The in vitro assay included four toxin binder treatments: 100% bentonite (BEN), 100% yeast cell wall (YEAST), 50% bentonite + 50% yeast cell wall (BEN+Y), and 47.5% bentonite + 47.5% yeast cell wall + 3% curcumin + 2% cinnamon (COMB). The toxin binder ingredients used in this study were calcium bentonite (PT. Fenanza Putra Perkasa, Jakarta, Indonesia) with 61.5% SiO2 and 17.1% Al2O3, yeast cell wall (PT. Behn Meyer Chemical, Tangerang, Indonesia) with 30% β-glucan and 17% mannan, curcumin (Curcuma xanthorrihiza) dry extract (PT. Phytochemindo Reksa, Jakarta, Indonesia) with 3.26% curcuminoid, and cinnamon (Cinnamomum burmani) dry extract (PT. Phytochemindo Reksa, Jakarta, Indonesia) with 15% cinnamaldehyde.

An in vitro procedure was modified according to the procedure used by Kong et al. (2014)(Kong et al., 2014) and Hojati et al. (2021)(Hojati et al., 2021). A total of 1 mL of aflatoxin standard solution (Biopure, Tulln, Austria) at a concentration of 2000 ppb was diluted with 7 mL of acetonitrile to obtain an AFB₁ standard solution of 250 ppb. Then, 2.5 mL of the AFB₁ standard solution was mixed with 1.25 mL of phosphate buffer (pH 6.8) and 9.1 mg of toxin binder. Subsequently, 150 μL of 1 M HCl was added, and the mixture was incubated for 2 hours at 39 °C in a shaker incubator. Following this, 500 μL of phosphate buffer (pH 6.8) and 150 μL of 1 M NaOH were added, and the mixture was incubated again for 4 hours in the shaker incubator. The mixture was then centrifuged at 2490 xg for 10 minutes, and the resulting supernatant was collected to determine the residual AFB₁ concentration using the AgraQuant^® enzyme-linked immunosorbent assay (ELISA; Romer Labs, Singapore). Each essay on binding capacity for toxin binders had four replicates. The binding capacity of the toxin binder was calculated using the following equation:

Binding capacity = [(Initial AFB₁ concentration - Final AFB₁ concentration) / Initial AFB₁ concentration] x 100%

The experimental procedures were approved by the Research Ethics Committee of the Faculty of Veterinary Medicine, Universitas Gadjah Mada, Yogyakarta, Indonesia (No. 00013/EC-FKH Eks./2021). A total of 288 one-day-old male broiler chicks (New Lohmann Indian River, MB 202 Platinum) from a commercial hatchery (PT. Widodo Makmur Unggas Tbk, Yogyakarta) were used in this study. The day-old chicks were vaccinated against Newcastle disease (ND) and infectious bursal disease (IBD). The birds were weighed and randomly distributed into 36 pens (1 m × 0.50 m).

This study involved six dietary treatments: CTRL (basal diet), AF (basal diet + 300 ppb AFB₁), CTRL-A (basal diet + BEN+Y), CTRL-B (basal diet + COMB), AF-A (AF + BEN+Y), and AF-B (AF + COMB). Each treatment consisted of six replicates, with eight birds per replicate. Aspergillus flavus FNCC 6122 was cultured on potato dextrose agar (PDA) medium according to the procedure of Anas et al. (2022)(Anas et al., 2022). Rice media were prepared for fungal propagation to produce crude AFB₁. Fungus-colonized substrate was dried at 55 °C, and the toxin concentration was quantified using an ELISA kit (Romer Labs, Singapore). AFB₁ analysis was conducted on a 2 g sample of contaminated rice, which was extracted with 10 mL of 70% methanol (v/v=1:5), vortexed for 3 min, and filtered through Whatman No. 1 paper. The filtrate was analyzed using an ELISA reader, and the resulting AFB₁ concentration was used to calculate the proportion of contaminated rice required to obtain a dietary AFB₁ level of 300 ppb. Dietary treatments were formulated according to Aviagen (2022)(Aviagen, 2022) recommendations (Table 1) and administered from days 1 to 10 (starter) and days 11 to 21 (grower).

Broiler management followed the guidelines of the Indian River Broiler Management Handbook(Aviagen, 2018). The broiler house was fumigated with 4% formalin one week before the trial to ensure cleanliness. The temperature was maintained at 30 °C for the first 3 days and then gradually reduced by 2.5 °C per week until it reached 22 °C (day 21). Birds received 23 hours of light (30–40 lux) and 1 hour of darkness (<0.4 lux) to ensure optimal early feed intake, drinking activity, growth, health, and welfare from days 0 to 7. From day 7 onward, the lighting program was adjusted to include approximately 5 hours of darkness (4–6 hours), which has been shown to prevent abnormal eating and drinking behaviors associated with sleep deprivation. Light intensity was reduced to 5–10 lux during this period.

Feed intake was measured by subtracting the residual feed (g) from the total feed offered (g) before each feeding. Feed intake per bird was calculated by dividing the total feed intake of each replicate by the number of birds in that replicate. Feed intake was recorded for days 1–11, 12–21, and 1-21. Individual body weights were measured weekly before morning feeding on days 11 and 21. Average daily gain (ADG) was calculated as the difference between initial and final BW divided by the number of days.

On day 21, one bird per replicate (a total of 36 birds) with body weights close to the median for each group was selected. Birds were euthanized by decapitation, followed by cutting of the jugular vein after a two-hour fasting period. Blood was collected in plain tubes and then centrifuged to obtain serum for analysis of biochemical parameters. Tissue samples collected from the jejunum were taken in a 6 cm section of the jejunum from Meckel’s diverticulum, with 2 cm segments immediately fixed in a 10% buffered formalin solution for histomorphology analysis. Another 2 cm segment was placed in microtubes, flash-frozen in liquid nitrogen, and stored at -80 °C until further molecular analysis.

Blood analysis was conducted using the enzymatic photometric method with a Microlab 300 photometer (ELITech Group Co., Dieren, Netherlands). The biochemical parameters included serum alanine aminotransferase (ALT), aspartate aminotransferase (AST), uric acid (UA), total protein (TP), albumin (ALB), total cholesterol (TC), and triglycerides (TG).

Jejunal samples were fixed in 10% paraformalde-hyde, decalcified in 10% ethylenediaminetetraacetic acid (EDTA) solution with 1% sodium hydroxide, dehydrated in ethanol, and embedded in paraffin. Sections were cut into three well-oriented pieces using a Leica RM 2235 microtome (Leica Microsystems Ltd.), deparaffinized with xylene, and stained with hematoxylin and eosin (H&E). Histomorphology images were captured at 40× magnification using a Nikon microscope (Tokyo, Japan) and measured using Image-Pro Plus 6.0 software.

Gene expression analysis was performed by extracting RNA from 30 mg jejunum samples using the Quick-RNA Miniprep Kit (Zymo Research Corp., Irvine, CA). The extraction followed the manufacturer’s instructions. RNA purity and concentration were determined using a Nanodrop Spectrophotometer (Maestrogen Inc., Hsinchu City, Taiwan). The extracted total RNA was then used as a template for cDNA synthesis using the ReverTrace qPCR RT Master Mix (Toyobo Co., Ltd., Osaka, Japan). Relative gene expression was analyzed using a QuantStudio 3 Real-Time PCR system (Thermo Fisher Scientific, Waltham, MA) and Thunderbird SYBR qPCR Mix (Toyobo Co., Ltd., Osaka, Japan; Cat No. QPX-201). The qPCR reactions were carried out in a total volume of 20 μL, consisting of nuclease-free water, 2 μL of diluted cDNA, 6 pmol of forward primer, 6 pmol of reverse primer, 0.04 μL of ROX reference dye, and 10 μL of qPCR Mix, which were loaded into qPCR tubes (Thermo Fisher Scientific, Waltham, MA). The specific primers used for mRNA level analysis are CLDN1, ZO-1, TNF-α, and IL-18 (Table 2). Following the initial denaturation at 95 °C for 2 min, qPCR amplification was performed for 40 cycles consisting of denaturation at 95 °C for 1 s and combined annealing/extension at 60 °C for 30 s. A melt curve analysis was performed at the end of the reaction to confirm the specificity of the amplified products. Each group consisted of eight samples, with each sample analyzed in duplicate. Relative mRNA expression levels were calculated using the 2⁻ΔΔCt method(Livak & Schmittgen, 2001), normalized to β-actin as the reference gene, and expressed as fold changes relative to the control group.

All experimental data were subjected to a one-way analysis of variance (ANOVA) using IBM SPSS Statistics (version 26.0, IBM Corp., released 2017). The results were considered statistically significant when p<0.05. Differences among treatments were determined using an orthogonal contrasts test.

The binding capacity of the toxin binder that consisted of bentonite–yeast cell enriched with curcumin and cinnamon is illustrated in Figure 1. Bentonite alone (BEN) exhibited the highest AFB₁ binding capacity (92.96%), whereas yeast cell wall (YEAST) had the lowest value (35.36%) (p<0.05). The toxin binders containing bentonite and yeast cell wall (BEN+Y) or enriched with curcumin and cinnamon (COMB) did not differ significantly in AFB₁ binding ability (p>0.05). Although the AFB₁ binding capacity of BEN+Y and COMB was slightly lower compared to that of BEN, both maintained binding efficiencies above 80%.

Table 3 presents the effects of toxin binder supplementation on daily feed intake (DFI), average daily gain (ADG), and feed conversion ratio (FCR) in broilers fed with AFB₁. Dietary treatments had no significant effect on DFI (p>0.05). However, ADG in all periods (d 1-21) differed significantly among treatments (p<0.05). Orthogonal contrast analysis showed that the AFB₁-contaminated diet (AF) reduced ADG compared with the control (CTRL vs. AF; p = 0.035). Broilers fed AFB₁-contaminated diets supplemented with toxin binders (AF-A and AF-B) exhibited higher ADG compared to the AF group (AF vs. AF-A, AF-B; p = 0.003), but similar to the control (CTRL vs. AF-A, AF-B; p = 0.495). The addition of toxin binders in diets without contamination (CTRL-A and CTRL-B) had no significant effect on ADG (p = 0.751). Additionally, the curcumin-and cinnamon-enriched toxin binder (COMB) showed no significant difference compared with the binder without phytogenic enrichment (p = 0.221).

Dietary treatments had no significant effect on FCR during the starter phase (d 1–10). Although FCR values did not differ significantly from the control group, the addition of toxin binders improved FCR during the grower phase (d 11–21) compared with AF, both in contaminated and uncontaminated diets (p<0.05). Orthogonal contrast analysis further indicated that broilers fed AFB₁-contaminated diets with toxin binder supplementation had better FCR compared to the AF group (AF vs. AF-A, AF-B; p = 0.009). In all periods (d 1–21), toxin binder supplementation in diets without contamination (CTRL-A and CTRL-B) resulted in better FCR compared with CTRL and AF groups (p<0.05). At the same time, supplementation in AFB₁-contaminated diets (AF-A and AF-B) also improved FCR (p = 0.009).

The effect of toxin binder supplementation in AFB₁-contaminated diets on the blood biochemical profiles of broiler chickens is shown in Table 4. The dietary treatment did not significantly affect blood parameters, including AST, ALT, uric acid, total protein, albumin, and total cholesterol (p>0.05).

Table 5 shows the effects of bentonite–yeast toxin binders enriched with curcumin and cinnamon on jejunal histomorphology of broiler chickens fed AFB₁-contaminated diets. Dietary treatments significantly affected villus height (VH), villus width (VW), and the villus height-to-crypt depth ratio (VH:CD) (p<0.05), whereas crypt depth (CD) was not influenced (p>0.05). Orthogonal contrast analysis revealed that AFB₁ contamination did not alter VH (p = 0.819) but increased VW (p = 0.006). Supplementation of toxin binders in the non-contaminated diets tended to decrease VH (CTRL vs CTRL-A, CTRL-B; p = 0.051). In contrast, toxin binder addition to AFB₁-contaminated diets reduced VW (AF vs AF-A, AF-B; p = 0.002).

Contamination of AFB₁ markedly reduced the VH:CD ratio (p = 0.005). However, the inclusion of toxin binders restored this ratio to levels comparable to those in the control (CTRL vs AF-A, AF-B; p = 0.198). Moreover, the VH:CD ratio in AFB₁-contaminated diets supplemented with toxin binders was significantly higher compared to the AFB₁ group without toxin binder addition (AF vs AF-A, AF-B; p = 0.033). Conversely, toxin binder supplementation in non-contaminated diets slightly reduced the VH:CD ratio (CTRL vs CTRL-A, CTRL-B; p = 0.017).

The effects of bentonite–yeast toxin binders enriched with curcumin and cinnamon on tight junction (TJ) and inflammatory cytokine gene expression in broilers exposed to AFB₁ are presented in Table 6. AFB₁-contaminated diets significantly downregulated the expression of CLDN-1 (p = 0.001) and ZO-1 (p = 0.003), and upregulated TNFα (p<0.001) and IL-18 (p<0.001). Supplementation with toxin binders markedly enhanced CLDN-1 and ZO-1 gene expression compared with the AFB₁-contaminated diet (p<0.001), where TNFα and IL-18 levels were lower. Orthogonal contrast analysis showed that the supplementation of toxin binders in AFB₁-contaminated diets effectively restored CLDN- 1 and ZO-1 expression to levels comparable with the control group (CTRL vs AF-A, AF-B; p = 0.237). The enriched formulation containing curcumin and cinnamon (COMB) did not significantly differ from the non-enriched bentonite–yeast binder (BEN+Y). Furthermore, supplementation of toxin binders in both non-contaminated and AFB₁-contaminated diets markedly reduced the expression of the pro-inflammatory cytokines TNFα and IL-18 (p<0.001).

The inclusion of AFB₁ at 300 ppb in the diet resulted in a significant decrease in ADG and FCR, while feed intake was not significantly affected (Table 3). This shows that AFB₁ is related to nutrient utilization efficiency rather than feed intake. Under these conditions, most dietary energy and protein are used to maintain physiological functions and repair tissues induced by oxidative stress and inflammation. The stress leads to the accumulation of malondialdehyde and other ROS. Moreover, at high doses or long-term exposure, oxidative stress results in hepatocellular injury and impaired hepatic function, and it also leads to disruption of nutrient metabolism and detoxification. However, lower oxidative stress generally affects cellular redox balance and metabolic efficiency without causing structural liver damage(Moloi et al., 2024);(Ma et al., 2021);(Rotimi et al., 2019).

Supplementation with toxin binders, either the BEN+Y formulation or COMB, effectively mitigated AFB₁-induced impairments in ADG and FCR (Table 3). In toxin-free groups administered toxin binders (CTRL-A and CTRL-B), a modest improvement (p<0.05) in FCR was observed despite relatively stable ADG values (Table 3). This result aligns with the hypothesis that toxin binders demonstrate physiological advantages despite non-toxic environments, potentially through stabilization of the intestinal environment and positive interactions with the gut microbiota and mucosal architecture.

The COMB and BEN+Y treatments showed no statistically significant differences between growth performance and intestinal parameters, suggesting that curcumin and cinnamon supplementation did not show a more positive impact during the 21-day experimental period. This is primarily associated with the limited bioavailability of the active compounds or suboptimal dosing below their effective thresholds.

The AFB₁ exposure in blood presented no statistically significant changes in alanine aminotransferase (ALT), aspartate aminotransferase (AST), total protein, albumin, uric acid, or total cholesterol concentrations (Table 4). Consistent with this, no significant changes in serum ALT and AST in the present study suggest that a dietary AFB₁ concentration of 300 ppb over 21 days was insufficient to induce overt hepatocellular damage, presented by elevated liver enzyme biomarkers. Although ALT and AST are sensitive indicators of hepatocyte necrosis induced by toxins, at subclinical exposure levels such as those in this study, AFB₁ is probably metabolized by hepatic cytochrome P450 enzymes into reactive epoxide intermediates in limited amounts, insufficient to cause structural liver injury. The AFB₁ is bioactivated by CYP450 enzymes to form aflatoxin B₁-8,9-epoxide (AFBO), a highly reactive compound for binding nucleic acids and proteins, thereby causing cellular damage and mutagenesis(Saba & Seal, 2022). Thus, the toxicological effects of AFB₁ were likely limited to the intestinal mucosa, as indicated by gene expression and histomorphology data, which demonstrated significant gastrointestinal impacts.

Furthermore, consistent concentrations of total protein and albumin indicate that hepatic protein synthesis remained intact. Albumin, a main hepatic product and a sensitive indicator of liver function, maintains stable levels, suggesting that hepatic biosynthetic and metabolic functions were not clinically compromised. This result aligns with previous reports indicating that changes in these parameters occur only at aflatoxin doses ≥500 ppb or with exposure durations exceeding 28 days(Spinella et al., 2016). In addition, no significant differences were shown in cholesterol or uric acid concentrations, which indicate lipid profiles and purine metabolism. These findings imply that the endocrine and metabolic systems of broilers remained capable of maintaining homeostasis under short-term, subclinical toxin exposure conditions implemented in this study.

The effect of AFB₁ in the diet damages the integrity of the jejunum mucosa, through changes in gut microbiota composition and degeneration of the intestinal villus and crypt structures(Choi et al., 2025). These morphological damages decrease the absorptive area and increase intestinal epithelial permeability, thereby impairing the absorption of nutrients such as amino acids, glucose, and minerals(Ibrahim et al., 2025). The decreased villus height to crypt depth (VH:CD) ratio in the jejunum (Table 5) is a key indicator of impaired absorptive capacity affecting increased epithelial turnover due to enterocyte stress.

Supplementation with toxin binders containing bentonite and yeast cell wall components markedly reduced villus width, indicating restoration of villus structure(Zhao et al., 2021). The toxin binder will adsorb aflatoxins in the gastrointestinal tract, thereby minimizing their intestinal absorption and cytotoxic effects(Feizy et al., 2025). Moreover, inclusion of toxin binders with curcumin and cinnamon enhanced this protective effect(Emam et al., 2023) as antioxidant and anti-inflammatory compounds to maintain mucosal integrity and protect enterocytes against oxidative damage(Hafez et al., 2022). As a result, the inclusion of a high-level compound toxin binder in AFB₁-contaminated diets improved the VH:CD ratio(Ahmadabadi et al., 2025). The enhancement of VH:CD ratio represents longer villi and shorter crypts, which indicate enhanced nutrient absorption and faster mucosal regeneration(Long et al., 2021).

In this study, toxin binder supplementation without containing AFB₁ showed a reduction in the VH:CD ratio, which is related to the adsorptive capacity of the binders, which could sequester specific nutrients in the absence of toxins. Nevertheless, this reduction was minor compared with the substantial improvement observed in the AFB₁-challenged birds. Overall, these results suggest that the combination of bentonite, yeast cell wall, curcumin, and cinnamon effectively mitigates the negative effects of AFB₁ on intestinal morphology by preserving villus structure, maintaining enterocyte integrity, and supporting mucosal recovery.

At the molecular level, there was a downregulated expression of CLDN-1 and ZO-1, two important tight junction proteins responsible for maintaining mucosal barrier integrity against pathogens and toxins. Moreover, the upregulation of pro-inflammatory cytokines TNF-α and IL-18 (Table 6) indicates activation of inflammatory signaling pathways within the intestinal tissue, affecting nutritional homeostasis and obstructing muscle tissue growth.

The improvement in jejunal histomorphology observed in broilers fed AFB₁-contaminated diets supplemented with toxin binders in this study was associated with the restoration of tight junction (TJ) integrity and modulation of inflammatory cytokine expression. Tight junctions (TJ) are critical structural complexes that maintain intestinal barrier integrity by regulating epithelial permeability(Suzuki, 2020). Those main components, including claudins (CLDN), occludin (OCLN), zonula occludens (ZO) proteins, and junctional adhesion molecules (JAM), form a dynamic network linking epithelial cells to the actin cytoskeleton(Hossain et al., 2025).

The AFB₁ downregulated CLDN-1 and ZO-1 expression in the jejunum, indicating a disruption of intestinal barrier function with increased intestinal permeability and mucosal injury due to epithelial cell apoptosis and oxidative damage(Lai et al., 2022);(Kuo et al., 2021). Supplementation of bentonite and yeast as toxin binders in the diet effectively restores CLDN-1 and ZO-1 expression(Deng et al., 2023). This protective effect is related to the adsorptive capacity of bentonite and yeast cell wall β-glucans and mannans, limiting AFB₁ absorption and mitigating intestinal oxidative stress (Feizy et al., 2025). The polyphenolic and antioxidant compounds in curcumin and cinnamon further support TJ restoration, protect enterocytes, and modulate signaling pathways(Emam et al., 2023);(Xu et al., 2024). Furthermore, AFB₁ also triggered inflammatory responses with upregulation of TNFα and IL-18, two key pro-inflammatory cytokines involved in intestinal inflammation and tissue injury(Jin et al., 2021);(Wang et al., 2022). TNFα will activate immune cells and promote cytokine cascades, compromise barrier function, and contributes to the amplification of inflammatory signaling(Ihim et al., 2022).

Supplementation with toxin binders significantly downregulated TNFα and IL-18 expression, demonstrating an anti-inflammatory effect to reduce toxin absorption and improve mucosal antioxidant defense(Hafez et al., 2022). This finding accords with the findings of previous studies, which have shown that yeast and phytogenic additives mitigate AFB₁ induced intestinal inflammation by inhibiting NF-κB activation and the transcription of pro-inflammatory cytokines(Milby-Blackledge et al., 2024). Collectively, these findings indicate that bentonite and yeast toxin binders, with or without curcumin and cinnamon enrichment, effectively maintain intestinal barrier integrity and suppress AFB₁ induced inflammatory responses. This suggests their potential as feed additives that protect epithelial tight junctions while modulating intestinal immune homeostasis.