**Background:** Deoxynivalenol (DON) is a prevalent food-associated mycotoxin produced by Fusarium fungi that frequently contaminates cereals. Chronic exposure causes intestinal toxicity, immunotoxicity, and disrupts gut barrier function. While previous studies have shown DON alters gut microbiota composition, the functional consequences and post-exposure recovery dynamics—including the potential role of prebiotic interventions—remain poorly understood.
**Methods:** Seventy-two female BALB/c mice were assigned to 9 groups (n=8 each). Three groups received 14 days of: (1) purified water (CK1), (2) 2 mg/kg bw/day DON (LD1), or (3) 5 mg/kg bw/day DON (HD1). Six additional groups underwent 14 days of DON or water followed by 14 days of recovery: spontaneous recovery (CK2, LD2, HD2) or inulin-supplemented recovery (5% inulin in diet; CK3, LD3, HD3). Cecal contents were collected for metagenomic sequencing using Illumina HiSeq PE150. Taxonomic assignment used DIAMOND BLAST against NR database; functional annotation used KEGG (v2018.01), eggNOG (v4.5), and CAZy (v2018.01) databases. Statistical significance was assessed using two-sided unpaired Student's t-test or one-way ANOVA with p<0.05 considered significant.
**Key Results:** High-dose DON (5 mg/kg bw/day) significantly decreased body weight (p<0.05) and caused diarrhea, while low-dose DON did not affect weight. Metagenomic sequencing yielded 1,356,874 ORFs, with 87.0% annotated as bacterial. PCoA based on Bray-Curtis distance showed clear separation of high-dose DON groups from controls. DON exposure dose-dependently increased Verrucomicrobia (phylum level), with Akkermansia muciniphila showing significant increases in both LD1 and HD1 vs. CK1. At the species level, 841 species differed significantly between CK1 vs. LD1, and 969 between CK1 vs. HD1; 245 species were significantly changed in both. DON increased A. muciniphila, Bacteroides vulgatus, Hungatella hathewayi, and Lachnospiraceae bacterium 28-4, while decreasing Mucispirillum schaedleri, Pseudoflavonifractor sp. An85, Faecalibacterium prausnitzii, Firmicutes bacterium ASF500, Flavonifractor plautii, Oscillibacter sp. 1-3, and uncultured Flavonifractor sp. After 2 weeks of spontaneous recovery, only 138 of 841 (LD2) and 117 of 969 (HD2) species remained significantly different from controls, indicating ~86-88% recovery. Inulin supplementation after low-dose DON (LD3) resulted in only 54 of 841 species still significantly different vs. CK3, suggesting enhanced recovery. However, after high-dose DON with inulin (HD3), 651 of 969 species remained significantly different vs. CK3—substantially worse than spontaneous recovery. Gene numbers in HD3 were significantly reduced (p<0.01) compared to all other groups. Functional analysis showed that inulin supplementation normalized gene function after low-dose DON (no significant KEGG pathway differences between LD3 and CK3), but after high-dose DON, 31 functional gene clusters at KEGG level 2 remained significantly different in HD3 vs. CK3, compared to only 8 in HD2 vs. CK2.
**Clinical Implications:** This study demonstrates that DON-induced gut microbiome disruption is largely reversible upon toxin withdrawal, particularly at lower exposure levels. The finding that inulin—a commonly used prebiotic—exacerbates dysbiosis after high-dose DON exposure is clinically important, as it suggests that prebiotic interventions during recovery from severe mycotoxin exposure may be counterproductive. The dose-dependent increase in A. muciniphila, a mucin-degrading bacterium with known beneficial metabolic effects, may represent a compensatory host response to intestinal barrier damage. These findings highlight the need for careful evaluation of dietary interventions during gut microbiome recovery from toxic insults.