Prebiotic and Chemopreventive Potential of Digested Cooked Huitlacoche (Ustilago maydis) in Human HT‐29 Colon Cancer Cells

ABSTRACT

Huitlacoche (Ustilago maydis) is a popular and widely consumed edible fungus that infects corn, inducing gall formation rich in phenolic compounds. During consumption, huitlacoche‐based phenolic compounds are affected by gastrointestinal conditions, thereby altering their bioaccessibility and, in turn, their prebiotic and chemopreventive potential. This study investigated the in vitro chemopreventive effects of cooked huitlacoche on human colon cancer cells (HT‐29) and its prebiotic potential. Cooked huitlacoche underwent in vitro gastrointestinal digestion, followed by colonic fermentation, to evaluate the bioaccessibility of phenolic compounds and antioxidant capacity at each stage. The prebiotic potential was monitored by Lactobacillus gasseri growth in the presence of the nondigestible fraction of huitlacoche (NDF‐H). Additionally, the chemopreventive effect of the colonic fermentation extract of huitlacoche (FE‐H) was evaluated on HT‐29 colon cancer cells. The cooking process increased (p < 0.05) total dietary fiber (+9.60%) while reducing other micronutrients. The concentrations of selected phenolic compounds increase during the gastric phase (benzoic acid: 265‐fold; and 4‐hydroxybenzoic acid: 19.13‐fold). During in vitro human‐gut colonic fermentation, acetic, propionic, and butyric acids were detected (up to 3.06–7.08 mmol/L). FE‐H induced cytotoxic effects in HT‐29 cells (LC₅₀: 33.34% v/v) due to necrosis and caspase‐3/7 activation, without being cytotoxic in human non‐tumoral Hs 738 St./int. Moreover, digested huitlacoche promoted L. gasseri growth. Overall, cooked huitlacoche components are bioaccessible during in vitro gastrointestinal digestion and suggest chemopreventive and prebiotic potential.

Practical Applications

The consumption of cooked huitlacoche may have beneficial effects, supporting its classification as a functional food with chemopreventive and prebiotic properties. These properties stem from the dietary fiber content and antioxidant compounds.

Abbreviations

4‐DiHBenz

4‐hydroxybenzoic acid

Acetic

acetic acid

ANOVA

analysis of variance

ATCC

American Type Cell Culture Collection

Benz

benzoic acid

Buty

butyric acid

CE

(+)‐catechin equivalents

CTs

condensed tannins

DF

digestible fraction

DiHBenz

dihydroxybenzoic acid

DMEM

Dulbecco's modified Eagle's medium

DPPH

1,1‐diphenyl‐2‐picrylhydrazyl

Epic

epicatechin

FBS

fetal bovine serum

FE‐H

colonic fermentation extract of huitlacoche (Ustilago maydis)

FRAP

ferric reducing antioxidant power

GAE

gallic acid equivalents

Gal

gallic acid

HEPES

10 mM 4‐(2‐hydroxyethyl‐1‐piperazineethanesulfonic acid

HPLC–DAD

high‐performance liquid chromatography coupled to a diode array detector

IDF

insoluble dietary fiber

LC₅₀

half‐lethal concentration

LC₅₀‐FE‐H

the 48 h‐half‐lethal concentration of the fermented extract of huitlacoche (Ustilago maydis)

LDH

lactate dehydrogenase

LS‐NDF

1% solid + liquid NDF

MRC

maximum release control (for LDH assay)

MTT

3‐(4,5‐dimethylthiazol‐2‐yl)‐2,5‐diphenyltetrazolium bromide

NDF‐H

nondigestible fraction of huitlacoche

NDH

non‐digested huitlacoche (raw)

OD

optical density

PC

principal component

PCA

principal components analysis

PI

propidium iodide

Prop

propionic acid

RE

rutin equivalents

ROS

reactive oxygen species

SCFAs

short‐chain fatty acids

SDF

soluble dietary fiber

TDF

total dietary fiber

TE

Trolox equivalents

TF

total flavonoids

TPC

total phenolic compounds

1. Introduction

Huitlacoche (Ustilago maydis) is a fungus that infects corn by forming galls in cobs, stems, and leaves. Even though huitlacoche has been valued as an edible mushroom since pre‐Hispanic times, it has recently gained international attention (Villagran et al. 2023). Huitlacoche contains bioactive compounds that can enhance its nutraceutical potential and culinary relevance. Those compounds include polyphenols, dietary fiber, amino acids, essential fatty acids (arachidic, oleic, palmitoleic, oleic, linolenic, palmitic, 11‐eicosenoic, linoleic, among others), and additional compounds (ustilipid A, sesquiterpenoids, ergothioneine, ustilagol A, and ustilagomaydisin B) that exert antioxidant, immunoregulatory, anticancer, and prebiotic properties (Aydoğdu and Gölükçü 2017; Villagran et al. 2023). Regarding polyphenolic compounds, ferulic acid is one of the major compounds (54.4 mg/100 g), followed by quercetin (4.5 mg/100 g) (Valdez‐Morales et al. 2016). Other relevant compounds found in U. maydis include phytosterols, particularly campesterol‐3‐glucoside, Δ7‐stigmasterol, Δ7‐avenasterol, and ergosterol, which are known for their lipid‐lowering and antioxidant properties. Additionally, ustilagomaydisin, ergothioneine, sesquiterpenes, and ustilipids have been tested under pharmacology‐relevant conditions, with ustilagomaydisins A–C demonstrating cytotoxic activity against multidrug‐resistant leukemia cells and showing dose‐dependent antimicrobial activity (Wang et al. 2014).

Collective evidence suggests that many of the bioactive compounds found in U. maydis can reduce the risk of colon cancer, which is the third most common cancer and the second leading cause of cancer‐related deaths worldwide (Half and Arber 2009; Ma et al. 2018; Zhao et al. 2018). As incidence and mortality are rapidly increasing in Mexico, as in other Latin American countries, with a 48% increase from 1990 to 2020 (Andrade‐Meza et al. 2023), there is an urgent need for additional research investigating factors, including diet, that contribute to reducing colon cancer risk. However, it has been found that these components must be bioaccessible and reach the tissue of interest to exert health effects. Studies on the bioaccessibility of huitlacoche bioactive compounds are limited, and only raw ethanolic extracts have been evaluated, showing that the bioaccessibility of total phenolic compounds and antioxidant capacity are maintained and increased during digestion; however, the colonic step has not been included (López‐Martínez et al. 2022). As huitlacoche is primarily consumed cooked, health effects assessments should be conducted on ready‐to‐eat material. Furthermore, cooking affects the chemical and nutraceutical content of huitlacoche (Valdes‐Morales et al. 2010).

A new approach to improve gut health and prevent noncommunicable diseases (NCDs) is the growing recognition of edible mushrooms as a rich source of prebiotic polysaccharides (Ayimbila and Kaewsompong 2025). Evidence for huitlacoche effects on the colon is limited, including modulation of the microbiota and anticancer effects; however, the polysaccharide content of huitlacoche (up to 60 g/100 g of dietary fiber) may reverse microbiota dysbiosis (Israel et al. 2025) and act as a prebiotic. Prebiotic components are of particular interest in colon cancer prevention. According to Gibson and Rastall (2006), for a food or ingredient to be classified as a prebiotic, it must meet the following specific criteria: (1) resist digestion, enzymatic hydrolysis, and gastrointestinal absorption; (2) be fermentable by colonic bacteria; and (3) selectively stimulate the growth and activity of probiotic bacteria (Carlson et al. 2018), such as Lactobacillus and Bifidobacteria, while promoting gut health by maintaining eubiosis in the gut microbiota (Davani‐Davari et al. 2019; Roberfroid 2007). Lactobacillus gasseri is a human‐sourced lactic acid bacterium of interest in food and health; it has probiotic effects, including promoting vaginal health and gastrointestinal homeostasis, treating obesity, regulating immunity, and displaying antioxidant activity (Huang et al. 2025).

Huitlacoche contains β‐glucans (20–100 mg/100 g), a group of compounds with prebiotic properties (Villagran et al. 2023) that promote the synthesis of short‐chain fatty acids (SCFAs) and support colon cancer prevention; however, studies on SCFAs production from huitlacoche are scarce. β‐Glucans modulate colonic microbiota, promoting Roseburia, Bifidobacterium, Bacteroidota increases, among others, linked to increased SCFAs concentrations, especially butyrate, compounds presumably responsible for their colon cancer protection by increasing apoptosis of tumor cells (Shoukat and Sorrentino 2021). Therefore, huitlacoche may have chemopreventive effects; however, other compounds in huitlacoche, such as phenolic compounds and components of its dietary fiber, may also exhibit these effects and warrant further study.

Given that huitlacoche is an edible mushroom endemic to Mexico with potential functional health properties, this study evaluated its in vitro prebiotic and chemopreventive potential in human HT‐29 colon cancer cells after its bioactive compounds reach the colonic stage of digestion.

2. Materials and Methods

2.1. Plant Material

Huitlacoche was obtained from a local producer in Santiago de Querétaro (Mexico). It was harvested 20 days after inoculation of white maize (variety 75–73). The material was washed and divided into two parts: raw and cooked. Samples were slow‐cooked for 15 min at about 65°C, simulating the preparation process for consumption across different regions of the country (Valdez‐Morales et al. 2010). The raw and cooked samples were dehydrated using forced‐air drying for 10 h at 55°C (IN 450, Natural Convection Incubator, Memmert, Germany). Samples were ground to a particle size of 0.5 mm and stored in the dark.

2.2. Chemical and Nutraceutical Characterization

The chemical composition of huitlacoche was determined according to the AOAC (2002) methodology: moisture (925.23), ash (945.46), fat (920.39), and protein (920.15). Carbohydrates were calculated by difference. Total dietary fiber (TDF), soluble dietary fiber (SDF), and insoluble dietary fiber (IDF) were determined by an enzymatic‐gravimetric assay described by Shiga et al. (2003) and following the reported 991.43 method from AOAC.

Phenolic compounds were extracted according to the methodology proposed by Domínguez‐Hernández et al. (2022). The samples were extracted with 80% v/v methanol (50 mg in 800 µL) under constant stirring for 15 min, followed by centrifugation (3500 × g for 10 min at 4°C) (323 K centrifuge, Hermle Labortechnik GmbH, Wehingen, Germany). The supernatant was separated from the precipitate to obtain the free phenolic compounds. The pellet was mixed with 500 µL of 2 M NaOH, incubated (90°C for 1 h), and then adjusted to pH 2 with 2 M HCl. The bound phenolic compounds were recovered in 600 µL ethyl acetate (250 rpm for 10 min at 25°C). The organic phase was separated, rinsed twice, and evaporated for 24 h at room temperature to resuspend the bound phenolic compounds in methanol.

Total phenolics were quantified according to the method of Singleton et al. (1999) for free and bound fractions and are expressed as mg gallic acid equivalents/g sample (mg GAE/g). Total flavonoids were determined by the modified microplate spectrophotometric assay as described by Oomah et al. (2005) using a microplate spectrophotometer (MultiSkan Go, ThermoScientific, Waltham, MA, USA) and expressed as mg rutin equivalents per g sample (mg RE/g). Condensed tannins (CTs) were quantified according to Feregrino‐Pérez et al. (2008) and expressed as mg of (+)‐catechin equivalents/g sample (mg CE/g).

2.3. Identification and Quantification of Free and Bound Phenolic Compounds by High‐Performance Liquid Chromatography Coupled to a Diode Array Detector (HPLC–DAD)

Previous extracts (free and bound phenolic compounds) were subjected to HPLC–DAD on an Agilent 1100 series HPLC instrument (Agilent Technologies, Palo Alto, CA, USA) according to the methodology described by Sánchez‐Quezada et al. (2023). A Zorbax Eclipse XDB‐C18 column (Agilent Technologies, 4.6 × 250 mm, 5 µm of granule size) was used. The column temperature was controlled by a thermostat at 35°C ± 0.6°C, with a flow rate of 1 mL/min. The mobile phase consisted of two solvents: (1) HPLC‐grade water adjusted to 0.1% acetic acid, and (2) 100% acetonitrile. A linear gradient was used as follows: 80%–83% A for 7 min, 83%–60% A for 5 min, 60%–50% A for 1 min, and finally 50%–85% A for 2 min. Detection was performed at 280 and 320 nm (for phenolic acids and flavonoids) with a 1 s acquisition speed. A 20 µL volume was injected, and samples were analyzed in triplicate. Quantification was performed using the external standard method with commercial standards of (+)‐catechin, rutin, and quercetin as representative flavonoids and chlorogenic, caffeic, p‐coumaric, ferulic, and gallic as representative phenolic acids (Figure S1).

2.4. Antioxidant Capacity Measurement by the 1,1‐Diphenyl‐2‐Picrylhydrazyl (DPPH) and the Ferric Reducing Antioxidant Power (FRAP) Methods

Antioxidant capacity was measured using the stable DPPH radical, following the method proposed by Fukumoto and Mazza (2000). The FRAP method was performed as described by Benzie and Strain (1996). Antioxidant capacity results were expressed as micromoles of Trolox equivalents per gram of the sample (µmol Trolox/g).

2.5. In Vitro Gastrointestinal Digestion

Gastrointestinal digestion was simulated in vitro using cooked huitlacoche, following the methodology of Campos‐Vega et al. (2015) with modifications, which resembles the standardized INFOGEST 2.0 method (Brodkorb et al. 2019) but includes separation of the digestible (DF) and nondigestible (NDF) fractions.

For the oral phase, four healthy volunteers were recruited. The participants (four females, aged 24–35) fasted for at least 90 min, sipped 5 mL of purified water, and performed a 30 s mouth rinse. The collected saliva was mixed. Saliva solution (10 mL) was mixed with huitlacoche sample (1 g) and incubated (2 min at 37°C). The oral phase solution was submitted to the gastric phase using 8 mL of the following gastric secretion: KCl, KH₂PO₄, NaHCO₃, MgCl₂(H₂0)₆, (NH₄)₂CO₃, NaCl, 0.005 mL CaCl₂·2(H₂O)₂, pepsin (0.055 g dissolved in 0.94 mL 20 mM HCl) (≥2500 U/mg, Sigma‐Aldrich, St. Louis, MO, USA), and 0.535 mL distilled water. Additionally, 60 µL of 120 U/mL gastric lipase 1000 U/mg, Sigma‐Aldrich) was added to the solution. The gastric phase was adjusted to pH 2 with HCl (2 M), vortexed, and incubated under continuous stirring for 2 h at 37°C (Campos‐Vega et al. 2015; Brodkorb et al. 2019).

For the intestinal stage, 5 mL of Krebs‐Ringer buffer [118 mM NaCl, 4.7 mM KCl, 1.2 mM MgSO₄, 1.2 mM KH₂PO4, 25 mM NaHCO₃, 11 mM glucose, and 2.5 mM CaCl₂, pH 6] was prepared and used to dissolve 3 mg bovine bile (Oxgall dehydrated fresh bile, Becton Dickinson, Franklin Lakes, NJ, USA) and 2.6 mg of pancreatin (8×USP, Sigma‐Alrich), and this mix was added to the resulting gastric phase solution. Additionally, a 7‐cm nitrocellulose membrane bag, knotted on both sides, containing 5 mL of the Krebs‐Ringer solution, was added to each tube, and the tubes were incubated at 37°C for 60 min. At the end of the incubation, the fraction outside the membrane bag was collected and named the NDF of huitlacoche (NDF‐H), and the fraction inside the membrane bag was named the DF.

2.6. In Vitro Colonic Fermentation

The effects of NDF in the colon were estimated using the human gut microbiota fermentation method proposed by Campos‐Vega et al. (2009). Fresh fecal samples were collected from a healthy female donor (aged 24 years, approximately 60 kg). The donor had no history of intestinal disease and had not received any antibiotics in the past 3 months. Fecal samples (3 g) were vortexed for 2 min with 27 mL of PBS buffer (NaH₂PO₄ 0.05 M and Na₂HPO₄ 0.05 M), pH 7. A culture medium was prepared (9 mL per tube) and mixed with 1 mL of fecal solution, 250 mg of solid NDF, and 250 µL of liquid NDF. Samples, a positive control (100 mg raffinose) (R0514, Sigma Aldrich), and a negative control (intestinal solution) were incubated for 24 h at 37°C. An aliquot was collected every 6 h for pH measurement and the following assays.

For the assays involving human subjects, all participants provided written consent, which included a written protocol where participants were informed about what was needed from them (saliva and fecal samples), how their data was going to be stored and processed, who would have access to the data, and the possibility of using this data for research purposes, including the publication of a research manuscript. The study and the protocol design were approved by the Universidad Autónoma de Querétaro Human Research Internal Committee (CBQ23/061) and complied with the ethical principles of the Declaration of Helsinki.

2.7. Bioaccessibility Measurement

The bioaccessibility of phenolic compounds during in vitro gastrointestinal digestion and colonic fermentation simulation was determined using the formula (D'Antuono et al. 2015):

$$B=\frac{C_f\times 100}{C_0}$$

where B indicates the bioaccessibility percentage of polyphenols, C ₀ is the concentration value of the undigested sample, and C_f is the final concentration of the compound after a certain incubation period.

2.8. SCFAs Quantification in Colonic Fermentation

SCFAs were extracted following the Díaz‐Corona et al. (2023) method. Briefly, 500 µL of K₂HPO₄ buffer (20 mM, pH 2) and 500 µL of methanol were added to 1 mL of colonic fermented extract of huitlacoche (FE‐H), and the mixture was incubated for 10 min. Solutions were transferred to vials suitable for HPLC assays.

Chromatographic analysis was performed on an Agilent 1100 series HPLC instrument (Agilent Technologies, Palo Alto, CA, USA) following the method of Díaz‐Corona et al. (2023) with slight modifications. A Polaris 5 column (C18‐A, 150 × 4.6 mm) was used at 40°C and a 15‐min elution gradient with acidified water mobile phase (H₂SO₄ 0.1%). The mobile phase consisted of 5 mM sulfuric acid (Solvent A) and acetonitrile (Solvent B), with the following gradient elution program: 0–4 min: 100% A and 0% B, 4–4.20 min: 98% A and 2% B, 4.2–11.9 min: 98% A and 2% B, 11.9–15 min: return to 100% A and 0% B.

A sample (20 µL) was injected, and the UV–VIS absorbance was measured at 210 nm using a diode array detector (DAD). For SCFAs identification and quantification, calibration curves (0–500 mmol/L) were generated using formic, propionic, acetic, and butyric acid standards (cat. 06547, cat. 94425, cat. 1005706, cat. 19215, Sigma‐Aldrich, USA) (Figure S2). As the fermented extract from huitlacoche after 24 h of fermentation (FE‐H) contained the highest levels of total phenolic compounds and SCFAs, this extract was used for the cell assays.

2.9. Cell Assays

2.9.1. HT‐29 and Hs 738 St./Int Cell Culture

Human HT‐29 cells (American Type Cell Culture Collection, ATCC) (ATCC HTB‐38) (passage: 18) were handled in a biosafety chamber where they were cultured in Dulbecco's modified Eagle's medium (DMEM) (ATCC) supplemented with heat‐inactivated fetal bovine serum (FBS) (10%, Gibco, Grand Island, NY), 1% v/v antibiotics‐antimycotics (100×, Gibco), and sodium bicarbonate (2.2 g/L, Sigma‐Aldrich). As a non‐tumoral control cell line, human Hs 738 St./Int (ATCC CRL‐7869) fibroblasts from the stomach/intestine were used (passage: 3). The cells were cultured using DMEM (Sigma‐Aldrich) supplemented with 10% FBS, 1% antibiotics‐antimycotics (100×, Gibco), 0.11 g/L sodium pyruvate (Sigma‐Aldrich), 3.7 g/L sodium bicarbonate (Sigma‐Aldrich), and 10 mM 4‐(2‐hydroxyethyl)‐1‐piperazineethanesulfonic acid (HEPES, Sigma‐Aldrich).

The culture medium was replaced with fresh medium every second (Hs t38 St./Int) or third day (HT‐29) until confluence (80%) was reached. The cells were maintained in a humidified 5% CO₂ atmosphere at 37°C in an incubator with automatic gas chambers (Forma Series II Water‐Jacketed CO₂ Incubator, Thermo Scientific, USA).

2.9.2. Evaluation of the Cytotoxic Effect of the Fermented Extract From the NDF of Huitlacoche (FE‐H) on HT‐29 and Hs 738 St./Int Cells by the MTT Assay

HT‐29 cells were seeded in 96‐well plates (3 × 10⁴ cells/well) in DMEM culture medium (ATCC) supplemented with FBS (10%) and antibiotic–antimycotic (1%). After 24 h of incubation, the medium was replaced with DMEM supplemented with 2% FBS and containing different concentrations (6%–50%) of the FE‐H (obtained after 24 h fermentation of NDF‐H) (100 µL/well), previously sterilized by filtration using 0.20 µm sterile syringe filters (Merck, Rahway, NJ, USA), and incubated for a further 24 or 48 h.

For the Hs 738 St./Int cells, only the 48 h‐LC₅₀ (half‐lethal concentration) FE‐H dose, obtained from the HT‐29 assays, was tested (Figure S3). Cytotoxicity was analyzed using the 3‐(4,5‐dimethylthiazol‐2‐yl)‐2,5‐diphenyltetrazolium bromide (MTT) assay (Invitrogen). The medium was removed, and a filtered MTT/PBS solution (0.5 mg MTT/mL; 200 µL/well) was added to each well prior to cell incubation (37°C, 1 h). The MTT/PBS solution was removed, and the insoluble MTT formazan crystals were dissolved in DMSO at room temperature. Finally, the absorbance of the whole microplate was measured at 595 nm using a microplate spectrophotometer (MultiSkan Go, Thermo Scientific, USA), and LC₅₀ was determined by plotting the growth inhibition rate. The negative control corresponded to untreated cells, whereas the positive control was a mild 50 µM H₂O₂ treatment to avoid excessive cell death. The blank of the 24 h fermentation of NDF‐H and the lysis buffer used for the Caspase 3/7 and lactate dehydrogenase (LDH) assays were also tested.

2.9.3. Impact of Treatments on LDH Activity

LDH activity was measured using an LDH Cytotoxicity Detection Kit (11644793001, Roche, Basel, Switzerland) according to the manufacturer's instructions. Briefly, cells were seeded for the MTT assay (3 × 10⁴ cells/well) and treated for 48 h with LC₅₀‐FE‐H (33.34% v/v, 200 µL/well). Then, 30 min before the 48‐h limit, selected wells (4 wells) were treated with a commercial lysis buffer (Invitrogen, Thermo Fisher Scientific), containing 75 µL Tris, pH 7.5, 2 M NaCl, 20 mM ethylenediaminetetraacetic acid, and 0.2% TRITON X‐100 per 1 mL of lysis buffer. Then, 150 µL of the cells’ supernatant was transferred to sterile test tubes and centrifuged (250 × g) to separate the cells’ debris, and 100 µL of the supernatant was transferred to a clear 96‐well plate. Afterward, 100 µL of a previously prepared LDH assay kit solution was added to each well, and the wells were incubated for 30 min. The absorbance was measured in a microplate reader at 492 nm, and the results were expressed as a percentage of the negative control. Untreated cells (2% FBS–DMEM) were considered the negative control, whereas cells treated with 50 µM H₂O₂ were considered the positive control. The blank of the fermentation was also evaluated, using the same concentration as in LC₅₀‐FE‐H (33.34% v/v).

2.9.4. Assessment of Caspase 3/7 Activity

The EnzCheck Caspase‐3 Assay kit #2 (Invitrogen, Thermo Fisher Scientific) was used to assess apoptosis by quantifying caspase‐3 and other Asp‐Glu‐Val‐Asp (DEVD)‐specific protease activities, such as caspase‐7, in cells exposed to the LC₅₀‐FE‐H, following the manufacturer's conditions and slight modifications to obtain enough cell protein to obtain measurable enzyme levels (Soto et al. 2022). Briefly, 1.05 × 10⁴ cells/cm² were seeded in 6‐well plates until reaching confluence (24 h). Then, the media were replaced with either 2 mL of LC₅₀‐FE‐H (33.34% v/v), 2 mL of 2% FBS–DMEM only (negative control), the blank from the colonic fermentation (dissolved in 2% FBS–DMEM), or a positive inducer of apoptosis (50 µM H₂O₂) for 48 h. Subsequently, the media was removed, and the cells were washed with 1× PBS, trypsinized with 0.25% Trypsin‐EDTA (Gibco) for 15 min, and 2% DMEM was added to stop trypsin activity. The solutions were then centrifuged (150 × g) to obtain the intact cellular pellets. Cell lysis was induced by adding 50 µL of the manufacturer's lysis buffer, and activity was further enhanced by freezing the pellets at −80°C for 20 min. The lysates were then centrifuged at 4000 × g for 10 min at 4°C, and the supernatants were used to quantify Caspase 3/7 activity by fluorescence (excitation/emission at 496/520 nm), with results expressed relative to the negative control. The same procedure and treatments were applied for the Hs 738 St./Int cells. The results were expressed as a percentage of the negative control. Untreated cells (2% FBS–DMEM) were considered the negative control, whereas cells treated with 50 µM H₂O₂ were considered the positive control. The blank of the fermentation was also evaluated, using the same concentration as in LC₅₀‐FE‐H (33.34% v/v).

2.10. Prebiotic Potential Assay

The prebiotic effect was assessed by testing the growth of L. gasseri bacteria (isolated from Swanson Probiotics L. gasseri; Swanson Health Products, USA). Bacteria were activated and inoculated in Man, Rogosa, and Sharpe (MRS) broth with dextrose (BD DifcoTM, Becton, Dickinson and Company, Sparks, MD, USA) and incubated (Shel Lab model 2100 incubator, Sheldon Manufacturing, Inc., Cornelius, OR, USA) at 37°C for 32 h under anaerobic conditions. After incubation time, the tubes were stored at 4°C until use.

The prebiotic potential of huitlacoche was evaluated following Pacheco‐Ordaz et al. (2018), with modifications. A dilution of the stock culture was performed using MRS without dextrose until a previously determined absorbance (OD₆₀₀) corresponding to 1 McFarland standard unit (1 × 10⁸ colony‐forming units). The developed method was performed as follows: culture medium, bacterial inoculum, and the corresponding NDF‐H concentration (1% solid + liquid NDF or LS‐NDF, 1% solid NDF‐H, and 2% solid NDF‐H). Experimental cultures were added to 1.5 mL microcentrifuge tubes and filled with 1.5 mL to secure an anaerobic environment. The NDF‐H was replaced with dextrose (2%) as a positive control for the experiment.

Inoculated tubes were incubated (37°C, 14 h) with aliquots taken every 2 h. Supernatant was collected from each tube, and absorbance was monitored at 600 nm using a microplate spectrophotometer (MultiSkan Go, Thermo Scientific, USA). Results were expressed as the average difference between the positive and negative controls in a triplicate assay. Optical density measurements were fitted to the Baranyi and Roberts (1994) model to estimate bacterial growth parameters, including lag phase duration, maximum population, and maximum growth rate (defined as the maximum slope of the fitted growth curve).

2.11. Statistical Analysis

At least two independent experiments, each performed in triplicate, were conducted. Data were screened for normality and homoscedasticity using Shapiro–Wilk's and Bartlett's tests. The effect of cooking on chemical and phytochemical composition was compared using Student's t‐test (p < 0.05). Analysis of variance (ANOVA) was used to evaluate changes in the bioaccessibility of phenolic compounds during digestion and the effect of colonic‐fermented NDF extract from huitlacoche on colon cancer cell survival. Both were followed by means comparison using the Tukey–Kramer test for dose comparison (p < 0.05). For the prebiotic effect, optical density (an indicator of the bacterial population) for the different NDF‐H doses was compared using ANOVA followed by Tukey–Kramer post hoc tests. A principal components analysis (PCA) was conducted to relate the chemical composition across all digestion stages to the observed antioxidant capacity. The Minitab 17 statistical software was used.

3. Results and Discussion

3.1. Effects of Cooking on the Chemical and Phytochemical Composition of Huitlacoche

The composition of huitlacoche was compared before (raw) and after cooking (cooked) (Table 1). Cooking reduced the lipid and protein contents of huitlacoche by 25% and 5%, respectively. Carbohydrates were the major component in both raw and cooked huitlacoche, with TDF ranging from 45.2% to 49.5% in raw and cooked huitlacoche, respectively. Observed results for IDF increase could be explained by their proportion increase, other than the amount, as reports for several fungi have shown that in the absence of water and depending on the fungus species, certain soluble fractions, such as  α‐ and β‐glucans, can be reduced (Roncero‐Ramos et al. 2017). Colorectal cancer can be prevented by consuming adequate amounts of fermentable and nondigestible carbohydrates, such as fiber. On the basis of our results, cooked huitlacoche (100 g on a dry basis) contributes ∼30 g TDF to the diet. Some studies suggest that higher IDF levels can improve microbial diversity and promote the activity of butyrogenic bacteria (Zhang et al. 2024).

TABLE 1.

Composition and antioxidant capacity of raw and cooked huitlacoche.

	Raw	Cooked
Chemical composition (%)
Moisture	7.50 ± 0.11	9.50 ± 0.25*
Ash	5.30 ± 0.16*	5.20 ± 0.04
Lipids	1.60 ± 0.03*	1.20 ± 0.01
Protein	10.60 ± 0.11*	10.10 ± 0.16
Carbohydrates	74.90 ± 0.06*	74.00 ± 0.13
Total dietary fiber	45.20 ± 2.29	49.50 ± 1.35*
Soluble dietary fiber	7.60 ± 2.21*	4.10 ± 0.75
Insoluble dietary fiber	37.60 ± 0.14	45.50 ± 0.62*
Phytochemical composition
Total phenolic content (mg GAE/g)
Free	1.09 ± 0.14*	0.36 ± 0.02
Bound	1.83 ± 0.27	1.52 ± 0.30
Total	2.92 ± 0.33*	1.88 ± 0.28
Total flavonoid content (mg RE/g)
Free	0.45 ± 0.05*	0.20 ± 0.02
Bound	1.33 ± 0.11*	1.13 ± 0.11
Total	1.78 ± 0.13*	1.33 ± 0.11
Total condensed tannins (mg CE/g)
Free	0.19 ± 0.02*	0.07 ± 0.01
Bound	0.52 ± 0.02	0.49 ± 0.13
Total	0.71 ± 0.01*	0.56 ± 0.05
Antioxidant capacity (TE/g)
DPPH
Free	8.08 ± 0.48*0	5.30 ± 0.3000
Bound	6.19 ± 0.3300	7.77 ± 0.63*
Total	14.27 ± 1.05*	13.08 ± 0.52
FRAP
Free	0.76 ± 0.03*	0.55 ± 0.06
Bound	7.55 ± 0.97	7.28 ± 1.04
Total	7.98 ± 1.04	7.82 ± 1.08

Note: Data are expressed as mean ± standard deviation of three independent assays.

Abbreviations: CE, (+)‐catechin equivalents; DPPH, 2,2‐diphenyl‐1‐picrylhydrazyl; FRAP, ferric reducing antioxidant power; GAE, gallic acid equivalents; RE, rutin equivalents; TE, Trolox equivalents.

*Indicate significant differences per line (Student's t‐test, p < 0.05).

Regarding the phytochemical composition of raw and cooked huitlacoche, bound total phenolic compounds showed higher concentrations (12%–22% decrease after cooking) than free compounds in both raw and cooked samples, suggesting that most of these compounds are bound to macromolecules, such as polysaccharides and proteins, present in the huitlacoche cell wall (Tu et al. 2021). The observed differences in total phenolic compounds in this study are like those reported by Valdez‐Morales et al. (2016). These results may vary depending on the region of origin of huitlacoche, the type of maize used, the cooking process, and the extraction method. In addition, the levels of total phenolics (1.88–2.92 mg GAE/g) and flavonoids (1.33–1.78 mg RE/g) found in huitlacoche are consistent with those reported for other edible fungi (Cantharellus symoensii) and fall within known ranges (Assemie and Abaya 2022).

Importantly, these results align with the current literature, which reports that U. maydis contains polysaccharides, such as β‐(1→3)‐glucans, with biological activity. Recent studies demonstrated that β‐glucans extracted from U. maydis display antioxidant, immunomodulatory, and cytotoxic properties against human cells, with an LC₅₀ of ∼291 µg/mL in endothelial cells and significant reductions in reactive oxygen species (ROS) levels in vitro (Luna‐Cruz et al. 2025). These findings support that the phytochemicals and cell‐wall–bound components of huitlacoche not only contribute to its antioxidant capacity but may also underlie the cytotoxic effects observed in colon cancer cell assays.

Although the thermal sensitivity might explain reductions in the phytochemical composition (Antony and Farid 2022), the presence of phenolics, along with β‐glucans and other polysaccharides, reinforces the potential of cooked huitlacoche as a functional food with antioxidant, anticancer, and prebiotic properties (Luna‐Cruz et al. 2025; Villagrán et al. 2023).

The antioxidant capacity of huitlacoche, measured by DPPH and FRAP methods, decreased slightly in cooked compared to raw samples (8% lower in DPPH and 2% lower in FRAP) (Table 1). Antioxidant capacity results agree with previous studies, which showed that thermal processing can reduce antioxidant capacity due to degradation of phenolic compounds (Yang et al. 2019). Antioxidant capacity quantified by the FRAP method differed only in the free fraction, which was higher in the raw samples, likely due to the loss of phenolic compounds from the free fraction during cooking (Antony and Farid 2022).

3.2. In Vitro Bioaccessibility of Phenolic Compounds in Huitlacoche and Its Antioxidant Capacity

Phenolic compounds (phenolic acids, flavonoids, and CTs) are known for their beneficial health properties. In this study, bioaccessibility of these compounds was quantified at the different stages of gastrointestinal digestion (Figure 1). During simulated digestion, the concentration of total phenolic compounds and CTs was higher at the oral stage compared to the non‐digested huitlacoche (NDH or raw samples) (2.4 vs. 1.58 mg/g for phenolics and 7.83 vs. 0.56 mg/g for tannins, respectively) (Figure 1A,B). This increase could be potentially attributed to the action of α‐amylase, which hydrolyzes starch and other complex carbohydrates, releasing phenolic compounds previously bound to the food matrix (Tu et al. 2021). However, because α‐amylase activity was not quantified in the saliva samples used in this study, further investigation is needed to validate these claims.

FIGURE 1.

Bioaccessibility of (A) total phenolic content, (B) total flavonoid content, (C) total condensed tannins, and (D) antioxidant capacity (DPPH and FRAP assay) of cooked huitlacoche during in vitro gastrointestinal digestion. Values are presented as the mean ± standard deviation of at least two independent assays. Different letters indicate significant differences between the digestion stage (Tukey, p < 0.05). CE, (+)‐catechin equivalents; DPPH, 2,2‐diphenyl‐1‐picrylhydrazyl; FRAP, ferric reducing antioxidant capacity; GAE, gallic acid equivalents; NDH, non‐digested huitlacoche; RE, rutin equivalents; TE, Trolox equivalents.

For flavonoids, the highest concentration was observed at the gastric stage (4.48 mg/g), likely due to pepsin activity under acidic conditions (pH 2), which facilitates their release from protein complexes (Wang et al. 2023). The primary flavonoid detected in this phase was epicatechin (3.0 mg/g). Interestingly, this compound is not typically detected in fungal sources. However, as U. maydis is a parasitic fungus, the phenolic profile of the galls also depends on that of the inoculated maize host (Valdez‐Morales et al. 2016).

Total phenolic compounds were undetected in the gastric and intestinal stages. This suggests potential interactions with fiber components such as starch and cellulose in huitlacoche, forming noncovalent bonds that reduce their solubility and detection (Domínguez‐Avila et al. 2017). These compounds are again detected in the colonic stage, where microbial fermentation breaks down dietary fiber, releasing bound phenolics (Cortés‐Martín et al. 2020).

Interestingly, CTs were undetected during colonic fermentation at 6 and 12 h. This absence may be due to microbiota‐mediated degradation. Tannins, particularly flavan‐3‐ol dimers (A and B), undergo microbial cleavage of the C‐ring, yielding simpler phenolic metabolites, such as dihydroxyphenylpropionic and dihydroxyphenylacetic acids (Sallam et al. 2021). A similar process occurs with flavonoids: the colonic pH decreases during fermentation, affecting compound solubility and stability, while microbial activity degrades them into simpler phenolics. This degradation is influenced by the compound concentration in the initial sample, the duration of fermentation, and the composition of the microbiota (Reddivari et al. 2016).

The antioxidant capacity, primarily attributed to phenolic compounds, was also assessed along the digestive process using the DPPH and FRAP methods (Figure 1D). However, the antioxidant effects of huitlacoche may also be attributed to other bioactive components, such as polysaccharides, depending on their solubility, sugar ring structure, molecular weight, presence of positively or negatively charged groups, protein moieties, and covalently linked phenolic compounds (Fernandes and Coimbra 2023; Villagrán et al. 2023). Other compounds found in huitlacoche exhibiting antioxidant capacity include ergothioneine, which is highly accumulated in tissues susceptible to oxidative damage and can prevent the formation of free radicals, directly scavenge ROS, or interact with other natural antioxidant defenses by activating antioxidant pathways involving the mitogen‐activated protein kinases (MAPKs) and regulating the levels of peroxidases and superoxide dismutases (Fu and Shen 2022).

DPPH antioxidant capacity significantly increased (11% higher) during the gastric stage than the oral stage. However, this was followed by a considerable reduction (62%) at the beginning of the intestinal stage. FRAP antioxidant capacity varied across intestinal time points, peaking at 30 min. During the colonic phase, differences were insignificant across fermentation times, although a reduction (up to 98%) occurred at the start of this stage compared to the intestinal phase. This reduction may be due to microbial degradation of antioxidant compounds (Luzardo‐Ocampo et al. 2017). These findings differ from those of López‐Martínez et al. (2022), who observed a 130% increase in antioxidant capacity at the gastric stage using both concentrated and unconcentrated huitlacoche ethanolic extracts. The discrepancy may stem from differences in digestion models and sample preparation, but more importantly, our study digested the whole food matrix for the first time.

Additionally, selected phenolic compounds were identified and quantified at each stage of huitlacoche digestion. Bioaccessibility of these compounds increased during the oral stage (Table 2). The identification of compounds at the digestion stages other than the raw samples (e.g., dihydroxybenzoic acid or ferulic acid at the 24‐h‐colonic stage) indicates their release following enzymatic or chemical degradation of other complex compounds under simulated digestive conditions (Kasprzak‐Drozd et al. 2024).

TABLE 2.

Bioaccessibility of phenolic compounds during in vitro gastrointestinal digestion of cooked huitlacoche.

Digestion phase	Benzoic acid	Dihydroxybenzoic acid	4‐Hydroxybenzoic acid	Ferulic acid	Gallic acid	Ellagic acid	Epicatechin
Raw	0.003 ± 0.00	—	0.01 ± 0.00	—	0.2 ± 0.01	—	3.0 ± 0.02
Oral	—	0.2 ± 0.00b	0.5 ± 0.01a	—	0.3 ± 0.0b	—	4.1 ± 0.0c
Gastric	0.8 ± 0.02ab	—	0.2 ± 0.00b	—		—	10.5 ± 0.2a
Small intestine (min)
15	—	—	0.1 ± 0.00c	—	0.8 ± 0.00b (400)	—	5.7 ± 0.00b
30	0.6 ± 0.03b	—	0.1 ± 0.00c	—	1.0 ± 0.00b (500)	—	5.0 ± 0.00d
60	1.0 ± 0.08a	—	0.1 ± 0.00c	—	0.9 ± 0.00b (450)	—	2.1 ± 0.00e
Colon (h)
6	—	0.5 ± 0.00a	—	—	0.003 ± 0.00d	—	—
12	—	0.3 ± 0.00b	—	—	0.01 ± 0.00c	—	—
24	—	0.1 ± 0.00c	—	1.8 ± 0.00	0.004 ± 0.00d	—	—

Note: Values represent the mean ± standard deviation of at least two independent experiments. Values are expressed as mg/g of sample. Bioaccessibility values (%) are presented in parentheses. Different letters indicate significant differences within each column and across digestion stages (Tukey, p < 0.05). Dashes in the table indicate that the compound was not detected.

Benzoic acid, dihydroxybenzoic acid, 4‐dihydroxybenzoic acid, gallic acid, and epicatechin were detected across most digestive stages. These phenolic compounds exert anticancer properties such as reduction of cancer cell metabolic activity, cell growth inhibition, and activation of apoptotic pathways, including activation of caspase‐3, cell cycle arrest in the G2/M phase, histone deacetylase enzymes, and induction of autophagy (Anantharaju et al. 2017; Chen et al. 2023; Forester et al. 2014; Gadelmawla et al. 2022). Additionally, both gallic and ferulic acids have demonstrated prebiotic potential by forming polysaccharide complexes, such as starch, which serve as substrates for probiotic bacteria like Lactobacillus rhamnosus and Bifidobacterium bifidum, thereby promoting their growth (Cai et al. 2024).

3.3. pH Change and SCFAs Production During In Vitro Colonic Fermentation

The colonic fermentation of NDF‐H induced pH variations (16%) after 24‐h fermentation, compared with a 37% decrease in the positive control with raffinose, a fermentable carbohydrate (Figure 2A). This reduction in pH suggests fermentative microbiota activity on the dietary fiber of huitlacoche, which favors the production of SCFAs (Figure 2B). This pH reduction was greater than that reported for other raffinose‐rich foods, such as common bean (Campos‐Vega et al. 2009).

FIGURE 2.

pH reduction and short‐chain fatty acids production during in vitro colonic fermentation of the nondigestible fraction from huitlacoche. Values are presented as the mean ± standard deviation of two independent measurements (Tukey, p < 0.05). In (A), the capital letters indicate significant differences (p < 0.05) between treatments, whereas the noncapital letters indicate significant differences (p < 0.05) between fermentation times. The differences were assessed using Tukey–Kramer's test. Control (−), negative control (intestinal solution); control (+), positive control (100 mg raffinose); NDF‐H, nondigestible fraction of huitlacoche. In (B), different letters indicate significant differences in fermentation time within each treatment.

Under low pH conditions (4–6), probiotic bacteria not only inhibit the growth of pathogenic bacteria but also produce postbiotic metabolites, such as SCFAs (propionic, acetic, and butyric acids) (Wang et al. 2020; Xie et al. 2023). The production and ratio of these SCFAs depend on the available substrate type, impacting their physiological functions. The analysis of SCFAs produced during colonic fermentation (Figure 2B) showed acetic, propionic, and butyric acids concentrations of 2.68, 5.20, and 0.23 mmol/L, respectively, in the first hours of fermentation (6 h). Both propionic and acetic acid decreased, whereas butyric acid concentration increased linearly over the fermentation period, following the equation Y = 0.159x − 0.708 (r ² = 0.9998). This can be attributed to cross‐fermentation, in which butyrogenic bacteria convert acetic acid (produced by acetic bacteria) into acetyl‐CoA, which is then enzymatically converted into butyric acid (Bedu‐Ferrari et al. 2022).

Moreover, after 24 h of fermentation, the concentration of SCFAs increased (3.06, 7.08, and 3.11 mmol/L for acetic, propionic, and butyric, respectively), with the largest increase in propionic acid. These values are within the range, or even higher for propionic and butyric acids, when the amount of fermented sample is taken into account, as reported by Hu et al. (2023), who evaluated a polysaccharide extract (100 mg) from straw mushroom (Volvariella volvacea). These authors also report similar amounts of these SCFAs from a fructo‐oligosaccharide supplement, with slight variations.

Previous studies have demonstrated that higher concentrations of certain fatty acids are attributable to factors such as substrate composition, the specific microbiota present, and environmental conditions, including pH (Ashaolu et al. 2021). β‐Glucans from mushrooms are able to modulate colonic microbiota, particularly Lactobacillus spp. and Bifidobacterium spp., which are producers of those SCFAs (Moniruzzaman et al. 2025). Furthermore, although the amounts of SCFAs produced seem low, only 5.10, 0.68, and 1.19 mmol/mL of acetic, propionic, and butyric acids are needed to inhibit the survival of human colon cancer cells in vitro (Campos‐Vega et al. 2012).

Acetic acid is converted to acetyl‐CoA, which plays a critical role in energy distribution and acts as a metabolic regulator, influencing gene transcription and protein function (Moffett et al. 2020). Propionic acid helps suppress lipogenesis, reduces hepatic triglyceride levels, improves insulin sensitivity, and participates in the formation of complex fatty acids (Weitkunat et al. 2016). Butyric acid is mainly an energy source for intestinal cells. It promotes cell differentiation, induces apoptosis, improves epithelial barrier function, increases mucin expression, and maintains colonocyte integrity (Banasiewicz et al. 2020).

These metabolites play a key role in maintaining intestinal health and regulating metabolism, highlighting the importance of colonic fermentation and SCFAs production in human health. The results of pH reduction and SCFA production indicate that NDF‐H is fermentable by the colonic microbiota.

3.4. Effect of FE‐H on the Survival and LDH and Caspase 3/7 Activity of HT‐29 Cells

The survival of human colon cancer cells (HT‐29) was evaluated by exposing them to different doses of 24 h FE‐H. Initially, cells were treated with concentrations of 5%–35% and showed morphological changes after 24 h of incubation compared to untreated cells. However, when the cells were tested using the MTT assay, the highest FE‐H concentration significantly decreased cell survival, but only 7% (Figure 3A), while the lowest concentration (5%) improved cell survival (14%) compared to the control. The concentration and exposure time of the cells to the extract were increased based on this result. This time, FE‐H had a cytotoxic effect on HT‐29 cells at 6%–50% FE‐H concentrations for 48 h (Figure 3B). Metabolic activity was reduced by up to 72% at the highest concentration compared with the control.

FIGURE 3.

Impact of the colonic fermentation extract on the metabolic activity of HT‐29 cells at (A) 24 h and (B) 48 h of exposure. Effect of the 48 h half‐lethal concentration of the colonic extract (LC₅₀‐FE‐H) on lactate dehydrogenase (LDH) activity in (C) Hs 738 St./Int and (D) HT‐29 cells; impact of the 48 h half‐lethal concentration of the colonic extract (LC₅₀‐FE‐H) on caspase 3/7 activity in (E) Hs 738 St./Int and (F) HT‐29 cells. The values are presented as the mean ± standard deviation of at least two independent assays. Different letters indicate significant differences between FE‐H doses (A and B) or treatments (C–F) (Tukey, p < 0.05). Control (−), untreated cells (cells grown in 2% FBS + DMEM); control (+), positive control (50 µM H₂O₂); FE, fermented extract of huitlacoche; MRC, maximum LDH release control (provided lysis buffer from the caspase 3/assay kit (75 µL Tris, pH 7.5, 2 M NaCl, 20 mM ethylenediaminetetraacetic acid, 0.2% TRITON X‐100 for each 1 mL of lysis buffer).

The LC₅₀ of FE‐H was 33.34%. Despite the challenge of translating observed in vitro doses to in vivo or clinical trials, previous research from our group suggests that in vitro doses may translate to similar effects in vivo. For instance, 1 g of digested Moringa oleifera leaves displayed pro‐apoptotic, necrotic, and cytotoxic effects in HT‐29 cells (LC₅₀: 5.9%–71.70% v/v) (Caicedo‐Lopez et al. 2021). When doses ranging from 0.25 to 1.0 g of powdered leaves were administered to azoxymethane/dextran sodium sulfate‐induced colorectal cancer in CD‐1 mice, observable macroscopic and genomic effects were observed in the animals, with the 1.0 g dose being the most effective at inducing apoptosis in cancer cells, reducing polyps, and improving the overall health of the animals (Cuellar‐Núñez et al. 2021). Moreover, the conducted digestion and colonic fermentation model effectively recapitulates several physiological effects to which samples are usually subjected during digestion, potentially producing more similar physiological and biological effects when assayed in in vitro models.

The metabolic activity at the lowest FE‐H dose (6% and 47%) was similar to the inhibition at LC₅₀. The metabolic activity fluctuated with increasing dose, suggesting that FE‐H was cytotoxic at both low and high concentrations, a pattern indicative of hormesis. This concept implies that the cytotoxic response depends not only on the dose but also on the exposure time, suggesting that although there is an initial disruption of cellular homeostasis, cells undergo an adaptive or reparative process that restores their activity at doses below the LC₅₀ (Calabrese 2008).

The FE‐H used in this treatment contains compounds that can affect HT‐29 metabolic activity and proliferation. Among them are SCFAs, such as propionic acid (2.36 mmol/L) and butyric acid (1.04 mmol/L); dihydroxybenzoic acid (0.1 mg/g); ferulic acid (1.8 mg/g); and CTs (0.01 mg rutin equivalent/g/g) present in LC₅₀‐FE‐H. These compounds have previously been shown to exert cytotoxic, pro‐apoptotic, and anti‐inflammatory effects in HT‐29 cells at doses similar to those found in FE‐H, which explains our results (Comalada et al. 2006; Ribas et al. 2021). Cytotoxicity results suggest that, given the composition of FE‐H, some huitlacoche compounds may be released upon reaching the colon stage. These compounds may also promote the production of SCFAs and possibly other substances, which can synergistically exert cytotoxic effects, likely through potential apoptosis, on human colon cancer HT‐29 cells. This suggests a chemopreventive effect by reducing their proliferation. However, additional studies are needed to investigate the type of cell death induced by FE‐H in these cells, its effects on noncancerous colon cell lines, and its mechanism of action.

To elucidate the impact of FE‐H on specific cell death mechanisms, the activities of LDH (Figure 3C) and caspase 3/7 (Figure 3D) were measured. A non‐tumoral cell line was included in the assays, but a previous MTT assay was conducted in both Hs 738 Int./St and HT‐29 cells, to discard cytotoxic effects from the 48‐h LC₅₀‐FE‐H and its blank (Figure S1), where none of the treatments was cytotoxic (metabolic activity >80%). As observed in the results, a low cytotoxic effect was observed in the non‐tumoral cells, with the blank and the LC₅₀ showing LDH (Figure 3C) and caspase 3/7 (Figure 3E) activities comparable to those of the untreated cells (negative control). A different behavior was observed in HT‐29 cells, with significantly higher LDH (Figure 3E) and caspase‐3/7 (Figure 3F) activities at the 48‐h LC₅₀‐FE‐H, reaching levels comparable to the positive control (H₂O₂). The observed results suggested that colonic metabolites from huitlacoche induced LDH release to the extracellular medium, suggesting a necrotic effect, yet it has been reported that the LDH measurement is not a complete indicator of necrosis, and additional assays, such as Annexin V‐Propidium iodide (PI) staining, would render better results in identifying late‐stage apoptosis or necrotic cells (Cummings and Schnellmann 2021). Additionally, caspase 3/7 activity, although widely used as a biochemical indicator of apoptosis, does not fully characterize the type of apoptotic process; however, its combination with LDH provides information on cell death mechanisms. As observed in this research, increased caspase 3/7 and LDH activities are both indicators of secondary necrosis, suggesting loss of plasma integrity and release of intracellular components even after the apoptotic program has been activated (Galluzzi et al. 2018).

Although the effects of the colonic metabolites of U. maydis on colon cancer cells have not been investigated, several of the compounds found as digested metabolites have been associated with necrotic or pro‐apoptotic effects. For instance, gallic acid is a common metabolite in Leccinum vulpinum, Ramaria patagonica, or Pleurotus ostreatus (Yu et al. 2025), and reports from our research group have found that this phenolic acid, together with other compounds found in this report, such as ferulic acid, ellagic acid, and epicatechin, when present as colonic metabolites, are effective LDH and pro‐apoptotic inducers in HT‐29 cells (Caicedo‐Lopez et al. 2021).

3.5. Effect of NDF on Lactobacillus gasseri Growth

NDF was used as a substrate to evaluate L. gasseri growth. Figure 4 shows bacterial growth in the presence of different NDF concentrations: 1% NDF (solid + liquid fractions, LS‐NDF), 1% NDF, and 2% NDF (solid fraction), with dextrose as a positive control for reference. OD increased after 6 h in all three cases, similar to the dextrose control. The 2% NDF treatment showed the highest OD (p < 0.05), followed by the 1% NDF and solid–liquid NDF treatments, with no difference (p > 0.05) between the latter two.

FIGURE 4.

Assessment of the prebiotic effect of the nondigestible fraction of huitlacoche (NDF‐H) on Lactobacillus gasseri growth. The values are presented as the mean ± standard deviation of at least two independent assays. Different letters indicate significant differences between treatments at each time. Control (+), positive control (dextrose); LS‐NDF, liquid and solid fraction of the nondigestible fraction of huitlacoche; NDF‐H, nondigestible fraction of huitlacoche (liquid fraction); OD, optical density.

In addition, the growth curve was fitted to the mathematical model proposed by Baranyi and Roberts (1994) to determine three bacterial growth parameters: lag phase duration (the time required for the bacteria to adapt to the substrate), maximum growth rate (the maximum rate at which the bacteria grow during the exponential phase), and final growth value (the maximum number of bacteria, represented by the highest O.D., before growth ceases). Table 3 shows the results for these three parameters across the three NDF concentrations, and no significant differences (p > 0.05) were found between treatments during the evaluation period. Although the NDF 1% treatment showed greater variability in maximum growth rate, this was attributed to one replicate with an atypically elevated value and was considered experimental variation rather than a consistent effect. Overall, these findings suggest that L. gasseri growth is not affected by increased substrate availability or by experimental conditions that led to metabolic saturation, indicating that the bacteria reached their maximum potential for substrate metabolism within the evaluated time frame (Hellweger 2018). However, additional assays or modifications to the experiment, such as longer exposure times or changes in cultivation techniques, are needed to confirm this.

TABLE 3.

Bacterial growth parameters for Lactobacillus gasseri.

	LSNDF	NDF (1%)	NDF (2%)
Lag phase duration	6.532 ± 0.08a	6.688 ± 0.42a	6.233 ± 0.43a
Final growth value	0.115 ± 0.00a	0.110 ± 0.00a	0.123 ± 0.00a
Maximum growth rate	0.032 ± 0.00a	0.107 ± 0.08a	0.033 ± 0.00a

Note: Data are expressed as the mean ± standard deviation of three independent assays. Different letters indicate significant differences per line (Tukey, p < 0.05).

Abbreviations: LSNDF, liquid and solid nondigestible fraction; NDF (1%), 1% nondigestible fraction of huitlacoche; NDF (2%), 2% nondigestible fraction of huitlacoche.

L. gasseri growth is stimulated by fructooligosaccharides, galactooligosaccharides, and inulin (Pan et al. 2020). Although not assessed in this study, huitlacoche contains raffinose‐family oligosaccharides that can potentially serve as fermentable substrates for L. gasseri (Beas et al. 2011). Furthermore, Lactobacillus acidophilus, a related species, exhibits enhanced growth in the presence of dietary fiber (soluble/insoluble) and phenolic compounds, both of which are components of huitlacoche. Combined with the observed growth kinetics and evaluated parameters, these findings demonstrate that L. gasseri can utilize huitlacoche's NDF as a substrate.

Finally, in view of the criteria for defining food as prebiotic proposed by Gibson and Rastall (2006) and the results of this study, the following conclusion can be drawn: the components of huitlacoche that are resistant to digestion are fermentable by L. gasseri. This suggests that they can lower pH, metabolize compounds, produce new metabolites, and stimulate bacterial growth during the evaluated period, indicating a prebiotic potential. Nonetheless, additional research is required to determine whether NDF can selectively stimulate the growth of probiotic bacteria while reducing or inhibiting pathogen growth. Additionally, although a consortium of bacteria representative of the colonic microbiota could be used to further assess the potential of FE‐H to modulate probiotic bacteria, L. gasseri is a well‐established human‐origin probiotic bacterium that serves as an appropriate model organism for the initial screening of the prebiotic activity, as the selective stimulation of beneficial microorganisms is a defining characteristic of prebiotics (Gibson et al. 2017). This approach has been used by our research group to assess in vitro the effects of prebiotics from food matrices on the promotion of Limosilactobacillus reuteri, Limosilactobacillus helveticus, and L. rhamnosus (Luzardo‐Ocampo et al. 2025). These results were further confirmed in vitro, when the same food matrix modulated the abundance of bacterial genera associated with well‐known prebiotics, such as Lactobacillus and Ligilactobacillus (Barbosa et al. 2025).

3.6. PCA for the Release of Selected Phenolic Compounds and Antioxidant Capacity, Along With the Digestion Process

Figure 5 shows the PCA analysis of the spectrophotometric and selected free phenolic compounds in the digested samples, as well as their impact on antioxidant capacity. For this analysis, two PCAs were conducted (referred to as “conditions”): one for the oral‐gastric‐small‐intestinal conditions and one for colonic fermentation. This separation was made because additional variables not measured in the previous steps were considered in colonic fermentation, such as pH and SCFA quantification. The PCA analysis indicated that up to three components explained >80% of the total variation for both conditions (oral‐gastric‐small intestine and colonic fermentation) (Tables S1 and S2). Different clusters were observed in the PCA analysis for both conditions, as shown in the scatter and loading plots for all variables (Figure 5A,B). For the first condition (oral‐gastric‐small intestine) (Figure 5A), the oral conditions were mostly impacted by epicatechin, hydroxybenzoic acid, CTs, and total flavonoids. Gastric conditions were primarily affected by DPPH and benzoic acid, and the NDFs were primarily influenced by gallic acid and FRAP antioxidant capacity. For the second condition (colonic fermentation), samples at 6 h were primarily related to pH changes, CTs, and total flavonoids; samples at 12 h were influenced by gallic acid, FRAP, and 4‐hydroxybenzoic acid; finally, samples at 24 h varied in SCFAs content and total phenolic compounds.

FIGURE 5.

Principal components analysis (PCA) for the chemical composition of huitlacoche, along with the digestive process and the observed antioxidant capacity. Scatter and loading plots for the first and second components of the (A) oral‐gastric‐small intestine digestion; (B) colonic fermentation. (C) Spearman's correlation analysis between the chemical composition and the antioxidant capacity methods at the (A) oral‐gastric‐small intestine digestion and (B) colonic fermentation. 4‐DiHBenz, 4‐hydroxybenzoic acid; Acetic, acetic acid; Benz, benzoic acid; Buty, butyric acid; CT, condensed tannins; DiHBenz, dihydroxybenzoic acid; DPPH, 2,2‐diphenyl‐1‐picrylhydrazil; Epic, epicatechin; FRAP, ferric reducing antioxidant power; Gal, gallic acid; HPLC‐DAD, high‐performance liquid chromatography coupled to a diode array detector; PC, principal component; Prop, propionic acid; SCFAs, short‐chain fatty acids; TF, total flavonoids; TPC, total phenolic compounds.

4. Conclusion

The thermal treatment of huitlacoche increased IDF and antioxidant capacity, while reducing total phenolic compounds. Notably, epicatechin was identified as the major compound in both raw and cooked huitlacoche, a finding not previously reported. So far, the bioaccessibility of phenolic compounds from huitlacoche has not been documented, nor has the impact of bioaccessibility on antioxidant capacity during in vitro gastrointestinal digestion. This study found that digestion increases the bioaccessibility of total phenolic compounds, flavonoids, and tannins during the oral stage, with a progressive decline in subsequent stages of digestion. Some phenolic compounds remained bioaccessible in the colon, where they continued to exert antioxidant capacity throughout gastrointestinal digestion. Additionally, the ability of huitlacoche to lower pH and stimulate the production of SCFAs—including propionic, acetic, and butyric acids, with propionic acid at the highest concentration—indicates that huitlacoche is fermentable by colonic microbiota. The FE‐H reduced the metabolic activity of HT‐29 colon cancer cells to the LC₅₀ (33.34%), likely due to the combined necrotic and caspase‐3/7‐activating effects of SCFAs combined with phenolic compounds. This interaction exhibited a hormesis‐like effect on HT‐29 cells. Moreover, certain components of huitlacoche appear resistant to digestion and fermentable by colonic microbiota, and they can stimulate L. gasseri growth, suggesting their prebiotic potential. The results serve as a foundation for further research into huitlacoche's potential to prevent colon cancer and other chronic degenerative diseases, while also highlighting its nutritional benefits and encouraging its consumption, as this is the first time huitlacoche has been evaluated as a potential functional food.

As a limitation of this study, during the intestinal phase an artificial membrane was used, which limited how compounds were absorbed in the small intestine, suggesting the need for more sophisticated models (e.g., ex vivo approaches or the polarized Caco‐2 cell model) to better mimic intestinal digestion. Moreover, the way dietary fiber fractions change during huitlacoche, suggesting not only changes in proportions but also potential molecular complexation, warrants further merit. Research investigating how gut microbiota composition may change during colonic fermentation in this model, or through in vivo or clinical trial approaches, may help elucidate microbiota modulation to produce beneficial metabolites that contribute to the observed biological effects. Finally, a deeper study of cell death mechanisms using proteomics or genomics approaches would better elucidate the specific pathways by which the biological compounds of huitlacoche act.

Author Contributions

Angélica Rebollar‐García: investigation, formal analysis, data curation, writing – original draft, visualization, methodology. Ivan Luzardo‐Ocampo: methodology, validation, visualization, writing – review and editing, formal analysis, supervision, resources, data curation, funding acquisition. Sofía Arvizu‐Medrano: writing – review and editing, methodology, resources, investigation, supervision. Héctor Pool: investigation, methodology, resources, writing – review and editing, supervision. Guadalupe Loarca‐Piña: resources, methodology, visualization, investigation, supervision. Dafne Velásquez‐Jiménez: investigation, methodology, validation, visualization, supervision. María Guadalupe Herrera‐Hernández: methodology, investigation, formal analysis, supervision, resources. Vanessa Sánchez‐Quezada: methodology, investigation, validation, formal analysis. Rocio Campos‐Vega: conceptualization, investigation, funding acquisition, writing – review and editing, project administration, supervision, methodology.

Ethics Statement

Ethical review committee approval was acquired.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supplementary Material: jfds71365‐sup‐0001‐SuppMat.docx