View Report Details

The impact of folate and its interaction with riboflavin on biomarkers of colorectal cancer risk

Project Code: N12002

30/12/2004

University of Newcastle
Mathers, P ; Williams, D; Welfare, M;
University of Sheffield
Powers, H

Bowel cancer (CRC) is among the top three cancers affecting men and women in the UK with about 35,000 new cases annually. Whilst it is clear that diet has a major influence on risk of CRC, it has proved difficult, using conventional epidemiology, to identify which factors in food have beneficial and adverse effects and to define dietary choices which will minimize risk. Research in this area is handicapped by the lack of proven surrogate endpoints (biomarkers of risk) which have been demonstrated to be responsive to nutrients and other food components. This project was designed to develop and validate novel biomarkers of bowel cancer risk and to determine the extent to which they are modifiable by changes in nutrient intake.

Approach

Folate (a B vitamin) was chosen as the model nutrient for several reasons. i) There is good epidemiological evidence that those with higher intakes of folate have reduced incidence of CRC. ii) There are plausible mechanisms which could explain the adverse effects of low folate status on the development of bowel cancer. Low folate status is likely to increase damage to DNA (the fundamental cause of tumorigenesis) through e.g. aberrant DNA methylation and misincorporation of the nucleic acid base uracil (instead of thymine) into newly synthesized DNA. iii) These forms of DNA damage are potential surrogate endpoints which could be used in surveys or in dietary intervention studies. iv) The hypothesis that there are interactions between nutrients in impact on CRC risk could be addressed through study of the interaction of riboflavin (another B vitamin) with folate.

Three groups of volunteers attending gastroenterology clinics for investigation of bowel symptoms were consented and recruited to the study viz. a) individuals in whom no evidence of neoplasia was detected (described as ‘Normals’), b) patients with colonic adenomatous polyps (‘Polyps’) and c) patients with CRC (‘Cancers’). Given that the risk of developing colorectal tumours is enhanced in those with polyps and even more so in those with a previous cancer, this provided 3 groups of subjects with increasing bowel cancer risk i.e. Normals < Polyps < Cancers. At baseline, blood samples and rectal mucosal biopsies were collected and habitual diet was assessed using a validated Food Frequency Questionnaire (FFQ). The blood samples were used for assessment of folate and riboflavin status and for genotyping volunteers for MTHFR C677T status. MTHFR encodes a key folate metabolizing enzyme (which uses a riboflavin derivative as a cofactor) and there is evidence that C677T status may influence risk of bowel cancer. The mucosal biopsies were assayed for folate concentration and for 3 molecular biological markers i.e. genomic (global) DNA methylation, methylation of CpG islands in the promoter regions of selected genes and uracil misincorporation into DNA. Normals and Polyps were randomized to a double-blind, placebo-controlled intervention trial in which they received i) placebo, ii) 400 mg folic acid/d, iii) 1200 mg folic acid/d, or iv) 400 mg folic acid/d plus 5mg riboflavin/d for 50 days. A further set of blood samples and mucosal biopsies was collected at the end of the intervention period and the biochemical and molecular biological measurements were repeated.

Key findings

Of the potential volunteers contacted, 339 attended for screening. Data from 95 Normals and 102 Polyp patients who participated in the intervention study were included in the final analysis. Forty three CRC cases were recruited through this screen to which were added biopsy samples from 31 CRC patients recruited via another study. Mean age of recruits was 62 years (range 40-87 years) with 55% being male, and mean body mass indices (BMI) were 26 and 25 kg/m² for males and females respectively. At baseline, overall folate status was good and there was the expected inverse relationship (P=0.009) between concentrations of folate and of homocysteine (tHcy) in plasma. Mucosal folate concentration was quantified successfully and correlated strongly (P<0.001) with plasma folate concentration. There were no significant (P>0.05) differences between patient groups in the proportions of MTHFR C677T variants. Methylation of the P16 gene promoter region increased significantly with age (P<0.001) and, after correction for potential confounders) was higher in those with Polyps and Cancer (12%) than in Normals (9%) (P=0.06).

Fifty days of supplementation with folate produced marked and dose-dependent increases in plasma and mucosal folate concentrations (P<0.001) and there were concomitant falls in plasma tHcy concentration (P<0.001). Those individuals given riboflavin supplements had significant (P<0.001) increases in plasma riboflavin concentration and improvements in EGRAC (a functional marker of riboflavin status). No change in riboflavin status was detected in response to folate supplementation alone. Methylation specific PCR detected methylation in the promoter regions of the P16, HPP1, MLH1 and APC genes in 83-100% of volunteers. In contrast only 3-12% of volunteers had detectable methylation of the DNA repair gene MGMT. There were no significant (P>0.05) effects of the interventions on DNA methylation (genomic or gene specific) or on misincorporation of uracil into DNA.

Technical evaluation/interpretation

To our knowledge, this is the largest study which has investigated effects of any dietary factor on biochemical and/or molecular biological markers of CRC risk in the large bowel mucosa of Normal volunteers. We have quantified folate status and several markers of DNA damage in colorectal mucosa. Our interventions evoked the expected change in biomarkers of folate and riboflavin status but there were no significant changes in molecular biological markers, possibly because of the relatively short intervention period. The demonstration that it is possible to recruit such volunteers in numbers and to obtain colorectal mucosal biopsies before and after a nutritional intervention opens the way to future studies using similar protocols for investigation of impact of food components on bowel health. We have also shown that it is possible to collect mucosal cells using a ‘brush biopsy’ which is a more acceptable, safer (for volunteers) and less expensive approach than the conventional surgical biopsy.

Main File

Some of the files on this site may be in a format that your computer can't read. However, you can download Readers and Viewers for the following document types below: