Scientific Report — Estrobolome: gut bacteria that metabolize estrogens via beta-glucuronidase

Date: 2026-05-15 Line: L1 — Gut-hormonal axis (estrobolome) Sub-topic: 1.1 — Estrobolome and microbial beta-glucuronidase Classification: Microbiome

Sources

1. Hu S, Ding Q, Zhang W, Kang M, Ma J, Zhao L (2023). "Gut microbial beta-glucuronidase: a vital regulator in female estrogen metabolism." Gut Microbes, 15(1). DOI: 10.1080/19490976.2023.2236749

2. Wang H, Shi F, Zheng L, Zhou W, Mi B, Wu S, Feng X (2025). "Gut microbiota has the potential to improve health of menopausal women by regulating estrogen." Frontiers in Endocrinology, 16. DOI: 10.3389/fendo.2025.1562332

3. Larnder AH, Manges AR, Murphy RA (2025). "The estrobolome: Estrogen-metabolizing pathways of the gut microbiome and their relation to breast cancer." International Journal of Cancer, 157(4):599-613. DOI: 10.1002/ijc.35427

4. Honda S, Tominaga Y, Espadaler-Mazo J, Huedo P, Aguiló M, Perez M, Ueda T, Sawashita J (2024). "Supplementation with a Probiotic Formula Having β-Glucuronidase Activity Modulates Serum Estrogen Levels in Healthy Peri- and Postmenopausal Women." Journal of Medicinal Food, 27(6). DOI: 10.1089/jmf.2023.k.0320

Main finding

The estrobolome is the set of gut bacterial genes whose enzymatic products are capable of metabolizing estrogens, either by reactivating them from inactive conjugated forms or by transforming precursors into bioactive estrogenic analogs. Its most studied enzymatic component — and the one with the greatest quantifiable impact on circulating levels of estradiol (E2) and estrone (E1) — is microbial β-glucuronidase (gmGUS, EC 3.2.1.31).

The mechanism is as follows: the liver conjugates free estrogens (mainly E2, E1 and estriol E3) with glucuronic acid via UDP-glucuronosyltransferases (UGTs), producing water-soluble and biologically inactive estrogen glucuronides. These conjugates are excreted in bile into the intestinal lumen. Under healthy microbiome conditions, bacteria carrying GUS genes deconjugate these glucuronides, releasing free estrogens that are reabsorbed through the intestinal mucosa into the portal circulation — the so-called enterohepatic cycle of estrogen. Hu et al. (2023) quantify that approximately 65% of E2, 48% of E1 and 23% of E3 secreted in bile are available for this microbial recycling. This recovered fraction represents a non-ovarian source of circulating estrogens that persists even when ovarian production declines during the perimenopausal transition.

The review by Larnder et al. (2025) broadens the picture beyond gmGUS, identifying four enzymatic classes with the ability to metabolize estrogens in the gut: (1) β-glucuronidase (deconjugation of glucuronides), (2) sulfatases/sulfotransferases (action on estrogen sulfates, a parallel and independent pathway), (3) 3β and 17β hydroxysteroid dehydrogenases (metabolism of androgenic precursors such as DHEA) and (4) β-glucosidases (activation of phytoestrogens from glycosides). This means the estrobolome is not an on/off switch but a functional continuum with multiple control points, each susceptible to diet composition and bacterial diversity.

The most clinically relevant finding to date comes from the RCT by Honda et al. (2024): after 12 weeks of supplementation with Levilactobacillus brevis KABP052 (a strain selected for the highest gmGUS activity in a screening of 84 strains), peri- and postmenopausal women in the probiotic group maintained E2 levels of 31.62 ± 7.97 pg/mL versus 25.12 ± 8.17 pg/mL in placebo, and E1 of 21.38 ± 8.57 pg/mL versus 13.18 ± 8.77 pg/mL. The placebo group showed a significant decrease in estrogens during the follow-up period; the probiotic group kept them stable. It is the first RCT to show that modulating intestinal gmGUS activity can measurably modify circulating estrogen levels in women without active ovarian production.

Molecular/endocrine mechanism

gmGUS-mediated enterohepatic cycle of estrogen

LIVER
  │
  ├─ Free E2/E1/E3
  │     │
  │     ▼ UGTs (UDP-glucuronosyltransferases)
  │     E2-glucuronide (inactive, water-soluble)
  │     │
  │     ▼ Bile → Small intestine/colon
  │
INTESTINAL LUMEN
  │
  ├─ gmGUS+ bacteria (Bacteroidetes, Firmicutes)
  │     │
  │     ▼ Microbial β-glucuronidase (EC 3.2.1.31)
  │     Free E2 (bioactive, lipophilic)
  │     │
  │     ▼ Passive absorption → portal vein
  │
SYSTEMIC CIRCULATION
  │
  ├─ Free E2 →  ERα receptor (ESR1) → target tissues
  │              ERβ receptor (ESR2) → ovary, CNS, bone
  │
  └─ Return to liver → new conjugation cycle

Parallel estrobolome pathways:

Phytoestrogens (diet)
  │ Daidzein, Genistein (isoflavones) → microbial β-glucosidase → Equol
  │ Lignans (flaxseeds) → β-glucosidase → Enterolactone/Enterodiol
  └ Equol: selective affinity for ERβ > ERα (potential cardioprotective effect)

Androgenic precursors (DHEA, androstenedione)
  │ Microbial 17β-HSD → conversion to testosterone/E1
  └ Relevant in postmenopause: main source of peripheral E1

Estrogen sulfates (E2-SO4, E1-SO4)
  │ Microbial sulfatases → free estrogens (pathway parallel to glucuronides)
  └ Sulfotransferases → re-conjugation (opposite effect)

Bacterial taxonomy with active GUS genes

Hu et al. (2023) identified functional β-glucuronidase genes in ~60 gut bacterial genera. Distribution by phylum is:

Note on activity vs. presence: the mere presence of GUS genes does not equal enzymatic activity. gmGUS expression depends on available substrate (glucuronide load in the lumen), luminal pH, and the transcriptional regulation of each strain. Larnder et al. (2025) identify that heterogeneity in clinical studies is due in part to this distinction: gene presence ≠ functional activity.

Scenario analysis

Scenario A — Robust estrobolome as a buffer for the hormonal transition

If a woman enters perimenopause with high microbial diversity and abundance of genera carrying active GUS genes (Bacteroides, Roseburia, Bifidobacterium, Lactobacillus), the enterohepatic cycle could partially compensate for the decline in ovarian E2 production by maximizing intestinal recovery of circulating estrogens.

The RCT by Honda et al. (2024) supports this hypothesis directly: the strain Levilactobacillus brevis KABP052 — chosen for its higher gmGUS activity among 84 candidates — maintained significantly higher E2 and E1 at 12 weeks vs. placebo in women without relevant ovarian production. The effect is modest in absolute terms (~6 pg/mL difference in E2), but potentially clinically relevant in the context of postmenopausal women whose basal E2 is 10-20 pg/mL.

Limitations of Scenario A: The Honda et al. RCT has a small N and a duration of only 12 weeks. The KABP052 strain was expressly designed for high GUS activity — it is not extrapolable to any commercial probiotic. In addition, the amount of estrogens recoverable via the enterohepatic cycle is bounded by what the liver excreted, which in turn depends on baseline ovarian/adrenal levels. The estrobolome amplifies but does not create estrogens ex nihilo.

Scenario B — Dysbiosis as an accelerator of estrogen decline

Wang et al. (2025) document a consistent pattern in postmenopausal women: significant reduction in Lactobacillus, Bifidobacteria and Roseburia — all genera with confirmed GUS activity — and proportional increase in Enterobacter, Tolumonas and other bacteria with scarce or no gmGUS activity. This dysbiosis reduces the estrobolome’s intestinal recovery capacity, accelerating estrogen clearance.

The problem becomes bidirectional: estrogens modulate microbiome composition (via ER receptors in enterocytes and mucosal immune cells), and the microbiome regulates estrogens via gmGUS. When ovarian production falls, the microbiome loses one of its main regulators → dysbiosis worsens → gmGUS activity falls → circulating estrogens fall further → the microbiome is disrupted further. It is a negative feedback loop.

Critical limitation: most studies are observational and cross-sectional. Temporal causality has not been established: does dysbiosis precede the estrogen decline or is it a consequence of it? A longitudinal study following women from premenopause to postmenopause with frequent microbiome and serum estrogen measurements would be necessary to resolve this ambiguity. That design does not yet exist in the published literature.

Scenario C — Connection with xenoestrogens and endocrine disruptors (via microbiome)

The estrobolome also metabolizes xenoestrogens (bisphenol A, phthalates, organochlorine pesticides) and dietary phytoestrogens through the same enzymatic pathways it uses for endogenous estrogens. This opens a third axis: diet not only provides substrates for the estrobolome, but the environmental xenoestrogen load may compete with endogenous E2 for available gmGUS capacity.

In regions with high exposure to agricultural pesticides and frequent consumption of processed foods containing bisphenols — as is the case across much of Latin America — this competition for available enzymatic capacity could be an unrecognized determinant of circulating estrogen availability in perimenopausal women. It is an environmental hypothesis that links the chemical load of the surroundings to microbiome-mediated hormonal regulation.

Individual variability

The core of the phenomenon is that the same estrogen and the same food can produce different hormonal responses depending on each woman's microbiome. The factors that explain individual variability in estrobolome function are:

Note on LATAM: Wang et al. (2025) and Larnder et al. (2025) base their findings primarily on European and North American cohorts. The estrobolome composition in Latin American women — with dietary patterns high in legumes, nixtamalized corn, chili, and differential consumption of fermented foods — has not been characterized. It is an active knowledge gap in the published literature.

Limitations and future directions

Most available studies on the estrobolome are observational or cross-sectional. The bidirectional nature of the estrogen-microbiome relationship makes it difficult to establish causality: it remains unresolved whether dysbiosis precedes the estrogen decline or is a consequence of it. Randomized controlled trials in this space are scarce — Honda et al. (2024) is one of the first — and rely on small samples. Resolving this causal ambiguity will require longitudinal studies that follow the same women across the perimenopausal transition, with frequent, simultaneous measurements of microbiome and serum estrogens.