Cancer-associated adipocytes promote peritoneal metastasis of colorectal signet ring cell carcinoma via FABP4 induction

Abstract

Background

Colorectal signet ring cell carcinoma (SRCC) is a rare and aggressive subtype with a high propensity for peritoneal metastasis, yet the underlying mechanisms remain poorly understood.

Methods

We isolated cancer-associated adipocytes (CAAs) from omental tissue adjacent to SRCC peritoneal metastases and examined their morphological and metabolic features compared to normal adipocytes (NAs). Co-culture systems, patient-derived organoids (PDOs), transcriptomic/metabolomic profiling, and peritoneal metastasis mouse models were employed to assess the functional impact of CAAs. The role of fatty acid binding protein 4 (FABP4) and its regulation via CAA-derived exosomes was also investigated.

Results

CAAs exhibited a dedifferentiated phenotype, enhanced free fatty acid secretion, and upregulation of matrix metalloproteinases. Co-culture with CAAs significantly promoted SRCC PDO proliferation, stemness, and peritoneal metastasis, accompanied by a metabolic shift toward fatty acid utilization. Among fatty acid metabolism–related genes, FABP4 was markedly upregulated in peritoneal metastases and associated with poor prognosis. Functional assays confirmed that FABP4 promoted fatty acid oxidation (FAO), stemness, and metastasis in PDOs, while FABP4 knockdown abrogated these effects. Mechanistically, CAA-derived exosomes induced FABP4 expression in PDOs, and inhibition of exosome release reversed the pro-tumorigenic phenotypes.

Conclusions

CAA-derived exosomal signaling promotes SRCC aggressiveness through FABP4-mediated fatty acid metabolic reprogramming, identifying FABP4 as a potential therapeutic target for peritoneal metastasis in colorectal SRCC.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12964-025-02569-2.

Introduction

Colorectal signet ring cell carcinoma (SRCC) is a rare and aggressive subtype of colorectal cancer, accounting for less than 1% of all cases [1, 2]. It is characterized by diffuse infiltration, abundant intracytoplasmic mucin, and a lack of glandular formation. Clinically, SRCC is notorious for its high propensity to develop peritoneal metastasis at early stages, contributing to poor prognosis and limited therapeutic responsiveness [3–5]. Unlike hematogenous or lymphatic dissemination, peritoneal metastasis involves transcoelomic seeding of tumor cells into the peritoneal cavity, where they encounter and interact with the adipose-rich microenvironment of the peritoneum [6, 7].

The peritoneum is anatomically composed of mesothelial layers supported by a stroma rich in immune cells, fibroblasts, and notably, adipocytes—particularly in structures such as the omentum and mesentery [8]. These adipocyte-enriched regions provide not only structural scaffolding but also abundant lipid-derived nutrients, cytokines, and extracellular vesicles that can influence tumor behavior. Emerging evidence in other malignancies, such as ovarian, gastric, and pancreatic cancers, suggests that adipocyte-derived signals in the peritoneal niche actively promote tumor growth, metabolic reprogramming, immune evasion, and metastatic colonization [9].

Upon interaction with tumor cells, adipocytes can undergo phenotypic and functional transformation into so-called cancer-associated adipocytes (CAAs), characterized by loss of lipid content, dedifferentiation, and increased secretion of free fatty acids and inflammatory mediators [10]. CAAs have been shown to support tumor progression by promoting invasion, resistance to therapy, and metabolic adaptation [11–13]. Despite this, the role of CAAs in colorectal SRCC peritoneal metastasis remains poorly understood, and the molecular pathways underlying their protumorigenic effects have not been systematically studied.

In this study, we focus on the interplay between CAAs and colorectal SRCC in the context of peritoneal metastasis. We hypothesize that CAAs within the peritoneal adipose niche promote SRCC progression by driving lipid metabolic reprogramming in tumor cells. Using patient-derived organoids (PDOs) and matched CAAs from peritoneal metastatic tissues, we explore how CAAs influence SRCC growth, stemness, and dissemination. Through integrative transcriptomic, metabolomic, and functional analyses, we identify fatty acid–binding protein 4 (FABP4) as a key mediator linking exosome-driven lipid signaling from CAAs to tumor metabolic reprogramming. Our findings uncover a previously underappreciated mechanism through which the peritoneal lipid-rich environment facilitates SRCC aggressiveness, with implications for therapeutic targeting of metabolic vulnerabilities in peritoneal metastasis.

Materials and methods

Isolation of cancer-associated and normal adipocytes

Primary adipocytes were isolated from omental tissue adjacent to peritoneal metastatic lesions of SRCC patients (cancer-associated adipocytes, CAAs) and from non-tumorous omental tissue (normal adipocytes, NAs) based on previous published protocol [14]. Briefly, omental tissue samples were obtained with patient consent under approved institutional protocols. Tissues were kept moist with PBS during transfer to minimize ischemic damage. For adipocyte isolation, tissue samples were finely minced and digested using collagenase I solution (0.2% w/v in adipocyte culture medium) at 37 °C with gentle shaking for 1 h. Digested samples were filtered through a nylon mesh (250 μm) to remove undigested tissue, and the filtrate was centrifuged to separate adipocytes. Floating adipocytes were carefully collected and washed in adipocyte culture medium to remove residual collagenase and other contaminants. Cell viability was verified using Calcein-AM staining, ensuring an above 90% viability rate for further experiments.

Lipid droplet staining and flow cytometric quantification

Cells were stained with BODIPY 493/503 (Thermo Fisher, Cat# D3922; 1 µg/mL, 15 min, room temperature, protected from light), washed with PBS, and analyzed by flow cytometry (e.g., BD LSRFortessa; 488 nm excitation, 530/30 nm emission). Dead cells were excluded using PI or Hoechst. Gating: FSC/SSC to remove debris → singlets → live cells. Neutral lipid content was reported as mean fluorescence intensity (MFI) and normalized to the NA control. Unless otherwise indicated, data represent n = 3 independent experiments.

Cell size measurement

Brightfield images were acquired with a 10× objective (phase-contrast enabled) under identical exposure settings. For each group, ≥ 5 random fields were imaged. Cell size was quantified in ImageJ (NIH) after spatial calibration using a stage micrometer; the diameter (Feret or equivalent circular diameter) of at least 100 non-overlapping adipocytes per group was measured.

Human colorectal cancer samples

SRCC tissue samples were collected from patients at the Fudan University Shanghai Cancer Center (FUSCC). These tissues were used for gene and protein expression analyses and tissue microarray preparation. All participants provided informed written consent in accordance with the regulations of the Institutional Review Boards of FUSCC.

Establishment of SRCC-PDO

Based on the methods we reported previously, SRCC-PDO was successfully insolated and cultured [15]. Tumor tissues were finely minced and enzymatically digested using a collagenase solution to achieve single-cell dissociation. The dissociated cells were then embedded in Matrigel, forming three-dimensional droplets that were plated and overlaid with a specialized culture medium optimized for intestinal organoids. The organoids were maintained in a humidified incubator at 37 °C with 5% CO₂, with medium changes every 2–3 days. For passaging, organoids were mechanically dissociated into smaller fragments and re-seeded in fresh Matrigel to establish continuous cultures.

Quantification of triacylglycerol and neutral lipids

For the quantitative estimation of triglycerides in cells, a Triglyceride Assay Kit (ab65336, Abcam, Cambridge, UK) was used in accordance with the manufacturer’s protocols. The lipophilic fluorescence dye BODIPY 493/503 (D3922, Invitrogen) was applied to stain the neutral lipid droplets, and flow cytometry (MoFlo XDP, Beckman Coulter, Pasadena, CA, USA) was conducted to quantify the neutral lipid content.

FAO quantification

The FAO rate in CRC cells with and without PTPRO knockdown and overexpression was assessed using a fatty acid oxidation assay kit (ab118183, Abcam) in accordance with the manufacturer’s instructions.

Establishment of peritoneal metastasis model

To evaluate the peritoneal metastatic potential of SRCC-PDOs, we utilized RAG1 gene-knockout immunodeficient mice (6–8 weeks old, female). Under sterile conditions, each mouse received an intraperitoneal injection of 5 × 10⁶ GFP-labeled SRCC-PDO cells suspended in 200 µL of sterile PBS. Following injection, the mice were housed under standard laboratory conditions. Peritoneal metastasis formation was monitored weekly for up to 4 weeks using an in vivo imaging system (e.g., IVIS Spectrum) to visualize tumor progression and dissemination. Imaging was initiated on day 7 post-injection and conducted at weekly intervals thereafter.

Metabolomic profiling of tissue samples

A combination of gas chromatography-time-of-flight mass spectrometry (GC-TOFMS, LECO Corp., St Joseph, MI, USA) and ultra-performance liquid chromatography tandem mass spectrometry (UPLC-MS/MS, Waters Corp., Milford, MA, USA) was used to quantify small molecule metabolites in the tissue samples. Metabolomics assays were conducted by Metabo-Profile Inc. (Shanghai, China) using previously published methods [16]. The metabolites were identified by comparison with an internal library built using standard reference chemicals.

Exosome isolation and characterization

Exosomes were isolated from the conditioned media of CAAs and NAs following a standard differential ultracentrifugation protocol. Briefly, CAAs and NAs were cultured in exosome-depleted medium (Gibco™, #A2720801) for 48 h. The collected supernatant was subjected to sequential centrifugation steps to remove cells and debris: 300 × g for 10 min, 2,000 × g for 20 min, and 10,000 × g for 30 min at 4 °C. The clarified supernatant was then ultracentrifuged at 100,000 × g for 90 min at 4 °C using an Optima™ L-100XP ultracentrifuge (Beckman Coulter) with a Type 70 Ti rotor. The exosome-containing pellet was washed in phosphate-buffered saline (PBS) and ultracentrifuged again at 100,000 × g for 90 min to ensure purity.The final exosome pellet was resuspended in PBS and stored at − 80 °C for further experiments. Exosome concentration was determined by bicinchoninic acid (BCA) protein assay (Thermo Fisher, #23225), and particle size distribution and concentration were assessed using Nanoparticle Tracking Analysis (NTA, NanoSight NS300, Malvern Instruments). Western blot analysis was performed to confirm the expression of classical exosome markers CD63 (Abcam, #ab216130) and CD81 (Abcam, #ab79559).

For functional assays, equal amounts of exosomal protein (20 µg per well) were added to SRCC-PDO cultures in the transwell system or organoid medium, as specified in experimental designs. For inhibition experiments, exosome secretion was blocked using GW4869 (Selleckchem, #S7609) at a final concentration of 10 µM for 24 h.

In vivo imaging

SRCC-PDOs (5 × 10⁶ cells) were injected intraperitoneally into RAG1–/– mice (n = 5 per group). On day 28, fluorescence signals were acquired using the IVIS Spectrum system (PerkinElmer), and abdominal region signals were quantified using Living Image software. The relative fluorescence level was calculated by normalizing each value to the mean fluorescence of the CAAs– group.

Immunohistochemistry (IHC) staining

IHC staining was conducted as described in our previous article [17]. FABP4 expression was evaluated on TMAs comprising 114 SRCC samples by two pathologists independently who were blinded to the clinicopathological information. The scores of 0 (negative), 1 (weak), 2 (medium), and 3 (strong) were used for staining intensity quantification, while the scores of 0 (< 5%), 1(5–25%), 2 (26–50%), 3 (51–75%) and 4 (>75%), which were based on the percentages of the positive staining areas in relation to the whole cancerous area, were used to evaluate the extent of staining. Then, the immunoreactivity score (IRS) was calculated by multiplying scores of staining intensity and percentages of positivity. Cases with a final staining score of ≤ 4 were deemed as low expression and those with IRS of >4 were reckoned as high expression.

Survival analysis

Overall survival (OS) was analyzed in 114 patients from the TMA cohort who had available clinical follow-up and FABP4 IHC data. Patients were classified into FABP4-positive (n = 76) and FABP4-negative (n = 38) groups using the IRS threshold of 4. Survival curves were plotted using the Kaplan–Meier method and compared with log-rank test. Follow-up time ranged from 3 to 63 months (median 30 months).

Western blotting

Briefly, equal quantities of cellular proteins were resolved using sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred onto polyvinylidene difluoride membranes, and subjected to immunoblotting using a primary antibody for β-actin (1/1,000 dilution; Abcam, Cambridge, UK) and incubating the membranes overnight at 4 °C. After incubation with the secondary antibody, the blots were visualized using ECL (Pierce, Thermo Scientific, USA), and the ECL intensity was detected using a BioImaging System. Band intensities were quantified using ImageJ software (NIH). The integrated density of each protein band was normalized to its corresponding β-actin loading control. Results were plotted as relative expression levels in bar graphs.

Statistical analysis

All described results are representative of at least three independent experiments. Data are presented as means ± standard deviations (SD). Statistical analyses were performed using SPSS 22.0 software (SPSS) or GraphPad Prism 99 software (GraphPad). Student’s t-test or one-way analysis of variance (ANOVA) was used to evaluate the significance of differences between groups. Statistical significance was defined as P < 0.05.

Results

Cancer-associated adipocytes exhibit dedifferentiated phenotypes and activated metabolic state

CAAs were isolated from omental tissue adjacent to peritoneal metastases of colorectal SRCC, whereas normal adipocytes (NAs) were obtained from healthy omental fat. Morphologically, CAAs displayed a dedifferentiated appearance characterized by reduced cell size and diminished lipid droplet content, as revealed by BODIPY staining (Fig. 1A). Gene expression analysis showed a significant downregulation of mature adipocyte markers, including hormone-sensitive lipase (HSL), adiponectin (APN), resistin, and aP2 (Fig. 1B–F), indicating a loss of terminal adipocyte differentiation. In contrast, expression levels of matrix metalloproteinases (MMP2, MMP9, and MMP11), which are implicated in extracellular matrix remodeling and tumor invasion, were markedly increased in CAAs (Fig. 1G–J). Furthermore, untargeted metabolomic profiling demonstrated that CAAs secreted substantially higher levels of free fatty acids and other lipid-related metabolites compared to NAs, reflecting a metabolically reprogrammed and activated state (Fig. 1K). This phenotypic shift aligns with observations in other cancers, such as ovarian and breast cancer, where CAAs promote tumor progression by creating a nutrient-rich microenvironment [18].

Fig. 1.

Characterization of CAAs derived from SRCC peritoneal metastases. A Representative BODIPY 493/503 staining of lipid droplets in CAAs and NAs isolated from omental tissue adjacent to SRCC peritoneal metastases and healthy omentum, respectively. Neutral lipid content was quantified by flow cytometry as mean MFI normalized to NAs, and cell size was measured from brightfield images by ImageJ (≥ 100 adipocytes per group). CAAs exhibited significantly lower neutral lipid levels and smaller cell size than NAs. B-E Quantitative RT-PCR analysis of adipocyte differentiation markers, including HSL, APN, resistin, and aP2, showing significant downregulation in CAAs compared to NAs. F Relative protein expression levels of mature adipocyte markers (HSL, APN, resistin, and aP2) were significantly downregulated in CAAs. Western blot analyses were performed in three independent biological replicates, and representative results are presented. G-I Matrix metalloproteinase expression analysis demonstrating significant upregulation of MMP2, MMP9, and MMP11 in CAAs compared to NAs. J Representative Western blot showing MMP2, MMP9, and MMP11 expression in NAs and CAAs and densitometric quantification of protein bands normalized to β-actin, showing elevated expression in CAAs. Western blot analyses were performed in three independent biological replicate. K Heatmap of representative lipid metabolites detected in conditioned medium from CAAs) and NAs (n = 3 per group). Each row represents an individual patient-derived sample. Data were z-score transformed prior to visualization All quantifications are shown as mean ± SD from three independent biological replicates (n = 3). *p < 0.05 by unpaired t-test

CAAs promote SRCC proliferation and peritoneal metastasis potentially through altering lipid metabolism

To investigate the functional impact of CAAs on tumor progression, we co-cultured SRCC patient-derived organoids (SRCC-PDOs) with CAAs and observed a marked increase in organoid proliferation and spheroid formation compared with those co-cultured with NAs (Fig. 2A–B). In a peritoneal metastasis model using RAG1-deficient mice, PDOs preconditioned with CAAs produced significantly more—and larger—peritoneal nodules, confirming the tumor-promoting role of CAAs in vivo (Fig. 2C). Transcriptomic and metabolomic profiling of PDOs after CAA co-culture revealed substantial accumulation of free fatty acids (Fig. 2D) and strong enrichment of fatty-acid–metabolism pathways, together with gene signatures linked to stemness and metastasis (Fig. 2E–F). Consistently, key regulators of fatty-acid uptake (e.g., CD36), intracellular transport (e.g., FABPs), β-oxidation (e.g., CPT1A), and lipid storage (e.g., DGAT2) were robustly up-regulated (Fig. 2G).

Fig. 2.

CAAs enhance SRCC-PDO proliferation, metastasis, and metabolic pathway activation. A Schematic of co-culture system and analyzing procedure in this study. B Comparison of SRCC-PDO proliferation and spheroid formation following co-culture with CAAs. B Representative images and quantification of spheroid formation assays indicating enhanced spheroid formation capacity of PDOs co-cultured with CAAs. C In vivo imaging of peritoneal dissemination of SRCC-PDOs co-cultured with or without CAAs. n = 5 mice per group. Fluorescence intensity was quantified in the abdominal region using the IVIS Spectrum system, and normalized to the median intensity of the CAAs– group. Data are presented as mean ± SD. D Metabolomic analysis revealing increased levels of free fatty acids in PDOs co-cultured with CAAs compared to NAs. E-F Transcriptomic analysis showing activation of fatty acid metabolism pathways and enhanced enrichment of tumor stemness and metastasis-related pathways in PDOs co-cultured with CAAs. G Heatmap of key genes involved in fatty acid uptake, transport, oxidation, and storage, highlighting significant upregulation in PDOs co-cultured with CAAs. All quantifications are shown as mean ± SD from three independent biological replicates (n = 3). *p < 0.05 by unpaired t-test

Taken together, these results show that CAAs stimulate SRCC-PDO growth and peritoneal dissemination while driving a pronounced shift toward lipid-based metabolic programs. Given the well-established role of lipid metabolism in maintaining tumor stemness and facilitating metastatic spread, we speculate that CAAs may promote SRCC peritoneal metastasis by reprogramming fatty-acid metabolism in tumor cells.

FABP4 is highly expressed in peritoneal metastases and predicts poor prognosis in colorectal SRCC

To explore lipid metabolism–related genes involved in SRCC peritoneal metastasis, we performed quantitative PCR analysis on paired primary and peritoneal metastatic tumor samples from 12 colorectal SRCC patients. Among the fatty acid metabolism–related genes analyzed, fatty acid–binding protein 4 (FABP4) showed the most prominent upregulation in peritoneal metastatic lesions compared to their matched primary tumors (Fig. 3A–B).

Fig. 3.

FABP4 is upregulated in peritoneal metastases and correlates with poor prognosis. A-B Quantitative PCR analysis of FABP4 expression in paired SRCC primary and peritoneal metastatic (SRCC-PM) tissue samples (n = 12), showing significant upregulation of FABP4 in metastatic tissues. C Representative IHC images of FABP4 expression in colorectal SRCC tissue microarray samples. In total, 114 patient samples were analyzed. FABP4 positivity was defined as score >4, based on staining intensity and the proportion of positive tumor cells. Insets show high-magnification views. D–E Kaplan–Meier curves showing OS and DFS of colorectal SRCC patients stratified by FABP4 expression status (positive: n = 76; negative: n = 38). Patients were from the TMA cohort in Fig. 3C. p-value calculated by log-rank test. All quantifications are shown as mean ± SD from three independent biological replicates (n = 3). *p < 0.05 by unpaired t-test

To further assess the clinical relevance of FABP4, we conducted immunohistochemical staining on a colorectal SRCC tissue microarray comprising 114 patients (Fig. 3C). The detailed clinical pathological features of these patients were described in Table S1. Further survival analysis demonstrated that high FABP4 expression was significantly associated with shorter overall survival (Fig. 3D) and disease-free survival (Fig. 3E), suggesting that FABP4 may serve as a prognostic biomarker and potential mediator of peritoneal dissemination in SRCC.

FABP4-mediated fatty acid metabolic reprogramming promotes stemness and metastasis in SRCC

To investigate the functional significance of FABP4 upregulation in SRCC progression, we examined its role in mediating lipid metabolic reprogramming and tumor aggressiveness. Western blot analysis confirmed that FABP4 protein levels were markedly elevated in PDOs co-cultured with CAAs compared to controls (Fig. 4A). Further functional characterization revealed that co-culture with CAAs led to a significant increase in intracellular triglyceride levels in PDOs (Fig. 4B), accompanied by lipid droplet accumulation as shown by BODIPY staining (Fig. 4C). These effects were largely abrogated upon FABP4 knockdown, which restored triglyceride levels and reduced lipid deposition. In line with this, FABP4 silencing markedly reduced FAO activity, as evidenced by decreased acetyl-CoA production (Fig. 4D), reduced FAO rate (Fig. 4E–F), and diminished ATP generation (Fig. 4G), indicating impaired energy output from fatty acid catabolism. Importantly, FABP4 knockdown reversed the CAA-induced upregulation of stemness markers (CD44, CD133) and epithelial–mesenchymal transition (EMT)-related changes, including increased N-cadherin and reduced E-cadherin expression (Fig. 4H–I). These results underscore the role of FABP4-mediated lipid metabolism in maintaining the stem-like and invasive phenotype of SRCC cells. Functionally, FABP4 knockdown significantly suppressed PDO proliferation (Fig. 4J), impaired spheroid formation capacity (Fig. 4K), and reduced peritoneal metastasis in vivo (Fig. 4L), highlighting the essential role of FAO in supporting the metabolic demands of aggressive SRCC phenotypes.

Fig. 4.

CAAs induced FABP4 expression regulates lipid metabolism, stemness, and peritoneal metastasis in SRCC. A Western blot analysis confirming increased FABP4 protein levels in PDOs co-cultured with CAAs compared to NAs. Western blot analyses were performed in three independent biological replicates, and representative results are presented. B Quantification of triglyceride levels in PDOs co-cultured with CAAs, showing significant upregulation, which was reversed by FABP4 silencing. C BODIPY staining showing increased lipid accumulation in PDOs co-cultured with CAAs, abrogated by FABP4 silencing. D-G Functional assays demonstrating reduced acetyl-CoA production, FAO rates, and ATP generation in FABP4-silenced PDOs compared to controls. H-I qRT-PCR analyses showing reversal of CAA-induced EMT (E-cadherin and N-cadherin expression) and stemness marker expression (CD44, CD133) upon FABP4 silencing. J-K Proliferation and spheroid formation assays showing suppression of PDO growth and stemness in FABP4-silenced cells. L In vivo peritoneal metastasis model demonstrating reduced metastatic capacity of FABP4-silenced PDOs compared to controls. All quantifications are shown as mean ± SD from three independent biological replicates (n = 3). *p < 0.05 by unpaired t-test

FABP4 expression induction by CAA-derived exosomes promotes stemness and peritoneal dissemination in SRCC

Adipocyte-derived exosomes are recognized as key mediators of intercellular communication, capable of regulating gene expression in tumor cells. To investigate whether exosomal signaling contributes to FABP4 upregulation, we isolated exosomes from CAAs and NAs (Fig. 5A-B) and treated SRCC-PDOs with equal quantities of each. PDOs exposed to CAA-derived exosomes exhibited a marked increase in FABP4 expression compared to those treated with NA-derived exosomes (Fig. 5C). To confirm the role of exosomes in this regulatory process, we inhibited exosome secretion using GW4869. Treatment with GW4869 significantly suppressed FABP4 upregulation in PDOs co-cultured with CAAs (Fig. 5D), supporting the notion that CAAs modulate FABP4 expression via exosomal delivery. Further analyses revealed that GW4869 treatment also reversed the CAA-induced upregulation of tumor stemness markers (CD44, CD133) (Fig. 5E and F) and EMT-associated genes, including N-cadherin and E-cadherin (Fig. 5G-H). Functionally, inhibition of exosome secretion significantly impaired PDO spheroid formation (Fig. 5I), reduced cell proliferation (Fig. 5J), and diminished peritoneal metastasis in vivo (Fig. 5K), suggesting that exosome-mediated FABP4 induction plays a critical role in sustaining SRCC aggressiveness.

Fig. 5.

CAA-derived exosomes mediate FABP4 induction and promote SRCC stemness and metastasis. A Transmission electron microscopy (TEM) image of exosomes isolated from CAAs and NAs. B Western blot analysis of exosomal protein markers CD63, CD81, and TSG101 in EVs isolated from NAs and CAAs. Marker expression confirmed the identity of the isolated EVs. C Western blot showing FABP4 upregulation in PDOs treated with CAA-derived exosomes. Western blot analyses were performed in three independent biological replicates, and representative results are presented. D FABP4 induction was blocked by exosome release inhibitor GW4869. Western blot analyses were performed in three independent biological replicates, and representative results are presented. E–H GW4869 treatment reversed CAA-induced upregulation of stemness and EMT markers. I–K GW4869 significantly suppressed CAA-induced PDO spheroid formation, proliferation, and peritoneal metastasis. L Schematic depiction of the promotion on peritoneal metastasis by exosomal FABP4 derived from CAAs in SRCC. All quantifications are shown as mean ± SD from three independent biological replicates (n = 3). *p < 0.05 by unpaired t-test

Discussion

In this study, we provide compelling evidence that cancer-associated adipocytes (CAAs) within the peritoneal fat microenvironment critically contribute to SRCC peritoneal metastasis by facilitating lipid metabolic reprogramming via exosomal FABP4 induction (summarized in Fig. 5K). This model deepens our understanding of how the adipocyte-rich peritoneal niche actively promotes SRCC dissemination, a question that has been understudied despite the clinical prevalence of peritoneal metastases in this aggressive subtype.

Our data show that CAAs undergo dedifferentiation and metabolic activation, consistent with prior reports in breast, ovarian, and colon cancers where adipocytes near tumors become metabolically reprogrammed, enhancing tumor progression and invasion [19]. CAAs secreted high levels of FFAs and drove FAO–related gene expression in SRCC PDOs, mirroring mechanisms observed in breast and ovarian cancer models. Notably, our study extends these findings specifically to colorectal SRCC, which has unique pathological and metastatic features.

Among fatty acid metabolism genes, FABP4 emerged as a key mediator, corroborating its established role in other cancer types. FABP4 has been shown to support invasion and metastasis in colon cancer by ferrying FFAs [20, 21], and in ovarian cancer, adipocyte-induced FABP4 enhances tumor aggressiveness and chemoresistance [12]. Our clinical analysis confirms that FABP4 expression is significantly higher in peritoneal metastatic lesions of SRCC and correlates with poorer prognosis, suggesting its potential as a biomarker and therapeutic target.

Mechanistically, we demonstrate that CAA-derived exosomes upregulate FABP4 in SRCC PDOs—an effect reversed by the exosome inhibitor GW4869. Exosome-mediated adipocyte-tumor cell communication has been implicated in supporting fatty acid metabolism and EMT in melanoma, breast, and ovarian cancers [22]. Our findings thus place FABP4 at the nexus of metabolic and metastatic pathways, highlighting exosomal induction of FABP4 as a novel mechanism in peritoneal metastasis. Functionally, FABP4 knockdown impaired FAO, lipid accumulation, stemness marker expression, and EMT, culminating in reduced proliferation, spheroid formation, and in vivo metastasis [22]. This aligns with yeast studies indicating that FAO provides metabolic intermediates that support cancer stem cell traits and metastatic competence [23].

Despite these insights, our study has several limitations. First, while we establish a causal role for FABP4 in metabolic reprogramming and metastasis, the specific exosomal cargo (e.g., miRNAs, proteins) driving FABP4 upregulation remains unidentified. Future proteomic and RNA sequencing of exosomes will help pinpoint these mediators. Second, though we focused on lipid metabolism, other metabolic pathways—such as glucose metabolism or glutaminolysis—might also be influenced by CAAs and warrant investigation. Third, validation in larger SRCC patient cohorts and the development of FABP4-targeted interventions (e.g., pharmacological inhibitors, antibody-based therapies) are needed to translate these findings.

In conclusion, our study identifies a metabolic axis in which peritoneal CAAs promote SRCC stemness and metastasis by enhancing FABP4-driven FAO through exosomal signaling. Disrupting this lipid-mediated communication—either by blocking exosome release or inhibiting FABP4—may offer new therapeutic strategies to prevent peritoneal dissemination in SRCC.

Authors’ contributions

R.W. had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. W.D., W.L. and H.F. contributed equally to this study. Study concept and design: R.W., S.Z., Y.C. and G.C. Acquisition, analysis, or interpretation of data: All authors.Drafting of the manuscript: W.D., W.L. and H.F. Critical revision of the manuscript for important intellectual content: R.W., S.Z., Y.C and G.C. Statistical analysis: H.F. and R.G.

Funding

This work was supported by the Grant of National Natural Science Foundation of China (No. 82103554, No. 82003317, No. 82173133, No. 82472874).

Data availability

Source data and reagents are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.