- MRC Dunn Nutrition Unit
- Start date
- End date
- Phytoestrogens are metabolites of plants that can induce biological responses and
mimic or modulate the action of endogenous oestrogens. In previous studies, it has been
shown that these compounds can produce a biological effect in humans.1 However, results
from epidemiological studies have been ambigous and it is not known whether these compounds
are beneficial or detrimental to human health. As phytoestrogens are present in
most plant-based foods consumed, it is important to assess their biological activity and the
exposure to these compounds.
The aim of this project was to analyse the phytoestrogen content of foods commonly consumed in the UK and to investigate their biological activity – in particular with regard to breast and prostate cancer – in a case-control study in the EPIC Norfolk cohort.
- Development of a robust and reliable analytical method to quantify phytoestrogens in foods using 13C–labelled internal standards and LC/MS.
- Sampling and analysis of approximately 500 foods commonly consumed in the UK.
- Incorporating the phytoestrogen content determined in this study into the food composition database of EPIC Norfolk.
- Investigation of relationships between phytoestrogen intake and cases of breast and prostate cancer in 400 cases and 1600 matched controls in EPIC Norfolk.
- More information
- Key findings:
- A sensitive and robust LC/MS/MS method for the analysis of twelve phytoestrogens (Biochanin A, Daidzein, Formononetin, Glycitein, Genistein, Matairesinol, Secoisolariciresinol, Shonanin, Coumestrol, Enterodiol, Equol, Enterolactone) in foods has been developed. This method uses multiple liquid extraction steps to achieve an extraction yield of up to 90%. Phytoestrogens were released from their carbohydrate conjugates using enzymatic hydrolysis, in particular cellulase, glucosidase and glucuronidase.The precision of this method was better than 15%; the limit of quantification 1.5 μg/100 g dry weight. To ensure the quality of the method, a quality control sample, consisting of equal amount of celery, red cabbage and orange, was analysed with each batch.
- 509 different foods were analysed for their phytoestrogen content. For most foods, five different samples from different sources were purchased, prepared and pooled prior to analysis. Each sample was analysed in triplicate. For more than 60 foods, the effect of cooking on total phytoestrogen content was also investigated.
- The phytoestrogen content in most foods analysed was well below 100 μg/100 g wet weight with a median content of 21 μg/100 g; only 5% of all foods contained more than 700 μg/100 g phytoestrogens. Lignans are the main contributors to total phytoestrogen content of most foods. The highest amount of phytoestrogens (mainly isoflavones) was found in soya flour (125,000 μg/100 g); other soya products contained considerably less phytoestrogens (approximately 20,000 μg/100 g). High amounts of isoflavones were also found in vegetarian pâte (1500 μg/100 g; chickpea based). Linseed (19,000 μg/100 g) and brazil nuts (900 μg/100 g) were the foods with the highest lignan content and beansprouts contained the highest amount of coumestrol (361 μg/100 g; total content: 798 μg/100 g). Soya-based infant formula contained 19,000 μg/100 g phytoestrogens, 300-times more than milk based formula (59 μg/100 g).
- The highest amount of isoflavones (median: 51 μg/100 g) was found in legumes (Fabaceae) whereas Alliaceae such as garlic, leek and onions and Apiaceae such as carrots, fennel and parsnips contained the highest amount of lignans (median: 65 μg/100 g)
- Cooking reduced the phytoestrogen content significantly (n=61, p<0.05, Wilcoxon signed rank test), presumably because phytoestrogens leach into the cooking water which was discarded. The phytoestrogen content was not affected by stewing, suggesting that these compounds are stable during preparation.
- For the first time, the phytoestrogen content of foods of animal origin other than milk was determined in this study. Meat, fish, seafood, eggs and dairy products all contained phytoestrogens above the limit of quantification with an average content of 20 μg/100 g. Enterolignans and equol were the main types of phytoestrogens in cheese and a major contributor to total phytoestrogens in eggs and dairy products except for butter. Meat contained only small amounts of enterolignans and equol and fish and seafood did not contain any, but both contained isoflavones and lignans.
- The variability of phytoestrogens was investigated in nine foods from different sources. The total range was up to 3-fold with a coefficient of variation of 39% for isoflavones and 33% for lignans.
- Phytoestrogen intake was similar in cases and controls for breast and prostate cancer. Breast cancer risk was not strongly associated with phytoestrogen intake; a weak association was found for secoisolariciresinol in the age adjusted model, but this association was attenuated in the multivariate model.
- Prostate cancer risk was positively associated with intake of equol, enterolactone and
total enterolignans, but this association became marginal after adjustment for multiple
Find more about this project and other FSA food safety-related projects at the Food Standards Agency Research webpage.
- Funding Source
- Food Standards Agency
- Project number
- Food Preparation and Handling
- Prevention and Control