Status Occur Early in the Colorectal Adenoma-Carcinoma Sequence 255
Total Page:16
File Type:pdf, Size:1020Kb
254 Gut 1996; 38: 254-259 Changes of the mucosal n3 and n6 fatty acid status occur early in the colorectal adenoma-carcinoma sequence Gut: first published as 10.1136/gut.38.2.254 on 1 February 1996. Downloaded from F Fernandez-Bafiares, M Esteve, E Navarro, E Cabre, J Boix, A Abad-Lacruz, J Klaassen, R Planas, P Humbert, C Pastor, M A Gassull Abstract support the idea that dietary n6-PUFAs can Despite data favouring a role ofdietary fat promote carcinogenesis. Several animal studies in colonic carcinogenesis, no study has have shown that diets containing high propor- focused on tissue n3 and n6 fatty acid (FA) tions of n6-PUFAs increase the incidence of status in human colon adenoma-carci- chemically induced colonic tumours in rats,4A6 noma sequence. Thus, FA profile was whereas diets rich in n3-PUFAs tend to have measured in plasma phospholipids of an antipromotional effect.4 6-8 Fish oil seems patients with colorectal cancer (n=22), to influence early stages of colon experimental sporadic adenoma (n=27), and normal carcinogenesis, with a decrease in proliferative colon (n= 12) (control group). Addi- indices and focal areas of dysplasia in tionally, mucosal FAs were assessed in azoxymethane treated rats.9 These studies both diseased and normal mucosa of support that the fatty acid composition of the cancer (n= 15) and adenoma (n=21) diet is more important than its overall fat patients, and from normal mucosa of content in terms of colon cancer risk. Recent controls (n=8). There were no differences data also suggest a role for PUFAs on human in FA profile of both plasma phospho- colon carcinogenesis. Firstly, studies on the lipids and normal mucosa, between ade- mucosal fatty acid content in patients with noma and control patients. There were colorectal carcinoma have shown increased considerable differences, however, in FAs concentrations of both arachidonic (C20:4n6) between diseased and paired normal and docosahexaenoic (DHA; C22:6n3) acids mucosa of adenoma patients, with in tumoral mucosa.10 1 Nevertheless, these increases of linoleic (p=0.02), dihomo- abnormalities could be a metabolic con- gammalinolenic (p=0.014), and eico- sequence of the established cancer rather than sapentaenoic (p=0.012) acids, and a causative factor for colon carcinogenesis. http://gut.bmj.com/ decreases of a linolenic (p=0.001) and Secondly, n3-PUFA supplementation in arachidonic (p=0.02) acids in diseased healthy subjects,12 and in patients with mucosa. A stepwise reduction of eico- sporadic adenomatous polyps'3 14 reduces the sapentaenoic acid concentrations in rate of colonic epithelial cell proliferation, thus diseased mucosa from benign adenoma to decreasing the risk for colon cancer.15 the most advanced colon cancer was seen In this study, a different approach to assess (p=0.009). Cancer patients showed lower the possible role of PUFAs in the colorectal on October 2, 2021 by guest. Protected copyright. a linolenate (p=0.002) and higher carcinogenesis processes was used. We investi- dihomogammalinolenate (p=0.003) in gated the changes of the mucosal fatty acid diseased than in paired normal mucosa. content in the human adenoma-carcinoma In conclusion changes in tissue n3 and n6 sequence, using biopsy specimens from Departments of Gastroenterology and FA status might participate in the early patients with adenomas and carcinomas of the Biochemistry, phases of the human colorectal carcino- colon. Measurements were performed in both Research Unit, genesis. diseased and adjacent normal mucosa. The Hospital Universitari Germans Trias i Pujol, (Gut 1996; 38: 254-259) study of colorectal adenomas avoids the Badalona, Catalunya, possible influence on PUFA metabolism of an Spain Keywords: fatty acid composition, colorectal adenoma, established cancer. Fatty acid profile in plasma F Femrndez-Baniares colorectal carcinoma, fish oil, colon carcinogenesis. M Esteve phospholipids was also evaluated. E Navarro E Cabre J Boix Several studies have Methods A Abad-Lacruz epidemiological suggested J Klaassen that populations that consume high fat diets R Planas have an increased risk of colon cancer.1 The Inclusion criteria P Humbert intake of saturated fat seems to account for this Patients who were found to have either C Pastor M A Gassull association, whereas an association to poly- sporadic polyps (with size > 1 cm) or cancer of unsaturated fat intake (linoleate) has not been the colon during total fibreoptic colonoscopy Correspondence to: Dr M A Gassull, seen.2 The low colon cancer rate in Alaskan were selected if they were not taking drugs Department of Eskimos, however, has been related to the high (such as hypolipemiant, oral contracep- Gastroenterology, Hospital Universitari Germans Trias i consumption of fish products rich in n3 tives, non-steroidal anti-inflammatory drugs Pujol, Carretera del Canyet polyunsaturated fatty acids (n3-PUFAs),3 sug- (NSAIDs) or platelet antiaggregants) or s/n 08916 Badalona, Catalunya, Spain. gesting that fish oil may be a protective factor vitamin, mineral or fish oil supplements. Accepted for publication in colon carcinogenesis. Likewise, only omnivore subjects taking a 9 August 1995 On the other hand, experimental studies Western type diet were included. In addition, Changes of the mucosal n3 and n6fatty acid status occur early in the colorectal adenoma-carcinoma sequence 255 patients with associated acute or chronic specimens were obtained from suspected carci- diseases that could disturb plasma fatty acid noma lesions for histological examination. pattern, or history of previous colorectal ade- Venous blood was obtained the day after nomatous polyps or carcinoma were excluded. colonoscopy, before any therapeutic procedure The same inclusion criteria were required for was performed, after a 12 hour overnight fast patients with normal colon. period. Plasma was separated by centrifugation Gut: first published as 10.1136/gut.38.2.254 on 1 February 1996. Downloaded from All patients gave informed consent to the at 3000Xg for 10 minutes and was immedi- study, which was approved by the Hospital ately stored at -20°C. Research and Ethical Committee. Fatty acid assay Patients The plasma lipid extraction procedure Forty nine patients with colonic adenoma or has been previously described.'6 Plasma adenocarcinoma were studied. Twenty seven phospholipids were separated by thin layer patients (22 men, five women; mean (SEM) chromatography on silica gel G-60 (Merck, age 61.9 (2) years) had sporadic adenomatous Darmstadt, Germany) by using the solvent polyps (mean (SEM) size, 2.01 (0 20) cm). system described by Skipski and Barclay.'7 Most of them were tubular adenomas (n= 12) Direct transesterification of fatty acids (FAs) and tubulovillous adenomas (n= 14), and only was immediately carried out in methanol-ben- one was a villous adenoma. Eight adenomas zene 4:1 (v/v) with acetyl chloride according to had severe cellular atypia considered as in situ the procedure of Lepage and Roy.'8 The carcinoma. The sites ofthe polyps were rectum benzene extract was evaporated under a stream (n=9), sigmoid (n=11), descending (n=6), of nitrogen at 40°C to complete dryness. The and ascending colon (n= 1). residue was dissolved in 100 ,u of benzene and Twenty two patients (16 men, six women; a 1 pAl aliquot was injected in the chromato- mean (SEM) age 61 (2.7) years) had colorectal graph. FA methyl esters were quantified by adenocarcinomas. Cancer staging according to gas-liquid chromatography in a Perkin-Elmer Dukes's was: Dukes's B, 12; and Dukes's Autosystem chromatograph (Perkin-Elmer, C-D, 10. The location of the cancers were Norwalk, CT) using a 30 m capillary column, rectum (n= 10), sigmoid colon (n=6), and 0.25 mm internal diameter, impregnated with caecum plus ascending colon (n = 6). SP-2330 as stationary phase. The identifica- In addition, 12 patients (six men, six tion and quantification of FA methyl esters women; mean (SEM) age 57 (3.6) years) with were made by comparison with an external normal colon during a routine total fibreoptic standard (Sigma Chemical, St Louis, MO). colonoscopy to rule out colonic disease (haem- FAs from C16:0 to C24:0 were measured, orrhoidal bleeding, abdominal pain, irritable unidentified peaks accounting for <05°/o of bowel syndrome) were also studied, and con- the total FA. Saturated fatty acids (SFAs) were http://gut.bmj.com/ stituted the control group. There were no expressed as the sum of C 16:0 (palmitic), significant differences in sex and age between C18:0 (stearic), and C24:0 (lignoceric); and the three clinical groups. monounsaturated fatty acids (MUFAs) as the sum of C16:1n7 (palmitoleic), C18:1n9 (oleic), C20:1n9 (gadoleic), C22:1n9 (erucic), Biopsy specimens and blood collection and C24: 1n9 (nervonic). FAs in plasma phos- Biopsy specimens were taken with endoscopic pholipids were expressed as a percentage of the on October 2, 2021 by guest. Protected copyright. forceps. In patients with polyps or carcinoma, total FAs present. specimens were taken as paired samples, from Tissue samples (mean wet weight, 27 mg; the tumour and from normal looking mucosa range, 10 to 40 mg), were put in a 4:1 (v/v) at least 10 cm away from the lesion. If methanol-benzene solution and shaken for more than one polyp larger than 1 cm was about one minute in a vortex mixer. found, only the largest one was studied. In Afterwards they were homogenised by sonica- patients with normal colon, biopsy specimens tion in an ultrasound bath. Then, direct trans- were taken from the rectosigmoid colon. esterification was performed as described Immediately after removal, the biopsy speci- above. The procedure for FA assay did not mens were placed in cryovials, flash frozen in differ from that described for FA plasma phos- liquid nitrogen, and stored at -80°C. Polyps pholipids. Mucosal FAs were expressed as per- were removed by snare diathermy and retained centage of total FAs present.