Recent interest in natural estrogens found in plants has increased, particularly by women who are experiencing menopause. It has been speculated that plant-derived estrogens, or phytoestrogens, may be effective in preventing the negative symptoms associated with menopause and provide protection from cardiovascular disease and osteoporosis. In response to this consumer interest, many companies have developed dietary supplements of phytoestrogens aimed specifically at women. The most common food source of phytoestrogens is soybean. Soy contains a relatively high concentration of genistein and daidzein, which are estrogenic compounds called isoflavones. Whereas soy is a main component of the diet in Asia cultures, it is not typically consumed in the US. Acceptance of soy foods by American consumers is very low due to perceived flavor dislikes. Hence, manufacturers have separated the phytoestrogens from soybeans to allow consumers the potential benefit of soy without the need to make acceptable food products containing the whole soybean. The phytoestrogens in soy are associated with the protein fraction of the soybean. and there have been no rigorous studies to determine the bioavailability of phytoestrogens when consumed in food matrices other than the soybean.

To address this topic we measured the absorption and excretion of isoflavones in women fed muffins made with a soy isoflavone supplement (208 mg NovaSoy, 80 mg genistein, Archer Daniels Midland). The isoflavone supplement was mixed into muffins made with either corn meal or soy protein. The soy protein had been previously processed to remove the naturally occurring isoflavones. Women (ages 18-23) volunteers, recruited from the University of Missouri community, consumed one type of muffin for 5 days. Saliva and urine were collected on day 0, and on day 5 saliva and urine were collected 6 hours after muffin consumption. Each volunteer was then one month later, corresponding to the same phase of their menstrual cycle, with the other muffin. The volunteers were asked to record any physiological, emotional or physical effects they experienced during the 5 days of muffin consumption.

A novel aspect of this study was the development of saliva analysis of the isoflavones. In order to assess bioavailability, or the amount of an ingested compound that is digested, absorbed and transported into the body, it is necessary to quantify circulating concentrations of the compound. Urine provides a marker of the excreted compounds, however does not allow quantification of those compounds that are being utilized by tissues. Blood does provide information about circulating compounds, however blood sampling imposes limitations on experimental protocols. It is well-documented that saliva reflects blood concentrations of many compounds, in particular the steroid hormones. We therefore hypothesized that saliva would be a useful way and non-invasive way to assess bioavailability of the isoflavones in our subjects. Hence, we developed the analytical methodology to quantify the isoflavones in saliva, a technique that had not previously been available.

Prior to conducting the main experiment, we performed a pilot experiment to test the salivary genistein concentration over a period of 10 hours. A group of volunteers consumed corn meal muffins containing the isoflavone extract for 5 days. On day 5 the subjects collected saliva at 2 hour intervals beginning just before muffin consumption and ending 10 hours after muffin consumption. Saliva was collected while chewing on a piece of dental wax over a 5-10 minute period. Genistein, daidzein, glycitein, and equol were measured using coulometric array detection. As shown in the figure below, salivary genistein peaked 2 hours after consumption and decreased rapidly such that by 4 hours genistein was barely detectable in the saliva samples. The genistein measured in the saliva was the aglycone, meaning it was not conjugated with sugar molecules. Most of the genistein found in the soy supplement is conjugated with sugars, hence the salivary genistein may reflect the small amount of unconjugated genistein that is rapidly absorbed by the gastrointestinal tract. This has significant implications for bioavailability as manufactured isoflavone supplements are being developed that may contain more unconjugated isoflavones. Further studies are needed to determine the health effects of high unconjugated isoflavones in the circulation.

The goal of our main study was to determine if the isoflavone supplement would be differently absorbed when consumed in a muffin containing corn meal compared to soy protein. A non-statistically significant trend (p=0.1) for higher genistein in saliva afier consuming muffins made with corn meal compared to soy protein was observed (data not shown). The urinary excretion patterns for the isoflavones were similar to other reports. As shown in the figure below, daidzein was the most abundant isoflavone found in the urine. The figure shows isoflavone concentrations in urine samples obtained from the women on day 5 of muffin consumption, at the baseline (base) before consuming the muffin and at the post (6 hours after consuming the muffin). This is of interest because the ratio of genistein to daidzein in the isoflavone supplement is about 3:1, hence genistein appears to be utilized by the women differently than daidzein. It is possible that the genistein is less well absorbed, excreted more rapidly, metabolized or retained by the bodv more so than daidzein. The statistical analvsis of the data demonstrated that urinary daidzein was significantly greater (p=0.02) in women who consumed the corn meal muffin compared with soy protein. Similar, but non-significant, trends for higher levels of glycitein and genistein were observed with corn meal compared to soy. A metabolite of daidzein, equol, was also detected in the urine samples, although at much lower levels than daidzein. Mean equol excretion was low and not different among the groups.

Our results demnstrate two important observations. First, salivary collection may be used as a non-invasive, inexpensive way to assess bioavailability of soy isoflavones. Second, consumption of isoflavones in a food made with soy protein resulted in a different excretion pattern of isoflavones and metabolites conprared to when the isoflavones were consumed in a food made with corn meal. This suggests that the food matrix in which isoflavone supplements are provided will alter the amount of isoflavones that are digested, absorbed or metabolized. Further experiments are needed to define the specific food ingredients that affect the bioavailability of the isoflavones, as well as the health implications of the differential response to the supplements. This information is of importance to manufacturers of isoflavone supplements and to consumers who use these supplements.