, p38 MAPK and STAT3 in the LPS + MEL-treated group showed an extremely significant decline compared with those in the LPS-treated group. The results we obtained clearly demonstrate that MEL is a potential therapy method in GC. It is capable of efficiently suppressing the TLR4/p38 MAPK/STAT3 axis, markedly diminishing the inflammatory intensity. It is worth mentioning that the effects of MEL go far beyond that. It can also strongly suppress the activation processes of p-p38 and p-STAT3 induced by LPS stimulation, and comprehensively and multilaterally reduce the inflammation level of GC, thus providing a highly promising treatment approach and direction for alleviating this disease. THBS1 shows an extremely significant association with inflammation-related processes and cell apoptosis-related procedures. THBS1 has long been acknowledged as a vital factor in the processes of immune activation and inflammatory responses. Moreover, THBS1 has the ability to deeply participate in regulating cell differentiation and apoptosis processes by activating inflammatory cells, and thus prompt the upregulation of IL-1β, IL-6 and TNF-α in immune cells. In the present study, it can be clearly observed that MEL has not only successfully reversed the abnormally elevated expressions of THBS1 and IL-10 under inflammatory conditions but also effectively curbed the elevation of IL-1β, IL-6 and TNF-α. Particularly, following the down-regulation of THBS1 expression, the levels of the aforesaid four inflammatory factors, which had been remarkably elevated upon LPS stimulation earlier, were notably reduced. Correspondingly, the quantities of the TLR4/p38 MAPK/STAT3 axis were also down-regulated to some extent. THBS1 is significantly associated with inflammation-related processes and cell apoptosis54. As a multifunctional glycoprotein, THBS1 not only initiates immune activation by binding to receptors on immune cells but also regulates the release of inflammatory mediators and the chemotaxis of inflammatory cells. Consequently, it has long been recognized as a critical regulator of immune activation and inflammatory responses55,56. Beyond inflammation, THBS1 also plays a role in modulating cell differentiation and apoptosis by activating signaling molecules released from inflammatory cells, thereby creating a synergistic link between inflammation activation and cellular function regulation57. In contrast, MEL is a key hormone involved in the body’s physiological regulation network. It effectively scavenges free radicals and maintains intracellular redox balance58. During inflammatory processes, melatonin inhibits the activation of pro-inflammatory signaling pathways and reduces the synthesis and release of inflammatory mediators, thereby suppressing the initiation and progression of inflammatory responses59,60. In this study on an LPS-induced model of GC, the regulatory effect of MEL on THBS1 and inflammatory factors was investigated through multi-dimensional experiments. In vitro, proteomic analysis revealed that LPS stimulation significantly increased the protein expression of THBS1 in bladder epithelial cells, which was attenuated by MEL treatment. Consistent with the cellular findings, in vivo animal experiments showed that the protein expression of THBS1 was markedly upregulated in the bladder tissues of LPS-induced GC rats compared to normal controls. Following intraperitoneal injection of MEL, the expressions of THBS1 and the inflammatory cytokines IL-1β, IL-6, IL-10, and TNF-α were effectively suppressed. These consistent results from both cellular and animal models strongly suggest that MEL successfully reverses the upregulation of THBS1 expression and alleviates inflammation in the LPS-induced GC model through its potent antioxidant and anti-inflammatory properties. At the molecular level, we propose that THBS1 can directly bind to TLR4. This interaction enhances both the protein stability of TLR4 and its ligand-binding affinity, thereby sustaining and amplifying TLR4-mediated inflammatory signals, which ultimately exacerbates the inflammatory response in GC. This mechanism aligns fully with the established paradigm that “TLR4 is the canonical receptor for LPS” and does not subvert the classical signaling pathway. As a core pattern recognition receptor in innate immunity, TLR4 activation depends critically on two sequential molecular events: ligand-induced homodimerization of its extracellular domain and recruitment of intracellular adaptor proteins to initiate signal transduction. We hypothesize that THBS1 binding enhances either the ligand-binding affinity of TLR4 or facilitates homodimerization of its extracellular domain, thereby modulating the activation of the downstream p38 MAPK/STAT3 signaling axis. Mechanistically, TLR4 activation is a well-recognized upstream trigger for p38 MAPK phosphorylation and STAT3 nuclear translocation; phosphorylated p38 MAPK can further enhance the transcriptional activity of STAT3 through direct phosphorylation, forming a functional signaling cascade that drives the expression of inflammatory genes. To verify the functional association between THBS1 and the TLR4/p38 MAPK/STAT3 axis, we conducted THBS1-targeted knockdown experiments at the cellular level. Quantitative analysis results showed that THBS1 silencing significantly reduced the phosphorylation levels of p38 MAPK and STAT3 mediated by TLR4. These findings provide direct experimental evidence for the regulatory role of THBS1 in the “TLR4 activation-downstream p38 MAPK/STAT3 signal transduction” process, and further support the integrity of the “MEL-THBS1-TLR4-p38 MAPK/STAT3” regulatory network proposed in this study. Our data closely connects THBS1 with numerous key factors affecting the development process of GC, revealing from a new perspective the possibility that MEL is very likely to exert an intervention on the progress of GC by strongly inhibiting THBS1. Thus, it provides potential new ideas and targets for treatments of this complex disease. This study has achieved the following remarkable characteristics: Firstly, by means of continuous bladder perfusion of LPS for 8 weeks, a typical rat model of GC was precisely constructed. This model has good representativeness and stability, laying a solid foundation for subsequent research. Secondly, LPS can specifically activate the TLR4/p38 MAPK/STAT3 axis, ultimately leading to bladder injury61,62. And the axis undoubtedly plays an indispensable and crucial role as key hubs in the complex pathogenesis and evolution process of GC. The axis’ aberrant activation is tightly associated with the onset and progression of the disease. Thirdly, MEL effectively alleviates the symptoms related to GC in rats induced by chronic inflammatory stimuli through its delicate regulatory operation on THBS1. Starting from the source and key links of the inflammatory response, it exerts its potential intervention efficacy. However, it cannot be ignored that this study has limitations in the technical dimension. First, the current findings only focus on LPS-induced typical GC and lack exploration of enteric GC—a histological subtype of human GC characterized by intestinal metaplasia of bladder epithelium52,63. This omission results in gaps in the research scope, as enteric GC may involve distinct pathogenic mechanisms that differ from LPS-induced infectious GC. Some of the experiments in our study were only analyzed at the mRNA level (such as the TLR4 signaling pathway inhibition assay and the THBS1 rescue experiment), without further validation at the protein level by Western blot; verification by this method would provide more comprehensive validation for our research findings, and we are currently working to address these aspects. Second, a key limitation of this study lies in its reliance on an in vitro bladder epithelial cell model. This system lacks the complexity of the in vivo bladder microenvironment, including its critical immune-epithelial interactions, and systemic modulation. Moreover, the sustained single-factor LPS stimulation used here does not mirror the multifactorial context of human chronic inflammation. Consequently, the observed sustained activation of signaling molecules may not accurately reflect the human bladder’s pathophysiological state. Further validation with clinical samples is essential. Meanwhile, the LPS-induced GC model established in this study fails to fully capture the etiological heterogeneity of human GC, particularly non-infectious subtypes. Clinical studies confirm that human GC has diverse pathogenesis: beyond the infection-associated pathway emphasized herein, some patients have no confirmed infection history, with etiologies more likely linked to chronic chemical irritation, autoimmune dysregulation, or mechanical insults64. Our model only simulates the “infection-inflammation-glandular metaplasia” cascade and cannot replicate the pathogenesis of non-infectious GC, thus limiting its representativeness across the full etiological spectrum of human GC. To address these limitations, future research should focus on two directions: (1) Utilize non-infectious GC models (e.g., cyclophosphamide- or protamine sulfate-induced models) to verify whether the melatonin-THBS1 regulatory axis applies to non-infectious subtypes, thereby clarifying the etiological specificity of this mechanism; (2) Conduct stratified analyses of clinical GC patients, quantifying infection history, urinary pathogen load, and THBS1 expression to explore correlations among infection status, THBS1 levels, and disease severity. This will help delineate the target population for the clinical translation of our findings. Notably, while the LPS-induced model has limitations in etiological coverage, its value lies in focusing on infection-associated GC—a clinically significant subtype of human GC cases. Acknowledging these limitations not only enhances the transparency of this study but also provides a clear direction for future research to refine subtype-specific mechanisms of GC and advance the development of precision-targeted therapies. Therefore, before MEL can be effectively translated into clinical practice for GC patients, a considerable amount of challenging work remains to be thoroughly addressed and overcome. It is particularly important to emphasize that the specific molecular mechanisms through which MEL regulates THBS1 to exert its therapeutic effects in GC still require further in-depth investigation. Moreover, the relevance and efficacy of the current experimental results in the context of human clinical applications are inevitably constrained by certain limitations, necessitating additional comprehensive research and validation. Conclusion MEL is capable of suppressing the inflammatory progression within GC through the inhibition of the TLR4/p38 MAPK/STAT3 axis. At the same time, THBS1 is regarded as a prospective and potential therapeutic target for GC, holding remarkable value for future clinical investigations. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Table S1 Supplementary Table S1 Supplementary Information: Original Data Supplementary Information: Original Data Supplementary Figure S1 Supplementary Figure S1 Supplementary Figure S2 Supplementary Figure S2 Supplementary Figure Legends Supplementary Figure Legends