**3. Results**

#### *3.1. Change in Fecal Microbiota Over Isoflavone Supplementation*

After denoising, performing chimera checks, and trimming the reads by length (150– 400 bp), a mean of 4756 (±875) high quality sequences was obtained. Taxonomic analysis grouped the sequences mainly into five phyla: *Firmicutes, Actinobacteria, Bacteroidetes, Proteobacteria*, and *Verrucomicrobia*. Fifty-two genera were identified, as well as five groups of clostridia (*Clostridium* cluster IV, cluster XI, cluster XIVa, cluster XVIII, and *Clostridium sensu stricto*) and two taxa with family-associated *incertae sedis* (*inc. sed.*) members (*Erysipelotrichaceae inc. sed.* and *Lachnospiraceae inc. sed.*). Taxonomic groups presenting at an abundance of <0.1% were designated as "others". A mean of 1813 sequences per sample remained unclassified.

Considerable differences were observed between the bacterial communities at T0 (before isoflavone supplementation) and T1 (one month after supplementation). Differences were noted at the family and genus levels (Figure 1). At the genus level, a significant (*p* < 0.05) increase in the relative abundance of the genus *Slackia* was observed at T1 (0.67%) versus T0 (0.27%). Although sequences of this genus were not detected in all women, when they were detected (WC, WG, WE, WL, and WN), their relative proportion increased after supplementation with soy isoflavones.

The supplementation with isoflavones significantly reduced alpha diversity in terms of Sobs and Shannon Indexes (Figure 2). The Sobs index reflects the number of observed species or "richness", while Shannon index weights the numbers of species by their relative evenness.

#### *3.2. Differences in Microbial Groups Associated with the Equol Producer Status*

UniFrac β-diversity analysis was done to assess the extent of similarity between microbial communities. UniFrac-based PCoA plots revealed a clear clustering between equol producing and non-producing women after isoflavone supplementation, while no clustering was observed at baseline (Figure 3).

**Figure 1.** Changes in microbial composition with isoflavone supplementation. Microbial composition at the family (**A**) and genus (**B**) levels in fecal samples of eight menopausal women before (T0) and after one month (T1) of isoflavone supplementation presented as relative abundances (%).

**Figure 2.** Comparison of Sobs and Shannon indexes before (T0) and after one month (T1) of isoflavone treatment in eight menopausal women. The lines inside the rectangles indicate the medians and the whiskers indicate the maximum and minimum values. Analysis was done using the nonparametric Wilcoxon signed-rank test to determine differences between T0 and T1.

**Figure 3.** Weighted UniFrac principal coordinates analysis (PCoA) plots of fecal microbiota composition from the women in the study (*n* = 8) before soy isoflavone intervention (**A**), and after one month of daily supplementation (**B**). Subject color coding: red, equol non-producers (*n* = 5); blue, equol producers (*n* = 3).

Furthermore, comparison of fecal microbial composition between equol producers and non-producers revealed some differences. At T0, relative abundance (% sequences) of *Lachnospiraceae inc. sed.* taxa was significantly higher (*p* = 0.025) in the equol-producer group versus the non-producers (Table 1). While at T1, after one month of isoflavone consumption, the relative abundance of sequences belonging to the genera *Pseudoflavonifractor* and *Dorea* was greater in the equol-producing women.


**Table 1.** Fecal genera showing significant greater relative abundances (% sequences) in equolproducing women before (T0) and after (T1) the soy isoflavone intervention.

a Mean relative abundance ± standard deviation. b Mann–Whitney test.

#### *3.3. Differences in Fatty Acids (FAs) Associated with the Equol Producer Status*

Fecal FAs remained stable after 1 month of isoflavone supplementation, except for caproic acid, which increased significantly after the intervention (Table 2). Regarding differences associated with the equol producing status, all FAs analysed showed higher concentrations in the equol non-producing women, but only isovaleric acid reached statistical significance (Table 3).

**Table 2.** Fecal fatty acids' (FAs) concentration before and after the isoflavone treatment of the eight menopausal women of the study.


Key of statistical significance: \* *p* < 0.05 versus basal sample (t = 0), Wilcoxon test.

**Table 3.** Differences in fecal FAs between equol producers and non-producers after isoflavone supplementation.


Key of statistical significance: \* *p* < 0.05, Mann–Whitney test.
