Phytoestrogen content of cereals and cereal-based foods consumed in the UK.

Dietary phytoestrogens may be involved in the occurrence of chronic diseases. Reliable information on the phytoestrogen content in foods is required to assess dietary exposure and disease risk in epidemiological studies. However, there is little information on isoflavone, lignan, and coumestrol content of cereals and cereal-based foods, leading to an underestimation of intake. This is the first study of phytoestrogens (isoflavones: biochanin A, daidzein, formononetin, genistein, glycitein; lignans: matairesinol, secoisplariciresinol; coumestrol) in a comprehensive selection of 101 cereals and cereal-based foods-including breads, breakfast cereals, biscuits, pasta and rice-consumed in the UK using a sensitive LCMS technique with 13C-labelled internal standards. Phytoestrogens were detected in all foods analyzed; bread contained the highest amount of phytoestrogens-many as isoflavones-with an average content of 375 +/- 67 microg/100 g wet weight (excluding soya-linseed bread with 12,000 microg/100 g). Most other foods contained less than 100 microg/100 g, many as lignans. Our study shows that all foods analyzed contained phytoestrogens, with the highest amount found in breads, making them one of the main sources of dietary phytoestrogens in the UK. These results will allow a more accurate estimation of exposure to dietary phytoestrogens.

Phytoestrogen content of foods of animal origin: dairy products, eggs, meat, fish, and seafood.

Dietary phytoestrogens may be involved in the occurrence of chronic diseases. Reliable information on the phytoestrogen content in foods is required to assess dietary exposure and disease risk in epidemiological studies. However, existing analyses have focused on only one class of these compounds in plant-based foods, and there is only little information on foods of animal origin, leading to an underestimation of intake. This is the first comprehensive study of phytoestrogen content in animal food. We have determined the phytoestrogen content (isoflavones: biochanin A, daidzein, formononetin, genistein, and glycitein; lignans: secoisolariciresinol and matairesinol; coumestrol; equol; enterolactone; and enterodiol) in 115 foods of animal origin (including milk and milk-products, eggs, meat, fish, and seafood) and vegetarian substitutes using liquid chromatography-mass spectrometry (LC-MS) with (13)C-labeled internal standards. Phytoestrogens were detected in all foods analyzed; the average content was 20 microg/100 g of wet weight (isoflavones, 6 microg/100 g; lignans, 6 microg/100 g; equol, 3 microg/100 g; and enterolignans, 6 microg/100 g). In infant soy formula, 19 221 microg/100 g phytoestrogens were detected (compared to 59 microg/100 g in non-soy formula). Our study shows that all foods analyzed contained phytoestrogens and most foods (except for fish, seafood, and butter) contained mammalian phytoestrogens (enterolignans and equol). This is the first comprehensive study of phytoestrogen content of foods of animal origin and will allow for a more accurate estimation of exposure to dietary phytoestrogens.

Phytoestrogen content of beverages, nuts, seeds, and oils.

Phytoestrogens are secondary plant metabolites that have received increasing attention for their bioactivity, in particular due to their structural and functional similarity to 17beta-estradiol. Although urinary and plasma phytoestrogens can be used as biomarkers for dietary intake, this is often not possible in large epidemiological studies or in the assessment of general exposure in free-living individuals. Accurate information about dietary phytoestrogens is therefore important, but there are very limited data concerning food contents. In this study was analyzed a comprehensive selection of tea, coffee, alcoholic beverages, nuts, seeds, and oils for their phytoestrogen content using a newly developed sensitive method based on LC-MS incorporating (13)C 3-labeled standards. Phytoestrogens were detected in all foods analyzed, although the contents in gin and bitter (beer) were below the limit of quantification (1.5 microg/100 g). Lignans were the main type of phytoestrogens detected. Tea and coffee contained up to 20 microg/100 g phytoestrogens and beer (except bitter) contained up to 71 microg/100 g, mainly lignans. As these beverages are commonly consumed, they are a main source of dietary lignans. The results published here will contribute to databases of dietary phytoestrogen content and allow a more accurate determination of phytoestrogen exposure in free-living individuals.

Detection and quantification of sucrose as dietary biomarker using gas chromatography and liquid chromatography with mass spectrometry.

Several epidemiological studies suggest a link between the intake of refined sugars and an increased risk for colorectal, breast, pancreatic and endometrial cancer. However, other studies failed to confirm these conclusions and the reason for this may be the ambiguity of dietary assessment methods - mainly self-reporting - employed. Sucrose is an established biomarker for sugars intake, allowing the objective assessment of dietary sucrose. So far, urinary excretion of sucrose was mainly determined using an enzyme assay. However, this method is time-consuming and labour-intensive. In this study, we present a mass spectrometric method for the determination of sucrose in urine using liquid chromatography with mass spectrometry (LC/MS) which can be used for large-scale epidemiological studies.

Urinary potassium is as reliable as urinary nitrogen for use as a recovery biomarker in dietary studies of free living individuals.

Twenty-four-hour urinary nitrogen (UN) is commonly used to validate dietary assessment methods. Potassium is more widespread in food than nitrogen, but the role of 24-h urinary potassium (UK) as a biomarker has been less studied and characterized. To investigate the performance of UK as a recovery biomarker compared with UN in subjects consuming their normal diet, 7 males and 6 females consumed known amounts of food based on their habitual varying diet (assessed beforehand from 4 consecutive 7-d food diaries) for 30 d in a metabolic suite. All daily urine samples and dietary duplicates were collected, and N and K were measured. Stool K was determined in 5-d pooled samples. Thirty-day mean analyzed K intake was 121.3 +/- 25.1 (mean +/- SD) mmol/d. Overall, 77 +/- 6.7% of K in the diet was excreted in urine and 18 +/- 5% in stool. Dietary K was correlated with UK (r = 0.89; P < 0.001). UN was 77.7 +/- 6.6% of N intake and was correlated with N in the diet (r = 0.87; P < 0.001). When 16 d of intake and 8 d of urine-collection measurements were randomly selected from the 30-d measurements, correlations were significant for both K (r = 0.86; P < 0.001) and N (r = 0.92; P < 0.001). The high correlations between UK and K in the diet show that UK is a reliable recovery biomarker for use in studies of dietary measurement error. Factors for use of urinary N as a recovery biomarker are also confirmed.

Urinary sucrose and fructose as biomarkers for sugar consumption.

The use of 24-hour urinary sucrose and fructose as potential biomarkers for sugars consumption was investigated in two studies of 21 healthy participants living in a volunteer suite where dietary intake was known and all specimens collected. The dose-response was assessed in 12 males using a randomized crossover design of three diets containing constant levels of 63, 143, and 264 g of sugars for 10 days each. Both sugars and sucrose intake were significantly correlated with the sum of sucrose and fructose concentration in urine (0.888; P < 0.001). To assess effects with volunteers consuming their habitual varying diets, seven males and six females were fed their usual diet (assessed beforehand from four consecutive self-completed 7-day food diaries) for 30 days under controlled conditions in the volunteer suite. The mean (+/-SD) calculated total sugars intake was 202 +/- 69 g/d, 41% from sucrose. Mean (+/-SD) urinary sucrose and fructose were 36.6 +/- 16.6 and 61.8 +/- 61.3 mg/d, respectively. The sum of sucrose and fructose in urine was significantly correlated with sugars (0.841; P < 0.001) and sucrose intake (0.773; P = 0.002). In the regression, 200 g of sugars intake predicted approximately 100 mg of sucrose and fructose in urine. The correlation between individual means of randomized 16 days of sugars intake and 8 days of sugars excretion data (as used in validation studies) remained as high as that obtained with the means of 30-day measurements and the regression estimates were very similar. Twenty-four-hour urinary sucrose and fructose could be grouped into a new category of biomarkers, predictive biomarkers, that can be used in studies determining the structure of dietary measurement error in free living individuals and to relate sugars intake to disease risk.