Dietary eicosapentaenoic and docosahexaenoic acids decrease prostanoids in the ileum of rats treated with irinotecan and 5-fluorouracil

Abstract

BACKGROUND

Irinotecan (CPT-11) and 5-fluorouracil (5-FU) are chemotherapeutic agents that disrupt the gastrointestinal tract. Prostanoids derived from arachidonic acid (AA) disrupt colonic water and electrolyte equilibrium and are implicated in chemotherapy-induced diarrhea. However, research assessing prostanoids in the ileum and effect of anti-inflammatory omega-3 polyunsaturated fatty acids are lacking.

AIM

To investigate effects of CPT-11 + 5-FU and dietary eicosapentaenoic acid (EPA) and docosahexaenoic acids (DHA) on phospholipids and oxylipins in ileum tissue.

METHODS

Ward colon tumor was implanted and grew for approximately 2 weeks in Fischer rats fed a semi-purified standard diet prior to providing CPT-11 + 5-FU. On the same day as injecting CPT-11 (day 0), half of the rats were switched to a diet containing EPA and DHA. Rats were euthanized on day 0 (baseline) and following chemotherapy on days 2, 4, and 8 (n = 4-5/group).

RESULTS

CPT-11 + 5-FU increased oxylipins by day 2 derived from n-6 fatty acid (P = 0.002) and AA (P = 0.013), specifically 6-keto-prostaglandin (PG)-F_1α (P < 0.0001), 8-hydroxyeicosatetraenoic acid (P = 0.006), and leukotriene B4 (P = 0.024). Feeding EPA + DHA increased EPA and DHA in ileal phospholipid by 40-fold and 2-fold from baseline to day 8, respectively, corresponding with a 10-fold increase in EPA metabolites and a 2-fold increase in DHA metabolites. EPA metabolites 15-hydroxyeicosapentaenoic acid (HEPE), 18-HEPE, and 17,18-dihydroxy-eicosatetraenoic acid; DHA metabolite 16-hydroxydocosahexaenoic acid; and AA metabolite lipoxin A4 increased by day 8 in animals fed the EPA + DHA diet compared to the control diet while 6-keto-PGF_1α and PGD₂ decreased below baseline.

CONCLUSION

Dietary EPA and DHA are incorporated into ileal membrane phospholipid, counteracting the chemotherapy-induced proinflammatory oxylipin profile, disrupting synthesis of prostanoids, and increasing synthesis of anti-inflammatory oxylipins.

Key Words: Ileum; Oxylipins; Phospholipid fatty acids; Rat; Tumor; Chemotherapy; Colorectal cancer; N-3 fatty acids; Fish oil; Irinotecan

Core Tip: The present study is the first to investigate n-6- and n-3-derived oxylipin levels from ileum tissue after irinotecan + 5-fluorouracil and modification by dietary eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). The arachidonic acid (AA) metabolite 6-keto-prostaglandin F_1α implicated in inhibiting water absorption and electrolyte dysregulation increased after irinotecan + 5-fluorouracil. Dietary EPA and DHA markedly reduced 6-keto-prostaglandin F_1α and several other prostanoids while increasing protective and resolving mediators, lipoxin A₄ (AA-derived) and protectin-D1 (DHA-derived). Oxylipin potential to reduce the risk of chemotherapy-induced adverse effects, such as diarrhea, should be explored in models of multi-cycle chemotherapy.

INTRODUCTION

Colorectal cancer is commonly treated with the chemotherapy agents, irinotecan hydrochloride (CPT-11) and 5-fluorouracil (5-FU). These agents trigger production of mediators such as cytokines, chemokines, and oxylipins that alter intestinal function and result in clinically adverse effects. Chemotherapy-induced intestinal toxicities have a well-characterized increase in proinflammatory cytokines [tumor necrosis factor (TNF)-α, interleukin (IL)-6, IL-1β] and chemokines[1,2] as well as simultaneous elevated levels of prostaglandin (PG) E₂ and thromboxane (TX) A₂ following CPT-11[3,4]. Diarrhea (mild to severe) that occurs in 80% of patients receiving CPT-11 or 5-FU has been linked to PGE₂ and TXA₂ by stimulating hypersecretion of chloride into the distal colon from colonic epithelium as well as impairing water absorption[3,5].

Electrolyte and water balance in the small intestine are disrupted by PGF_2α and prostacyclin I₂ (PGI₂)[6,7]. Injection of PGE₂ into the peritoneal cavity triggers water accumulation in the ileum and colon, and initiates diarrhea through colonic propulsive movement[8]. On the other hand, a clinically relevant, single dose of CPT-11 + 5-FU increased several PGs and TXA₂ known for roles in water and electrolyte regulation and did not trigger diarrhea[9]. Indeed, animal studies with subsequent chemotherapy cycles using a dose and regimen identical to the present study report exacerbated effects in rats (reduced body weight and food intake and altered intestinal microbiota) after two cycles compared to the first[10]. The majority of studies evaluating oxylipins have assessed colonic mucosa which is known to be damaged by these chemotherapeutic agents and associated with late diarrhea[8,11,12]. However, approximately 90% of ingested water is absorbed by the small intestine[13] and all segments of the intestine are impacted by CPT-11 and 5-FU[14]. Chemotherapy agents known for inducing diarrhea may also stimulate prostanoid synthesis in the ileum but has not been investigated.

The use of marine-derived longchain omega (n)-3 polyunsaturated fatty acid (PUFA), eicosapentaenoic acid (EPA) (C20:5n-3) and docosahexaenoic acid (DHA) (C22:6n-3) for cancer prevention, anti-tumor effects, and chemosensitization is established, but EPA and DHA used in treating toxic effects of chemotherapy is emerging[15-19] and no oxylipins in ileum have been assessed. Oxylipins are bioactive mediators of PUFAs, synthesized via oxygenase enzymes or non-enzymatic reactions through release of fatty acids from membrane phospholipid. Oxylipins produced are, in part, determined by membrane phospholipid fatty acid composition. Three series PG and TX and 5 series leukotrienes (LT) are EPA-derived oxylipins via cyclooxygenase (COX) and lipoxygenase (LOX) enzymes, respectively, and have weak bioactivity compared to 2 series PG and TX and 4 series LT derived from arachidonic acid (AA) (C20:4n-6)[20,21]. Maresins, resolvins and protectins are a distinct group of oxylipins known as specialized pro-resolving mediators (SPMs) that are derived from EPA and DHA. SPMs have anti-inflammatory properties and actively resolve inflammation (Figure 1)[22]. A mix of oxylipins derived from AA vs EPA and DHA is essential for recovery from intestinal insults. Levels of EPA and DHA are very low in intestinal membrane phospholipid while AA is much higher in diets with low habitual fish intake.

Figure 1 Pathophysiology of irinotecan and 5-fluorouracil on ileal membrane phospholipid when enriched with eicosapentaenoic acid and docosahexaenoic acid. CPT-11: Irinotecan; 5-FU: 5-fluorouracil; AA: Arachidonic acid; PLA₂: Phospholipase A₂; LOX: Lipoxygenase; COX: Cyclooxygenase; HETE: Hydroxyeicosatetraenoic acid; LTB₄: Leukotriene-B₄; PG: Prostaglandin; TX: Thromboxane; DHA: Docosahexaenoic acid; EPA: Eicosapentaenoic acid; NE: Non-enzymatic; CYP: Cytochrome P450; HDoHE: Hydroxy-docosahexaenoic acid.

Provision of n-3 PUFAs reduces n-6 fatty acids in membrane phospholipid, thereby reducing substrates for synthesis of prostanoids. EPA and DHA obtained through diet rapidly displace AA from phospholipid in the intestinal mucosa[23]. Animal studies show dietary n-3 PUFAs variably attenuate chemotherapy-induced intestinal toxicity depending on the dose and type of n-3 PUFA applied as well as the chemotherapeutic agents used[4,18,24-26]. Usami et al[4] assessed changes in phospholipid fatty acids of ileal mucosa following intensive treatment with CPT-11 for four consecutive days and dietary enrichment with alpha-linolenic acid (ALA) (C18:3n-3), the precursor to EPA and DHA via elongation and desaturation reactions. Despite a significant increase in ALA and, to a lesser extent, EPA in the ileal mucosa, intestinal toxicity was not reduced, suggesting potential inadequacy of dietary ALA supplementation to curb intestinal toxicity vs direct supply of more biologically active n-3 PUFAs, EPA and DHA. Oxylipins were not assessed in the ileal mucosa and AA content remained stable. COX inhibitors, such as celecoxib and indomethacin, also reduced prostanoids following CPT-11 or 5-FU. Preclinical studies showed COX-2 inhibition or non-selective COX inhibition reduced PGE₂ and was associated with reduced severity of diarrhea[27-30]; however, these preclinical findings have not been replicated in humans[28].

To investigate the effect of CPT-11 + 5-FU and dietary EPA + DHA on phospholipid and oxidative lipid metabolites in ileum tissue we conducted the present experiment. It is hypothesized that chemotherapy stimulates production of pro-inflammatory oxylipins derived from n-6 fatty acids whereas dietary EPA and DHA mitigates these changes to restore balance of oxylipins within days after chemotherapy.

MATERIALS AND METHODS

Animal model

Animal use was approved by the University of Alberta Animal Care Committee and conducted in accordance with Guidelines of the Canadian Council on Animal Care (No. AUP00003572). Thirty-free female Fischer 344 rats (150-180 g and 11-12 weeks of age) were obtained from Charles River Laboratories International (Wilmington, MA, United States). Rats were housed two per cage for the first nine days of acclimation and fed rat chow. Rats were then separated into single housing for individual monitoring in a temperature-controlled room (22 °C) with positive air pressure maintained on a 12-hour light/dark cycle and were subsequently transitioned to experimental diets over one week. Cages had bedding (aspen chips), tubing and balls, and filter tops. Water and food were available ad libitum.

Diet

Rats were provided a powdered fat-free, semi-purified basal diet (80 g/100 g; Teklad TD 170570) to which commercially available fat (20 g/100 g) was added as previously published[9]. Lard, canola oil, corn oil, and olive oil were combined with fish oil (EPA + DHA diet; 4.3 g/100 g; carnivora cold water fish oil, Saskatoon, SK, Canada) or without fish oil (control diet). Fatty acid composition of the diets differed primarily in EPA (C20:5n-3) and DHA (C22:6n-3) content as confirmed by gas chromatography (Agilent 7890A, Agilent Technologies, Santa Clara, CA, United States) as previously published[9]. Both diets contained ALA (C18:3n-3) from canola oil. Proportions of saturated and monounsaturated fatty acids were similar between diets. The dietary polyunsaturated to saturated fatty acid ratio (0.39) in the control diet reflects that typically consumed by North Americans[31]. The isocaloric diets contained 40% of total energy from fat, 36% from carbohydrates, and 24% from protein.

Tumor injection and chemotherapy

Ward colorectal carcinoma (0.05 g; Dr. Rustum Y, Roswell Park Institute; Buffalo, NY, United States)[32] was transplanted subcutaneously into the left flank of the rats via trocar under slight isoflurane anesthesia. Tumor size was continuously evaluated until it reached approximately 2.3 cm³ (1.2% of body weight) at which time bolus CPT-11 + 5-FU was initiated. CPT-11 (50 mg CPT-11/kg body weight, intraperitoneal; Camptosar^TM, Pfizer, United States) was injected as a clinical formulation solution (day 0; Accord Healthcare Inc.; London, United Kingdom) followed by 5-FU (50 mg 5-FU/kg body weight, intraperitoneal) twenty-four hours later (day 1). Atropine (1 mg/kg body weight; Sandoz Group, Switzerland) was subcutaneously delivered before CPT-11 to alleviate early-onset cholinergic symptoms[25]. Weight loss, tumor volume, and subcutaneous ulceration over tumor were monitored daily. Diarrhea was absent in rats at all time points (data not shown) similar to what has previously been reported by studies in our lab using single- and double-cycle treatments of the same CPT-11 + 5-FU dosing and regimen as that used in the present study.

Experimental design

The study protocol was planned a priori as part of the ethics review process and laboratory procedure preparation based on a previously published, non-registered experimental design[9]. Briefly, thirty-three female Fischer rats (127 ± 18 g; 11-12 weeks of age) were acclimated for 1 week then thirty were injected with Ward colon tumor. After two weeks, animals not assigned to receive chemotherapy were euthanized (tumor-bearing only) and the remaining rats received chemotherapy consisting of CPT-11 + 5-FU on subsequent days. Tumor-bearing rats euthanized before chemotherapy represent the baseline group (n = 5) at day 0 to which chemotherapy groups were compared. Rats (n = 12) were randomized based on body weight and assigned to continue feeding of the control diet while another group of rats (n = 13) was switched to the EPA + DHA diet initiated on the same day as chemotherapy (day 0). Each group was further subdivided into three groups (n = 4-5/group) for planned termination on days 2, 4, 8. Rats without tumor and chemotherapy (healthy reference, n = 3) were handled similar to other groups, fed the control diet, and euthanized on day 7. Termination dates were staggered and equal numbers of animals per group were terminated on the same day and in random order according to time of day. The primary focus of this design was to determine acute effects of chemotherapy on ileum phospholipid composition and oxylipin production (control diet) and, secondly, to determine the extent a diet containing EPA + DHA modifies the lipidomic profile within the first week after chemotherapy.

Study termination and tissue collection

Rats were euthanized by carbon dioxide asphyxiation. The intestine was flushed and rinsed with ice cold physiological saline. Distal ileum segments (n = 4/group, 6 cm) nearest the cecum were immediately snap frozen in liquid nitrogen then stored at -80 °C until analysis.

Phospholipid fatty acid analysis

Phospholipid fatty acids were extracted from ground ileum tissue (approximately 16 mg) by a modified Folch procedure as previously described[33]. Tissue was homogenized in 0.8 mL of calcium chloride (0.025%) solution on ice. Chloroform: Methanol (2: 1, v/v) containing standards (No. 330731. Avanti Polar Lipids, Alabaster, AL, United States) was added to the homogenate, vortexed, flushed with nitrogen gas (N₂), and stored overnight at 4 °C. The bottom layer was concentrated under N₂ and reconstituted with mobile phase A composed of isopropanol/hexane/water (58/40/2, v/v) with 5 mmol/L ammonium acetate and 0.1% acetic acid.

Phospholipid extracts were separated by normal phase liquid chromatography (LC) (Agilent Zorbax RX-Sil column 3.0 mm × 100 mm, 1.8 μm particle size) using an Agilent 1260 Infinity LC System (Agilent Technologies; Santa Clara, CA, United States) as described previously[34,35]. Mobile phase B consisted of isopropanol/hexane/water (50/40/10, v/v) with 5 mmol/L ammonium acetate and 0.1% acetic acid.

All mass spectrophotometer (MS) measurements were obtained by triple-quadrupole LC/MS system (Agilent 6430, Agilent Technologies; Santa Clara, CA, United States) operating in multiple reaction monitoring (MRM) in negative mode. A library of theoretical precursor ions was generated for phospholipids (phosphatidylcholine, phosphatidylethanolamine, phosphatidylinositol, phosphatidylserine, and phosphatidylglycerol) of varying fatty acid carbon chain as previously described[35]. Data acquisition and analysis was carried out using the Agilent Mass Hunter software package.

Phospholipid fatty acids (ng) were expressed per miligram (mg) of protein in ileum tissue and as percentage of total phospholipid fatty acids (w/w%). Total protein concentration (mg/mL) of tissue homogenate was determined by bicinchoninic acid (BCA) assay (Pierce BCA protein assay kit, 23227, Thermo Fischer Scientific).

Oxylipin analysis

The extraction and analysis of oxylipins were performed as previously described[36]. Ground ileum tissue (approximately 50 mg) homogenized in ice cold Tyrode’s salt solution (potential of hydrogen = 7.6, 5.71 mL/mg of tissue) and a BCA assay was used to measure protein levels of ileum homogenate. Deuterated internal standards (Supplementary Table 1) (Cayman Chemical, MI, United States) were added to 200 μL of homogenate aliquots and subjected to analysis as described[9,36].

All MS measurements were obtained using an Agilent 6430 Triple-Quad LC/MS system (Santa Clara, CA, United States) operating in dynamic-MRM in negative mode. The transitions scanned is (M-H)^-. Details of the scanned oxylipins, deuterated internal standards, mass transition, and retention times for all deuterated standards are listed in Supplementary Tables 2 and 3. The quantitation was performed according to Deems et al[37]. Data acquisition and analysis was carried out using the Agilent Mass Hunter software package. Assignment of samples was unknown to the analyst during analysis of oxylipins and phosphopholipid fatty acid.

Enzyme-linked immunosorbent assay

Levels of ileum IL-6, IL-10, IL-1β, TNF-α, and fractalkine were measured in duplicate using commercially available enzyme-linked immunosorbent assay kits for each cytokine and chemokine (RayBiotech, Norcross, GA, United States) according to the manufacturer’s instructions. Concentrations for each analyte were calculated according to a standard curve constructed for each assay. Protein levels of the ileum homogenate were determined using a BCA assay.

Statistics analysis

Statistical analysis used GraphPad Prism version 10.2.3 for macOS (GraphPad Software, San Diego, CA, United States). Values are reported as mean ± SEM. Two samples for linoleic acid (LA)-derived oxylipin 13-hydroxyoctadecadienoic acid (HODE) were more than three standard deviations from the aggregated mean of rats fed the EPA + DHA diet. Thus, the outliers were replaced with the mean of the outlier’s treatment group as noted in Supplementary Table 4. Absolute fatty acids were expressed as relative percent of total phospholipid (w/w). Cytokines were log transformed before statistical analysis. Post-hoc power calculations indicated adequate study power (100%). Significance of treatment and diet effect was tested using two-way analysis of variance and Bonferroni post hoc analysis. Effect of chemotherapy was assessed by comparing baseline day 0 to the control diet at each timepoint, and the effect of diet was assessed by comparing the control diet and the EPA + DHA diet groups on the same day after chemotherapy. One-sample independent t-test assessed differences between healthy reference and tumor-bearing animals at baseline to determine the effect of tumor growth. P ≤ 0.05 were considered statistically significant.

RESULTS

Effects on phospholipid

The healthy reference and tumor-bearing animals were similar at baseline, suggesting tumor has a minimal impact on outcomes measured (data not shown). In healthy controls, ileal membrane phospholipids contained < 0.4% (w/w) of total n-3 fatty acids (predominantly DHA, with minimal ALA and trace EPA) and 14% (w/w) total n-6 fatty acids (10% AA, 4% LA) (Table 1). Chemotherapy acutely elevated AA by day 2 (P < 0.05), which subsequently declined by day 4 regardless of diet. By day 8, however, AA was significantly lower in the EPA + DHA group than in the control group (Table 1).

Table 1 Total phospholipid fatty acids in ileum tissue of rats treated with chemotherapy and fed eicosapentaenoic acid + docosahexaenoic acid diet, mean ± SEM.

Phospholipid fatty acids (% w/w)	Baseline	Control diet			EPA + DHA diet			Main effects		Interaction
	Day 0 (n = 5)	Day 2 (n = 4)	Day 4 (n = 4)	Day 8 (n = 4)	Day 2 (n = 5)	Day 4 (n = 4)	Day 8 (n = 4)	P value (diet)	P value (day)	P value (diet and day)
18: 2n-6	3.9 ± 0.38	3.8 ± 0.163	5.3 ± 0.17	4.0 ± 0.19	3.9 ± 0.31	5.0 ± 0.17	4.4 ± 0.22		0.0003
20: 4n-6	9.7 ± 0.28	11 ± 0.251,3	8.9 ± 0.47	10 ± 0.16	10 ± 0.133	8.0 ± 0.30	8.6 ± 0.332	0.001	< 0.001	0.031
Total n-6	14 ± 0.39	15 ± 0.16	14 ± 0.17	14 ± 0.20	14 ± 0.33	13 ± 0.20	13 ± 0.24	0.049	0.003
18: 3n-3	0.024 ± 0.009	0.002 ± Tr3	0.053 ± Tr	0.050 ± 0.011	0.027 ± 0.0203	0.057 ± 0.005	0.058 ± 0.014		0.002
20: 5n-3	Tr	0	0.002 ± 0.001	Tr	0.058 ± 0.0353	0.54 ± 0.102	0.80 ± 0.122	< 0.001	< 0.001	< 0.001
22: 6n-3	0.29 ± 0.01	0.27 ± 0.005	0.31 ± 0.003	0.30 ± 0.02	0.31 ± 0.0143	0.52 ± 0.032	0.69 ± 0.022	< 0.001	< 0.001	< 0.001
Total n-3	0.31 ± 0.018	0.27 ± 0.005	0.37 ± 0.005	0.36 ± 0.027	0.40 ± 0.0693	1.1 ± 0.142	1.5 ± 0.152	< 0.001	< 0.001	< 0.001

¹Denotes significant differences between baseline and day 2 in rats fed the control diet (chemotherapy effect).

²Denotes significant differences between diets on the same day (diet effect).

³Denotes significant differences between days within the same diet (time effect).

Tr: Trace; EPA: Eicosapentaenoic acid; DHA: Docosahexaenoic acid.

The EPA + DHA diet substantially enriched membrane composition, driving a > 40-fold increase in EPA (P < 0.0001), a 2-fold increase in DHA (P = 0.0001), and a nearly 4-fold increase in total n-3 fatty acids from baseline to day 8 (P < 0.0001). Consequently, EPA, DHA, and total n-3 fatty acid levels were significantly elevated in the EPA + DHA group compared to controls at both day 4 and day 8.

Effects on lipid signaling mediators

Baseline oxylipin profiles derived from n-3 and n-6 fatty acids were unaffected by tumor presence (data not shown). Following chemotherapy, n-6-derived oxylipins and the total sum of AA-derived oxylipins increased 2-fold. Specifically, three AA metabolites [6-keto-PGF_1α, 8-hydroxyeicosatetraenoic acid (HETE), and LTB₄] and the LA metabolite 9-HODE (data not shown) were elevated by day 2 post-chemotherapy (Table 2). In animals fed the control diet, the majority of n-6-derived oxylipins returned to baseline levels by day 8. Conversely, the EPA + DHA diet suppressed 6-keto-PGF_1α, PGD₂, PGF_2α, and TXB₂ (the stable metabolite of TXA₂) well below baseline levels by day 4. Notably, the anti-inflammatory AA metabolite lipoxin (LX) A4 was 5-fold higher (P = 0.018) on day 8 in the EPA + DHA group compared to the control group.

Table 2 Oxylipins derived from n-6 polyunsaturated fatty acid in ileum tissue of rats treated with chemotherapy and fed eicosapentaenoic acid + docosahexaenoic acid diet, mean ± SEM.

Oxylipin (ng/mg of protein)	Baseline	Control diet			EPA + DHA diet			Main effects		Interaction
	Day 0 (n = 5)	Day 2 (n = 4)	Day 4 (n = 4)	Day 8 (n = 4)	Day 2 (n = 5)	Day 4 (n = 4)	Day 8 (n = 4)	P value (diet)	P value (day)	P-diet × day
PGD2	88 ± 5.8	110 ± 166	66 ± 8.8	83 ± 7.9	92 ± 4.16	31 ± 6.0	36 ± 2.35	< 0.001	< 0.001	0.026
PGE2	10 ± 0.48	22 ± 1.1	20 ± 2.5	11 ± 1.7	25 ± 3.7	16 ± 9.7	5.9 ± 0.42		< 0.001
6-keto-PGF1α	19 ± 0.93	34 ± 3.14,6	19 ± 1.8	19 ± 1.5	32 ± 3.06	9.4 ± 1.55	9.2 ± 0.465	< 0.001	< 0.001	0.021
PGF2α	63 ± 2.3	89 ± 12	49 ± 8.5	58 ± 2.7	79 ± 196	21 ± 4.7	20 ± 2.3	0.007	< 0.001
1TXB2	170 ± 10	180 ± 27	110 ± 18	170 ± 18	150 ± 276	60 ± 12	97 ± 15	0.006	< 0.001
5-HETE	37 ± 6.6	65 ± 5.36	15 ± 3.6	50 ± 5.5	66 ± 9.86	15 ± 5.1	48 ± 11		< 0.001
8-HETE	17 ± 3.3	49 ± 164,6	9.3 ± 2.4	26 ± 1.4	44 ± 8.16	8.8 ± 2.4	23 ± 3.6		< 0.001
11-HETE	590 ± 97	1300 ± 4086	380 ± 112	750 ± 101	1000 ± 155	370 ± 107	630 ± 52		< 0.001
12-HETE	34 ± 6.3	56 ± 156	21 ± 3.1	36 ± 6.3	37 ± 8.5	10 ± 2.3	29 ± 5.5		0.004
15-HETE	300 ± 32	330 ± 566	150 ± 26	340 ± 12	380 ± 656	120 ± 39	220 ± 37		< 0.001
20-HETE	1.5 ± 0.52	2.7 ± 1.2	0.91 ± 0.77	1.9 ± 0.21	1.2 ± 0.33	0.16 ± 0.16	1.0 ± 0.49
LTB4	0.30 ± 0.045	0.92 ± 0.124	0.35 ± 0.086	0.31 ± 0.06	0.83 ± 0.196	0.09 ± 0.05	0.56 ± 0.34		< 0.001
LXA4	0.16 ± 0.049	0.36 ± 0.13	0.033 ± Tr	0.11 ± 0.02	0.49 ± 0.126	0.10 ± 0.02	0.57 ± 0.225	0.024	0.004
2Sum of AA-derived	1400 ± 120	2400 ± 4504,6	880 ± 170	1700 ± 140	2100 ± 1906	700 ± 180	1300 ± 50		< 0.001
3Sum of LA-derived	890 ± 160	1600 ± 2405,6	1800 ± 390	690 ± 88	1500 ± 2406	690 ± 1005	830 ± 190		0.006	0.038
Sum n-6-derived	2300 ± 220	4000 ± 6604	2700 ± 250	2400 ± 200	3500 ± 2506	1400 ± 250	2100 ± 230	0.03	< 0.001

¹Thromboxane (TX) B₂ is the inactive but stable product of the bioactive precursor TXA₂.

²Sum of arachidonic acid-derived: Prostaglandin (PG) D₂, PGE₂, 6-keto-PGF_1α, PGF_2α, thromboxane (TX) B₂, 5-hydroxyeicosatetraenoic acid (HETE), 8-HETE, 11-HETE, 12-HETE, 15-HETE, 8,9-dihydroxy-eicosatetraenoic acid (DiHETE), 11,12-DiHETE, 14,15-DiHETE, 20-HETE (trace, not shown), leukotriene-B_4, and LXA₂.

³Sum of linoleic acid-derived: 9-hydroxyoctadecadienoic acid (HODE), 13-HODE, 9,10-dihydroxyoctadecenoic acid (DiHOME), 12,13-DiHOME. Two sample assessed for 13-HODE were more than three standard deviations from the aggregated mean of rats fed the eicosapentaenoic acid + docosahexaenoic acid diet. Thus, the two outliers were replaced with the mean of each outlier’s treatment group.

⁴Denotes significant differences between baseline and day 2 in rats fed the control diet (chemotherapy effect).

⁵Denotes significant differences between diets on the same day (diet effect).

⁶Denotes significant differences between days within the same diet (time effect).

Tr: Trace; EPA: Eicosapentaenoic acid; DHA: Docosahexaenoic acid; PG: Prostaglandin; TX: Thromboxane; HETE: Hydroxyeicosatetraenoic acid; LT: Leukotriene; LX: Lipoxin; AA: Arachidonic acid; LA: Linoleic acid.

Interestingly, EPA-derived oxylipins were detectable at all time points in the control group (Table 3), despite trace or undetectable baseline membrane EPA levels (Table 1). Chemotherapy also induced a 2-fold increase in protectin D1, a SPM derived from DHA (P = 0.024). By day 8, the EPA + DHA diet significantly elevated total n-3 and EPA-derived oxylipins, all individual EPA metabolites (15-HETE, 18-HETE, and 17,18-dihydroxy-eicosatetraenoic acid), and the DHA metabolite 16-hydroxy-DHA compared to the control diet. This substantial accumulation of EPA and DHA metabolites by day 8 temporally reflects the peak availability of their respective fatty acid substrates within the ileal membrane phospholipids.

Table 3 Oxylipins derived from n-3 polyunsaturated fatty acid in ileum tissue of rats treated with chemotherapy and fed eicosapentaenoic acid + docosahexaenoic acid diet, mean ± SEM.

Oxylipin (ng/mg of protein)	Baseline	Control diet			EPA + DHA diet			Main effects		Interaction
	Day 0 (n = 5)	Day 2 (n = 4)	Day 4 (n = 4)	Day 8 (n = 4)	Day 2 (n = 5)	Day 4 (n = 4)	Day 8 (n = 4)	P value (diet)	P value (day)	P-diet × day
15-HEPE	7.7 ± 2.3	12 ± 1.3	2.9 ± 2.2	6.1 ± 1.3	13 ± 4.05	11 ± 2.4	29 ± 5.84	0.001	0.009	0.003
18-HEPE	1.1 ± 0.8	1.1 ± 1.1	Tr ± Tr	2.2 ± 1.3	5.0 ± 1.75	6.8 ± 1.8	64 ± 174	< 0.001	< 0.001	< 0.001
17,18-DiHETE	0.0 ± Tr	Tr ± Tr	0.06 ± 0.06	0.07 ± Tr	0.16 ± 0.065	0.43 ± 0.164	2.1 ± 0.194	< 0.001	< 0.001	< 0.001
1Sum EPA-derived	8.8 ± 3.0	13 ± 2.2	2.9 ± 2.2	8.4 ± 0.61	18 ± 5.05	18 ± 4.0	94 ± 204	< 0.001	< 0.001	< 0.001
4-HDoHE	3.9 ± 1.0	7.9 ± 1.3	2.1 ± 0.53	3.5 ± 0.43	9.2 ± 2.05	2.5 ± 0.67	11 ± 3.1	0.037	< 0.001
8-HDoHE	0.4 ± 0.1	1.1 ± 0.41	0.06 ± 0.06	0.2 ± 0.07	0.81 ± 0.31	0.68 ± 0.17	1.2 ± 0.14
13-HDoHE	4.0 ± 0.3	3.5 ± 1.0	2.8 ± 0.98	3.3 ± 0.52	5.7 ± 1.4	3.5 ± 0.96	7.4 ± 1.9	0.023
14-HDoHE	4.3 ± 0.8	4.2 ± 1.2	2.7 ± 0.50	3.7 ± 0.70	5.8 ± 1.4	2.7 ± 0.34	5.4 ± 1.7
16-HDoHE	154 ± 20	170 ± 32	66 ± 18	160 ± 8.3	240 ± 565	120 ± 34	340 ± 764	0.013	0.005
17-HDoHE	4.2 ± 0.3	5.7 ± 1.2	5.0 ± 1.0	3.8 ± 0.67	8.4 ± 1.0	10 ± 6.8	10 ± 1.5	0.044
20-HDoHE	24 ± 7.9	41 ± 9.2	12 ± 7.5	31 ± 2.7	33 ± 7.75	12 ± 2.7	58 ± 17		0.006
Protectin D1	0.71 ± 0.0	2.0 ± 0.403,4	0.54 ± 0.29	0.71 ± 0.11	1.8 ± 0.465	0.61 ± 0.22	1.6 ± 0.49		< 0.001
2Sum DHA-derived	196 ± 26	230 ± 39	91 ± 20	210 ± 8.1	300 ± 685	150 ± 38	430 ± 87	0.013	0.002
Sum n-3-derived	205 ± 28	250 ± 38	94 ± 20	220 ± 8.5	320 ± 725	170 ± 42	520 ± 804	0.002	< 0.001	0.018

¹Sum of eicosapentaenoic acid-derived oxylipins: 15-hydroxyeicosapentaenoic acid (HEPE), 18-HEPE, 17,18-dihydroxy-eicosatetraenoic acid (trace, not shown).

²Sum of docosahexaenoic acid-derived oxylipins: 4-hydroxy-docosahexaenoic acid (HDoHE), 8-HDoHE, 13-HDoHE, 14-HDoHE, 16-HDoHE, 17-HDoHE, 20-HDoHE, protectin-D1 (trace, not shown).

³Denotes significant differences between baseline and day 2 in rats fed the control diet (chemotherapy effect).

⁴Denotes significant differences between diets on the same day (diet effect).

⁵Denotes significant differences between days within the same diet (time effect).

Tr: Trace; EPA: Eicosapentaenoic acid; DHA: Docosahexaenoic acid; HEPE: Hydroxyeicosapentaenoic acid; DiHETE: Dihydroxy-eicosatetraenoic acid; HDoHE: Hydroxy-docosahexaenoic acid.

Effects on cytokines and chemokines

Ileal cytokines and chemokines were not affected by the presence of a tumor (data not shown). By day 8 following chemotherapy, TNF-α and IL-1β were highest while fractalkine was highest on day 4. There was little impact of diet on cytokines measured (Table 4).

Table 4 Cytokines and chemokine in ileum tissue of rats treated with chemotherapy and fed eicosapentaenoic acid + docosahexaenoic acid diet, mean ± SEM.

Immune signals (pg/mg of protein)	Baseline	Control diet			EPA + DHA diet			Main effects		Interaction
	Day 0 (n = 5)	Day 2 (n = 4)	Day 4 (n = 4)	Day 8 (n = 4)	Day 2 (n = 5)	Day 4 (n = 4)	Day 8 (n = 4)	P value (diet)	P value (day)	P-diet × day
TNF-α	74 ± 2.4	58 ± 6.61	60 ± 6.6	93 ± 1.6	62 ± 7.11	69 ± 2.8	100 ± 6.9		< 0.001
IL-10	88 ± 3.0	67 ± 3.6	74 ± 15	110 ± 18	77 ± 8.6	87 ± 10	110 ± 21
Fractalkine	710 ± 31	600 ± 311	930 ± 120	890 ± 76	610 ± 501	946 ± 68	860 ± 79		< 0.001
IL-1β	220 ± 28	110 ± 151	140 ± 22	300 ± 35	140 ± 301	146 ± 23	330 ± 69		< 0.001
IL-6	480 ± 7.6	480 ± 9.8	490 ± 3.7	500 ± 9.5	500 ± 5.7	490 ± 6.7	520 ± 9.4

¹Denotes significant differences between days on the same diet (time effect).

EPA: Eicosapentaenoic acid; DHA: Docosahexaenoic acid; IL: Interleukin; TNF: Tumor necrosis factor.

DISCUSSION

The present study is the first to analyze a broad range of oxylipins derived from ileum tissue fatty acids after CPT-11 + 5-FU and to report effects of dietary EPA and DHA on fatty acid content in ileal membrane phospholipid. Total oxylipins derived from AA and oxylipins derived from n-6 fatty acids increased after low-dose chemotherapy despite minimal change to substrate and proinflammatory cytokines. 6-keto-PGF_1α increased after chemotherapy, but dietary EPA and DHA markedly reduced several prostanoids that may have positive implications for managing effects of repeated chemotherapy cycles.

PGI₂ hydrolysis produces 6-keto-PGF_1α, a bioactive, stable, non-enzymatic product. Intravenous injection of PGI₂ into healthy subjects inhibited absorption of water, sodium, and chloride absorption the lumen of the jejunum for 200 minutes[7]. Given the short half-life of PGI₂ (2-3 minutes), 6-keto-PGF_1α is implicated in the inhibitory effect that extends long after a decline in PGI₂ levels. The clinical importance of post-chemotherapy increases in 6-keto-PGF_1α on the incidence of diarrhea remains to be elucidated. Diarrhea was not detectable in the present study attributed to using a single, low-dose CPT-11 + 5-FU regimen.

Dietary supplementation with EPA and DHA, initiated concurrently with chemotherapy, suppressed levels of 6-keto-PGF_1α, PGD₂, and PGF_2α, TXB₂ well below baseline from day 4 through the end of the study. While EPA and DHA displaced AA in membrane phospholipids, altering substrate availability for prostanoid synthesis, the prostanoid reduction was not driven by a deficiency in AA substrate. Upon study completion, AA comprised 10% of total ileal phospholipid fatty acids, five-fold higher than the relative abundance of EPA and DHA combined.

Our findings align with previous literature demonstrating that elevated EPA content reduces AA metabolite concentrations[38-40] in a dose-dependent manner[41], despite COX-2 preferentially oxidizing AA over EPA[42]. This reduction in AA-derived metabolites may be mediated by two distinct mechanisms. First, increased membrane EPA directly competes with AA for the active binding sites of the COX, LOX, and cytochrome P450 enzymatic pathways. Second, EPA acts as an inhibitor of AA oxidation via COX-1, the constitutively expressed COX isoform[42].

Low prostanoid levels may reduce effects of compounding toxicity from multiple cycles. Diarrhea, which can be dose limited, is a common side effect after treatment with multiple cycles of combination chemotherapy containing CPT-11 and 5-FU[43]. The present study applied a clinically relevant dose of CPT-11 + 5-FU and was not designed to induce diarrhea. Increasing EPA and DHA content in phospholipid through diet reduced prostanoids to levels lower than baseline. Notably, rats treated with chemotherapy reduced food intake by 75% within the first two days of injection and intake was not restored to baseline until day 4. Thus, observed prostanoid changes are occurring under very low dietary availability of EPA and DHA. Oral gavage feeding of EPA and DHA rather than dietary delivery may elevate EPA and DHA content in ileal membrane phospholipid before starting chemotherapy and subsequent cycles. A similar reduction of hepatic prostanoids to levels much lower than pre-CPT-11 + 5-FU was observed after treatment with dietary EPA and DHA while n-3 derived oxylipins were restored[44].

To date, intestinal phospholipid fatty acid profiles have not been evaluated across multiple chemotherapy cycles, an assessment which is critical to determining the clinical relevance of depleted baseline ileal EPA and DHA content over successive rounds of treatment. Neither the low-dose combination regimen used in the present study nor the intensive four-day CPT-11 protocol (60 mg/kg) evaluated by Usami et al[4] substantially altered essential fatty acid content in rat ileal mucosa. However, cumulative cycles of chemotherapy may exert more pronounced effects on these essential fatty acids. Entering subsequent treatment cycles with minimal ileal EPA and DHA reserves could facilitate robust prostanoid synthesis. Consequently, dietary provision of EPA and DHA to enrich ileal tissue may offer therapeutic benefits in clinical settings, particularly for patients undergoing multiple cycles of chemotherapy, suffering from treatment-induced anorexia or reduced oral intake.

By day 8, every evaluated AA metabolite was lower in the EPA + DHA group than in the control group, with the notable exception of the SPM LXA₄, which was elevated 5-fold. Synthesized via the 5-LOX and 12/15-LOX pathways, LXA₄ possesses potent anti-inflammatory properties that counteract the pro-inflammatory actions of other AA-derived eicosanoids. Mechanistically, LXA₄ inhibits neutrophil migration[45] and mucosal infiltration[46] while suppressing nuclear factor kappa-B activation in intestinal epithelial cells and macrophages[47,48]. Although dietary AA supplementation in a colitis model was previously shown to increase mucosal AA and LXA₄ levels without exacerbating PGE₂ production or disease severity[49], elevated AA in adipose tissue phospholipids is clinically associated with an increased risk of developing ulcerative colitis[50]. Dietary EPA and DHA supplementation therefore represents a safer strategy to selectively upregulate protective LXA₄ synthesis while avoiding risks associated with higher systemic AA availability[51].

Nearly all EPA and DHA metabolites were increased within 8 days through diet. Conversely, protectin-D1 increased two-fold by day 2 after chemotherapy while other SPMs maresin-1 and resolvins D1 and E1 were below detection limits (Supplementary Table 3). Protectin-D1 is produced from DHA by neutrophils and inhibits neutrophil infiltration through autocrine signaling[52]. Higher EPA and DHA content in ileal membrane phospholipid may enhance the natural protective effect of protectin-D1 in response to chemotherapy. LTB₄ is an AA-derived LOX metabolite that has opposite effects of protectin-D1 as a potent chemoattractant involved in inflammation and immune response[53]. Fractalkine is also a chemoattractant produced by epithelial and endothelial cells for leukocyte migration to the intestine[54,55]. Fractalkine remained stable while LTB₄ increased three fold after chemotherapy (Table 2), highlighting the responsiveness of oxylipins to chemotherapy compared to cytokines (Table 4). The opposing effects of protectin-D1 and LTB₄ underscore the potential to inhibit neutrophil infiltration by skewing the balance of dietary EPA and DHA vs AA. Determining how dietary EPA and DHA modify neutrophil presence via protein and lipid-derived mediators after chemotherapy treatment is an area for future exploration.

CPT-11 + 5-FU minimally impacted proinflammatory cytokines and oxylipins that have known regulatory effects on water and electrolyte balance in ileum tissue in the present study. A preclinical model using multiple cycles of CPT-11 and 5-FU to induce diarrhea would enrich interpretation of oxylipin changes in relation to a clinical endpoint. Moreover, testing the effect of EPA independent of DHA would elucidate mechanisms that consider the interplay of elongation and desaturation reactions of n-3 PUFAs. Activation of peroxisome proliferator-activated receptor-γ and inhibition of nuclear factor kappa-B by EPA and DHA need to be explored in future studies along with possible anti-inflammatory effects mediated through G-protein coupled receptor 120. Joint therapies such as EPA and DHA with COX-2 inhibitors or EPA and DHA with supplementary amino acids[15,56] may have superior protective effects against CPT-11 and 5-FU.

CONCLUSION

Dietary supplementation with EPA and DHA may provide a safe and effective therapeutic strategy to mitigate prostanoid production implicated in impaired water absorption and electrolyte dysregulation that frequently complicate intensive or multi-cycle chemotherapy regimens. Beyond suppressing pro-inflammatory eicosanoids, dietary EPA and DHA offer dual benefits of upregulating LXA₄ and protectin D1, both of which drive the active resolution of acute inflammation. This concurrent reduction in prostanoids and enrichment of protective n-3 and n-6 lipid mediators represent a promising therapeutic avenue to alleviate severe intestinal toxicities associated with aggressive or repeated CPT-11 + 5-FU regimens.

ACKNOWLEDGEMENTS

The authors of the present manuscript would like to thank Abha Dunichand-Hoedl, Peter O Isesele, and MD Monirujjaman for their technical assistance and valuable support in the animal experiment.