Lua Labs Report — LATAM Fermented Foods and Hormonal Microbial Diversity

Date: 2026-05-18 Researcher: Lua Labs (Scientist) Classification: Microbiome Line: L1 — Gut-hormonal axis (estrobolome) Sub-topic: 1.4 — LATAM fermented foods and hormonal microbial diversity (literature gap)

External sources

Chacón-Vargas, K., Torres, J., Giles-Gómez, M., Escalante, A., Gibbons, J.G. (2020). Genomic profiling of bacterial and fungal communities and their predictive functionality during pulque fermentation by whole-genome shotgun sequencing. Scientific Reports / Systematic and Applied Microbiology, PMID 33045640.
Rocha-Arriaga, C., Espinal-Centeno, A., Martinez-Sanchez, S., Caballero-Pérez, J., Alcaraz, L.D., Cruz-Ramirez, A. (2020). Deep microbial community profiling along the fermentation process of pulque, a biocultural resource of Mexico. Systematic and Applied Microbiology 43(6):126138.
Reyes-Sánchez, F.J., Páez-Lerma, J.B., Soto-Cruz, N.O., et al. (2022). Microbial community structure, physicochemical characteristics and predictive functionalities of the Mexican tepache fermented beverage. Systematic and Applied Microbiology / Food Microbiology.
Romero-Luna, H.E., Hernández-Mendoza, A., González-Córdova, A.F., Peredo-Lovillo, A. (2022). Bioactive peptides produced by engineered probiotics and other food-grade bacteria: A review. Review applicable to Mexican fermented foods.
Honda, H., Gibson, G.R., et al. (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):454-463. PMID 38742994.
Wang, S., Hu, P., et al. (2023). Gut microbial beta-glucuronidase: a vital regulator in female estrogen metabolism. Gut Microbes 15(1):2236749.
Ávila-Reyes, S.V., Camacho-Díaz, B.H., et al. (2022). Gut metabolites produced during in vitro colonic fermentation of the indigestible fraction of a maize-based traditional Mexican fermented beverage, Tejuino. Foods / Nutrients, PMC9723516.
Wacher, C., Díaz-Ruiz, G., Tamang, J.P. (2017–2023). Consolidated reviews on pozol, sour atole, tejuino. Annals of Microbiology / World Journal of Microbiology and Biotechnology.
Filannino, P., Gobbetti, M., et al. (2025). β-Glucosidase Activity of Lactiplantibacillus plantarum: A Key Player in Food Fermentation and Human Health. Foods 14(9):1451 (PMC12072041).
Pessione, A., Pessione, E., et al. (2010). Synthesis of isoflavone aglycones and equol in soy milks fermented by food-related lactic acid bacteria. Journal of Agricultural and Food Chemistry 58(19):10428–10434. PMID 20822177.
Marrón-Ponce, J.A., Flores, M., Cediel, G., Monteiro, C.A., Batis, C. (2022). Ultra-processed foods consumption reduces dietary diversity and micronutrient intake in the Mexican population. Journal of Human Nutrition and Dietetics. PMID 35279889.
HCHS/SOL investigators (2024). Dietary Acculturation Is Associated With Altered Gut Microbiome, Circulating Metabolites, and Cardiovascular Disease Risk in US Hispanics and Latinos. Circulation.

Baseline knowledge (what I know before searching)

Fermented foods are living microbial ecosystems. The probiotic industry has reduced them to "Lactobacillus + Bifidobacterium" — a caricature. Each traditional fermented food is a community with taxonomic hierarchies, temporal succession, production of exopolysaccharides (EPS), organic acids, bioactive peptides and, crucially for our program, glycosidase and sulfatase enzymes that can interact with the enterohepatic cycle of steroid hormones.

Before searching the specific literature, my baseline is as follows:

Estrobolome vs. fermented foods — a non-obvious relationship. The estrobolome (L1.1) is dominated by β-glucuronidase (gmGUS). The classic lactobacilli in fermented foods (L. plantarum, L. mesenteroides, L. lactis) are NOT major producers of β-glucuronidase — on the contrary, a desirable characteristic for commercial probiotics is the ABSENCE of gmGUS because it has historically been associated with activation of procarcinogens. The case of Levilactobacillus brevis KABP052 is the documented exception: a specific strain with high gmGUS that does deconjugate biliary estrogens. This creates an interesting tension: the "ideal anti-cancer probiotic" and the "ideal pro-estrogen menopausal probiotic" are biochemically opposite. Strain-level selection rather than genus-level selection matters more than the commercial literature admits.

β-glucosidase: the enzyme that is truly abundant in LATAM fermented foods. L. plantarum, dominant in pulque, pozol, tejuino, tepache and jocoque, is a consistent producer of β-glucosidase. This enzyme hydrolyzes glycosidic bonds in phytoestrogens (isoflavone glycosides → aglycones → equol; lignans → enterolactone). Estrogenic activity in humans does not come so much from "reactivating one's own E2" but from releasing plant aglycones with ERβ affinity. This repositions the LATAM nutritional role: the traditional diet was not rich in soy (as in Asia), but it was rich in lignans from corn, black beans, chia seeds and nopal, which require microbial β-glucosidase to become bioactive.

Sulfatases and the progesterobolome. Recalling L1.2: the progesterobolome depends on microbial sulfatases (anSME). Acid-tolerant anaerobic bacteria from low-oxygen fermented foods (pozol buried in banana leaves, pulque in leather wineskins) could sustain strains with an enzymatic profile different from European kefir/yogurt. Eggerthella lenta (key to the progesterobolome) has NOT been systematically characterized in LATAM fermented foods, but its ecological niche (strict anaerobe, O₂-sensitive) is compatible with traditional fermentations under reducing conditions.

Nixtamalization: forgotten as a microbial modulator. Treating corn with Ca(OH)₂ is not only a source of calcium. It solubilizes cell-wall pentosans, releases ferulates esterified to arabinoxylans, and increases retrograded RS3 resistant starch when the dough cools. RS3 is a preferred substrate for Faecalibacterium prausnitzii, Roseburia intestinalis and Bifidobacterium adolescentis — butyrate-producing bacteria. And here is the connection no one makes explicit: butyrate regulates estrogen receptor β expression in colon and uterus via HDAC inhibition. More butyrate → greater sensitivity to circulating estrogens at the same serum levels. Nixtamalization is, mechanistically, an intervention in estrogenic sensitivity via butyric epigenetics.

Lactobacillus in pulque, Acetobacter in tepache, Weissella in pozol. Strains whose genomics have only been characterized in the last 5-7 years. Hormonal functionality: open hypothesis.

Findings from recent papers

Pulque (Chacón-Vargas 2020, Rocha-Arriaga 2020)

Shotgun metagenomics of pulque identifies 6 central bacterial genera (Acinetobacter, Lactobacillus, Lactococcus, Leuconostoc, Saccharomyces, Zymomonas) and 10 species throughout fermentation. Dominant strains include: Lactobacillus sanfranciscensis (relevant: this lactobacillus is associated with European sourdough and produces opioid peptides; in pulque its hormonal role has not been characterized), Acinetobacter boissieri, A. nectaris, Lactococcus lactis, Leuconostoc citreum, L. gelidum, Zymomonas mobilis, Saccharomyces cerevisiae. Total diversity reaches 2,855 bacterial OTUs and 1,494 fungal species — exceeding the diversity of commercial kefir (~30-50 species) by orders of magnitude. Predicted metagenomic functionality: biosynthesis pathways for essential amino acids, B-group vitamins (folate, limited B12, riboflavin) and EPS production by Leuconostoc with documented prebiotic activity in intestinal Bacteroides.

Tepache (Reyes-Sánchez 2022)

At 72h, the community is dominated by Lactobacillus, Leuconostoc, Acetobacter, Lactococcus, with the yeasts Saccharomyces, Zygosaccharomyces, Candida, Meyerozyma. Production of ethanol (3.39 g/L), acetic acid (0.54 g/L), L-lactate (8.77 g/L). The late acetogenic phase is unique among the fermented foods studied: acetic acid is a cross-feeding substrate for Faecalibacterium prausnitzii and Roseburia — butyrate producers. The Lactobacillus + Acetobacter pair could have a biphasic effect: it provides aglycones via β-glucosidase AND a butyrate precursor via cross-feeding. Not documented in the literature, but biochemically predictable. Alert: a 2025 study (PMC12714347) detects antibiotic-resistance genes in homemade tepache — safety flag for mass recommendations.

Pozol and tejuino (Wacher 2017–2023; Ávila-Reyes 2022)

Pozol (Yucatán/Tabasco): ground nixtamal spontaneously fermented in banana leaves. Microbiota dominated by Lactococcus, Leuconostoc, Lactobacillus plantarum, L. confusus (Weissella). Succession: initial Lactococcus/Leuconostoc → late L. plantarum. Tejuino (Jalisco/Nayarit): nixtamalized corn + piloncillo, 24-72h fermentation. Liquid dominated by Lactobacillus; solid by Weissella. The Ávila-Reyes 2022 paper performs in vitro colonic fermentation of the indigestible fraction of tejuino and demonstrates significant production of butyrate and propionate in human fecal culture — the first direct functional evidence of tejuino as a substrate for SCFA-producing microbiota. The responsible substrate is the resistant starch from nixtamalized corn modified by LAB.

Jocoque (Durango)

Product made from fermented raw milk (artisanal) and pasteurized milk with selected cultures (commercial). The artisanal version contains a mixture of LAB with a profile similar to kefir but with less yeast. Mexican studies (2018-2024) document immune modulation (IL-10 ↑, IgA ↑) in animal models. Fine taxonomic characterization remains incomplete. Regulatory risk: the unpasteurized raw version cannot legally be sold in the formal supply chain — a regulatory gap the probiotic industry ignores.

β-glucuronidase: the real enzymatic map

Screening of 84 LAB strains for gmGUS (Honda 2024) found that only one strain of Levilactobacillus brevis (KABP052) and three strains of Lacticaseibacillus rhamnosus deconjugate estrogens. L. plantarum, dominant in ALL analyzed LATAM fermented foods, does NOT deconjugate estrogens efficiently. This means: LATAM fermented foods do not act mainly through the direct estrobolome. Their hormonal effect must pass through other pathways (β-glucosidase on phytoestrogens, SCFAs on nuclear receptors, immune modulation, support for endogenous Parabacteroides/Coriobacteriaceae).

Mexican nutritional transition

Marrón-Ponce 2022 quantifies: 30% of current Mexican dietary energy comes from ultra-processed foods. Significant inverse association between ultra-processed food consumption and dietary diversity (-0.42, p<0.001). HCHS/SOL 2024: Latinas in the U.S. with greater dietary acculturation show a microbiome with lower abundance of Prevotella copri (ancestral Mesoamerican signature) and increased Western Bacteroides spp. — a pattern associated with cardiovascular risk and possibly altered estrogen metabolism (not directly measured, gap).

Complete molecular/endocrine mechanism

LATAM fermented food (pulque, tepache, pozol, tejuino, jocoque)
       │
       ├── LAB (Lactobacillus plantarum, Leuconostoc, Weissella)
       │       │
       │       └── microbial β-GLUCOSIDASE
       │               │
       │               └── Plant phytoestrogens (isoflavones, lignans
       │                   from corn, beans, chia, nopal) glycosylated
       │                       │
       │                       └── FREE AGLYCONES (daidzein, genistein,
       │                           enterolactone, equol via L. plantarum)
       │                               │
       │                               └── ERβ affinity >> ERα
       │                                       │
       │                                       └── Selective modulation
       │                                           (anti-proliferative
       │                                           in breast/endometrium,
       │                                           CNS-neuroprotective)
       │
       ├── EPS (Leuconostoc, Lactobacillus)
       │       │
       │       └── Prebiotic substrate for Bacteroides spp.,
       │           Bifidobacterium adolescentis
       │
       ├── Resistant starch (nixtamalization in pozol, tejuino)
       │       │
       │       └── Colonic fermentation by Faecalibacterium prausnitzii,
       │           Roseburia intestinalis
       │               │
       │               └── BUTYRATE (major producer in colon)
       │                       │
       │                       ├── HDAC inhibition
       │                       │       │
       │                       │       └── Histone hyperacetylation in
       │                       │           ESR2 promoter → ↑ expression
       │                       │           of estrogen receptor β
       │                       │
       │                       ├── GPR43/GPR109a → intestinal anti-inflammation
       │                       │
       │                       └── Colonocyte energy substrate
       │                           → mucosal barrier integrity
       │                           → LESS systemic LPS
       │                                   │
       │                                   └── (L1.3 connection: less
       │                                       TLR4/NF-κB in granulosa
       │                                       → less oocyte apoptosis)
       │
       ├── Acetate (Acetobacter in tepache, late acetogenic phase)
       │       │
       │       └── Cross-feeding with Faecalibacterium → more butyrate
       │
       └── low pH + lactic acid
               │
               └── Inhibition of pro-inflammatory Enterobacteriaceae
                   (Capnocytophaga, linked to low AMH in L1.3)

Parallel pathway: nixtamalization per se

Dry corn → Ca(OH)₂ + heat → Pentosan solubilization
                                         │
                                         ├── Release of ferulic acid
                                         │   esterified to arabinoxylans
                                         │       │
                                         │       └── Substrate for hydrolases
                                         │           from Bifidobacterium longum
                                         │           and Faecalibacterium
                                         │
                                         └── Gelatinized starch → cooled
                                              → Retrograded RS3
                                                  resistant starch
                                                      │
                                                      └── Colonic
                                                          fermentation → SCFAs

Cross-synthesis with previous findings

Reinterpretation of L1.1 in light of L1.4

L1.1 established that the estrobolome lives on β-glucuronidase, and the first human RCT (Honda 2024) used a European strain (KABP052) with high gmGUS. Today we discovered that LATAM fermented foods are not rich in gmGUS — they are dominated by L. plantarum with β-glucosidase (not β-glucuronidase). This means that the naive hypothesis "LATAM fermented foods = Latin American estrobolome" is false or at least incomplete. The true effect of traditional fermented foods passes through:

β-glucosidase on phytoestrogens (selective ERβ pathway, not through enterohepatic recycling of one's own E2).
SCFA production via nixtamalized resistant starch (epigenetic modulation of hormone receptors, not hormone levels).
Support for the anaerobic ecosystem where Eggerthella, Parabacteroides live (progesterobolome — L1.2).

This retrospectively explains why Asian cohorts (high soy, high β-glucosidase, high ERβ) have fewer vasomotor symptoms despite similar serum E2 levels: it is not more estrogen, it is greater sensitivity via ERβ saturated with phytoestrogens. The pre-industrial Mexican diet would have replicated this pattern with lignans from corn/chia/beans instead of soy isoflavones.

Reinterpretation of L1.2 (progesterobolome)

Eggerthella lenta, key to the progesterobolome, is a strict anaerobe sensitive to O₂. Traditional LATAM fermentations under reducing conditions (pulque in wineskins, wrapped pozol, closed tepache) do not introduce E. lenta directly (it has not been isolated from these products), but they create a sustained anaerobic colonic terrain that favors its endogenous niche. Hypothesis 1 of L1.2 (antibiotics + low-fiber predicts early symptomatic perimenopause) is complemented: traditional LATAM diet with fermented foods + nixtamal + legumes protected the progesterobolome regardless of the exact strain composition of the fermented food. What protected it was the complete dietary matrix, not a specific probiotic.

Reinterpretation of L1.3 (dysbiosis accelerates menopause)

The "compensatory microbiome" (Hypothesis 4) had a missing component: what builds it? Probable answer: sustained diversity over decades. The traditional LATAM diet with 8-12 different prebiotic foods per week (beans, corn, nopal, squash, tomato, chili, herbs, fermented foods, seeds, cacao) builds a microbiome with functional redundancy. The transition to ultra-processed foods (Marrón-Ponce 2022: 30% of current energy) destroys this redundancy in one generation. The menopausal acceleration observed in urban Mexican cohorts (50.1 years average vs 51.4 years in 1990s rural cohort studies) could be partially attributable to this loss — testable retrospectively with NHANES vs. ENSANUT data.

Emerging connection with L2 (HPA-HPO, future)

SCFAs produced by fermentation of LATAM fiber negatively regulate the HPA axis via GPR41/43 receptors in enteroendocrine cells → PYY/GLP-1 → vagal → paraventricular nucleus → CRH reduction. This means that the traditional LATAM diet modulates the stress axis DIRECTLY, not only the ovarian axis. L2 should inherit this loop.

Lua Labs hypotheses

Hypothesis 5: "The ancestral LATAM effect is via ERβ, not via the estrobolome"

Statement: The protective hormonal effect of the traditional Mexican diet does NOT operate mainly through enterohepatic recycling of endogenous estradiol (classic estrobolome), but through an increase in systemic ERβ tone via phytoestrogen aglycones released by β-glucosidase from Lactobacillus plantarum abundant in regional fermented foods, combined with epigenetic upregulation of ESR2 via butyrate derived from nixtamalized resistant starch.

Proposed mechanism: Two converging pathways:

(a) L. plantarum (pulque, pozol, tejuino, tepache) hydrolyzes glycosylated phytoestrogens from black beans, chia, flaxseed, nopal and pigmented corn → aglycones with ERβ:ERα affinity >20:1 → sustained partial occupation of ERβ in colon, hypothalamus, bone and breast → endogenous SERM-like modulation.
(b) Resistant starch from nixtamal + Leuconostoc EPS → colonic butyrate → HDAC inhibition in ESR2 promoter → ↑ receptor transcription → greater sensitivity to circulating estrogens (endogenous + phytoestrogens) at equal serum levels.

The net effect in perimenopause is greater functional estrogenic tone without elevated serum levels — explaining why rural Mexican cohorts historically report fewer vasomotor symptoms with E2 levels similar to European/North American cohorts.

Confidence level: Medium.

High in the underlying biochemistry (L. plantarum β-glucosidase and equol production are established; butyrate-HDAC-ESR2 is documented).
Medium in the magnitude of the clinical effect (there is no Mexican epidemiological study directly correlating ancestral fermented foods with Greene Score and AMH).
Low in directional causality (there may be confounding by co-variables: physical exercise, sun exposure, family structure).

How to validate:

With a formal study: 12-week RCT, n=80, perimenopausal women, intervention arm = 250mL/day of fresh non-commercial white pulque + 1 daily serving of pozol/tejuino, control arm = unfermented beverage. Outcomes: urinary equol, fecal butyrate, optional ESR2 expression in rectal biopsy, Greene Score.

Limitations:

L. plantarum has extremely heterogeneous strains — β-glucosidase activity varies 100× across isolates. Without strain-level characterization, the population effect may be diluted.
Equol is efficiently produced by only ~30% of women (depends on endogenous Slackia isoflavoniconvertens). In equol non-producers, the effect is reduced to daidzein alone.
Homemade pulque and tepache can have high microbiological variability and contamination risk.

Hypothesis 6: "The double estrobolome + progesterobolome pincer of the ancestral Mexican diet"

Statement: The specific combination nixtamalized corn + beans + traditional lactic fermented food (pozol/sour atole/jocoque) + acetogenesis (tepache/pulque) creates the only four simultaneous conditions necessary to sustain BOTH sub-ecosystems (estrobolome and progesterobolome) in a single typical meal: (1) resistant starch → butyrate; (2) exogenous acetate → cross-feeding; (3) estrogenic aglycones → ERβ tone; (4) low colonic pH + accelerated transit → less pro-LPS dysbiosis.

Proposed mechanism: No isolated component replicates the effect. The synergy is structural: beans provide secoisolariciresinol lignans that require β-glucosidase to be released; nixtamal provides RS3 that feeds Roseburia; pozol/jocoque provides L. plantarum with active β-glucosidase; tepache provides acetate for Faecalibacterium. The modern industrial diet (sugary soda + industrial tortilla without real nixtamal + canned beans + absence of fermented food) breaks all 4 conditions simultaneously.

Confidence level: Medium-low in magnitude, high in direction.

How to validate:

With a formal study: controlled crossover feeding study, n=40, 2 weeks reconstructed ancestral Mesoamerican diet vs 2 weeks current urban Mexican diet (matched calories). Measure: fecal SCFAs, microbiome Bray-Curtis dissimilarity, urinary equol/enterolactone, Greene Score, digital mood.

Limitations:

Difficult deconvolution of which component does what (that is precisely the point — synergy is the mechanism).
Enormous variability in preparation (artisanal vs industrial lime, black vs pinto beans, 24h vs 72h fermentation).

Candidate formulation: "Mesoamerican Ancestral Buffer" (concept, not product)

Compounds (foods, not extracts): 100% nixtamalized tortilla from pigmented corn (blue/purple, higher anthocyanin) + whole black beans + 1 daily serving of traditional LATAM fermented food (preferably pozol or tejuino — safer than homemade pulque) + cooked nopal + ground chia.

Target population: Carmen (47, perimenopause) and urban Mexican women aged 40-55 with moderate symptom score.

Complementary mechanisms:

Resistant starch from nixtamal → butyrate → ↑ ESR2.
Lignans from beans/chia + β-glucosidase from fermented food → ERβ-selective aglycones.
Acetate from tepache/sour atole → Faecalibacterium cross-feeding.
Nopal mucilage → substrate for Akkermansia muciniphila → mucosal integrity.
Anthocyanins from pigmented corn → CYP1B1 inhibition → less 4-OH-E1 (procarcinogenic catechol).

Regulatory status: Common foods with documented historical consumption (de facto GRAS). They do not require approval.

Individual variability

Genetics relevant to the fermented-food-phytoestrogen effect

Equol producers vs non-producers: depends on the presence of endogenous Slackia isoflavoniconvertens and/or Adlercreutzia equolifaciens. ~30% of women produce equol; the rest only daidzein. Producers benefit more from the ERβ effect of the diet. Not directly measurable without fecal metagenomics, but proxy: mother/grandmother with history of late + asymptomatic menopause.
UGT1A1, UGT2B15 polymorphisms: rapid vs slow estrogen glucuronidators. Slow glucuronidators recycle more E2 through the enterohepatic route; rapid glucuronidators depend more on exogenous phytoestrogens for ERβ tone.
COMT Val158Met: Met/Met (slow methylator) accumulates catechol-estrogens; combined with low intake of fermented foods and nixtamal, breast risk rises. Met/Met + ancestral diet = possibly more protection than Western diet + Val/Val.
MTHFR C677T: T/T compromises methylation of catechol-estrogens; ancestral diet provides natural folate from beans/quelites that partially compensates.
Histamine tolerance (DAO, ABP1): fermented foods can produce histamine; women with low DAO poorly tolerate kombucha/pulque/aged cheese. Variable to capture (symptoms after consumption).

Environmental variability

Birth type and lactation (inherited from L1.3): C-section + formula = lower probability that the adult woman has the anaerobic ecosystem receptive to nutrients from fermented foods.
Childhood antibiotics: if the ancestral microbiota was eradicated before age 12, adult fermented foods partially repopulate but do not restore diversity.
Migration / dietary acculturation: HCHS/SOL 2024 shows loss of Prevotella copri (Mesoamerican signature) with each additional year in the U.S. Second-generation Mexican immigrant woman = mostly Westernized profile.