LC-MS/MS-Based Targeted Lipidomic Analysis of Obesity-Related Colorectal Cancer: Potential Roles of the CYP Eicosanoid Pathway

Abstract

Objective:

The mechanisms by which obesity increases colorectal cancer (CRC) risks are not well understood. Eicosanoids, lipid signaling molecules generated by cyclooxygenase (COX), lipoxygenase (LOX), and cytochrome P450 (CYP) enzymes, regulate inflammation, immune responses, and tumorigenesis. While recent clinical studies support a role for the COX pathway in obesity-associated CRC, the roles of other pathways are unclear.

Methods:

We performed LC-MS/MS–based lipidomics, capable of quantifying >100 eicosanoid metabolites, to profile colonic eicosanoids in an azoxymethane (AOM)/high-fat diet (HFD) model of obesity-related CRC.

Results:

Lipidomics identified changes in established CRC-associated eicosanoids, including increased COX-derived prostaglandin E₂ in obese CRC mice. Surprisingly, CYP-derived fatty acid epoxides were among the most markedly altered metabolites, exhibiting a 40–76% reduction in obese CRC mice compared with lean controls. Gene expression analysis supported the lipidomics results: CYP2C/2J isoforms responsible for epoxide production, such as Cyp2c38, Cyp2c39, Cyp2c65, Cyp2c70, and Cyp2j13, were reduced by ~70–96%, while expression of soluble epoxide hydrolase (sEH), which degrades fatty acid epoxides, increased by ~43% in the colons of obese CRC mice.

Conclusions:

These findings demonstrate that the CYP eicosanoid pathway is profoundly dysregulated in obesity-related CRC, providing a basis for exploring the roles of this pathway in the development of obesity-related CRC.

Introduction

Colorectal cancer (CRC) is the third most common cancer and the second leading-cause of cancer-related death in US ¹: every year there are ~130,000 new cases and ~50,000 fatalities from CRC in US ¹. It is well established that obese individuals have higher risks of developing CRC and late-stage CRC ^2,3. Indeed, some studies suggested that obese individuals have 30–60% higher risk of developing CRC ^2,3. Considering the obesity epidemic and the potential lethal consequence of CRC, obesity-related CRC is a serious health problem. However, the mechanism by which obesity increases the risks of CRC are not well understood, and there are few effective strategies to prevent obesity-related CRC ⁴.

Eicosanoids refer to a group of endogenous lipid-based signaling molecules generated through the enzymatic metabolism of arachidonic acid (ARA, 20:4ω−6) ^5,6. Upon cellular stimulation such as infection or inflammation, esterified ARA in membrane phospholipids is released by phospholipase A₂ or related enzymes, leading to the formation of intracellular, free-form ARA ^5–7. ARA is subsequently metabolized by various metabolic enzymes to form a wide array of lipid mediators termed eicosanoids ^5,6. There are three main pathways involved in eicosanoid biosynthesis: (i) the cyclooxygenase (COX) enzymes which produce prostaglandins and thromboxanes ^5,8; (ii) the lipoxygenase (LOX) enzymes which generate leukotrienes and hydroxyl fatty acids ^5,9; and (iii) the cytochrome P450 (CYP) monooxygenases which produce epoxy fatty acids ^6,10–15. Besides ARA, other polyunsaturated fatty acids (PUFAs), such as α-linolenic acid (ALA, 18:3 ω−3), eicosapentaenoic acid (EPA, 20:5 ω−3), and docosahexaenoic acid (DHA, 22:6 ω−3), could also be metabolized by these enzymes, leading to the formation of corresponding PUFA metabolites ^5,6. Many of these PUFA metabolites have potent effects to regulate various biological processes including inflammation, immune responses, and tumorigenesis ^5,6. Therefore, eicosanoid signaling pathways are therapeutic targets of many drugs on the market ^5,6.

Eicosanoids play a critical role in CRC pathogenesis, and emerging evidence suggests their involvement in obesity-associated CRC ¹⁶. Among various eicosanoid pathways, the COX pathway is the best characterized. Human studies consistently demonstrate that COX inhibitors, such as aspirin and other nonsteroidal anti-inflammatory drugs which block prostaglandin biosynthesis, are among the most effective agents for CRC prevention ¹⁷. For example, in the CAPP2 randomized clinical trial, daily aspirin use resulted in a 63% reduction in CRC incidence compared with placebo ¹⁸. Notably, the CAPP2 trial also revealed that obesity markedly increases CRC risk in individuals with Lynch syndrome (an inherited genetic condition that increases risks of developing certain cancers such as CRC), but this elevated risk is eliminated in those taking daily aspirin ¹⁹. These findings highlight a potential role of the COX eicosanoid pathway in obesity-related CRC. However, beyond the COX pathway, the contributions of other eicosanoid pathways to obesity-associated CRC remain poorly understood.

To investigate the role of eicosanoids in obesity-related CRC, we used a LC-MS/MS–based lipidomics, which can quantify >100 metabolites derived from COX, LOX, and CYP enzymatic pathways, as well as products of non-enzymatic oxidation, to analyze how eicosanoids are dysregulated in obesity-related CRC ²⁰. We applied this approach to a well-established model of obesity-associated CRC: azoxymethane (AOM)–induced, high-fat diet (HFD)–enhanced colorectal tumorigenesis ^21–25. Our findings demonstrate that, in addition to the well-characterized COX pathway, other eicosanoid pathways, most notably the CYP pathway, are significantly altered in obesity-related CRC. This study provides a foundation to explore the functional roles of these pathways in the development of obesity-related CRC.

Materials and Methods

Animal experiment

The animal experiment was performed in accordance with protocols approved by the Institutional Animal Care and Use Committee of the University of Massachusetts Amherst. Male C57BL/6 mice (Charles River Laboratories) were housed under standard conditions with free access to food and water. After adaptation, mice were randomly divided into two groups and fed either a low-fat diet (LFD; 10% kcal from fat, Research Diets D12450J) or a high-fat diet (HFD; 60% kcal from fat, Research Diets D12492). We selected the D12450J/D12492 diet pair because D12450J is the purified low-fat counterpart to D12492. These diets are matched for protein source, carbohydrate source, fiber type and amount, and micronutrient composition on a per-calorie basis, differing primarily in fat content (diet formulations are listed in Table S1) ²⁶. These two diets are widely used for obesity research ^27–31. From week 0 to week 5, all mice received weekly intraperitoneal injections of AOM (dose = 10 mg/kg; Sigma-Aldrich) for six consecutive weeks, as previously described ²¹. Mice were maintained on their respective diets for 33 weeks and sacrificed for tissue collection. Colons were longitudinally opened, rinsed in PBS, and examined under a dissection microscope for macroscopic tumors. Tumor size was calculated as π × d² / 4, where d represents tumor diameter.

Histological and immunohistochemical analysis

Histological and immunohistochemical (IHC) analyses were performed as previously described with minor modifications ³². Briefly, formalin-fixed, paraffin-embedded colon sections (5 μm) were stained with hematoxylin and eosin (H&E) for histopathological evaluation. For IHC, antigen retrieval was performed in 0.01 M citrate buffer (pH 6.0) at 95 °C for 10 min, followed by incubation with primary antibodies against proliferating cell nuclear antigen (PCNA; Dako, M0879) or non-phospho (active) β-catenin (Ser33/37/Thr41; Cell Signaling, 8814) overnight at 4 °C. After incubation with HRP-conjugated secondary antibodies, immunoreactivity was visualized using a DAB substrate kit (Abcam). Sections were counterstained with hematoxylin and analyzed using light microscopy. Quantification of positively stained cells was performed using ImageJ software.

Flow cytometry analysis of immune cell infiltration in colon tissues

Flow cytometry was performed as previously described with minor modifications ³². Briefly, distal colon tissues were dissected, washed with cold PBS, and digested in Hank’s balanced salt solution (Lonza) supplemented with 1 mM dithiothreitol and 5 mM EDTA overnight at 4 °C. Released epithelial cells were passed through a 70 μm cell strainer (BD Biosciences) to obtain single-cell suspensions. Cells were stained with APC-conjugated anti-mouse CD45 (BioLegend, 1031112), PE-conjugated anti-mouse F4/80 (BioLegend, 123110), APC-conjugated anti-mouse CD25 (BioLegend, 302642), FITC-conjugated anti-mouse CD4 (BD Pharmingen, 553651), APC-conjugated anti-mouse CD80 (Invitrogen, 2036646), PE-conjugated anti-mouse CD8 (Invitrogen, 2016851), FITC-conjugated anti-mouse Gr1 (Invitrogen, 2088455), PE-conjugated anti-mouse RORγt (Invitrogen, 2014781), Alexa Fluor 647-conjugated anti-mouse T-bet (BioLegend, 644804),and corresponding isotype control antibodies. Dead cells were excluded using Zombie Violet^™ Fixable Viability dye (BioLegend) according to the manufacturer’s protocol. Samples were analyzed on a BD LSRFortessa^™ flow cytometer (BD Biosciences), and data were processed using FlowJo software (FlowJo LLC). Cell doublets were excluded using FSC-H vs. FSC-A gating, and debris was removed using FSC-A vs. SSC-A. Leukocytes were defined as CD45⁺ cells, and neutrophils as CD45⁺Gr1⁺ cells. T cell subsets were gated as follows: CD4⁺ helper T cells (CD4⁺CD8⁻), cytotoxic T cells (CD4⁻CD8⁺), regulatory T cells (Tregs; CD4⁺CD25⁺), Th1 cells (CD4⁺T-bet⁺), and Th17 cells (CD4⁺RORγt⁺). M1 macrophages were defined as F4/80⁺CD80⁺ cells.

LC-MS/MS analysis of eicosanoids in colon tissues

Eicosanoid analysis was performed at UC Davis as previously described with minor modifications ²⁰. Briefly, mouse colon tissues (15–35 mg) were homogenized with 20 μL of isotope-labeled internal standards and 10 μL of antioxidant solution (0.2 mg/mL butylated hydroxytoluene and 0.2 mg/mL triphenylphosphine in methanol), followed by 400 μL methanol containing 0.1% acetic acid and 0.1% butylated hydroxytoluene. After overnight storage at −80 °C, samples were centrifuged, and the supernatants were subjected to solid-phase extraction using Oasis HLB columns (60 mg, 3 cc; Waters). Analytes were eluted with methanol and ethyl acetate, evaporated under vacuum, and reconstituted in 50 μL methanol for LC–MS/MS analysis. Chromatographic separation was achieved on an Agilent ZORBAX Eclipse PLUS C18 column (2.1 × 150 mm, 1.8 μm) using an Agilent 1200SL HPLC system coupled to a 6500 QTRAP tandem mass spectrometer. Metabolites were identified based on retention time and specific multiple reaction monitoring transitions of authentic standards and quantified using calibration curves.

Quantitative real-time-PCR (qRT-PCR) analysis

Total RNA was extracted from frozen colon tissues using TRIzol reagent (Ambion) according to the manufacturer’s instructions. RNA purity and concentration were determined using a NanoDrop spectrophotometer (Thermo Scientific). Complementary DNA was synthesized using a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems). Quantitative PCR was performed using Maxima SYBR Green Master Mix (Thermo Fisher Scientific) on a DNA Engine Opticon system (Bio-Rad). Primer sequences are listed in Table S2. Relative mRNA expression was calculated using the 2^−ΔΔCt method, with Gapdh serving as the internal reference gene.

Statistical analysis

Data were expressed as mean ± SEM. Comparisons between two groups were performed using Student’s t test, and P < 0.05 was considered statistically significant. Statistical analyses were conducted using GraphPad Prism 10 (GraphPad Software, San Diego, CA). For LC–MS/MS–based targeted lipidomic analysis, multiple-testing correction was performed using the Benjamini–Hochberg false discovery rate (FDR) method, and FDR-adjusted P-values were reported for all metabolites. To facilitate cross-sample comparison and visualization in heatmaps (Fig. 2D), z-score normalization. To facilitate cross-sample comparison and visualization in the heatmaps (Figs. 2D), z-score normalization was applied to each metabolite across all samples. Specifically, z-scores were calculated using the formula z = (x – μ)/σ, where x is the concentration of a given metabolite, μ is the mean, and σ is the standard deviation of that metabolite across the dataset.

Figure 2. CYP-derived eicosanoids are decreased in colons of obese CRC mice compared with lean CRC mice.

(A) Principal component analysis (PCA) shows clear separation of eicosanoid profiles between LFD and HFD groups. (B) Volcano plot of colonic eicosanoids. Fold change reflects metabolite levels in HFD-fed mice relative to LFD-fed mice. Statistical significance was determined using FDR-adjusted P values (Benjamini–Hochberg correction). (C) Schematic of the CYP/sEH pathway: polyunsaturated fatty acids (PUFAs) are metabolized by CYP monooxygenases (primarily CYP2C/2J isoforms) to generate fatty acid epoxides, which are further converted into fatty acid diols by soluble epoxide hydrolase (sEH). (D) Heatmap of CYP-derived fatty acid epoxides and their corresponding diols (LFD, n = 12; HFD, n = 13). Abbreviations: CRC, colorectal cancer; CYP, cytochrome P450; LFD, low-fat diet; HFD, high-fat diet.

Data availability

All data supporting this study are provided in the article and supporting information and can also be obtained from the corresponding author upon request.

Results

AOM/HFD-induced mouse model of obesity-related CRC

We used a well-established AOM and HFD–induced CRC model to study obesity-related CRC (Fig. 1A) ^21–25. AOM-treated mice were fed either LFD (10% kcal from fat) or HFD (60% kcal from fat) for 33 weeks. HFD-fed mice gained significantly more body weight than LFD-fed controls (Fig. 1B). At the end of the experiment (t = week 33), the relative body weights of HFD- and LFD-treated mice were 234.6 ± 8.1% and 170.3 ± 4.2%, respectively (mean ± SEM; baseline weight at week 0 defined as 100%; P < 0.0001), illustrating successful induction of diet-induced obesity (Fig. 1B).

Figure 1. AOM/HFD-induced mouse model of obesity-related CRC.

(A) Schematic overview of the animal study design. (B) Body weight of mice fed either LFD or HFD. (C) Representative images of colons and quantification of tumor burden (LFD, n = 12; HFD, n = 13). (D) Representative H&E staining and immunohistochemistry for PCNA and active β-catenin in colon tissues (n = 12 per group). (E) Quantification of immune cell infiltration in colon tissues (n = 11–12 per group). Data are presented as mean ± SEM. Abbreviations: AOM, Azoxymethane; LFD, low-fat diet; HFD, high-fat diet; CRC, colorectal cancer; PCNA, proliferating cell nuclear antigen.

By week 33, 9 of 13 obese CRC mice (AOM-stimulated, HFD-fed for 33 weeks) developed colon tumors, whereas only 1 of 12 lean CRC mice (AOM-stimulated, LFD-fed for 33 weeks) showed tumor formation (Fig. 1C), illustrating the establishment of the obesity-related CRC model. Histological and immunohistochemical analysis revealed elevated expression of the tumorigenic markers, including PCNA and active β-catenin in the colons of HFD-fed mice (Fig. 1D), together with increased infiltration of immune cells, including leukocytes (CD45⁺), macrophages (CD45⁺ F4/80⁺) and other immune cells (Fig. 1E). These results demonstrate that obesity promotes colon tumorigenesis in AOM-treated mice.

LC-MS/MS targeted lipidomic analysis of obesity-related CRC

We performed LC-MS/MS–based targeted lipidomics, capable of quantifying more than 100 eicosanoid metabolites, to compare eicosanoid profiles in the colons of obese and lean CRC mice ²⁰. A total of 59 metabolites were detected, while others were below the detection limit (see Table S3 for a summary and the raw data in the supplemental Excel file). Principal component analysis (PCA) revealed a clear separation between lean and obese CRC groups, indicating distinct metabolic profiles (Fig. 2A). Among the altered metabolites, several well-established CRC-related lipid mediators were dysregulated in obesity-related CRC. Notably, prostaglandin E₂ (PGE₂), a COX-derived pro-inflammatory lipid and a well-characterized regulator of CRC ⁵, was significantly elevated in the colons of obese CRC mice compared with lean controls (Table S3).

CYP–derived eicosanoids are reduced in the colons of obese CRC mice

Volcano plot analysis showed that CYP-derived eicosanoids were among the most dramatically altered metabolites in obesity-related CRC (Fig. 2B). CYP monooxygenases (primarily CYP2C/2J isoforms) catalyze the conversion of PUFAs to fatty acid epoxides, which are subsequently converted to the corresponding fatty acid diols by soluble epoxide hydrolase (sEH) (see metabolic pathway in Fig. 2C) ⁶. Heatmap analysis revealed that multiple fatty acid epoxides and diols were decreased in the colons of obese CRC mice compared with lean CRC mice (Fig. 2D).

Representative concentrations of fatty acid epoxides and diols are shown in Fig. 3. Notably, LC-MS/MS revealed substantial reductions in several fatty acid epoxides—including epoxyoctadecadienoic acids (EpODEs) derived from ALA, epoxyeicosatrienoic acids (EETs) from ARA, epoxyeicosatetraenoic acids (EpETEs) from EPA, and epoxydocosapentaenoic acids (EpDPEs) from DHA—in the colons of obese CRC mice (Fig. 3A). For instance, colonic 11,12-EET levels were 1,150.32 ± 179.54 pmol/g in obese CRC mice versus 2,794.14 ± 500.54 pmol/g in lean CRC mice (FDR-adjusted P-value= 0.020), representing an approximately 60% reduction (Fig. 3A). The downstream metabolite, 11,12-dihydroxyeicosatrienoic acid (DHET), was also reduced: 29.26 ± 1.49 pmol/g in obese CRC mice compared with 48.18 ± 6.12 pmol/g in lean CRC mice (FDR-adjusted P-value= 0.023) (Fig. 3B). Together, these data demonstrate that CYP-derived metabolites, including fatty acid epoxides and their corresponding diols, are significantly decreased in the colons of obese CRC mice relative to lean CRC mice.

Figure 3. Colonic levels of representative CYP-derived fatty acid epoxides and diols. (A) Fatty acid epoxides.

(B) Fatty acid diols. The data are mean ± SEM, n=12 for LFD group and n=13 for HFD group. Statistical significance was determined using FDR-adjusted P values (Benjamini–Hochberg correction). Abbreviations: CYP, cytochrome P450; LFD, low-fat diet; HFD, high-fat diet.

The ratio of fatty acid diol to fatty acid epoxide, a surrogate marker of epoxide hydrolase, is increased in obese CRC mice

We further analyzed the fatty acid diol-to-fatty acid epoxide ratio, a surrogate marker of sEH activity ⁶. We found that the diol-to-epoxide ratio was significantly increased in the colons of obese CRC mice compared with lean CRC mice (Fig. 4). For example, the ratio of 15,16-DiHODE (a diol metabolite of ALA) to its corresponding epoxide, 15,16-EpODE, was 0.23 ± 0.04 (mean ± SEM) in obese CRC mice versus 0.10 ± 0.02 in lean CRC mice (P = 0.0052) (Fig. 4). Similar increases in diol-to-epoxide ratios were observed for other PUFA-derived metabolites (Fig. 4). Together, these findings suggest that sEH activity may be elevated in the colons of obese CRC mice.

Figure 4. The diol-to-epoxide ratio, a surrogate marker for epoxide hydrolase activity, is elevated in obese CRC mice compared with lean CRC mice.

The data are mean ± SEM, n=12 for LFD group and n=13 for HFD group. Abbreviations: CRC, colorectal cancer; LFD, low-fat diet; HFD, high-fat diet.

Gene expression of CYP monooxygenases is decreased, while expression of sEH is increased, in the colons of obese CRC mice

We next examined the colonic expression of CYP monooxygenases (primarily CYP2C/2J isoforms) and Ephx2 (which encodes sEH). Consistent with the LC-MS/MS results (Figs. 2–3), qRT-PCR revealed that several CYP2C/2J isoforms, including Cyp2c38, Cyp2c39, Cyp2c65, Cyp2c70, and Cyp2j13, were reduced by approximately 70–96% in the colons of obese CRC mice compared with lean CRC mice (Fig. 5). In agreement with the increased diol-to-epoxide ratios (Fig. 4), Ephx2 expression was significantly elevated in obese CRC mice (Fig. 5). These findings suggest that obesity-related CRC alters the colonic expression of CYP monooxygenases and sEH, thereby reshaping the levels of CYP-derived fatty acid epoxides and diols.

Figure 5. Colonic expression of CYP2C/2J monooxygenase is reduced, while expression of Ephx2 is increased, in obese CRC mice compared with lean CRC mice.

Data are presented as mean ± SEM, n = 12 per group. Abbreviations: CRC, colorectal cancer; CYP, cytochrome P450; LFD, low-fat diet; HFD, high-fat diet.

Discussion

The mechanism by which obesity increases the risks of CRC are not well understood, and there are few effective strategies to prevent obesity-related CRC ⁴. In this study, we aimed to study how colonic eicosanoid profiles are altered in lean CRC mice versus obese CRC mice. To do so, we employed LC-MS/MS–based targeted lipidomics, which can measure >100 eicosanoid metabolites generated by COX, LOX, and CYP enzymes, to systematically examine how eicosanoids are altered in obesity-related CRC. We observed changes in well-established CRC-associated eicosanoids, such as COX-derived PGE₂. Strikingly, CYP-derived fatty acid epoxides were among the most dramatically altered metabolites, showing a 40–76% reduction in obese CRC mice compared with lean CRC controls. Previous studies indicate that fatty acid epoxides plays key roles in multiple obesity-associated disorders ^{30,31,33–41}, as well as in colonic inflammation and tumorigenesis ^{30,31,42–44}, suggesting their potential involvement in the pathogenesis of obesity-related CRC. This study provides a foundation to explore the roles of eicosanoid signaling in obesity-related CRC.

Using LC-MS/MS and qRT-PCR analyses, we found that the CYP eicosanoid pathway is markedly dysregulated in obesity-related CRC. The expression of CYP2C/2J monooxygenases was dramatically reduced in the colons of obese CRC mice compared with lean CRC mice, with qRT-PCR showing decreases of ~70–96% for genes including Cyp2c38, Cyp2c39, Cyp2c65, Cyp2c70, and Cyp2j13. Consistent with this, LC-MS/MS analysis revealed significantly lower colonic concentrations of multiple CYP-derived fatty acid epoxides in obese CRC mice. In contrast, the expression of sEH, the major enzyme responsible for converting fatty acid epoxides to fatty acid diols ⁶, was increased by ~50% in obese CRC mice. This upregulation likely contributes to both the reduced epoxide levels and the increased diol-to-epoxide ratio observed. Together, these data indicate that obesity-related CRC is characterized by suppressed production and enhanced degradation of fatty acid epoxides due to reduced CYP monooxygenase expression and increased sEH expression, resulting in markedly decreased colonic levels of fatty acid epoxides. In our previous work, we conducted LC-MS/MS–based lipidomics analyses examining either the effects of diet alone (C57BL/6 mice fed LFD versus HFD) ^30,31 or the effects of CRC alone (healthy C57BL/6 mice versus AOM/dextran sulfate sodium-induced CRC mice) on colonic profiles of eicosanoids ⁴⁵. Here, we integrate both variables—diet and CRC—to determine how obesity modifies colonic eicosanoid profiles in the context of tumorigenesis.

Our observations of changes in CYP monooxygenase and sEH expression are largely consistent with findings reported in previous studies. For CYP monooxygenases, in our earlier work using a HFD–induced obesity model, we found that both the expression of CYP2C/2J monooxygenases and the levels of CYP-derived eicosanoids were reduced in adipose tissue of obese mice ⁴⁶. For sEH, substantial studies have shown that the expression and/or activity of sEH was increased in various tissues of HFD-induced obese mice, including adipose ^47,48, liver ^48–50, and kidney ⁵¹. Furthermore, previous studies also support that sEH is upregulated in the tissues of genetically induced obese animals ⁵². Human studies further support dysregulation of this pathway in obesity: compared with non-obese individuals with atherosclerotic cardiovascular disease, obese patients show reduced plasma concentrations of EETs and an increased DHET-to-EET ratio, suggesting an increased sEH activity ⁵³. Collectively, these findings align with our observations in the obesity-related CRC model and highlight the importance of the CYP/sEH pathway in obesity and its associated diseases, including CRC.

Fatty acid epoxides are important lipid mediators that regulate inflammation, immune responses, and a wide range of other important biological processes ⁶. Pharmacological inhibitors of sEH, which stabilizes fatty acid epoxides, have been evaluated in multiple human clinical trials ^14,15. Notably, the sEH inhibitor EC5026 is currently in clinical development and has completed Phase 1B trials ¹⁵. Experimental studies using pharmacological inhibition or genetic deletion of sEH demonstrate beneficial effects of stabilized fatty acid epoxides in various obesity-related disorders, including endoplasmic reticulum stress, metabolic syndrome, insulin resistance, fatty liver disease, hepatic steatosis, systemic inflammation, and endothelial dysfunction ^33–41. Our own work further shows that sEH inhibition or deletion prevents obesity-induced colonic inflammation, activation of pro-tumorigenic Wnt signaling, and intestinal barrier dysfunction ^30,31. Collectively, these studies underscore the protective roles of fatty acid epoxides in obesity-associated pathologies, including colonic inflammation. Thus, the reduced levels of fatty acid epoxides observed in the colons of obese CRC mice may contribute to the development of obesity-related CRC.

Further studies are needed to define the functional roles of the CYP pathway in obesity-related CRC. There are several challenges to study the actions of the CYP-derived eicosanoids in cancer. First, CYP-derived eicosanoids are thought to act through specific G-protein–coupled receptors, but the relevant receptors remain largely unidentified, making it difficult to perform mechanistic studies ⁵⁴. Second, previous studies have demonstrated complex and context-dependent effects of CYP/sEH-derived eicosanoid metabolites on tumorigenesis. In models of colonic inflammation and CRC, inhibition or genetic deletion of sEH has been shown to attenuate disease severity: previous studies demonstrated that loss or inhibition of sEH reduces DSS- and IL-10–deficiency–induced colitis and CRC progression ^42–44. Consistent with this, our own work shows that sEH inhibition or deletion markedly suppresses obesity-induced colonic inflammation and intestinal barrier dysfunction ^30,31. These data suggest that increasing fatty acid epoxides (through reduced sEH activity) may protect against colonic inflammation and CRC. However, in other tumor models, overexpression of CYP monooxygenases or inhibition of sEH enhances angiogenesis and promotes tumor growth and metastasis, indicating that elevated fatty acid epoxides can also worsen cancer progression ⁵⁵. Therefore, additional investigation is required to determine how these pathways influence obesity-related CRC.

There are some limitations of this work. First, in interpreting the lipidomic and transcriptomic differences between dietary groups, it is important to consider the potential influence of tissue heterogeneity. Because measurements were performed using whole-colon homogenates, variations in tumor burden and immune-cell infiltration—both of which differed between HFD- and LFD-fed mice—may contribute to the observed metabolic and gene-expression patterns. Tumor-rich and inflamed tissues have distinct cellular compositions and metabolic activities, which could influence the abundance of eicosanoid metabolites and related transcripts. Therefore, the observed differences should be interpreted as reflecting combined changes across multiple cellular compartments rather than cell type–specific effects. This study cannot retrospectively isolate epithelial, stromal, and immune compartments, and accordingly, conclusions regarding obesity-associated alterations are made at the tissue level. Future studies employing cell-type–resolved or spatial molecular approaches will be important for disentangling these effects more precisely. Second, daily food intake was not directly measured in the HFD/AOM mouse experiment. However, body weight was monitored throughout the study, and mice fed the HFD exhibited significantly greater weight gain than those fed the LFD, indicating successful induction of diet-induced obesity.

In summary, our findings demonstrate that the previously underappreciated CYP/sEH eicosanoid pathway is profoundly altered in obesity-related CRC. Previous studies show that this pathway plays key roles in multiple obesity-associated disorders ^{30,31,33–41}, as well as in colonic inflammation and tumorigenesis ^{30,31,42–44}, suggesting that dysregulation of CYP/sEH signaling may contribute to the pathogenesis of obesity-related CRC. This study provides a foundation to explore eicosanoid signaling in the pathogenesis of obesity-related CRC.

Abbreviations

ALA

α-linolenic acid

AOM

azoxymethane

ARA

arachidonic acid

CRC

colorectal cancer

COX

cyclooxygenase

CYP

cytochrome P450

DHA

docosahexaenoic acid

DHET

dihydroxyeicosatrienoic acid

EETs

epoxyeicosatrienoic acids

EPA

eicosapentaenoic acid

EpETEs

epoxyeicosatetraenoic acids

EpDPEs

epoxydocosapentaenoic acids

EpODEs

epoxyoctadecadienoic acids

FDR

false discovery rate

H&E

hematoxylin and eosin

HFD

high-fat diet

IHC

immunohistochemical

LFD

low-fat diet

LOX

lipoxygenase

PCNA

proliferating cell nuclear antigen

PGE₂

prostaglandin E₂

PUFAs

polyunsaturated fatty a

qRT-PCR

quantitative real-time-PCR

sEH

soluble epoxide hydrolase