Macrophage-derived IL-6 reprograms lipid metabolism to promote colorectal cancer development through USP14-mediated FASN deubiquitination.

Background
Colorectal cancer (CRC) represents a substantial global public health challenge because of its high morbidity and mortality rates [1]. The pathogenesis of CRC is multifaceted and involves a range of factors, such as advanced age, genetic mutations, aberrant epigenetic modifications, and inflammation related to cancer [2]. Among these diseases, inflammatory bowel disease (IBD) is a critical contributor, significantly exacerbating the morbidity and mortality associated with CRC. Individuals with long-standing IBD face a 2–3-fold increased risk of developing a specific subtype of CRC, termed colitis-associated colorectal cancer (CAC) [3]. Tumor-associated inflammation, recognized as the seventh hallmark of cancer, plays a crucial role in the pathogenesis of malignant tumors [4, 5]. Pathological studies indicate that gastrointestinal malignancies are characterized by significant immune cell infiltration, with macrophages constituting a major component [6]. Notably, the degree of macrophage infiltration in intestinal tissues is greater in colorectal cancer (CRC) patients than in healthy individuals, highlighting the critical involvement of macrophages in CRC progression [7]. Despite the well-established importance of macrophages in tumorigenesis, the underlying regulatory mechanisms governing their role remain largely unexplored.
The reprogramming of lipid metabolism is a key adaptive response that fuels cancer progression. This reprogramming is context-specific and fulfills distinct cellular demands. Enhanced de novo synthesis of lipids (driven by enzymes like ACLY, ACC, and FASN) is frequently upregulated to provide essential building blocks for membrane production and bioactive signaling molecules, supporting rapid proliferation. Moreover, increased fatty acid oxidation is often activated under metabolic stress (e.g., nutrient scarcity) to generate ATP and maintain redox balance, promoting cell survival. In cancer, lipid metabolism is dynamically regulated rather than uniformly “high turnover.” The coordinated yet flexible shift between anabolic and catabolic pathways equips tumors with growth and survival advantages, making key regulatory enzymes critical targets. Fatty acids play dual roles as energy sources and signaling molecules, meeting the diverse demands of signaling pathways and lipid membrane synthesis. In contrast to typical normal cells, which primarily acquire fatty acids from exogenous dietary sources, cancer cells predominantly rely on de novo fatty acid synthesis [8]. Notably, serum and tissue lipid levels exhibit abnormal deviations in CRC, where dysregulated lipid metabolism, has emerged as a hallmark intricately linked to CRC pathogenesis and progression [9]. The inhibition of lipid synthesis within CRC cells effectively curtails proliferation, invasion, and angiogenesis and increases chemosensitivity [10]. At the molecular level, numerous genes associated with lipogenesis are highly expressed in cancer cells [11]. The key enzymes include fatty acid synthase (FASN), ATP citrate lyase (ACLY), and acetyl-CoA carboxylase (ACC). Elevated FASN expression has been reported to drive cancer cells into S phase, thereby promoting their proliferation [12]. The literature has reported the regulation of lipid metabolism by factors such as hypoxia, lactate accumulation, and lipid deficiency [13]. Furthermore, a recent study reported the regulatory impact of macrophages on lipid biosynthesis in hepatocytes, suggesting a potential link between macrophages, lipid metabolism reprogramming, and the inflammatory microenvironment in cancer [14].
In this study, we determined the pivotal role of macrophage infiltration in driving lipogenesis in various colon cancer cell lines and mouse models of CRC. Our investigation revealed that macrophage-secreted interleukin-6 (IL-6) plays a pivotal role in increasing the expression of ubiquitin-specific processing protease 14 (USP14), which, in turn, acts as a key mediator of FASN deubiquitination, which is crucial for the regulation of the lipid biosynthesis pathway. In addition, our findings identify 6-gingerol as a potential therapeutic agent for CAC- and USP14-driven diseases. This study not only deepens our comprehension of the intricate molecular mechanisms underpinning inflammation-associated tumorigenesis but also offers a promising approach for the discovery of more effective therapeutic strategies for CAC.

Results

DSS or LPS induction promoted cellular lipid metabolism reprogramming in CRC mouse models
To elucidate the role of chronic inflammation in CRC, we utilized the AOM/DSS-induced colitis-associated colorectal cancer (CAC) model. A separate CRC model induced solely by AOM without chronic inflammation served as a control (Fig. 1A). As anticipated, an additional 15-day DSS treatment markedly accelerated tumor progression (Fig. 1B and C). Notably, compared with that in the CRC model, macrophage infiltration within the colons of the CAC model mice was significantly greater (Fig. 1D). Furthermore, the proportion of CD86⁺CD206⁻ M1-polarized macrophages among CD45⁺F4/80⁺ macrophages was elevated in the CAC group relative to that in the normal control or CRC group, while the proportion of CD86⁻CD206⁺ M2-polarized macrophages was reduced (Figure S1). Metabolic reprogramming, particularly alterations in lipid metabolism, has been recognized as a pivotal instigator in tumorigenesis. Our analysis of lipid levels revealed that the concentrations of fatty acids and triglycerides were substantially greater in the tumor tissues of CAC model mice than in those of CRC model mice (Fig. 1E and F). These insights were further substantiated in a CT26 allograft model. Allograft experiments revealed significantly accelerated tumor growth and elevated lipid levels in C57BL/6J mice subjected to LPS stimulation compared with those in the PBS control group (Fig. 1G-J). To determine whether macrophage infiltration was responsible for the observed increase in lipid levels, we depleted macrophages with clodronate liposomes. Notably, this intervention resulted in the failure of LPS to induce a significant increase in tumor burden and lipid levels (Fig. 1G-J). These findings suggest that macrophage infiltration within tumor tissues promotes tumor growth, an effect that is associated with the reprogramming of lipid metabolism.

FASN is upregulated by LPS-induced macrophage-derived medium in vitro
FASN-catalyzed de novo lipogenesis is a prominent feature of lipid metabolism in tumors [15]. Therefore, we explored the potential role of FASN dysregulation in cancer cells in the context of macrophage-mediated tumor growth. Elevated FASN expression was observed in tumor tissues from LPS-treated allograft and CAC model mice. Additionally, the protein levels of ACLY, ACC, and SCD1 were assessed, revealing no significant alterations. The administration of clodronate liposomes to depleted macrophages resulted in a reduction in LPS-induced FASN expression levels (Fig. 2A-D). Additionally, we collected conditioned medium (CM) from LPS-treated macrophages and added it to HCT-8 cells. Notably, the CM significantly increased lipid accumulation (Fig. 2E). Furthermore, incubation of CM from macrophages upregulated FASN expression in multiple CRC cell lines (Fig. 2F and G). To examine this further, we used a nontargeted lipidomics approach to measure the effects of the conditioned medium on lipids and found that 191 out of the 675 identified lipid species differed significantly, with 95 elevated and 95 decreased lipid species in conditioned medium-treated HCT-8 cells. Most of the greatest increases were in glycerolipids (Fig. 2H-K). These results demonstrate that macrophages may influence CRC progression by modulating FASN-mediated lipogenesis.

Macrophage-secreted IL-6 enhances lipid synthesis via FASN regulation
LPS stimulation led to increased secretion of proinflammatory cytokines and chemokines, including tumor necrosis factor-α (TNF-α), interleukin-1beta (IL-1β), IL-6, IL-12, IL-23, CCL2, CCL5 and CXCL10, in the conditioned medium (Fig. 3A). Prior to incubation with macrophage-derived medium, the cells were treated with neutralizing antibodies. Among these, only the anti-IL-6 neutralizing antibody effectively inhibited the increase in both FASN protein levels and lipid content induced by the macrophage-derived medium (Fig. 3B and C). Moreover, IL-6 markedly upregulated FASN expression (Fig. 3B). Collectively, these results suggest that IL-6 is the primary factor responsible for the upregulation of FASN by macrophages. We further substantiated the role of IL-6 in vivo and demonstrated that IL-6 markedly facilitated tumor progression in CT26 allograft mice (Fig. 3D and E). IL-6 upregulated FASN expression and increased the cellular levels of fatty acids and triglycerides (Fig. 3F-I). Additionally, IL-6 neutralization abolished the LPS-induced promotion of tumor growth and increase in lipid levels in the allograft mice (Figure S2). To elucidate the involvement of FASN in the IL-6-mediated promotion of lipogenesis, we stably knocked down FASN in HCT-8 cells via short hairpin RNA (shRNA). Subsequent cell growth assays and Oil Red O staining revealed that FASN knockdown significantly attenuated the IL-6-induced increase in cell proliferation and lipid droplet accumulation (Fig. 3J and K). Finally, pharmacological inhibition of FASN with C75 in the allograft CRC model significantly reduced tumor size, weight and levels of free fatty acid (FFA) and triglyceride (TG), with an extent comparable to IL‑6 neutralization (Figure S3A-D).

IL-6 regulates FASN via a ubiquitin-dependent pathway
We proceeded to investigate the expression pattern of FASN and observed elevated protein levels in cells treated with IL-6 (Fig. 4A and B). Notably, the FASN mRNA level remained unchanged (Fig. 4C). To further elucidate the impact of IL-6 on FASN stability, we treated cells with cycloheximide (CHX), a protein synthesis inhibitor. Our results revealed an increase in FASN stability in IL-6-treated cells (Fig. 4D and E). Additionally, the IL-6-induced increase in FASN protein levels was effectively blocked by the proteasome inhibitor MG132, but not by the lysosomal inhibitor Bafilomycin A1 (Fig. 4F and G, Figure S4A and B). These findings strongly suggest that IL-6 impedes FASN degradation via the proteasome pathway. Given that FASN undergoes deubiquitination during hepatic lipogenesis in the context of hepatosteatosis, we investigated the potential role of ubiquitination in the IL-6-mediated stabilization of FASN. Our results demonstrated that IL-6 treatment led to a significant reduction in the ubiquitination of FASN in HCT-8 cells. Furthermore, MG-132 markedly increased the accumulation of ubiquitinated FASN, an effect that was attenuated by IL-6 (Fig. 4H). These observations suggest that IL-6 may promote FASN activation through a deubiquitination mechanism. To confirm that FASN degradation regulated by IL-6 occurs in a ubiquitin-dependent manner, we analyzed FASN degradation in the presence of an inhibitor of the endogenous ubiquitination system (TAK243). As expected, IL-6-induced FASN upregulation and BODIPY staining were abolished by TAK243, confirming that the ubiquitination system is involved in IL-6-induced FASN expression (Fig. 4I-L).

USP14 deubiquitinates and stabilizes FASN in IL-6-stimulated CRC cells
E3 ubiquitin ligases and deubiquitinating enzymes play crucial roles in regulating protein stability through ubiquitin-mediated degradation. Within the deubiquitinating enzyme family, ubiquitin-specific proteases (USPs) are known to contribute to protein stabilization [16]. To investigate the relationship between the IL6 level and ubiquitin regulatory proteins, we utilized TIMER (https://cistrome.shinyapps.io/timer/) to analyze correlations with 57 E3 ubiquitin ligases and 52 USPs in a cohort of 457 cases of colon adenocarcinoma. Among these genes, RNF43 and TRIM24 expression was significantly negatively correlated with IL-6 expression. In contrast, the expression levels of 22 USPs were significantly positively correlated with IL-6 expression (Fig. 5A and B). We further examined the mRNA levels of RNF43, TRIM24, and the four USPs that presented the strongest positive correlations with IL-6 (USP14, USP15, USP18, and USP44). Notably, USP14 was highly expressed at both the mRNA and protein levels following IL-6 stimulation (Fig. 5C-E). FASN was detectable in anti-USP14 immunoprecipitates, indicating an endogenous interaction between USP14 and FASN (Fig. 5F). Furthermore, USP14 overexpression led to a notable increase in FASN protein levels without a corresponding increase in FASN mRNA expression, suggesting that USP14 regulates FASN expression at the posttranslational level. Notably, the protein abundance of other lipogenic genes, including SREBP-1c, SCD-1, and ACC1, remained unaffected by USP14 overexpression, underscoring the specificity of the role of USP14 in regulating FASN (Fig. 5G-I). An investigation of FASN’s half-life in HCT-8 cells treated with CHX revealed that overexpression of USP14 significantly prolonged the half-life of FASN, confirming the role of USP14 in stabilizing the FASN protein (Fig. 5J and K). Double immunofluorescence staining further confirmed the colocalization of USP14 and FASN in HCT-8 cells (Fig. 5L). Transfection with shUSP14 accelerated the degradation of FASN (Fig. 5M and N). Additionally, IL-6 failed to induce changes in ubiquitination and protein expression levels in USP14-knockdown cells (Fig. 5O-Q). Collectively, these findings underscore the role of IL-6 in inhibiting the degradation of ubiquitinated FASN through USP14.

IL-6-induced STAT3 phosphorylation activates USP14 transcription
IL-6 enhances the activity of various signaling pathways, including the PI3K/AKT, ERK, STAT3, and NF-κB pathways [17]. Notably, the application of the STAT3 inhibitor stattic led to a significant reduction in FASN levels (Fig. 6A and B). Furthermore, an increase in phosphorylated STAT3 in response to IL-6 stimulation was accompanied by elevated levels of USP14 and FASN (Fig. 6C and D). These findings suggest the involvement of STAT3 in the upregulation of FASN. We subsequently used siRNA to confirm that STAT3 is involved in IL-6-induced FASN upregulation. Knockdown of STAT3 repressed both the protein and mRNA levels of USP14 (Fig. 6E-G). These findings led us to further explore the relationship between STAT3 and USP14. The transcription inhibitor actinomycin D effectively decreased the expression of USP14 and FASN in colon cancer cells treated with IL-6, suggesting the necessity of the transcriptional activation of USP14 for FASN expression (Fig. 6H and I). By performing ChIP‒qPCR analysis, we confirmed that the promoter of USP14 was significantly enriched in the STAT3-immunoprecipitated complex. IL-6 increased USP14 enrichment, whereas STAT3 knockdown attenuated this increase (Fig. 6J). Furthermore, IL-6 significantly increased the luciferase activity of the USP14 promoter, whereas knockdown of STAT3 significantly suppressed luciferase activity under IL-6 induction (Fig. 6K). These data demonstrated that STAT3 acts as an activator to promote USP14 transcription. We also used an overexpression plasmid for STAT3, which led to increases in FASN expression and lipid levels. However, these effects were completely abrogated by USP14 knockdown (Fig. 6L-N). Furthermore, based on the CPTAC colorectal cancer cohort data retrieved via the LinkedOmics Suite (https://www.linkedomics.org), analysis of phosphoproteomic and proteomic datasets revealed that the phosphorylation level of STAT3 at Y705 is significantly positively correlated with USP14 expression (Pearson correlation: 0.563, P: 1.076e-9), indicating a strong association between STAT3 activation and USP14 expression (Fig. 6O). The expression levels of IL-6, p-STAT3, USP14, and FASN were markedly elevated in the colon tumor tissues of AOM/DSS mice compared with those of AOM mice (Fig. 6P). Collectively, these findings suggest that STAT3 activation leads to the upregulation of USP14 expression, which is a critical step in the IL-6-mediated stabilization of FASN and the subsequent promotion of tumor growth.

USP14 inhibition suppresses the growth of CRC
Given the modulation of FASN stability by USP14, we further investigated the expression level of USP14 in human CRC via transcriptome data from the TCGA database. As expected, USP14 was overexpressed in CRC tumors (Fig. 7A). We subsequently investigated whether inhibiting USP14 could abrogate tumor growth in vivo. Knockdown of USP14 in allograft models resulted in reduced tumor volume and tumor weight (Fig. 7B and C). Additionally, USP14 inhibition downregulated lipid levels and suppressed FASN expression (Fig. 7D-F), suggesting the potential of USP14 as a novel therapeutic target for CRC treatment.

Molecular docking analysis was further performed to identify potential USP14 inhibitors, with 6-gingerol (Fig. 7G) emerging as a notable candidate because of its binding energy of -5.9 kcal/mol to USP14. 6-gingerol bound to USP14 through visible hydrogen bonds and strong electrostatic interactions. Moreover, the hydrophobic pocket was successfully occupied by 6-gingerol (Fig. 7H). The interaction between USP14 and 6-gingerol was confirmed through CETSA, suggesting that 6-gingerol specifically targets USP14 (Fig. 7I). Importantly, 6-gingerol significantly decreased FASN protein levels in five CRC cell lines (Fig. 7J and K). Furthermore, 6-gingerol induced apoptosis and suppressed cell proliferation in HCT-8 cells, and these effects were abrogated in FASN-knockdown (shFASN) cells (Fig. 7L-O). The loss of 6-gingerol’s effects in FASN-knockdown cells confirms that its anti-tumor activity is predominantly mediated through targeting and downregulating FASN. Consistent with a USP14-dependent mechanism, the established USP14 inhibitor IU1 also attenuated IL‑6‑induced FASN upregulation (Figures S5A and B). Moreover, the activity of 6-gingerol was abolished in USP14‑knockdown cells, supporting its role as a USP14 inhibitor (Figures S5C and D). Consistent with the in vitro findings, the tumor burden, as well as the protein levels of FASN and USP14, were markedly reduced in AOM/DSS mice following treatment with 6-gingerol (Fig. 7P-T). Additionally, these mice presented lower lipid and IL-6 levels (Fig. 7U-W). Collectively, these results provide compelling evidence that 6-gingerol holds significant promise for the treatment of CAC.

Discussion
The role of macrophages in cancer development is complex and deserves further study. M1-polarized macrophages exert anticancer effects by enhancing tumor immunity within tumor tissues [18]. However, our research demonstrated that, within the CAC model, there was notable progression of colorectal tumors, characterized by increased inflammation and increased macrophage infiltration, in comparison with those in the CRC mouse model. Furthermore, we observed an increased ratio of M1-polarized macrophages in the CAC group compared to both the normal control and CRC groups. In contrast, the ratio of M2-polarized macrophages was downregulated in the CAC model mice. Additionally, in the context of the transplanted tumor model, LPS significantly promoted the growth of the transplanted tumors. Conversely, the depletion of macrophages with clodronate liposomes significantly inhibited LPS-induced tumor growth. These findings collectively substantiate the role of M1 macrophages in facilitating the initiation and progression of colorectal cancer. Consistent with our results, Chao Liu et al.. demonstrated that LPS-induced macrophages promote colorectal cancer development both in vivo and in vitro [19]. Indeed, macrophages and their monocytic precursors represent the largest fraction of leukocytes in most solid tumors and serve as critical drivers of cancer-associated inflammation [20, 21]. A plausible explanation is that activated inflammatory macrophages produce potentially mutagenic reactive nitrogen species (RNS) and reactive oxygen species (ROS) and secrete cytokines, including TNF-α and interleukins (e.g., IL-1β, IL-6, and IL-12). This creates a conducive environment for the initiation and progression of chronic inflammation-associated cancers [22]. Given this distinct shift towards an M1-dominant phenotype in CAC model, we focused our subsequent investigations on elucidating the role of M1-polarized macrophages in regulating CRC growth.
Alterations in the lipid metabolic phenotype are directly driven by extracellular tumor microenvironment (TME) factors such as hypoxia, acidosis, and nutritional alterations [23–25]. However, how the inflammatory microenvironment regulates lipid metabolism in tumors remains unclear. In this study, colorectal cancer tissues from a colitis-associated cancer (CAC) mouse model demonstrated increased macrophage infiltration, accompanied by a significant elevation in lipid levels. In the allograft model, the LPS-induced upregulation of FASN and subsequent tumor progression were inhibited by clodronate liposome administration. Moreover, colorectal cancer cells treated with macrophage supernatant presented an increase in lipid content. Collectively, these findings suggest that macrophages play a regulatory role in the lipid metabolism of colorectal cancer cells. In this study, THP-1 cells, as a human monocytic cell line, were used for in vitro studies modeling tumor-associated macrophages. The THP-1 cell line is a widely utilized, human-derived, monocytic cell line that serves as a cornerstone in vitro model for studying human macrophage biology, innate immune responses, and the role of monocytes/macrophages in disease pathogenesis. However, this model also presents significant limitations, such as inherent biological differences from primary macrophages and an inability to recapitulate the complex three-dimensional architecture, multicellular interaction networks, and dynamic metabolic milieu of the in vivo tumor microenvironment. Furthermore, the specific functions of macrophages, either independently or in conjunction with other immune cells, warrant further investigation in human colorectal cancer samples. Additionally, whether macrophages maintain similar functions in other solid tumors needs to be clarified in future studies.
Cytokines produced by immune cells serve as primary mediators of intercellular communication within the TME. Among them, IL-6 has been extensively studied for its functions in cell cycle regulation, apoptosis, angiogenesis and immune escape [26–30]. However, research on the involvement of IL-6 in lipid metabolism reprogramming remains limited. Earlier studies by Brass et al.. demonstrated that IL-6 enhances lipid synthesis in primary cultures of rat hepatocytes [31]. Furthermore, in IL-6 knockout mice subjected to a high-fat diet, the administration of recombinant IL-6 directly resulted in the upregulation of lipid synthesis enzymes, including ACC, FASN, and SCD1, thereby promoting lipid production [32]. On the other hand, IL-6-mediated chronic inflammation contributes to fat loss in cancer cachexia by modulating lipolysis and white-to-brown transdifferentiation of white adipose tissue [33]. In our investigation, IL-6 was identified as the pivotal factor in macrophage-induced FASN expression using neutralizing antibodies. Our lipidomic analysis revealed extensive lipid remodeling. We therefore profiled key lipogenic enzymes, including FASN, ACC, SCD1 and ACLY, in colorectal tumors from both AOM-induced and syngeneic mouse models. However, only FASN exhibited consistent protein-level upregulation across models, pinpointing FASN-driven lipogenesis as a likely core mechanism of inflammation-induced metabolic rewiring. An unresolved question is whether additional regulatory layers contribute to the observed lipidomic landscape. Notably, our findings revealed that IL-6 does not alter the transcription level of FASN but regulates FASN expression exclusively at the protein level. This observation implies that IL-6 may affect FASN protein expression via posttranslational modification. Increased O-GlcNAc glycosylation of FASN can enhance its interaction with ubiquitin-specific protease 2a (USP2A), which stabilizes FASN by deubiquitinating it [34]. Additionally, the E3 ligase COP1 facilitates FASN ubiquitination and subsequent degradation by binding to FASN via SH2 adaptor proteins [35]. Liu et al.. demonstrated that FASN is a specific substrate for USP14 and that overexpression of USP14 in the liver stabilizes the FASN protein, thereby promoting lipid accumulation [36]. In our study, we observed that IL-6 induces the dissociation of FASN from ubiquitin, reversing the elevated level of FASN ubiquitination induced by the proteasome inhibitor MG-132. These findings suggest that IL-6 may reduce FASN ubiquitination via a deubiquitinase, thereby stabilizing FASN protein expression. Notably, compared with other deubiquitinases, IL-6 significantly upregulated the expression of USP14 at the mRNA level. This upregulation can be reversed by the STAT3 inhibitor stattic. Furthermore, silencing USP14 via shRNA reversed the IL-6-induced increase in FASN protein expression. Co-IP experiments confirmed the interaction between endogenous USP14 and FASN. Furthermore, IL-6 did not reverse the increase in FASN ubiquitination induced by shUSP14. Therefore, we conclude that USP14 stabilizes the FASN protein through deubiquitination and plays a regulatory role in the IL-6-induced increase in FASN protein expression. However, a recent study indicated that USP14 inhibits FASN protein expression in human prostate cancer, breast cancer, and lung cancer cells [37]. These findings suggest that USP14 may play diverse roles in different tumor types, which warrants further investigation.
IL-6 can activate various signaling pathways, including the JAK, Src, STAT3, Ras/Raf/MAPK, PI3K/Akt, and SHP2 pathways, among others [17]. Overactivation of STAT3 can upregulate the expression of the key enzymes ACC and FASN in the de novo lipid synthesis pathway and downregulate the mRNA level of the rate-limiting enzyme ACOX1 in fatty acid beta-oxidation [38]. Moreover, IL-6-STAT3 signaling can stimulate INDY expression, increase the influx of citrate into the cytoplasm, and increase intracellular lipid synthesis [39]. Our research indicated that treatment with the STAT3 inhibitor stattic reversed the IL-6-induced increase in FASN. Furthermore, the transcription inhibitor Act D attenuated the IL-6-mediated upregulation of USP14, suggesting that the increased expression of USP14 is dependent on its transcriptional activation. ChIP‒qPCR experiments revealed that p-STAT3 specifically binds to the promoter region of USP14. Consequently, we conclude that macrophage infiltration, coupled with the secretion of IL-6, results in the activation of STAT3. This activation subsequently enhances USP14 transcription, which stabilizes FASN protein expression via deubiquitination, thereby facilitating de novo lipid synthesis in colorectal cancer. This preliminary finding indicates that macrophage infiltration within the inflammatory microenvironment regulates lipid synthesis metabolism through USP14, a mechanism that has been linked to the occurrence and progression of various tumors [40–43]. However, whether inhibiting USP14 can ultimately suppress the development of CAC requires further validation. To address this, we implemented both a cell line-derived xenograft model and a CAC mouse model. Consistent with the in vitro findings, both models exhibited substantial reductions in tumor growth following USP14 inhibition via shRNA or 6-gingerol. Notably, 6-gingerol, a USP14 inhibitor, significantly inhibited the expression of FASN and IL-6. These findings indicate that 6-gingerol suppresses CAC development primarily by targeting USP14 to downregulate FASN-mediated lipogenesis, and secondarily by inhibiting the IL-6/STAT3 signaling axis. These findings underscore the promising therapeutic potential of targeting USP14 in CRC and offer valuable perspectives for advancing 6-gingerol as a potential anticancer drug.

Conclusions
In summary, macrophage-derived IL-6 induces a metabolic switch to lipogenesis in colorectal cancer within an inflammatory environment. IL-6 activates STAT3, which subsequently binds to the promoter region of USP14, resulting in its upregulation. USP14 interacts with FASN by inhibiting its ubiquitination and subsequent degradation, thereby positively regulating lipogenesis. Notably, 6-gingerol, a USP14 inhibitor, significantly impaired FASN stabilization in CRC cells, leading to a reduction in de novo lipid biosynthesis (Fig. 8). Consequently, this study revealed a critical role of the IL-6-mediated STAT3-USP14-FASN axis in lipid metabolism reprogramming in CAC.

Methods

Cell culture
The human colorectal cancer cell lines HCT-8, HCT-116, LoVo, SW480, and HT-29; the mouse colorectal cancer cell line CT26-WT; and the human acute monocytic leukemia cell line THP-1 were all purchased from the Cell Bank of the Shanghai Institute of Biochemistry and Cell Biology (Shanghai, China). All the cell lines were authenticated via STR profiling and confirmed to be free of mycoplasma contamination via PCR. The cells were cultured in F12K (Cat#KGM21129N-500, KeyGEN BioTECH, Nanjing, China), Leibovitz’s L-15 (Cat#KGM41300N-500, KeyGEN BioTECH), McCoy’s 5 A (Cat#KGM4892N-500, KeyGEN BioTECH), RPMI-1640 (Cat#KGM31800N-500, KeyGEN BioTECH) or DMEM (Cat#KGM12800N-500, KeyGEN BioTECH) supplemented with 10% fetal bovine serum (FBS, Cat#AUS-01E-02, Cell-Box Biological, Hong Kong, China) and 1% penicillin‒streptomycin (Cat#ST488S, Beyotime, Shanghai, China) in a 5% CO2 atmosphere.

Animal models
Male 8-week-old C57BL/6J mice were purchased from Vital River Laboratories (Beijing, China) and housed in individually ventilated cages on a standard chow pellet diet with access to water and a 12-hour light/12-hour dark cycle. All animal experiments performed in this study were conducted under the National Research Council’s Guide for the Care and Use of Laboratory Animals and approved by the Ethics Committee for Animal Experiments of Anhui Medical University (Approval No. LLSC20200068). Investigators were not blinded during data collection or analysis.
For the chemically induced murine CRC models, sixteen 8-week-old male C57BL/6J mice were randomly divided into two groups of eight: a CRC model group and a colitis-associated CRC (CAC) model group (8 mice per group). The CAC model was induced via AOM (Cat#5486, Sigma‒Aldrich, St. Louis, USA) and three cycles of DSS (Cat#60316ES76, Yeasen Biotechnology, Shanghai, China) in the drinking water [44]. Another CRC model was induced with AOM alone [45]. The mice received intraperitoneal injections of 15 mg/kg AOM weekly for four weeks. For treatment with 6-gingerol (Cat#FY1342, FEIYUBIO, Jiangsu, China), 6-gingerol was administered orally at 100 mg/kg every other day throughout the CAC modeling period. For the allograft model, 1 × 107 CT26 cells were injected subcutaneously into 8-week-old male C57BL/6 mice, which were then randomly grouped (at least 5 mice per group). For lipopolysaccharide (LPS, Cat# L2630, Sigma‒Aldrich) treatment, the mice received 10 µg of LPS or PBS intraperitoneally every 6 days for 30 days post-inoculation. For macrophage depletion, 100 µL of clodronate or PBS liposomes (Cat# CP-005-005, LIPOSOMA B.V., Amsterdam, Netherlands) was administered intraperitoneally every 5 days. For IL-6 (20 ng/g, Cat# 216 − 16, PeproTech, Rocky Hill, USA) treatment, the mice were intratumorally injected with IL-6 every other day for 30 days. For anti-IL-6 neutralizing antibody (200 ng/g, Cat# 16-7061-81, Invitrogen, Carlsbad, USA) treatment, the mice received an intratumoral neutralizing antibody injection for 30 days. For C75 (10 µg/g, Cat# 218137-86-1, MedChemExpress, Monmouth Junction, USA) treatment, the mice received C75 intraperitoneally every other day. Tumors were evaluated every 3 days, and tumor volume was calculated via the following formula: tumor volume = 0.5 × length × width2.

Hematoxylin and eosin (H&E) staining and immunohistochemistry
Colon tissues were fixed in 4% (w/v) paraformaldehyde, dehydrated, embedded in paraffin and cut into 5 μm sections. For H&E staining, the sections were stained with H&E (Cat# C0105S, Beyotime). For immunohistochemistry, the slices were incubated with antibodies against IL-6 (Cat#AF-206-NA, R&D Systems, Minneapolis, Minnesota, USA), p-STAT3 (Cell Signaling Technology, Danvers, Massachusetts, USA), USP14 (Proteintech, Wuhan, Hubei, China) and FASN (Cat#10624-2-AP, Proteintech) and then treated with an HRP-labeled secondary antibody (Proteintech). After the final round of staining, DAPI (Cat#sc-3598, Santa Cruz Biotechnology, Texas, USA) solution was added. Photomicrographs were obtained via a light microscope (Leica, Germany).

Lipid accumulation assay
Cellular lipid staining was performed with Oil Red O (Cat# 1320-06-5; Alladin Biochemical, Shanghai, China) and a BODIPY 493/503 stain kit (Cat# GC42959; GLPBIO, California, USA). The cells were gently washed with PBS, fixed in 4% paraformaldehyde for 30 min and stained with freshly prepared Oil Red O or BODIPY 493/503 staining solution for 30 min at 37 °C. For Oil Red O staining, the cell nuclei were stained with hematoxylin for 2.5 min. For BODIPY 493/503 staining, the nuclei were stained by adding DAPI for 30 min. Similarly, quantitative determination of lipids was performed with triglyceride (Cat#ab65336, Abcam, Cambridge, UK) and free fatty acid quantification kits (Cat#ab176768, Abcam) and measured with a microplate reader (PerkinElmer, USA) according to the manufacturer’s instructions.

Western blot analysis
Proteins were extracted via RIPA buffer (Cat# P0013C, Beyotime) and assayed via a BCA protein assay (Cat# KGB2101, KeyGEN BioTECH). An equal amount of protein from each sample was loaded per lane for 4–12% SDS‒PAGE. The proteins were transferred to PVDF membranes (Cat#03010040001; Merck Millipore, Massachusetts, USA). The membranes were blocked in TBST containing 5% nonfat milk for 1 h at room temperature and incubated overnight with primary antibodies against FASN (Cat#10624-2-AP, Proteintech), USP14 (Cat#14517-1-AP, Proteintech), SREBP1 (Cat#14088-1-AP, Proteintech), ACC1 (Cat#21923-1-AP, Proteintech), SCD1 (Cat#28678-1-AP, Proteintech), ACLY (Cat#15421-1-AP, Proteintech) and β-actin (Cat#66009-1-Ig, Proteintech), followed by incubation with an HRP-labeled secondary antibody. The protein bands were visualized with Immobilon Western HRP substrate (Cat# WBKLS0100, Merck Millipore) and captured via a Tanon Scanning System (Shanghai, China).

Quantitative real-time PCR (qRT‒PCR)
Total RNA was extracted from cells via TRIzol Reagent (Cat#15596026CN; Invitrogen) and reverse transcribed via the HiScript Ⅱ RT SuperMix for qPCR Kit (Cat#R222-01; Vazyme Biotech, Nanjing, Jiangsu, China) following standard protocols. Real-time PCR was performed via a QuantStudio™ 5 Real‑Time PCR (Applied Biosystems, Taxes, USA) instrument with Taq Pro Universal SYBR qPCR Master Mix (Cat#Q712-02, Vazyme Biotech). The expression levels of the samples were determined via the comparative CT (ΔΔCT) method. The primer sequences are listed in Table 1.

Preparation of macrophage-conditioned medium (CM)
THP-1 monocytes at a density of 5 × 105 cells/mL in RPMI 1640 medium were treated with 100 nM phorbol 12-myristate 13-acetate (PMA; Cat# P1585; Sigma‒Aldrich) for 24 h and then incubated with 20 ng/mL IFN-γ (Cat# 300-02; PeproTech) and 100 ng/mL LPS for another 48 h. Then, the culture medium was collected and stored at − 80 °C for further study.

Cell proliferation assay
Cell proliferation rates were assessed via a Cell Counting Kit-8 (Cat# CK04, Dojindo, Kumamoto, Japan) following the standard protocol. Briefly, cells were seeded into 96-well plates at a density of 1 × 104 cells per well. After the indicated treatment, the cells were incubated in 100 µL of 10% (v/v) CCK8/complete medium mixture for 1–4 h at 37 °C and 5% CO2. The cell viability was subsequently evaluated by monitoring the absorbance at 450 nm via a microplate reader (Tecan, Mannedorf, Switzerland).

Immunofluorescence
The cells cultured on microscope coverslips were fixed in 4% PFA for 15 min, washed with PBS three times, permeabilized with 0.2% Triton X-100 (Cat#T8200; Solarbio, Beijing, China) for 5 min and then blocked with 3% BSA for 1 h at room temperature. The coverslips were incubated overnight at 4 °C with primary antibodies, followed by incubation with the corresponding secondary antibodies for 1 h at room temperature in the dark. To stain the nuclei, the coverslips were incubated with DAPI for 10 min. Images were observed via Zeiss LSM880 confocal microscopy at the same voltage and analyzed via ZEN software.

Coimmunoprecipitation (co-IP)
The cells were lysed in RIPA buffer. The 10% of the cell lysate was used as input. The cell lysates were incubated with endogenous antibody or control IgG (Cat#30000-0-AP; Proteintech) overnight at 4 °C, followed by incubation with Protein A/G agarose beads (Cat#HY-K0202; MedChemExpress) for 4 h at 4 °C. The beads were subsequently washed three times with PBST and boiled in 2×SDS loading buffer for 10 min. The bead-bound proteins were subjected to SDS‒PAGE and subsequent Western blot analysis.

Enzyme-linked immunosorbent assay (ELISA)
Cytokines from the macrophage-conditioned medium were determined by specific ELISA kits for murine IL-6 (Cat#ab222503, Abcam), interleukin-1β (IL-1β, Cat#ab197742, Abcam), tumor necrosis factor-α (TNF-α, Cat#ab208348, Abcam), IL-12 p70 (Cat#ab223592, Abcam), IL-23 (Cat#ab221837, Abcam), CCL2 (Cat#ab179886, Abcam), CCL5 (Cat#ab174446, Abcam), and CXCL10 (Cat#ab83700, Abcam) production. The assays were performed according to the manufacturer’s directions.

Lentiviral infection and plasmid transfection
The lentiviral-based short hairpin RNAs (shRNAs) or overexpression plasmids used to silence or overexpress the FASN, USP14 and STAT3 genes were purchased from GeneChem Company (Shanghai, China). Puromycin (1 µg/ml, Cat# HY-B1743A, MedChemExpress) was used to select the transfected cells 48 h after lentiviral transfection. For small interfering RNA (siRNA) transfection experiments, cells were seeded at 60–70% confluency and allowed to attach overnight. The STAT3 or control-siRNA was transfected into cells via Lipofectamine 3000 reagent (Cat# L3000008, Invitrogen) following the manufacturer’s instructions. The targeting sequences of the shRNAs and siRNAs are listed in Table 2.

Chromatin immunoprecipitation (ChIP)
The cells were incubated with 1% (v/v) formaldehyde for 10 min at room temperature. Thereafter, the crosslinked cells were washed twice and lysed in SDS buffer. The nuclear extracts were fragmented into 1 kb fragments via sonication. The harvested supernatants were diluted in ChIP dilution buffer and precleared with protein A agarose. Chromatin samples were immunoprecipitated with normal rabbit IgG (Cat# 2729 S; Cell Signaling Technology) as the negative control and with an antibody specific for STAT3 (Cat# 9139; Cell Signaling Technology). IP was subsequently performed, and the immunoprecipitated DNA was analyzed via qPCR as described above. The sequences of the primers specific for the USP14 promoter are listed in Table 1.

Luciferase reporter assay
Luciferase assays were conducted with the Dual Luciferase Reporter Assay System (Cat# E1910; Promega, Madison, WI, USA). HCT-8 cells were cotransfected with 800 ng of the USP14-promoter-Luc plasmid and 6 ng of the Renilla luciferase plasmid with or without IL-6 using Lipofectamine 2000 transfection reagent (Cat# 11668019, Invitrogen). si-STAT3 was used to inhibit STAT3 signaling. After 36 h of transfection, cell lysates were prepared, and firefly and Renilla luciferase activities were measured via a BD MonoLight 3010 luminometer (BD Biosciences). Transcriptional activity was represented as the F-luc/R-luc ratio.

5-Ethynyl-2’-deoxyuridine (EdU) incorporation assay
For the EdU incorporation assay, cells cultured under different conditions were independently treated with EdU (10 µmol/L, Cat# C0081S, Beyotime) for 40 min. Then, the cells were harvested for fixation, permeabilization, and Click-iT reactions via Azide 647. DAPI was used for DNA count following the manufacturer’s instructions (Cat#P0131, Beyotime). Next, the cells were analyzed via a Beckman Coulter CytoFLEX flow cytometer.

Apoptosis assay
The percentage of apoptotic cells was analyzed via an Annexin V-FITC Apoptosis Detection Kit (Cat# C1062M; Beyotime). The cells were harvested and resuspended in binding buffer containing Annexin V-FITC and PI according to the manufacturer’s instructions. The samples were analyzed with a Beckman Coulter CytoFLEX flow cytometer.

Lipidomics
Sample preparation: The samples were dissolved in 1.5 mL of chloroform/methanol (2/1, v/v solution), vortexed for 1 min, and then ultrasonicated for 30 min at 4 °C or below. Centrifugation was performed at 3000 rpm for 10 min. The organ phase was transferred into another EP tube for blowing drying with nitrogen gas, and 400 µL of methanol/isopropanol (1/1, v/v solution) and 5 µL of LPC (12:0) internal standard with a concentration of 0.14 mg/mL were added. Finally, the mixture was centrifuged at 12,000 rpm for 10 min, and 200 µL of the supernatant was collected for detection.
LC/MS instrument analysis platform: LC‒MS (Ultimate 3000LC, Q Exactive, Thermo); Column: Hypersil GOLD™ C18 (2.1 × 100 mm, 3 μm, 25003–102130, Thermo), Chromatographic separation conditions: column temperature, 40 °C; flow rate, 0.3 mL/min; mobile phase composition, A: acetonitrile: water (6:4, V/V), solution containing 0.77 g ammonium formate, B: acetonitrile: isopropanol (1:9, V/V); injection volume, 8 µL; autosampler temperature, 4 °C; equilibration time, 3 min; and mass spectrometry parameters: electrospray ionization (ESI)-positive and negative ion modes were used. Scan mode: Heater temperature, 300 °C; Sheath gas flow rate, 45 arb; Aux gas flow rate, 15 arb; Sweep gas flow rate, 1 arb; spray voltage, 3.0 kV (positive mode) or 3.5 kV (negative mode); capillary temperature, 350 °C; and S-Lens RF level, 50%. The scan range was 200–1500.
Statistical analysis: The data were subjected to feature extraction and preprocessed with MSDIAL software. The total peak area extracted by MSDIAL software was normalized to the data after the data were preprocessed via UV scaling and multidimensional statistical analysis, including unsupervised principal component analysis (PCA). Unidimensional statistical analysis included Student’s t test and variation fold analysis, R software to draw volcano maps, and hierarchical clustering analysis.

Molecular docking analysis
To analyze the binding affinities and modes of interaction between the drug candidates and their targets, AutodockVina 1.2.2, an in silico protein–ligand docking software, was used [46]. The molecular structure of 6-gingerol was retrieved from PubChem (https://pubchem.ncbi.nlm.nih.gov/) [47]. The 3D coordinates of USP14 (PDB ID, 2AYN; resolution, 3.2 Å) were downloaded from the RCSB PDB (http://www.rcsb.org/pdb/home/home.do). For docking analysis, the protein and molecular files were converted into PDBQT format with all water molecules excluded and polar hydrogen atoms added. The grid box was centered to cover the domain of each protein and to accommodate free molecular movement. The grid box was set to 30 Å × 30 Å × 30 Å, and the grid point distance was 0.05 nm. Molecular docking studies were performed via AutoDock Vina 1.2.2 (http://autodock.scripps.edu/).

Cellular thermal shift assay (CETSA)
HCT-8 cells (1 × 10^7) were collected and washed with ice-cold PBS three times. One milliliter of ice-cold PBS with complete EDTA-free protease inhibitor cocktail (Cat#04693132001, Roche, Basel, Switzerland) was then added to resuspend the cells, followed by three freeze‒thaw cycles consisting of 1 min in liquid nitrogen and then at 25 °C until thawing and 15 s of vortexing after each thaw. The cell lysates were then centrifuged at 20,000 × g for 20 min at 4 °C to pellet the cellular debris. To determine the melting curves, the cell lysates were divided into two aliquots. One group was treated with 6-gingerol, and the other was treated with the corresponding concentration of DMSO. After 30 min of incubation at room temperature, the lysates were divided into smaller (35 µL) aliquots and heated individually at different temperatures (45, 49, 53, 57, 61 and 65 °C) for 3 min, followed by cooling for 3 min at room temperature. The heated lysates were centrifuged at 20,000 × g for 20 min at 4 °C to separate the soluble fractions from the precipitates. The supernatants were analyzed via western blotting.

Statistical analysis
The data are presented as the means ± standard deviations (means ± S.Ds.). All of the statistical analyses were performed via SPSS (version 25.0, IBM Corp., Armonk, NY, USA) and GraphPad Prism (version 8, GraphPad Software, La Jolla, CA, USA) software. After the homogeneity of variance was tested, Student’s t test and one-way ANOVA were used for comparisons between and among different groups, respectively. For the tumor growth curve, two-way ANOVA was used. A p value of less than 0.05 was considered statistically significant. The sample size for the in vivo experiments was based on previous similar experiments in our laboratory. For all other studies, the data were replicated more than three times to ensure the robustness and reliability of the findings in line with rigorous scientific standards. No blinding of the investigators was performed.

Supplementary Information
Below is the link to the electronic supplementary material.