**4. Discussion**

Diet modulates the composition of the intestinal microbiota [32] and, in turn, gu<sup>t</sup> microbiota metabolism can determine the final metabolites produced, and thus the corresponding effects on human health. Most studies, however, have focused on the effect of fat and fiber [33,34], while dietary microcomponents, like polyphenols, have received less attention [35]. Certainly, little is known about the influence of isoflavones on the microbial populations of the gu<sup>t</sup> [15–19].

The use of high throughput sequencing techniques allows for the determination of gu<sup>t</sup> members whose culture requirements are still unknown or are uncultivable—having estimated that they are 80% of the bacterial species found by molecular tools [36]. As previously suggested, different bacteria may contribute towards equol production [9,37], but these might be present in the gu<sup>t</sup> in low abundance, making their detection difficult by other techniques. In this study, with the aim of identifying changes in gu<sup>t</sup> microbiota composition associated with the ingestion of isoflavones, and related to the equol-producing metabotype, we selected and made use of fecal samples from eight menopausal women receiving daily isoflavone supplementation over one month. Among these women, we selected three equol-producers and five non-producers for comparative purposes.

In the present work, the abundance of the genera *Slackia* significantly increased after the isoflavone supplementation. This genus, belonging to the family *Coriobacteriaceae*, includes described equol-producing species and strains [38–40] and has been associated in vivo with isoflavone metabolism [17]. Additionally, bacteria belonging to the family *Lachnospiraceae* (*Dorea* and *inc. sed.*) increased in the postmenopausal women with an equol-producing metabotype. *Lachnospiraceae inc. sed.* has previously been reported to increase significantly with isoflavone treatment in a case report of an equol-producing woman [19], while *Dorea* has been associated with isoflavone metabolism in humans in several studies [17,41]. Enrichment of some of these bacteria belonging to the *Lachnospiraceae* family, as well as *Pseudoflavonifractor*, has also been seen in in vitro fecal cultures with isoflavones [42]. The family *Lachnospiraceae* has a very large presence in the human gu<sup>t</sup> and has been linked to the production of butyric acid [43], a compound with beneficial effects on the gastrointestinal epithelium [44].

Supplementation with isoflavones for one month was shown to cause a decrease in the number of species (Sobs index) as well as in the species evenness (Shannon index). These effects have previously been observed with the use of other culture-independent techniques [18]. It has been suggested that isoflavones could provide a chemical environment that selects a subset of the initial bacterial communities [17,45]. Alternatively, isoflavones might have antimicrobial effects on certain intestinal bacterial populations, as has recently been reported on pure cultures of intestinal species [46]. When considering the two different metabotypes studied, no effect in the alpha diversity indexes was observed (data not shown). However, UniFrac analysis indicated a greater similarity of the microbial communities from equol-producing women after one month of isoflavone supplementation, suggesting that isoflavones enriches the gu<sup>t</sup> with microbial species involved in the degradation of isoflavones and equol production.

The production of FAs (relevant gu<sup>t</sup> bacterial metabolites) was carried out to determine their relationship with the consumption of isoflavones and the production of equol. Butyric, acetic, and propionic acids are the main short-chain fatty acids (SCFAs). They are produced in the proximal colon by bacterial fermentation of non-digestible carbohydrates [47] and exert anti-inflammatory and anticarcinogenic activities [48]. In contrast, medium-chain fatty acids (MCFAs), including caproic acid (CA), by favoring TH1 and TH17 differentiation [49], could antagonize the anti-inflammatory activities of SCFAs. Branched-chain fatty acids, such as isobutyric and isovaleric acids, are often associated with protein breakdown and have been less studied.

The current data reveal an increase in CA production after isoflavone supplementation, which indicates differential microbial activity leading to the production of this compound. CA derives from chain elongation reactions in which SCFAs are converted to MCFAs mainly using ethanol or lactate as an electron donor [50]. The elongation process is mediated by microorganisms through the reverse β-oxidation pathway. Whether *Slackia*, the bacterial species found to be increased after the consumption of isoflavones in this study, produces CA is not currently known. Alternatively, isoflavone consumption could stimulate the production of CA by other intestinal microorganisms. These possibilities, however, would require further study. Although studies are still controversial, CA has been related to inflammation-regulating effects. In some studies, diminishing of the production of inflammatory cytokines by CA has been reported [51], while inflammatory effects have been reported in others [49,52].

The concentration of isovaleric acid was higher in samples from the equol nonproducing group (*n* = 5). This result partially agrees with the effect of isoflavones observed previously in fecal anaerobic batch cultures [42], where isovaleric acid was reported to increase in cultures inoculated with feces from equol producers (*n* = 3). This suggests that, regardless of the equol producing status, consumption of isoflavones might stimulate the production of this FA.

In this work, although limited to the small sample size, the description of specific gu<sup>t</sup> microbial and FA changes with the ingestion of isoflavones is provided, contributing to the understanding of the modulation of the gu<sup>t</sup> microorganisms and their activity by these polyphenols. However, more studies with greater numbers of people, and even different populations, are needed to confirm the effects of isoflavone intake on the gu<sup>t</sup> ecosystem.
