Metabolism of isoflavones and lignans by the gut microflora: a study in germ-free and human flora associated rats

https://doi.org/10.1016/S0278-6915(02)00324-1Get rights and content

Abstract

We have investigated the metabolism of isoflavones and lignans in germ-free (GF) rats and rats associated with human faecal bacteria (human flora associated [HFA] rats), in order to provide unequivocal evidence for the role of the gut microflora in the absorption and metabolism of these phytoestrogens. Furthermore, we have investigated whether certain metabolic characteristics (high equol-producing and low equol-producing status) of human intestinal floras can be transferred to GF rats. Germ-free rats fed a soy-isoflavone containing diet excreted large quantities of daidzein and genistein in urine indicating that the gut microflora is not required for the absorption of isoflavones. The isoflavone metabolites equol, O-desmethylangolensin and the lignan enterolactone were not detectable in urine from the GF rats, but were present in HFA rat urine, indicating that they were products of gut microflora activity. Colonization of GF rats with a faecal flora from a human subject with the capacity to convert daidzein to equol, resulted in the rats excreting substantial amounts of the metabolite. In contrast, equol was undetectable in urine of HFA rats associated with a faecal flora from a low equol-producing subject. The results therefore show that the inability of some subjects to produce equol is a consequence of the lack of specific components of the gut microflora.

Introduction

The isoflavones genistein and daidzein are found almost exclusively in soy and soy products, usually in the form of glucosides (genistin and daidzin), and can reach concentrations up to 1 mg/g wet weight. (Coward et al., 1993, Wang & Murphy, 1994, Reinli & Block, 1996, Mazur & Adlercreutz, 1998). The lignans are more widely distributed and are found in cereals, especially rye, oats, a few fruits and in berries as well as in wine and tea and, in much larger quantities, in flaxseed. (Thompson et al., 1991) The major lignans are syringiresinol, pinoresinol, lariciresinol, isolariciresinol, matairesinol and secoisolariciresinol, usually found as glucosides and diglucosides (Axelson et al., 1982, Heinonen et al., 2001).

Both isoflavones and lignans have structural similarities to the human female hormone, 17β-estradiol, and other steroid hormones (Setchell and Adlercreutz, 1988) and can bind to the oestrogen receptor in many tissues and exert weak estrogenic activity (Setchell, 1995a, Setchell, 1995b, Wang & Kurzer, 1998). There is evidence that they have potential health benefits in humans particularly against hormone-dependent diseases such as breast and prostate cancer and osteoporosis.(Adlercreutz & Mazur, 1997, Bingham et al., 1998).

One of the important aspects of isoflavones and lignans is that they are metabolized, in some cases extensively, in the human body (Xu et al., 1995, Heinonen et al., 1999). There is considerable evidence that hydrolysis of the isoflavone glucosides is necessary for their absorption from the gut since the glucosides have not been detected in plasma (Andlauer et al., 2000, Setchell et al., in press). The site or sites of hydrolysis have not been fully elucidated, although it is known that glucosidase enzymes are produced in the small intestine and by members of the gut microflora (Day et al., 1998, Rowland et al., 2003). The aglycones undergo further metabolism, with genistein being converted to p-ethyl phenol and 4-hydroxyphenyl-2-propionic acid, while daidzein can be reduced to the isoflavan equol and O-desmethylangolensin (O-Dma) (Axelson et al., 1984, Kelly et al., 1993, Setchell, 1998, Coldham et al., 1999, Heinonen et al., 1999, Wiseman, 1999). There is considerable interindividual variation in production of certain metabolites, particularly equol and O-Dma. For example, only about 35% of subjects excrete substantial amounts of equol after consuming soy (Lampe et al., 1998; Rowland et al., 2000). The lignans also undergo extensive metabolism, with matairesinol being converted to enterolactone and secoisolariciresinol to enterodiol. Enterodiol may then be converted to enterolactone. These reactions have been shown to occur in in vitro incubations of lignans with human faeces (Borriello et al., 1985). Antibiotic treatment of human subjects decreases the excretion of the metabolites of the lignans and isoflavones, providing some evidence of the importance of the bacterial flora (Adlercreutz et al., 1986, Kilkkinen et al., 2002), but other evidence for the role of the microflora is scanty.

The role of gut bacteria in metabolism and disease is often complicated by the difference in gut flora between animals and man. In order to overcome these problems, human flora associated rats (HFA rats) have been developed (Mallett et al., 1987) whereby germ-free rats are colonized with a complete human faecal microflora. Such rats retain the bacteriological and metabolic characteristics of the human microflora (Rumney & Rowland, 1992, Hirayama et al., 1995).

In the present study we have investigated the metabolism of isoflavones and lignans in germ-free rats and rats associated with human faecal bacteria, in order to provide unequivocal evidence for the role of the gut microflora in the absorption and metabolism of phytoestrogens. Furthermore, we have also investigated whether the metabolic characteristics (high equol-producing and low equol-producing status) of human intestinal floras can be transferred to germ-free rats.

Section snippets

Identification of high and low equol-producing microfloras

Five human volunteers were asked to consume isoflavone-containing food products for at least 48 h. These consisted of a minimum of three items of phytoestrogen-rich foods (soya milk and textured vegetable protein products), per day, providing the volunteers with a minimum total isoflavone intake of 40 mg, measured as aglycones. The volunteers then collected a 24-h urine sample.

Ascorbic acid (approximately 0.1%, w/v) was added to the collection vessels as a preservative. The volume of the urine

Results

The five subjects that were screened for isoflavonoid and lignan excretion exhibited a wide range of equol and O-DMA values in urine (Table 2). Equol excretion ranged from 14 to 539 nmol/24 h and O-Dma excretion from 260 to 1653 nmol/24 h. Subject 2 was chosen as the low equol producer for the germ-free rat study. The amount of equol excreted by this subject (14 nmol/24 h) represented only 0.09% of the total isoflavonoid excretion. Subject 4 was chosen as the high equol producer; equol

Discussion

The interindividual variation in isoflavone metabolism reported by several authors (Lampe et al., 1998; Watanabe et al., 1998; Rowland et al., 2000) was evident in the present study, with only one out of five subjects excreting substantial amounts of the daidzein metabolite equol in urine. It is noteworthy that low equol excretion in the human subjects was not due to low isoflavone intake; total isoflavonoid excretion in urine was actually highest in the low excretor of equol (subject 2).

Acknowledgements

This study was carried out with financial support from the Commission of the European Communities, FAIR programme CT-95-0894. It does not reflect its views and in no way anticipates the Commission's future policy in this area. We thank Rob Gilbert and Kate Green for animal husbandry.

References (39)

  • K.D.R. Setchell et al.

    Nonsteroidal estrogens of dietary originpossible roles in hormone-dependent disease

    American Journal of Clinical Nutrition

    (1984)
  • Y. Zhang et al.

    Urinary disposition of the soybean isoflavones daidzein, genistein and glycitein differs among humans with moderate fecal isoflavone degradation activity

    Journal of Nutrition

    (1999)
  • H. Adlercreutz et al.

    Phytoestrogens and Western diseases

    Annals of Medicine

    (1997)
  • M. Axelson et al.

    The identification of the weak oestrogen equol in human urine

    Biochemical Journal

    (1982)
  • M. Axelson et al.

    Soya- a dietary source of the non-steroidal oestrogen equol in man and animals

    Journal of Endocrinology

    (1984)
  • S.A. Bingham et al.

    Phytoestrogenswhere are we now?

    British Journal of Nutrition

    (1998)
  • S. Borriello

    P, Setchell, K

    D.R., Axelson, M., Lawson, A.M., Production and metabolism of lignans by the human fecal flora. Journal of Applied Bacteriology

    (1985)
  • L. Coward et al.

    Genistein, daidzein and their b-glycoside conjugatesantitumor isoflavones in soybean foods from American and Asian diets

    Journal of Agriculture and Food Chemistry

    (1993)
  • S. Heinonen et al.

    In vitro metabolism of plant lignans: new precursors of mammalian lignans enterolactone and enterodiol

    Journal of Agricultural and Food Chemistry

    (2001)
  • Cited by (339)

    • Impact of polyphenols on human gut microbiome and associated biomarkers

      2022, Technologies to Recover Polyphenols from AgroFood By-products and Wastes
    • Diet-induced hypercholesterolemia in small laboratory animal models

      2022, Cholesterol: From Chemistry and Biophysics to the Clinic
    View all citing articles on Scopus
    View full text