HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mehta, S. P Articles by Rhodes, M. Search for Related Content PubMed PubMed Citation Articles by Mehta, S. P Articles by Rhodes, M. Agricola Articles by Mehta, S. P Articles by Rhodes, M.

Effect of n–3 polyunsaturated fatty acids on Barrett's epithelium in the human lower esophagus ^1,2,3

¹ From the Department of Upper Gastrointestinal Surgery (SPM, APB, JC, and MR) and the Department of Histopathology (VS), Norfolk and Norwich University Hospital, Norwich, United Kingdom, and the Institute of Food Research, Norwich, United Kingdom (SPM, APB, EKL, and ITJ)

² Supported by SLA Pharma, who donated the EPA capsules used for supplementation.

³ Reprints not available. Address correspondence to IT Johnson, Institute of Food Research, Norwich Research Park, Colney, Norwich NR4 7UA, United Kingdom. E-mail: ian.johnson{at}bbsrc.ac.uk.

	ABSTRACT

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Background: Epidemiologic studies suggest a reduced risk of esophageal adenocarcinoma in populations with a high consumption of fish, and n–3 fatty acids inhibit experimental carcinogenesis. One possible explanation is the suppression of eicosanoid production through inhibition of cyclooxygenase 2 (COX-2).

Objective: The objective was to determine the effects of dietary supplementation with the n–3 fatty acid eicosapentaenoic acid (EPA) on a number of biological endpoints in Barrett's esophagus.

Design: Fifty-two participants with known Barrett's esophagus underwent endoscopy. Biopsy samples were obtained from a recorded level within the area of Barrett's esophagus, and then 27 patients were randomly assigned to consume EPA capsules (1.5 g/d) for 6 mo or no supplement (controls). At the end of this period, patients again underwent endoscopy, and biopsy samples were collected at the same level. Tissue samples were analyzed for mucosal lipid, prostaglandin E₂, leukotriene B_4, COX-2 protein, and RNA concentrations. Cellular proliferation was also measured, by Ki-67 immunohistochemistry.

Results: The EPA content of esophageal mucosa increased over the study period in the n–3–supplemented subjects and was significantly different from the content in the controls (P < 0.01). There was also a significant decline in COX-2 protein concentrations (measured by immunoblotting) in the n–3 group, and the difference was significant from that in the controls (P < 0.05); no difference in COX-2 RNA concentrations was observed between groups. This change in COX-2 protein was inversely related to the change in EPA content (P < 0.05). There was no significant difference in the change in prostaglandin E₂, leukotriene B₄, or cellular proliferation between the 2 groups.

Conclusion: Supplementation with EPA significantly changed n–3 fatty acid concentrations and reduced COX-2 concentrations in Barrett's tissue.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
The incidence of esophageal adenocarcinoma (EAC) in Europe and the United Kingdom has increased considerably over the past 20 y (1), with an increase per annum greater than that of any other malignancy (2). This condition is diagnosed at an advanced stage in most patients, and EAC is usually associated with a poor prognosis and mean survival time of <1 y (3). Existing treatments for EAC have generally had limited success, and attention is therefore now focusing on identifying potential targets for prevention strategies or therapy that might arrest or delay the progression to adenocarcinoma. Many current trials looking at chemoprevention have targeted the metaplastic condition Barrett's esophagus, in which the normal squamous epithelium in the lower esophagus is replaced with columnar epithelium (4, 5). Barrett's esophagus is a premalignant condition from which EAC arises almost exclusively (6), although <5% of those affected will go on to develop EAC (7).

Both epidemiologic and experimental studies suggest that diet has an important influence on the prevalence of various types of cancer (8). One particularly interesting dietary association is the inverse relation between fish consumption and colorectal cancer risk across populations in Europe (9). Epidemiologic and experimental data implicate fish oils as inhibitors of the development and progression of a range of human cancers (9, 10), including EAC (11).

Oily fish contain relatively high concentrations of the essential n–3 polyunsaturated fatty acids (PUFAs) eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). Previous studies have shown that a diet high in fish oil can result in a reduced expression of the enzyme cyclooxygenase 2 (COX-2) in colonic mucosa and in a decrease in colonic tumor development in rats, compared with a high-fat corn oil diet (12). This is important because up-regulation of COX-2 has been shown to occur in both Barrett's esophagus and EAC (13). Furthermore, in vitro studies have shown that overexpression of COX-2 can reduce the rate of apoptosis, increase the invasiveness of malignant cells, and promote angiogenesis (14, 15). EPA and DHA can also compete with arachidonic acid for the COX-2 and lipooxygenase enzymes (16) and thus reduce the formation of the eicosanoids prostaglandin E₂ (PGE₂) and leukotriene B₄ (LTB₄), which have also been implicated in the carcinogenic process (17-20).

In the present study, we evaluated the effect of dietary supplementation with EPA for 6 mo in patients with Barrett's esophagus. Tissue eicosanoid (PGE₂ and LTB₄), COX-2 protein, and mRNA concentrations have been measured in endoscopically obtained biopsy samples. Additionally, the possibility that EPA may exert anticarcinogenic effects by down-regulating esophageal epithelial cell proliferation has been explored by measuring the expression of Ki-67 as a marker of proliferation in Barrett's epithelium before and after EPA supplementation.

	SUBJECTS AND METHODS

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Subjects and protocol
Participants known to have histologic evidence of Barrett's esophagus (intestinal metaplasia identified in previous surveillance biopsy samples, in addition to endoscopic evidence of >3 cm of the distal esophagus affected) and scheduled for a routine surveillance endoscopy within 1 y of the study were invited to participate in this study. Each participant was identified from the hospital endoscopy database and received an invitation letter and information leaflet 3–4 wk before the procedure. Participants were also asked to stop all proton pump inhibitor medication use 3–4 d before their endoscopy. Patients were excluded from the study if they were allergic to fish, pregnant or breastfeeding, or receiving immunosuppressive treatment or warfarin. The study was approved by the East Norfolk and Waveney Research Governance Committee and by the Norwich Research Ethics committee (ref. 04/Q0101/96).

A power calculation was performed to estimate the sample size required. From published data it was estimated that, for the primary outcome measure (COX-2 protein concentrations) and secondary outcome measures (PGE₂, LTB₄, COX-2 mRNA, and COX-1 concentrations), a sample size of 10 in the intervention group was required to observe biologically significant changes at a statistical power of 90%. To provide enough biopsy material for all of the experiments, 30 participants were recruited into the intervention (n–3) group and 20 participants into a matched control group in a randomized fashion as explained below.

Each participant underwent an initial gastroscopy performed by SPM or JC. Barrett's epithelium was identified by the presence of red columnar islands within squamous esophageal mucosa or by the presence of lighter squamous islands within a circumferential columnar mucosa, present above the endoscopically defined lower esophageal sphincter. Quadrantic paired biopsy samples were collected from the Barrett's mucosa at a designated level (1 cm proximal to the endoscopically defined lower esophageal sphincter) and also from normal squamous mucosa in the midesophagus.

Each participant was then randomly allocated to n–3 fatty acid supplementation or control groups by means of cards, prepared by using random-number generator software and sealed in envelopes before the commencement of the study. The supplement consisted of 1.5 g/d of unesterified (99%) EPA provided in 500-mg capsules (SLA Pharma, Watford, United Kingdom). EPA-supplemented participants were asked to take 1 capsule 3 times/d day and were issued with a diary sheet to assist them. After 3 mo, each participant was reviewed at the clinic and given another 3-mo supply of capsules. Control subjects received no supplementation. After 6 mo, all participants were invited to undergo another endoscopy. Again, quadrantic biopsy samples were collected from the lower esophagus, at the same level from the gastroesophageal junction as previously and from the midesophagus. All biopsy samples were either formalin-fixed for histopathologic assessment at the Norfolk and Norwich University Hospital or snap-frozen for subsequent experimental use.

All biopsy samples were coded to protect the patients' identity. Laboratory work was performed so that the investigators were blinded during the analysis of the results. In general, biopsy samples from 10 subjects from each of the n–3 and control groups were used for each of the described experiments, except for the COX-2 immunoblotting experiment, for which samples from 15 subjects from the n–3 group were used.

PGE₂ and LTB₄ measurements
Each biopsy tissue sample (5 mg) was weighed and homogenized (using plastic grinders) in 400 µL protein extraction buffer. Buffer components consisted of 50 mmol/L Tris (pH 7.5), 1% Tween 20, 150 mmol/L NaCl, 1% Nonidet P-40, 1% sodium deoxycholate, and 5 mmol/L EDTA. Protease inhibitor tablets were added just before solubilization (Mini Protease Inhibitor Cocktail tablets; Roche, Basel, Switzerland). The protein concentration of the homogenized tissue was determined by using a Bicinchoninic acid assay (BCA Protein Assay Kit; Pierce Biotech, Rockford, IL) according to the protocol. PGE₂ and LTB₄ were analyzed by using enzyme immunoassay kits (Assays Designs, Ann Arbor, MI). Briefly, eicosanoids were first extracted from homogenized protein samples by using 200 mg C₁₈ Sep-Pak columns (Waters, Milford, MA) under a slight positive pressure. Eluted samples were dried under nitrogen and homogenized in Tris-buffered saline. PGE₂ and LTB₄ concentrations in each sample were determined by enzyme immunoassay according to the manufacturer's protocol.

Lipid extraction, methylation, and gas chromatography
Lipid extraction was performed by the addition of chloroform-methanol (2:1) to partition the lipid and nonlipid constituents from biopsy samples. This method was previously described by Bligh and Dyer (21). Lipid esters were then methylated by adding anhydrous methanol in the presence of potassium hydroxide after drying down the lipid extract. Methyl esters were finally solubilized in hexane. Two microliters of the esterified lipid sample were separated by gas chromatography–mass spectrometry (Hewlett-Packard 5890; Hewlett-Packard, Palo Alto, CA), and the relative abundances (expressed as percentages) of individual lipid constituents were determined within each sample. Standards containing known amounts of pure fatty acid compounds were run under the same conditions and retention times, and the lipid samples were matched with those of the standards.

Measurement of COX-2 and COX-1 concentrations
Homogenized biopsy samples were centrifuged (13 000 rpm) at 4 °C, and the supernatant protein content was measured as described previously. Total protein (10 µg) was separated by 10% polyacrylamide gel electrophoresis by using a precast bis-tris gel system (NuPAGE system; Invitrogen, Carlsbad, CA). Gels were blotted by using a wet-blotting method (XCell II Blot module; Invitrogen) onto a Hybond ECL-nitrocellulose membrane (Amersham Pharmacia Biotech, Uppsala, Sweden). Membranes were then blocked for 1 h in 5% low-fat milk/0.1% Tween-20/TBS. After being washed for 5 min in 0.1% Tween-20/TBS, the membranes were probed for 2 h with a COX-2 mouse monoclonal antibody (Cayman Chemicals, Ann Arbor, MI) at 1:1000 at room temperature. After another wash (5 min), the membranes were incubated for 90 min with a horseradish peroxidase–linked anti-mouse secondary antibody at 1:2000 dilution (Amersham Pharmacia Biotech). The membranes were then washed 3 additional times for 5 min in 0.1% Tween 20/TBS. Protein bands were detected by using an enhanced chemiluminescence procedure (ECL-Plus detection kit; Amersham Pharmacia Biotech). Band intensity was quantified with a densitometer (Quantity One Software; Bio-Rad, Hercules, CA) and associated software. After quantification of the COX-2 bands, the membranes were stripped by submersion in stripping buffer (100 mmol/L 2-mercaptoethanol, 2% sodium dodecyl sulfate, 62.5 mmol/L Tris-HCl; pH 6.7) and incubated at 50 °C for 30 min. After blocking, each membrane was probed for 2 h with a COX-1 mouse monoclonal antibody (Cayman Chemicals) 1:1000 at room temperature. Membranes were washed and probed with the secondary antibody, and bands were detected and quantified as described above.

Immunohistochemistry for Ki-67
Formalin-fixed paraffin-embedded samples were cut into 4-µm sections. Sections were then washed and incubated with a Ki-67 rabbit monoclonal antibody (Labvision Corporation, Fremont, CA) at 1:200 for 30 min at room temperature. After being washed, sections were immersed in a polymer penetration enhancer containing 10% animal serum in Tris-buffered saline for 15 min and immersed in a polymer-horseradish peroxidase anti-rabbit immunoglobulin G reagent for 15 min. After a further wash, sections were exposed to the substrate chromogen 3,3'-diaminobenzidine to visualize the complex as a brown precipitate. Finally, sections were counterstained with 0.02% hematoxylin to visualize cell nuclei. Sections of colonic epithelium displaying strong levels of Ki-67 staining were used as positive controls.

Scoring of slides for Ki-67 staining was performed by a senior histopathologist (VS) with individual sections coded for anonymity. Sections were divided into areas of predominantly gastric, intestinal, or squamous appearance. Ki-67 staining was scored on an 8-point scale (0–7) for numbers of cells with nuclear Ki-67 staining within each area.

Determination of COX-2 RNA concentrations by RT-PCR
RNA was extracted from biopsy samples stored in RNALater (Ambion, Austin, TX) by using a column extraction kit (RNeasy Plus mini kit; Qiagen, Hilden, Germany). After extraction, total RNA was quantified by using spectrophotometry (Nanodrop ND-1000 spectrophotometer, Wilmington, DE). COX-2 mRNA concentrations were determined by using off-the-shelf COX-2 primers and probes (Assay ID Hs00153133_m1; Applied Biosystems, Foster City, CA). PCR amplification of 20 µg RNA from each sample was performed through a one-step technique (Applied Biosystems) with the Taqman 7300 system. The following cycling conditions were used: 1 cycle at 48 °C for 30 min (for conversion to cDNA), 1 cycle at 95 °C for 10 min, and 40 cycles at 95 °C for 15 s followed by 60 °C for 1 min. β-Actin was used as a housekeeper gene, and, as for COX-2, off-the-shelf primers and probes were used for this RT-PCR reaction (Assay ID Hs99999903_m1; Applied Biosystems). Relative concentrations of COX-2 mRNA were then calculated from Ct values by using a standard curve for a control sample run on each plate.

Statistical analysis
Changes in fatty acid composition and in COX-2, PGE₂, and LTB₄ concentrations between the start and end of the study, in tissue samples from the n–3 and control groups, were compared by using unpaired 2-tailed t tests with statistical significance accepted at the 5% level. Bonferroni correction for multiple comparisons was applied for the statistical analysis of the lipid data; multiple t tests were performed for several fatty acids. For Ki-67, statistical comparisons were made by using Wilcoxon's signed-rank test. These calculations were performed with the assistance of a computer statistics package (SPSS version 11; SPSS Inc, Chicago, IL)

	RESULTS

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
In total, 52 participants with previously diagnosed Barrett's epithelium, were recruited into the study from January to May 2005. Thirty-four participants were randomly assigned to consume EPA capsules (n–3 group) and 18 participants served as controls (control group). Six participants left the study and 3 rejected the capsules after a few days because of intolerance to the EPA capsules (either nausea or vomiting or abdominal pain). Tissues from these 3 participants were recovered and analyzed, but they were eventually excluded from the statistical analysis. By the end of the study period, 27 participants had successfully maintained a regular dose of EPA capsules for 6 mo, and 19 participants served as controls. The demographic details of each of these cohorts are shown in Table 1.

PGE₂ and LTB₄ concentrations in Barrett's mucosa
Changes in PGE₂ and LTB₄ over 6 mo were compared between the n–3 and control groups, but no significant differences were observed (n = 10 for each group; Figure 1).

View larger version (7K): [in this window] [in a new window]   FIGURE 1.. Mean (±SD) changes in prostaglandin E2 (PGE2) and leukotriene (LTB4) concentrations over 6 mo in Barrett's tissue from the n–3 supplemented and control groups, determined by immunoassay. n = 10. No significant difference in PGE2 (P = 0.38) or in LTB4 (P = 0.79) was found between groups.

Table 2

Figure 2

Figure 3

View larger version (9K): [in this window] [in a new window]   FIGURE 2.. Mean (±SD) changes in band volumes of cyclooxygenase (COX) 1 and COX-2 over 6 mo in Barrett's tissue from the n–3 supplemented (n = 15) and control (n = 10) groups, determined by immunoblotting. *Significantly different from the n–3 group, P = 0.05 (Student's t test). No significant difference was observed for COX-1 (P = 0.58).

View larger version (18K): [in this window] [in a new window]   FIGURE 3.. Typical gel illustrating the decrease in band intensity after 6 mo of supplementation with the n–3 polyunsaturated fatty acid eicosapentaenoic acid (EPA). From left to right, the first well is the Magic Mark protein standard. The following 5 lanes contain protein in Barrett's tissue from 5 subjects at baseline. The next 5 wells are the same subjects' tissue samples after 6 mo of n–3 supplementation.

COX-2 gene expression
COX-2 gene expression was lower at 6 mo than at 0 mo in both the control and n–3 groups, but differences between the 2 groups were not statistically significant (n = 10 from each group).

Immunohistochemistry for Ki-67
Each tissue section was separated into squamous, gastric-type or intestinal-type epithelial areas and scored from 0 to 7 in ascending degree of cellular nuclear staining. Scores for each participant at the end of 6 mo were subtracted from baseline scores, and these mean differences are shown in Table 3.

	DISCUSSION

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
To our knowledge there has been no previous investigation of the effects of supplementation with PUFAs on the fatty acid composition of the human esophageal mucosa. Previous studies, however, have shown that fish-oil supplementation can have significant effects on mucosal lipids in other tissues of the alimentary tract. For example, Hillier et al (23) showed a 7-fold increase in human colonic mucosal EPA after only 3 wk of dietary supplementation with 18 g fish oils. In an earlier study from this laboratory, we also observed a significant increase in EPA concentrations in the colorectal mucosa of patients given fish-oil capsules (2.4 g/d) for 12 d before surgery (24).

In the present study we used a commercially available concentrated n–3 formulation containing EPA in the form of free fatty acid, rather than as a triglyceride, because it was felt that this might offer an advantage with regard to the efficiency of delivery of EPA to the target tissue (25). However, it was also previously noted that patients are less tolerant of free fatty acids than of other forms of n–3 PUFAs, the main side effect being belching (26). To reduce the risk of any adverse symptoms, participants received their 1.5 g EPA/d in 3 doses, and they were encouraged to ingest the capsules after each meal and never on an empty stomach. It should be noted too that the capsules used in this study were coated to delay the release of their contents to minimize side effects (27). Although participants in this study were not blinded with regard to whether they were supplemented with EPA, we believe that blinding would have had a minimal effect on the results because the dosages were far in excess of what is commercially available. In addition, the primary endpoints were biological rather than clinical effects of EPA.

After 6 mo of dietary supplementation with EPA, we observed important changes in the fatty acid composition of Barrett's mucosal epithelium. In particular, there were significant increases in both EPA and docosapentaenoic acid concentrations in the n–3 group compared with the control group. Previous observations have noted that long-term supplementation with fish oils high in n–3 fatty acids can lead to a progressive decrease in arachidonic acid concentrations in gut mucosa and that this decrease might be the result of competitive incorporation of EPA (23, 28). In the present study, arachidonic acid concentrations decreased with supplementation, but not significantly so (P = 0.14). However, most previous work has been performed with colonic or rectal mucosa, which characteristically has a higher rate of turnover than does the esophagus and may incorporate and metabolize fatty acids at higher rates. Moreover, at baseline, the ratio of C20:4 to C20:5 was 8, so that arachidonate concentrations far exceeded those of eicosapentaenoate. Therefore, relative increases in EPA concentrations would be expected to have minimal effects on the large quantities of arachidonic acid already present.

After supplementation with EPA, we observed a significant decrease in COX-2 protein concentrations in Barrett's tissue compared with the control group. COX-2 is known to be a 72-kDa protein, but, in the present study, only a 62-kDa band was detected in most subjects—the exceptions being 2 patients with low-grade dysplasia in whom a 72-kDa band was observed in addition to an unusually strong 62-kDa band. The 62-kDa band was observed in previous work (22) and is thought to be either a degraded or cleaved form of the enzyme or, alternatively, COX-2 in a different glycosylation state. It is interesting that the only 2 subjects with evidence of a strong 72-kDa band also had dysplasia. Generally speaking, COX-2 is expressed at relatively low levels in Barrett's esophagus, but expression levels may increase considerably during the change from columnar metaplasia to dysplasia (13, 29).

COX-2 mRNA concentrations were found to be low in both Barrett's and squamous tissues, and there was no evidence of any change in COX-2 gene expression as a result of 6 mo of EPA supplementation. The significance of COX-2 gene expression levels in relation to the regulation of the COX-2 protein is questionable, however, because it is now known that COX-2 expression can be regulated through transcriptional and posttranscriptional mechanisms. Indeed, Dixon et al (30) identified an AU-rich element within the 3' untranslated region of COX-2 mRNA that can control both mRNA decay and protein translation. They have hypothesized that posttranscriptional regulation is modulated by AU-rich element binding factors, which form complexes with mRNA and can then act as a translational silencer. The results of the present study are consistent with this hypothesis, ie, there was a significant decrease in COX-2 protein concentrations after EPA supplementation as measured by immunoblotting, but no concomitant effect on COX-2 gene transcription.

Despite the significant decrease in COX-2 enzyme concentrations, we found that EPA supplementation had no effect on concentrations of PGE₂ in Barrett's mucosa. Grataroli et al (31) previously showed a significant reduction in PGE₂ production in gastric mucosa in rats fed salmon oil. However, they noted that there was an increase in phospholipase A₂ activity and concluded that the decrease in PGE₂ might be the result of substitution of arachidonic acid by n–3 PUFAs rather than any direct mitigating effect on phospholipase A₂ activity. In the present study we observed no significant alteration in arachidonic acid concentrations as a result of EPA supplementation, and concentrations of both PGE₂ and LTB₄ were also unaltered. Our results are also consistent with previous studies on neutrophils in human subjects, which have shown significant increases in neutrophil EPA content in response to fish-oil supplementation, but no evidence of suppression of LTB₄ synthesis (32, 33).

Previous studies have shown a suppression of crypt cell proliferation in human rectal tissue after prolonged supplementation with fish oil (34, 35), and this has been ascribed to the inhibitory effects of n–3 PUFAs on the COX-2 pathway and the consequent reduction in proinflammatory eicosanoids including PGE₂. We found generally greater Ki-67 labeling in the areas of intestinal metaplasia than in gastric-type and squamous areas, and this finding is consistent with previous studies on cell proliferation in esophageal metaplasia (36, 37). However, there was no clear evidence that EPA had any influence on proliferative levels, despite the observed reduction in COX-2. It should be noted, however, that, unlike the colon, where cell proliferation is higher and spatially localized in morphologically homogeneous crypts, an accurate measurement of proliferative activity in vivo in Barrett's biopsy samples is difficult to achieve. Cell proliferation in both squamous mucosa and Barrett's esophagus is relatively slow, and Barrett's mucosa is itself a complex aggregate of tissue subtypes, all of which have different levels of proliferation.

In conclusion, the present study showed that regular, relatively small doses of EPA result in significant incorporation of this metabolically active n–3 PUFA into Barrett's mucosa and a concomitant reduction in COX-2 protein concentrations. Many published studies support the possible chemopreventive effect of COX-2 inhibition in Barrett's tissue. For instance, a human intervention study showed that the COX-2 inhibitor Rofecoxib can reduce both COX-2 expression and cell proliferation in Barrett's esophagus (38). Furthermore, regular aspirin and other nonsteroidal anti-inflammatory drugs have also been shown to reduce the progression of Barrett's esophagus to esophageal adenocarcinoma (39). We were unable to show a significant decrease in epithelial cell proliferation, but the reduction in COX-2 protein concentrations that we observed suggests that n–3 PUFAs, perhaps at higher concentrations of supplementation, might have a role to play in chemoprevention in patients with dysplasia, in whom there is a higher baseline proliferation index with increased COX-2 expression. Moreover, the long-term maintenance of relatively high concentrations of n–3 PUFAs in the esophageal mucosa, which might be expected in populations with a habitually high consumption of oily fish, may inhibit the expression of COX-2 and thereby inhibit either the development or the progression of Barrett's esophagus in the community. Further studies on the potential chemopreventive effects of marine oils therefore seem warranted.

	ACKNOWLEDGMENTS

 
The authors' responsibilities were as follows—SPM (Principal Investigator): data collection, data analysis, and writing of manuscript; APB: data collection and analysis; JC: endoscopies, sample collection, and advice; VS: histopathology and immunohistochemistry of samples; EKL: study design, sample analysis, and writing of manuscript; ITJ: study design, sample analysis, and writing of manuscript; and MR (Chief Investigator): study design and analysis of results. None of the authors had any financial or personal relationships with SLA Pharma or any other conflicts of interest.

	REFERENCES

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Powell J, McConkey CC, Gillison EW, Spychal RT. Continuing rising trend in oesophageal adenocarcinoma. Int J Cancer 2002;102:422–7.[Medline]
2. Blot WJ, Devesa SS, Kneller RW, Fraumeni JF Jr. Rising incidence of adenocarcinoma of the esophagus and gastric cardia. JAMA 1991;265:1287–9.[Abstract/Free Full Text]
3. Jankowski JA, Perry I, Harrison RF. Gastro-oesophageal cancer: death at the junction. BMJ 2000;321:463–4.[Free Full Text]
4. Heath EI, Canto MI, Wu TT, et al. Chemoprevention for Barrett's esophagus trial. Design and outcome measures. Dis Esophagus 2003;16:177–86.[Medline]
5. Jankowski J, Sharma P. Review article: approaches to Barrett's oesophagus treatment-the role of proton pump inhibitors and other interventions. Aliment Pharmacol Ther 2004;19(suppl)1:54–9.
6. Naef AP, Savary M, Ozzello L. Columnar-lined lower esophagus: an acquired lesion with malignant predisposition. Report on 140 cases of Barrett's esophagus with 12 adenocarcinomas. J Thorac Cardiovasc Surg 1975;70:826–35.[Abstract]
7. Caygill CP, Watson A, Reed PI, Hill MJ. Characteristics and regional variations of patients with Barrett's oesophagus in the UK. Eur J Gastroenterol Hepatol 2003;15:1217–22.[Medline]
8. Johnson IT. New approaches to the role of diet in the prevention of cancers of the alimentary tract. Mutat Res 2004;551:9–28.[Medline]
9. Caygill CP, Charlett A, Hill MJ. Fat, fish, fish oil and cancer. Br J Cancer 1996;74:159–64.[Medline]
10. Bougnoux P. n–3 Polyunsaturated fatty acids and cancer. Curr Opin Clin Nutr Metab Care 1999;2:121–6.[Medline]
11. Chen H, Ward MH, Graubard BI, et al. Dietary patterns and adenocarcinoma of the esophagus and distal stomach. Am J Clin Nutr 2002;75:137–44.[Abstract/Free Full Text]
12. Singh J, Hamid R, Reddy BS. Dietary fat and colon cancer: modulation of cyclooxygenase-2 by types and amount of dietary fat during the postinitiation stage of colon carcinogenesis. Cancer Res 1997;57:3465–70.[Free Full Text]
13. Wilson KT, Fu S, Ramanujam KS, Meltzer SJ. Increased expression of inducible nitric oxide synthase and cyclooxygenase-2 in Barrett's esophagus and associated adenocarcinomas. Cancer Res 1998;58:2929–34.[Abstract/Free Full Text]
14. Liu XH, Yao S, Kirschenbaum A, Levine AC. NS398, a selective cyclooxygenase-2 inhibitor, induces apoptosis and down-regulates bcl-2 expression in LNCaP cells. Cancer Res 1998;58:4245–9.[Abstract/Free Full Text]
15. Tsujii M, Kawano S, Tsuji S, Sawaoka H, Hori M, DuBois RN. Cyclooxygenase regulates angiogenesis induced by colon cancer cells. Cell 1998;93:705–16.[Medline]
16. Rao CV, Reddy BS. Modulating effect of amount and types of dietary fat on ornithine decarboxylase, tyrosine protein kinase and prostaglandins production during colon carcinogenesis in male F344 rats. Carcinogenesis 1993;14:1327–33.[Abstract/Free Full Text]
17. Tsujii M, Kawano S, DuBois RN. Cyclooxygenase-2 expression in human colon cancer cells increases metastatic potential. Proc Natl Acad Sci U S A 1997;94:3336–40.[Abstract/Free Full Text]
18. Tsujii M, DuBois RN. Alterations in cellular adhesion and apoptosis in epithelial cells overexpressing prostaglandin endoperoxide synthase 2. Cell 1995;83:493–501.[Medline]
19. Gupta S, Srivastava M, Ahmad N, Sakamoto K, Bostwick DG, Mukhtar H. Lipoxygenase-5 is overexpressed in prostate adenocarcinoma. Cancer 2001;91:737–43.[Medline]
20. Nielsen CK, Ohd JF, Wikstrom K, et al. The leukotriene receptor CysLT1 and 5-lipoxygenase are upregulated in colon cancer. Adv Exp Med Biol 2003;525:201–4.[Medline]
21. Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Physiol 1959;37:911–7.[Medline]
22. Bishop-Bailey D, Pepper JR, Larkin SW, Mitchell JA. Differential induction of cyclooxygenase-2 in human arterial and venous smooth muscle: role of endogenous prostanoids. Arterioscler Thromb Vasc Biol 1998;18:1655–61.[Abstract/Free Full Text]
23. Hillier K, Jewell R, Dorrell L, Smith CL. Incorporation of fatty acids from fish oil and olive oil into colonic mucosal lipids and effects upon eicosanoid synthesis in inflammatory bowel disease. Gut 1991;32:1151–5.[Abstract/Free Full Text]
24. Gee JM, Watson M, Matthew JA, et al. Consumption of fish oil leads to prompt incorporation of eicosapentaenoic acid into colonic mucosa of patients prior to surgery for colorectal cancer, but has no detectable effect on epithelial cytokinetics. J Nutr 1999;129:1862–5.[Abstract/Free Full Text]
25. el Boustani S, Colette C, Monnier L, Descomps B, Crastes de Paulet A, Mendy F. Enteral absorption in man of eicosapentaenoic acid in different chemical forms. Lipids 1987;22:711–4.[Medline]
26. Beckermann B, Beneke M, Seitz I. [Comparative bioavailability of eicosapentaenoic acid and docasahexaenoic acid from triglycerides, free fatty acids and ethyl esters in volunteers. ] Arzneimittelforschung 1990;40:700–4 (in German).[Medline]
27. Belluzzi A, Brignola C, Campieri M, et al. Effects of new fish oil derivative on fatty acid phospholipid-membrane pattern in a group of Crohn's disease patients. Dig Dis Sci 1994;39:2589–94.[Medline]
28. Anti M, Marra G, Armelao F, et al. Effect of omega-3 fatty acids on rectal mucosal cell proliferation in subjects at risk for colon cancer. Gastroenterology 1992;103:883–91.[Medline]
29. Shirvani VN, Ouatu-Lascar R, Kaur BS, Omary MB, Triadafilopoulos G. Cyclooxygenase 2 expression in Barrett's esophagus and adenocarcinoma: ex vivo induction by bile salts and acid exposure. Gastroenterology 2000;118:487–96.[Medline]
30. Dixon DA, Balch GC, Kedersha N, et al. Regulation of cyclooxygenase-2 expression by the translational silencer TIA-1. J Exp Med 2003;198:475–81.[Abstract/Free Full Text]
31. Grataroli R, Leonardi J, Charbonnier M, Lafont R, Lafont H, Nalbone G. Effects of dietary corn oil and salmon oil on lipids and prostaglandin E2 in rat gastric mucosa. Lipids 1988;23:666–70.[Medline]
32. Prescott SM, Zimmerman GA, Morrison AR. The effects of a diet rich in fish oil on human neutrophils: identification of leukotriene B5 as a metabolite. Prostaglandins 1985;30:209–27.[Medline]
33. Strasser T, Fischer S, Weber PC. Leukotriene B5 is formed in human neutrophils after dietary supplementation with eicosapentaenoic acid. Proc Natl Acad Sci U S A 1985;82:1540–3.[Abstract/Free Full Text]
34. Bartram HP, Gostner A, Scheppach W, et al. Effects of fish oil on rectal cell proliferation, mucosal fatty acids, and prostaglandin E2 release in healthy subjects. Gastroenterology 1993;105:1317–22.[Medline]
35. Courtney ED, Matthews S, Finlayson C, et al. Eicosapentaenoic acid (EPA) reduces crypt cell proliferation and increases apoptosis in normal colonic mucosa in subjects with a history of colorectal adenomas. Int J Cardiol 2007;22:765–76.
36. Jankowski J, McMenemin R, Yu C, Hopwood D, Wormsley KG. Proliferating cell nuclear antigen in oesophageal diseases; correlation with transforming growth factor alpha expression. Gut 1992;33:587–91.[Abstract/Free Full Text]
37. Gray MR, Hall PA, Nash J, Ansari B, Lane DP, Kingsnorth AN. Epithelial proliferation in Barrett's esophagus by proliferating cell nuclear antigen immunolocalization. Gastroenterology 1992;103:1769–76.[Medline]
38. Kaur BS, Khamnehei N, Iravani M, Namburu SS, Lin O, Triadafilopoulos G. Rofecoxib inhibits cyclooxygenase 2 expression and activity and reduces cell proliferation in Barrett's esophagus. Gastroenterology 2002;123:60–7.[Medline]
39. Gupta RA, DuBois RN. Cyclooxygenase-2 inhibitor therapy for the prevention of esophageal adenocarcinoma in Barrett's esophagus. J Natl Cancer Inst 2002;94:406–7.[Free Full Text]

This article has been cited by other articles:

				P. Wertz Essential fatty acids and dietary stress Toxicology and Industrial Health, May 1, 2009; 25(4-5): 279 - 283. [Abstract] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mehta, S. P Articles by Rhodes, M. Search for Related Content PubMed PubMed Citation Articles by Mehta, S. P Articles by Rhodes, M. Agricola Articles by Mehta, S. P Articles by Rhodes, M.