Ya-Han-Lu-Duan decoction reshapes tumor microenvironment to treat colitis-associated colorectal cancer by regulating arachidonic acid metabolism

Abstract

BACKGROUND

Ya-Han-Lu-Duan (YHLD) decoction is a classical Dai ethnic prescription widely used to treat intestinal inflammatory diseases. However, its ability to prevent colitis-associated colorectal cancer (CAC), the molecular mechanisms responsible, and the active constituents of the formulation remain unknown.

AIM

To elucidate YHLD’s therapeutic mechanisms against CAC, focusing on tumor microenvironment (TME).

METHODS

We evaluated YHLD in a mouse model of CAC induced by azoxymethane and dextran sulfate sodium. To elucidate mechanisms, we applied transcriptomic and untargeted metabolomic analyses to colon tissues from treated and control animals. Key pathways identified by these omics approaches were validated by enzyme-linked immunosorbent assay, western blot, and immunofluorescence. We characterized YHLD’s chemical constituents using UPLC-Q-TOF-MS, then performed molecular docking to model interactions between major components and enzymes in arachidonic acid (AA) metabolism, providing a preliminary screen for potential molecular targets.

RESULTS

YHLD significantly attenuated body weight loss and colon shortening in CAC mice, alleviated colonic histopathological injury, and reduced tumor formation. Transcriptomic analysis showed that genes differentially expressed after YHLD treatment were enriched in TME-related pathways, including Cytokine-cytokine receptor interaction and extracellular matrix-receptor interaction. YHLD decreased the expression of pro-inflammatory and pro-tumor growth factors and reduced markers of cancer-associated fibroblasts and myeloid-derived suppressor cells. Untargeted metabolomic analysis further identified AA metabolism as a pathway co-enriched in both transcriptomic and metabolomic datasets. Consistently, YHLD suppressed the expression of key AA metabolism enzymes, PTGS2 and ALOX12. In addition, major components of YHLD, including Pentahydroxybufostane, Clematoside B, Psammosilenin A and Dulcoside A, showed high binding affinity to PTGS2 and ALOX12, respectively, suggesting that YHLD may coordinately regulate AA metabolism through multiple active components.

CONCLUSION

This study provides the first evidence that YHLD exerts a significant protective effect against CAC. The underlying mechanism likely involves inhibition of ALOX12-mediated and PTGS2-mediated AA metabolism, suppression of cancer-associated fibroblast activation and myeloid-derived suppressor cells infiltration, and remodeling of the immunosuppressive TME, thereby alleviating CAC progression. These findings offer experimental and mechanistic support for the development of YHLD as a preventive agent for CAC.

Key Words: Colitis-associated colorectal cancer; Ya-Han-Lu-Duan decoction; Arachidonic acid metabolism; Cancer-associated fibroblasts; Myeloid-derived suppressor cells

Core Tip: Ya-Han-Lu-Duan (YHLD) decoction exerts a significant protective effect against colitis-associated colorectal cancer. YHLD remodels the immunosuppressive tumor microenvironment by suppressing cancer-associated fibroblast activation and myeloid-derived suppressor cells infiltration. YHLD is likely to inhibit ALOX12-mediated and PTGS2-mediated arachidonic acid metabolism in cancer-associated fibroblasts and myeloid-derived suppressor cells through multiple active components.

INTRODUCTION

Colorectal cancer (CRC) is the third most common malignancy worldwide and arises from lesions in the colon and rectum[1]. In 2020, approximately 1.95 million new CRC cases and 950000 deaths were reported globally, accounting for 9.4% of all cancer-related mortality[2]. Despite advances in diagnosis and treatment, the 5-year survival rate for CRC remains below 20%, severely compromising patients’ quality of life[3]. Recent epidemiological projections estimate that by 2040, the global incidence and mortality of CRC will increase by 71.10% and 85.10%[4], respectively. Moreover, between 2020 and 2050, CRC is expected to impose an economic burden of up to 2.76 billion United States dollars worldwide[5]. Collectively, these trends establish CRC as a major and escalating global public health challenge.

CRC comprises biologically distinct disease entities, among which colitis-associated colorectal cancer (CAC) is a subtype driven by chronic intestinal inflammation[6,7]. Inflammatory bowel disease (IBD), including ulcerative colitis (UC) and Crohn’s disease, is characterized by persistent and recurrent gastrointestinal inflammation and is a well-established CRC risk factor[8]. Chronic inflammation critically contributes to multiple stages of tumorigenesis, including initiation, proliferation, metastasis, senescence, and apoptosis[9,10], thereby markedly increasing cancer risk[11]. Epidemiological studies indicate that the cumulative risk of CRC in patients with IBD reaches 2.6% after 10-20 years of disease and increases to 6.6% after more than 20 years of disease; notably, approximately 21% of patients with IBD may develop CRC within 20 years of disease onset[6,12]. In addition, patients with IBD have a threefold higher risk of missed CRC diagnosis compared with the general population[13], highlighting the urgent need for effective early prevention and intervention strategies for CAC.

During the development of CAC, a complex interplay among gut microbiota dysbiosis, host Toll-like receptor-mediated innate immune signaling, and oncogenic transcriptional reprogramming converges to remodel the tumor microenvironment (TME). Gut microbiota dysbiosis disrupts epithelial barrier integrity, enabling microbial products such as lipopolysaccharide to chronically activate Toll-like receptors on intestinal epithelial and immune cells, thereby establishing a self-perpetuating inflammatory milieu that promotes neoplastic transformation[14,15]. Concurrently, transcriptomic profiling of UC and CRC patient samples has identified a set of UC-critical genes that harbor genetic alterations in up to 72% of CRC cases and are functionally implicated in the UC-to-CRC transition[16]. Within this chronic inflammatory milieu, the TME is progressively driven toward a highly immunosuppressive state[17], in which myeloid-derived suppressor cells (MDSC) directly inhibit effector T cell function by secreting immunosuppressive cytokines such as transforming growth factor beta and interleukin (IL)-10, thereby enabling tumor cells to evade immune surveillance[18]. At the same time, non-immune stromal cells, particularly cancer-associated fibroblasts (CAF), undergo excessive activation in response to persistent inflammatory stimuli. Activated CAF secrete a wide range of pro-inflammatory and pro-fibrotic mediators, recruit MDSC, and interact with immune cells, vascular cells, and tumor cells, collectively forming a pro-tumorigenic microenvironment that promotes tumor growth, invasion, and metastasis[19,20]. Consequently, the coordinated inhibition of MDSC infiltration and CAF activation may disrupt the self-sustaining cycle of inflammation, fibrosis, and immunosuppression, representing a critical strategy for preventing CRC progression.

At present, radical surgical resection combined with radiotherapy and chemotherapy remains the mainstay of treatment for patients with CAC, while immunotherapy and targeted therapies are increasingly being explored as alternative options[21]. However, treatment efficacy remains limited by frequent adverse effects, tumor recurrence, and distant metastasis[22]. Therefore, the development of safer and more effective therapeutic strategies for CAC represents an urgent and unmet clinical need.

Traditional Dai medicine, with a documented history spanning over two millennia, is recognized as one of the four major ethnomedical systems in China (alongside Tibetan, Mongolian, and Uyghur medicine). It constitutes an integral component of both traditional Chinese medical culture and the Dai “Pattra-leaf” (Beiye) cultural heritage. Dai medicinal preparations are predominantly multi-herb compound formulas designed to exert synergistic, multi-target, multi-pathway regulatory effects – a pharmacological feature particularly relevant to the management of chronic inflammatory disorders and inflammation-driven carcinogenesis[23]. Ya-Han-Lu-Duan (YHLD), also known as Xieliling Capsule, is an empirical compound formula derived from the clinical tradition of traditional Dai medicine. It is composed of Gonglaomu (Mahoniae Caulis), Baihuashupi (the dried bark of Styrax tonkinensis), Guanyinzuolian (Sempervivum tectorum L.), Tieteng (Cyclea polypetala Dunn.), and Jinhuaguo (Dioscorea cirrhosa Lour.). In clinical practice, YHLD is widely used to treat intestinal inflammatory disorders, including abdominal pain, diarrhea, gastroenteritis, and bacterial dysentery. Although YHLD has demonstrated clear clinical efficacy, its direct therapeutic effects on CAC, the underlying molecular mechanisms, and the key bioactive components responsible for these effects remain largely undefined.

To address these gaps, we first established a CAC mouse model induced by azoxymethane (AOM) combined with intermittent administration of dextran sulfate sodium (DSS) in drinking water to systematically evaluate the in vivo therapeutic efficacy of YHLD. We then integrated transcriptomic and untargeted metabolomic analyses to comprehensively characterize YHLD-induced alterations in gene expression and metabolite profiles in colon tissues. Based on multi-omics integration, we focused on the regulatory effects of YHLD on arachidonic acid (AA) metabolism. In parallel, we identified the major chemical constituents of YHLD using UPLC-Q-TOF-MS and assessed their binding affinities to key enzymes involved in AA metabolism through molecular docking analysis. Together, these findings not only confirm the therapeutic potential of YHLD in CAC but also preliminarily elucidate its principal active components and molecular mechanisms, providing a strong experimental foundation for the further development and clinical application of YHLD as a preventive and therapeutic option for CAC.

MATERIALS AND METHODS

Drugs and reagents

YHLD (batch No. 250608) was obtained from the Dai Medicine Hospital of Xishuangbanna Dai Autonomous Prefecture. We used UPLC-Q-TOF-MS to qualitatively identify the chemical components of YHLD. Detailed analytical procedures and component information are provided in the Supplementary material. Information on all other reagent kits and experimental materials used in this study is also described in the Supplementary material.

Animals

We obtained specific pathogen-free (SPF) male C57BL/6 mice (8 weeks old, body weight 23 ± 2 g) from Beijing Huafukang Bioscience Co., Ltd. The Ethical Review Committee of Animal Experiments in Yunnan University of Chinese Medicine approved all animal experiments (No. YNUCM-XMSB-G-20250280), and we conducted the study in strict accordance with established guidelines for laboratory animal usage and care.

Model establishment, grouping, and drug administration

After one week of acclimatization, we established the CAC mouse model according to previously published protocols[24]. Briefly, a single intraperitoneal injection of AOM (10 mg/kg) was given to mice. One week later, the mice entered an intermittent DSS drinking water regimen consisting of 2.5% DSS for one week followed by two weeks of normal drinking water. We repeated this cycle three times, after which the mice received normal drinking water for an additional four weeks.

To evaluate the therapeutic effects of YHLD on CAC, we randomly and evenly assigned 120 mice into six groups: (1) Normal control (CON); (2) CAC model (CAC); (3) Sulfasalazine (SASP); (4) Low-dose YHLD (YHLD-L); (5) Medium-dose YHLD (YHLD-M); and (6) High-dose YHLD (YHLD-H). The CON group received normal drinking water throughout the experiment, whereas the remaining five groups underwent AOM/DSS-induced CAC modeling. Drug administration began after completion of the first DSS cycle. The SASP group received daily oral gavage of SASP (250 mg/kg), while the YHLD-L, YHLD-M, and YHLD-H groups received daily oral gavage of YHLD at doses of 0.234 g/kg, 0.468 g/kg, and 0.936 g/kg, respectively. We gavaged the CON and CAC groups with an equivalent volume of purified water. We calculated YHLD doses using an animal-to-human dose conversion formula, with the medium-dose group corresponding to the clinical human-equivalent dose. Figure 1A illustrates the detailed experimental design for CAC induction and drug administration.

Figure 1 Ya-Han-Lu-Duan treatment alleviates pathology in colitis-associated colorectal cancer mice. A: Establishment of the colitis-associated colorectal cancer mouse model and timeline of drug administration; B: Survival curves of mice in each group; C: Weekly body weight changes in each group of mice; D: Colon lengths in each group of mice; E: Tumor counts in each group of mice; F: Tumor burden in each group of mice; G: Representative hematoxylin and eosin staining of colon tissue sections from each group of mice. n = 20 in control group, n = 12 in colitis-associated colorectal cancer group, n = 18 in sulfasalazine group, n = 13 in low-dose Ya-Han-Lu-Duan (YHLD) group, n = 15 in medium-dose YHLD group, n = 18 in high-dose YHLD group for panel (B-G). Data are presented as the mean ± SD. ^aP < 0.05, ^bP < 0.01. CAC: Colitis-associated colorectal cancer; AOM: Azoxymethane; DSS: Dextran sulfate sodium; CON: Control; SASP: Sulfasalazine; YHLD-H: High-dose Ya-Han-Lu-Duan; YHLD-L: Low-dose Ya-Han-Lu-Duan; YHLD-M: Medium-dose Ya-Han-Lu-Duan; HE: Hematoxylin and eosin.

From the initiation of model induction, we recorded body weight and survival status of mice in each group on a weekly basis. The pharmacologic responses including diarrhea and bloody stools of mice were recorded. After 10 weeks of drug intervention, we anesthetized mice with sodium pentobarbital, harvested the colons, and measured colon length. We then rinsed the colon tissues three times with pre-cooled PBS. After longitudinal incision, we assessed tumor number and tumor burden (the sum of the longest diameters of tumor tissue in the colon)[24]. We collected a 1 cm segment of the distal colon for subsequent histological and biochemical analyses.

Pathological staining

We fixed mouse colon tissues in 10% formalin, followed by dehydration and paraffin embedding. After solidification, we sectioned the paraffin blocks into 3-5 μm slices and performed hematoxylin and eosin staining. We examined the stained sections under an optical microscope to evaluate histopathological changes in the colon tissues.

Transcriptomics

Novogene Co., Ltd. (China) performed transcriptomic sequencing according to previously published methods[24]. We selected colon tissues from the same anatomical location in six mice from each of the CON, CAC, and YHLD-H groups. We extracted total RNA using TRIzol reagent and conducted quality control prior to sequencing. We performed high-throughput sequencing using the Illumina platform. We identified differentially expressed genes (DEGs) using DESeq2 v1.20.0, with |log₂FC| > 1 and P value < 0.05 as the criteria for significant differential expression. We subsequently conducted KEGG pathway enrichment analysis based on the identified DEGs.

Untargeted metabolomics

Untargeted metabolomic analysis was performed to examine changes in metabolite profiles in colon tissues, following previously reported methods[25]. Colon tissues from the CON, CAC, and YHLD-H groups were homogenized, mixed with methanol by vortexing, and centrifuged at 15000 rpm for 20 minutes. We collected the supernatant and vacuum-dried it. The dried samples were reconstituted in 50% acetonitrile, and the supernatant obtained after centrifugation was subjected to LC-MS analysis. We pooled equal volumes of supernatant from each sample to generate quality control samples. Novogene Co., Ltd. (China) performed all metabolomic analysis procedures. We used MetaboAnalyst software to identify differential metabolites between the CAC and CON groups and between the YHLD-H and CAC groups. We defined differential metabolites using the criteria of fold change (FC) > 1.5 or < 0.67, P value < 0.05, and variable importance in projection > 1. We then conducted KEGG pathway enrichment analysis based on the identified metabolites.

Detection of cytokine levels

We measured the levels of CXCL15, CXCL2, IL-1β, tumor necrosis factor (TNF), SPP1, IL-11, CSF3, and IL-23 in mouse colon tissues according to the manufacturer’s instructions provided with the enzyme-linked immunosorbent assay kits. We determined protein concentrations using the BCA assay for normalization.

Immunofluorescence

We dehydrated, paraffin-embedded, sectioned, and rehydrated formalin-fixed colon tissues, followed by antigen retrieval in sodium citrate buffer. We blocked the sections with bovine serum albumin for 1 hour and incubated them with primary antibodies at 4 °C overnight. The following day, we incubated the sections with fluorescence-labeled secondary antibodies at 37 °C for 1 hour. We stained nuclei with DAPI at 37 °C in the dark for 10 minutes. Finally, we mounted the sections and examined them under a fluorescence microscope. We quantified fluorescence intensity using ImageJ software.

Western blot

We lysed mouse colon tissues and extracted total protein using RIPA lysis buffer. We determined protein concentrations with the BCA assay for normalization. We separated equal amounts of protein by SDS-PAGE and transferred them onto PVDF membranes. After blocking the membranes with 5% skim milk at 37 °C for 2 hours, we incubated them with primary antibodies at 4 °C overnight. On the following day, we washed the membranes and incubated them with HRP-conjugated secondary antibodies at 37 °C for 2 hours. We visualized protein bands using an enhanced chemiluminescence kit and quantified band intensities with ImageJ software. We used β-actin as an internal loading control for protein normalization.

Molecular docking

We obtained the three-dimensional structure of PTGS2 and ALOX12 from the Protein Data Bank and removed crystal water molecules and unrelated small ligands using PyMOL 2.4 software. We downloaded the molecular structure of the main compounds in YHLD from the PubChem database and converted them to pdbqt format using AutoDock Tools. We then performed molecular docking and visualized the docking results.

Statistical analysis

We performed statistical analyses using the SPSS Pro online data analysis platform. We present all data as mean ± SD. We assessed data normality using the Shapiro-Wilk test. One-way analysis of variance was used to evaluate differences between groups. When variance homogeneity was not satisfied, we applied Welch’s correction. We considered P values < 0.05 to indicate statistical significance.

RESULTS

Identification of the main components of YHLD via UPLC-Q-TOF-MS

We conducted qualitative profiling of the chemical constituents present in YHLD and YHLD-containing medicated serum using UPLC-Q-TOF-MS. Total ion chromatograms acquired in both positive and negative ionization modes were shown in Figure 2. Among these identified constituents, 30 compounds, including Psammosilenin A, Pentahydroxybufostane, Ganoderic acid DM, Protostemonine, Isomangiferolic acid, Citraurin beta, 1-Formyl-3-hydroxyneogrifolin, Dulcoside A, 3-Hydroxyneogrifolin, Neotriptophenolide, Cordycepeptide A, 1-Methyl-2-[(4Z,7Z)-4,7-tridecadienyl]-4(1H)-quinolone, 5beta-Cyprinol, Brevilin A, Lup-20(29)-ene-3alpha-acetoxy-24-oic acid, Deoxycapillartemisin, [Z,E]-4,8,12-Trimethyl-3,7,11-tridecatrienoate, Kleinioxanthrone 3, 4,8,12-Trimethyl tridecanoic acid, Asarumin B, Ricinoleic acid, 1-Formylneogrifolin, 1-Indanone, Gentianamine, (-)-Jasmonic acid, 1-Methyl-4-methylethenylcyclohexene, 2-Isopropyl-5-methylanisole, Benzoic acid, Clematoside B, Sodium tauropythocholate, were confirmed as prototype compounds directly originating from YHLD and entering the systemic circulation (Supplementary Tables 1 and 2).

Figure 2 UPLC-Q-TOF-MS analysis of Ya-Han-Lu-Duan extraction and plasma samples. A: Total ion chromatogram in ESI⁺ mode; B: Total ion chromatogram in ESI^- mode. YHLD: Ya-Han-Lu-Duan.

Therapeutic effects of YHLD on CAC mice

Analysis of body weight trajectories showed a steady increase in the CON group throughout the experiment. Following AOM/DSS administration, mice exhibited a rapid decline in body weight beginning on day 7 of the modeling period. Survival curves for all groups during the modeling and treatment period are presented in Figure 1B. At the end of the experiment (week 14), the survival rate of CAC mice was 60%. In contrast, survival rates in the SASP, YHLD-L, YHLD-M, and YHLD-H groups increased to 90%, 65%, 75%, and 90%, respectively. SASP and YHLD treatment attenuated this weight loss to varying extents, with the most pronounced protective effects observed in the SASP and YHLD-H groups (Figure 1C).

Compared with the CON group, CAC mice showed significant colon shortening accompanied by marked increases in tumor number and tumor burden. SASP and YHLD treatment progressively restored colon length and significantly reduced both tumor number and tumor load (Figure 1D-F). Histopathological examination by hematoxylin and eosin staining revealed that CAC mice displayed severe colonic damage, characterized by disorganized glandular architecture, reduced goblet cell content, high-grade intraepithelial neoplasia within tumor crypts, increased nuclear mitotic numbers in focal areas, tumor invasion into the mucosa, and extensive inflammatory cell infiltration surrounding tumor cells as well as within the mucosal muscle layer and serosa. In contrast, SASP and YHLD markedly alleviated these pathological alterations in colonic tissues (Figure 1G). Among the treatment groups, YHLD-H exerted the most robust protective effects, significantly improving survival rate, attenuating body weight loss, reversing colon shortening, suppressing tumor growth, and alleviating pathological symptoms in CAC mice. Based on these results, we selected the YHLD-H group for subsequent mechanistic investigations of YHLD in CAC.

YHLD decreases secretion and activation of MDSC and CAF-related factors

To elucidate the protective mechanisms of YHLD against CAC, we performed transcriptomic analysis. We identified DEGs between the CAC and CON groups and between the YHLD-H and CAC groups using the criteria of |log₂FC| > 1 and P value < 0.05, and we presented these results in volcano plots (Figure 3A). We then conducted KEGG pathway enrichment analysis separately for the DEGs from the CAC vs CON and YHLD-H vs CAC comparisons (Figure 3B). The analysis revealed significant enrichment of pathways associated with tumor growth, the TME, and metabolic regulation, including AA metabolism, Cell adhesion molecules, Cytokine-cytokine receptor interaction, extracellular matrix (ECM)-receptor interaction.

Figure 3 Effects of Ya-Han-Lu-Duan on transcriptomics of colon tissue in colitis-associated colorectal cancer mice. A: Differentially expressed genes between colitis-associated colorectal cancer vs control and high-dose Ya-Han-Lu-Duan vs colitis-associated colorectal cancer; B: KEGG pathway enrichment analysis of differentially expressed genes. n = 6 for A and B. CAC: Colitis-associated colorectal cancer; CON: Control; YHLD-H: High-dose Ya-Han-Lu-Duan; TNF: Tumor necrosis factor; ECM: Extracellular matrix.

Among the enriched KEGG pathways, we focused on the cytokine-cytokine receptor interaction and ECM-receptor interaction pathways and further analyzed the associated genes. We observed that most of these genes were markedly upregulated in CAC mice but were downregulated following YHLD intervention, including Cxcl15, Cxcl2, Il1b, Tnf, Spp1, Il11, Csf3, and Il23a (Figure 4A). Consistently, enzyme-linked immunosorbent assay results demonstrated that levels of CXCL15, CXCL2, IL-1β, TNF, SPP1, IL-11, CSF3, and IL-23 were significantly elevated in the colon tissues of CAC mice, whereas YHLD treatment significantly reduced these cytokines (Figure 4B). Western blot analysis further showed that YHLD intervention suppressed the protein expression of COL9A3 and BMP7 in colon tissues (Figure 4C). Nine studies have demonstrated that these factors, including CXCL15, facilitate tumor progression by promoting MDSC infiltration, CAF proliferation, and tumor immune evasion[26-34]. To further validate these findings, we performed immunofluorescence staining to assess MDSC-associated and CAF-associated markers. Compared with the CON group, CAC mice exhibited significantly increased positive expression of FAP and CD11b⁺Gr-1⁺ cells in colon tissues; YHLD treatment markedly reversed these changes (Figure 4D). Together, these results indicate that YHLD modulates the secretion and activation of MDSC-related and CAF-related factors within the tumor immune microenvironment, thereby alleviating CAC progression.

Figure 4 Effects of Ya-Han-Lu-Duan on secretion and activation of myeloid-derived suppressor cells-associated and cancer-associated fibroblasts-related factors. A: Heatmap of genes related to the cytokine-cytokine receptor interaction and extracellular matrix–receptor interaction pathways; B: Enzyme-linked immunosorbent assay detection of levels of CXCL15, CXCL2, interleukin (IL)-1β, tumor necrosis factor, SPP1, IL-11, CSF3, and IL-23 in colon tissue; C: Western blot detection of COL9A3 and BMP7 expression levels in colon tissue; D: Immunofluorescence observation of FAP and CD11b⁺Gr-1⁺ expression. n = 6 for A and D, n = 20 in control group, n = 12 in colitis-associated colorectal cancer group, n = 18 in high-dose Ya-Han-Lu-Duan group for B; n = 3 for C. CAC: Colitis-associated colorectal cancer; CON: Control; YHLD-H: High-dose Ya-Han-Lu-Duan.

YHLD regulates AA metabolism in colon tissues of CAC mice

Metabolites play a critical role in regulating the TME. Consistent with this concept[7], our transcriptomic analysis revealed significant enrichment of metabolic pathways, including AA metabolism, in the KEGG pathway analysis. Based on these findings, we hypothesized that the protective effects of YHLD against CAC may involve metabolic regulation. We therefore performed untargeted metabolomic analysis of colon tissues from the CON, CAC, and YHLD-H groups. Principal component analysis showed clear differences in metabolic profiles among the three groups (Figure 5A). Partial least squares discriminant analysis further demonstrated distinct separation between the CAC and CON groups and between the YHLD-H and CAC groups in the score plots. Both partial least squares discriminant analysis models exhibited R²Y values greater than 0.8, and Q² values from 200 permutation tests were all less than -0.2 (Figure 5B), indicating robust model stability and strong predictive performance. Using the screening criteria of FC > 1.5 or < 0.67, P value < 0.05, and variable importance in projection > 1, we identified differential metabolites between groups and subjected them to KEGG pathway enrichment analysis. The results revealed that the intersecting pathways between the CAC vs CON and YHLD-H vs CAC comparisons included AA metabolism, one carbon pool by folate, glyoxylate and dicarboxylate metabolism, taurine and hypotaurine metabolism, alanine, aspartate and glutamate metabolism, pyrimidine metabolism, inositol phosphate metabolism, and glycerolipid metabolism (Figure 5C). Notably, the AA metabolism pathway showed high significance and emerged as a shared pathway in both transcriptomic and metabolomic analyses.

Figure 5 Effects of Ya-Han-Lu-Duan on metabolomics of colon tissue in colitis-associated colorectal cancer mice. A: Principal component analysis of metabolic profiles of control (CON), colitis-associated colorectal cancer (CAC), and high-dose Ya-Han-Lu-Duan (YHLD-H) groups; B: PLSDA validation between CAC vs CON and YHLD-H vs CAC; C: KEGG pathway enrichment analysis of differential metabolites between CAC vs CON and YHLD-H vs CAC. n = 6 for A-C. PCA: Principal component analysis; CAC: Colitis-associated colorectal cancer; CON: Control; YHLD-H: High-dose Ya-Han-Lu-Duan.

We next examined the AA metabolism pathway in greater detail. By intersecting the DEGs identified in the CAC vs CON and YHLD-H vs CAC comparisons within the AA metabolism pathway, we observed that most DEGs were downregulated following YHLD intervention, including Alox12e, Cyp2b10, Alox12, Alox15, Pla2g2e, Pla2g2a, Ptgds, Gm11771, Ptgs2, Gm43621, and Ptgr1 (Figure 6A). In parallel, we integrated the differential metabolites identified in the CAC vs CON and YHLD-H vs CAC comparisons within the AA metabolism pathway and visualized their expression patterns using a heatmap. Compared with the CON group, the levels of AA, prostaglandin E2 (PGE2), prostaglandin B2, prostaglandin G2 (PGG2), 5-OxoETE, and 12(S)-HETE were markedly elevated in the CAC group, whereas YHLD treatment significantly reduced the abundance of these metabolites (Figure 6B). Notably, YHLD did not significantly alter the level of leukotriene E4 compared with the CAC group. Considering the regulatory effects of YHLD on key enzymes and metabolites involved in AA metabolism, as illustrated in the schematic diagram (Figure 6C), and given the established pro-inflammatory and pro-tumorigenic roles of this pathway in CAC progression[35,36], we hypothesize that the anti-tumor effects of YHLD are closely associated with its ability to modulate AA metabolism.

Figure 6 Ya-Han-Lu-Duan regulates arachidonic acid metabolism. A: Heatmap of differentially expressed genes involved in arachidonic acid (AA) metabolism; B: Heatmap of differential metabolites involved in AA metabolism; C: Schematic diagram of AA metabolism. n = 6 for A and B. AA: Arachidonic acid; CAC: Colitis-associated colorectal cancer; CON: Control; YHLD-H: High-dose Ya-Han-Lu-Duan.

YHLD regulates PTGS2 and ALOX12 in AA metabolism

Based on these findings, we proposed that PTGS2 and ALOX12 are key target proteins through which YHLD regulates AA metabolism. Western blot analysis confirmed this hypothesis, showing that YHLD significantly reduced PTGS2 and ALOX12 protein expression in the colon tissues of CAC mice (Figure 7A). To further identify the active components of YHLD responsible for its anti-CAC effects and to clarify their molecular targets within the AA metabolism pathway, we performed molecular docking to evaluate the binding interactions between the 30 major YHLD components and PTGS2 and ALOX12 (Supplementary Table 3). The docking results revealed that the top three compounds with the strongest binding affinities (i.e., lowest binding energies) to PTGS2 are Pentahydroxybufostane, Clematoside B, and Psammosilenin A (with binding affinities < -10 kcal/mol). For ALOX12, the top three compounds with the strongest binding affinities are Psammosilenin A, Clematoside B, and Dulcoside A (with binding affinities < -5 kcal/mol) (Figure 7B). These findings indicate the strong binding capabilities of these components to PTGS2 and ALOX12, suggesting that Psammosilenin A, Pentahydroxybufostane, Clematoside B, and Dulcoside A may regulate the AA metabolism through direct interactions with PTGS2 and ALOX12.

Figure 7 Ya-Han-Lu-Duan regulates PTGS2 and ALOX12 in arachidonic acid metabolism. A: Western blot detection of PTGS2 and ALOX12 expression levels in colon tissue; B: Molecular docking results of the main components of Ya-Han-Lu-Duan (including Pentahydroxybufostane, Clematoside B, Psammosilenin A and Dulcoside A) with PTGS2 and ALOX12. n = 3 for A. CAC: Colitis-associated colorectal cancer; CON: Control; YHLD-H: High-dose Ya-Han-Lu-Duan.

DISCUSSION

CAC, one of the most severe complications of IBD, arises from persistent intestinal mucosal inflammation that drives repeated epithelial injury, dysregulated regeneration, and ultimately malignant transformation[37]. Conventional treatment strategies, including surgical resection, chemotherapy, and targeted inhibitors, often fail to eliminate the underlying chronic inflammatory milieu that sustains tumor development[38]. This limitation has highlighted an urgent need for alternative therapeutic approaches. Increasing attention has therefore focused on natural medicines with multi-target regulatory properties, particularly those capable of concurrently suppressing inflammation, restoring intestinal homeostasis, and exhibiting favorable safety profiles. In this study, we systematically elucidated the mechanistic basis of the classic Dai medicinal formulation YHLD in the intervention of CAC, thereby providing experimental and mechanistic support for its potential clinical application.

Based on UPLC-Q-TOF-MS analysis, we identified 30 prototype components of YHLD that entered the systemic circulation, including Psammosilenin A, Pentahydroxybufostane, and Clematoside B, among others. Previous studies have confirmed the anti-inflammatory and anti-tumor activities of these components. As a multi-component formulation, YHLD may therefore exert synergistic therapeutic effects against CAC through the combined actions of these constituents.

We next systematically assessed the therapeutic efficacy of YHLD in an AOM/DSS-induced CAC mouse model. This model, which combines AOM administration with intermittent DSS exposure in drinking water, induces colonic tumors that closely recapitulate colorectal cancer arising from chronic intestinal inflammation in patients with IBD and is widely used for mechanistic studies of CAC interventions[39,40]. In this model, YHLD dose-dependently improved multiple clinical and pathological parameters, including attenuation of body weight loss, preservation of colon length and tissue architecture, and suppression of tumor initiation and growth. Histological analysis showed that YHLD alleviated intestinal epithelial damage, crypt deformation and atrophy, inflammatory cell infiltration, goblet cell loss, and glandular atypia. Notably, high-dose YHLD exerted the strongest protective effects, with improvements in CAC-associated phenotypes comparable to those observed with SASP, a clinically used positive control for IBD[24]. Collectively, these findings suggest that YHLD may delay the progression from IBD to CRC and provide experimental support for the development of Dai medicine-based strategies for CAC prevention and treatment. In our study, we used male mice to establish the CAC model because UC is more common and severe in males. Epidemiological data show that UC incidence is 42% higher in males than females[41], and males also have higher early mortality and require more IBD-related surgeries[42]. Animal studies further confirm that male mice develop more severe colitis than age-matched females, possibly due to the protective effects of estrogen in females[43,44]. However, this means we could not assess whether YHLD works differently in females. Since sex-related factors may affect drug metabolism and immune responses, future studies will include female mice to fully evaluate the general applicability and potential sex differences of YHLD treatment.

Transcriptomic sequencing of mouse colon tissues showed that the DEGs were mainly enriched in TME-related KEGG pathways, including Cytokine-cytokine receptor interaction and ECM-receptor interaction, which is consistent with the pathological features of CAC. YHLD treatment markedly downregulated the expression of multiple pro-inflammatory and pro-tumor genes, including Cxcl15, Cxcl2, Il1b, Tnf, Spp1, Il11, Csf3, and Il23a. These inflammatory factors are functionally linked to the core oncogenic network driving CAC. Among these, CXCL15, the murine homolog of IL-8, forms a CXCR2 Ligand cluster with CXCL2 and cooperatively recruits MDSC to inflammatory lesions[26,33,45]. Engagement of CXCR2 activates downstream oncogenic effectors, including nuclear factor kappa B (NF-κB) and signal transducer and activator of transcription 3, which in turn drive the transcription of c-Myc, cyclin D1, and Bcl-xL in intestinal epithelial cells, establishing a direct link between chemokine signaling and the proliferative and anti-apoptotic machinery of CAC[46,47]. As a major cellular component of the TME, MDSC promote tumor growth and metastasis by stimulating the release of pro-inflammatory cytokines such as TNF and IL-1β and by interacting with tumor cells[48]. Both cytokines are potent activators of the IKK/NF-κB pathway. Sustained NF-κB activation in pre-malignant epithelial cells couples inflammation-driven transcription to invasion and angiogenesis[49]. Previous studies have demonstrated that suppressing MDSC infiltration can reverse immunosuppression within the TME and delay tumor progression[50]. Consistently, our results showed that YHLD significantly reduced the levels of CXCL15, CXCL2, TNF, and IL-1β in the colon tissues of CAC mice, indicating that YHLD exerts anti-inflammatory and immunomodulatory effects during CAC progression.

In addition to immune cells, the ECM constitutes a critical component of the local cellular microenvironment and undergoes dynamic remodeling during tumor development, shifting from a tumor-suppressive to a tumor-promoting state[51]. This pathological remodeling disrupts tissue homeostasis and provides structural support for immune evasion, therapeutic resistance, and invasive metastasis. In tumor tissues from patients with CRC, ECM-related genes, including collagen COL9A3 and the glycoprotein SPP1, are markedly upregulated[52]. COL9A3, a key structural component of the ECM[28], promotes epithelial-mesenchymal transition (EMT) in tumor cells and enhances the proliferation of CAF[30]. CAF, which represent the earliest recruited and most abundant stromal cell population, arise from the activation of quiescent fibroblasts and continuously remodel the ECM while suppressing anti-tumor immunity through the secretion of cytokines such as IL-11[32,53,54]. SPP1, a phosphorylated and sialic acid-rich non-collagenous matrix protein, exhibits pro-inflammatory and pro-migratory tumor-promoting activities and plays an important role in ECM regulation[55]. Moreover, it is a transcriptional target of the transforming growth factor beta/SMAD and Wnt/β-catenin pathways, two central oncogenic cascades in CAC[56]. In addition, CSF3 and IL-23, which are elevated in patients with IBD[57,58], stimulate the proliferation and activation of myeloid cells, including neutrophils, thereby promoting the transition from chronic inflammation to tumorigenesis. In vivo neutralization or in vitro silencing of CSF3 has been shown to significantly inhibit tumor growth[31,59]. Although BMP7 contributes to tissue repair under physiological conditions, it is aberrantly overexpressed in poorly differentiated CRC[60] and promotes CAF activation and ECM deposition[61]. Currently, IL-11, IL-23, SPP1, CSF3, and BMP7 are considered potential therapeutic targets for CRC[62,63]. Our experimental results further demonstrated that YHLD significantly reduced the levels of these cytokines and suppressed the expression of the CAF marker FAP and the MDSC surface markers CD11B and GR-1. Collectively, these findings suggest that YHLD may alleviate immunosuppression and delay CAC progression by multi-target modulation of the TME, including inhibition of CAF activation and MDSC infiltration.

Furthermore, the marked enrichment of AA metabolism and other metabolic pathways warrants particular attention. Metabolic reprogramming is widely recognized as a critical driver of inflammation-associated cancers[64], including CAC. Epidemiological evidence indicates that CRC risk is positively associated with metabolic syndrome and its related factors, such as obesity, physical inactivity, and unhealthy dietary patterns[65]. In parallel, an increasing number of metabolic biomarkers, including triglycerides, linolenic acid, and saturated fatty acids, have been linked to CRC development[66]. Based on these observations, we applied untargeted metabolomic analysis to further characterize alterations in intestinal metabolite composition. Notably, AA metabolism emerged as a significantly enriched pathway in both transcriptomic and metabolomic analyses, highlighting its central role in the therapeutic effects of YHLD. Analysis of differential metabolites demonstrated that YHLD markedly reduced the levels of several pro-inflammatory AA-derived metabolites in the colon tissues of CAC mice, including AA and its downstream products PGE2, PGG2, prostaglandin B2, 5-OxoETE, and 12(S)-HETE. Previous studies have shown that AA is initially converted to PGG2 under the catalysis of PTGS2 and subsequently transformed into PGE2 by PGE2 synthase. Clinical evidence further indicates that PTGS2 is highly expressed in CRC tumor tissues[35]. As a major AA-derived mediator, PGE2 not only induces the release of inflammatory cytokines such as TNF and IL-1β but also directly promotes tumor cell proliferation, migration, and immunosuppression[36]. Importantly, PGE2 has been shown to recruit MDSC to CRC tissues[67], and inhibition of PTGS2 or blockade of PGE2 receptors can significantly attenuate the immunosuppressive activity of MDSC[35,68]. In addition, ALOX12 catalyzes AA to generate 12(S)-HETE[69]. During CAC progression, tumor cells produce high levels of 12(S)-HETE, which activates CAF, enhances their contractile properties, and promotes tumor invasiveness[70]. In this study, YHLD treatment significantly reduced the protein expression of PTGS2 and ALOX12 in the colon tissues of CAC mice, indicating that YHLD alleviates CAC by suppressing key enzymes involved in AA metabolism.

By integrating the prototype compounds of YHLD identified in plasma by UPLC-Q-TOF-MS with molecular docking results, we propose that Psammosilenin A, Pentahydroxybufostane, Clematoside B, and Dulcoside A represent the major bioactive constituents responsible for the therapeutic effects of YHLD. Previous studies have confirmed the anti-inflammatory and anti-tumor activities of these compounds. As a multi-component herbal formulation, YHLD may therefore exert synergistic anti-CAC effects through the combined actions of these constituents. Importantly, these compounds showed strong binding affinities to PTGS2 and ALOX12, further supporting the notion that YHLD regulates AA metabolism by targeting these enzymes. Together, these results provide preliminary molecular evidence for a multi-component, multi-target synergistic mechanism underlying the anti-CAC activity of YHLD. Nevertheless, CAC is a multifactorial pathological process driven by chronic inflammation, immune dysregulation, ECM remodeling, and metabolic reprogramming. Moreover, as a multi-component Dai medicine preparation, YHLD is expected to exert its effects through multiple targets and pathways simultaneously. In-depth investigations into other potential pathways or proteins involved and further in vivo and in vitro activity assays are required to clarify the active components of YHLD.

CONCLUSION

This study demonstrates that YHLD has significant therapeutic potential against CAC. Mechanistically, YHLD regulates PTGS2-mediated and ALOX12-mediated AA metabolism, suppresses CAF and MDSC activation, and reduces the secretion of associated pro-inflammatory and pro-tumor factors, thereby remodeling the TME and alleviating CAC progression (Figure 8). These findings provide a strong scientific basis for the modern development and clinical application of YHLD and offer a promising natural product-based strategy for the prevention and treatment of CAC.

Figure 8 Ya-Han-Lu-Duan regulates arachidonic acid metabolism by downregulating expression of PTGS2 and ALOX12, synergistically reducing myeloid-derived suppressor cells infiltration and cancer-associated fibroblast activation, improving the tumor microenvironment, and alleviating colitis-associated colorectal cancer. PGE2: Prostaglandin E2; PGG2: Prostaglandin G2; AA: Arachidonic acid; MDSC: Myeloid-derived suppressor cell; TNF: Tumor necrosis factor; IL: Interleukin; CAF: Cancer-associated fibroblast.