Bifidobacterium animalis subsp. animalis GY007 Mitigates High Fluoride Exposure-Induced Ileal Injury and Restores the Ileal Microbiota–Metabolome Imbalances

TL;DR

A probiotic strain, GY007, helps protect the ileum from damage caused by high fluoride exposure by reducing inflammation and restoring gut health.

Contribution

This study identifies GY007 as a probiotic that mitigates fluoride-induced ileal injury and reveals key metabolites involved in the process.

Findings

GY007 reduced intestinal permeability and oxidative stress in fluoride-exposed mice.

The probiotic restored ileal mucosal architecture and modulated inflammatory pathways like TLR9/NF-κb/IRF7.

Microbiome and metabolome analyses showed GY007 reversed microbial and metabolic dysregulation in the ileum.

Abstract

This study investigated the protective effects of Bifidobacterium animalis subsp. animalis GY007 against ileal injury resulting from fluoride exposure. Three cohorts of experimental mice were examined: a healthy control group (Ctrl), a fluoride exposed group (F), and a group co-treated with both fluoride and the probiotic GY007 (F+Bi). The findings demonstrated that GY007 administration significantly mitigated fluoride-induced ileal injury in mice. This protective effect was evidenced by reduced intestinal permeability, increased expression of tight junction proteins, restoration of ileal mucosal architecture, and decreased oxidative stress levels. In parallel, inflammatory cytokines and markers associated with the TLR9/NF-κb/IRF7 pathway were reduced (TLR9, Myd88, IRAK4, NF-κb, IRF7, TNF-α, IL-6, and IFN-α). Analysis of the microbiome and metabolome demonstrated that GY007 influenced the structure of the ileal microbial community and its associated metabolite profile. Our results suggest that GY007 may play a beneficial role in mitigating fluoride-induced ileal injury and shed light on the alterations in the gut microbial community and metabolome. This study provides new insights into how probiotic repairs the damaged ileum caused by endogenous fluoride exposure through interactions with the host organism.

Exposure to fluoride is strongly associated with impaired intestinal function. Probiotics are widely regarded as an effective strategy to maintain microbial homeostasis and to mitigate the progression of fluoride-induced intestinal injury. This study aimed to evaluate the measurable protective effects of the probiotic strain Bifidobacterium animalis subsp. animalis (B. animalis subsp. animalis) GY007 in reversing high fluoride-induced ileal injury. The results showed that GY007 (1 × 109 CFU/mL, once/daily) attenuated intestinal barrier disruption and alleviated ileal mucosal abnormalities in mice receiving fluoride (24 mg/kg) by gavage for eight consecutive weeks. GY007 attenuated elevated oxidative stress and modulated the inflammatory response associated with the TLR9/NF-κb/IRF7 signaling pathway. Microbiome and metabolomic analyses showed that GY007 reversed the dysregulation of the ileal microbial community structure and metabolite profiles. Spearman’s rank correlation analysis further supported a regulatory role for Bifidobacterium in this protective process and identified three key functional metabolites meriting further investigation: isocytosine (ISO), 7α,24S-dihydroxy-3-oxocholest-4-en-26-oic acid (OIC-7α), and sinapinic acid (SIA). Our findings demonstrate that GY007 protects against fluoride-induced ileal injury and elucidate the associated changes in the intestinal microbial community and metabolite profiles. This study provides new evidence clarifying the restorative effect of the probiotic GY007 on the ileum under environmental fluoride exposure, offering an integrative perspective on the interaction between microorganisms and their host.

Figures

Full text

1. Introduction

As the thirteenth most prevalent element in the Earth’s crust, fluorine is broadly distributed across natural and artificial environments [1]. Human exposure to fluorides occurs through several routes: 60% via water, 35% via food, and 5% via air [2]. Long-term excessive intake of fluoride has exposed more than 200 million people in more than 50 countries to high levels of fluoride, making it a global public health problem [3,4]. Fluoride pollution predominantly impacts populations in Africa (37–46% of the total population) and Asia (51–59% of the total population) [5]. Fluoride pollution is a widespread issue across 31 provincial-level regions in China, both north and south, impacting nearly 60 million individuals living in 80,011 villages [6]. It is estimated that in India, the onset of disease in approximately 62 million people is closely associated with exposure to fluoride-contaminated groundwater, of which about 6 million are children [7]. High fluoride intake is associated with dental and skeletal fluorosis and extra-skeletal toxicity (e.g., brain, kidney, and gastrointestinal tract), potentially leading to chronic pain, functional limitations, and reduced quality of life [8,9,10]. These health effects also impose economic burdens through increased healthcare costs, productivity losses, and long-term care needs; in a field survey from fluorosis-endemic regions in India, per capita expenditures attributable to fluorosis were INR 5454–14,815 [10,11]. Anthropogenic activities, including the production of fertilizers, operation of brick kilns, manufacture of glass, and smelting of aluminum, contribute to the accumulation and pollution of fluorides [12]. The fertilizer industry alone generates approximately 5.4 to 11.8 million tons of fluoride waste each year [12]. An analysis of 25 soil samples from the area surrounding an aluminum smelter in south-western China revealed fluorine concentrations of up to 4000 mg/kg, compared to a global average of 400 mg/kg in soil [13]. Fluorine is absorbed by plants from polluted soils and aquatic environments, subsequently entering the human food chain [5].

The digestive system is the primary route for the absorption of fluoride. The gut is generally one of the first areas to be impacted by its effects [14]. Based on previous reports, the exposure to fluoride was associated with structural changes in the intestinal tissue, notably the thinning of the muscularis layer and sparse microvilli [15,16,17]. Marked edema, villus shortening, and villus coalescence were observed in the small intestine, accompanied by increased circulating concentrations of d-lactic acid and diamine oxidase [18]. Significant factors that disrupt the intestinal homeostasis can stimulate the secretion of inflammatory cytokines, thereby leading to chronic inflammation [19]. Our prior research has demonstrated that fluoride exposure in drinking water for 70 days was associated with the dysbiosis of the murine ileal microbiome, heightened intestinal permeability, and aberrant inflammatory responses [20,21]. More recently, our team found that sodium fluoride (NaF) given by gavage with 24 mg/kg body weight (bw) for 8 consecutive weeks induced villus damage and shortening in the murine small intestine, accompanied by an evident decrease in the proportional relationship between villus height and crypt depth [22]. Further research revealed that the potential mechanism underlying ileal injury in mice exposed to 50 ppm NaF in drinking water for 56 consecutive weeks may be associated with the mucus barrier and immune response through single-cell sequencing analysis [14]. Moreover, antibiotic-induced disruption of the ileal microbiota was shown to attenuate ileal injury induced by oral gavage administration of NaF at 24 mg/kg bw for 8 weeks [14]. These findings suggest that the ileum warrants significant attention in both intestinal and systemic diseases induced by fluoride exposure, and that microbial community intervention may present a novel therapeutic approach for fluoride-related pathologies.

Current strategies to mitigate excessive fluoride exposure focus primarily on reducing intake at the source, notably by providing low-fluoride water and implementing defluoridation technologies (e.g., adsorption using activated alumina/bone char, precipitation-based approaches, ion exchange, and membrane/separation techniques) [23,24,25]. Furthermore, previous studies have reported that vitamin C and vitamin E can attenuate fluoride-induced damage in experimental animals [26,27]. Miao et al. found that selenium supplementation was associated with increased superoxide dismutase (SOD) activity, reduced fluoride-related toxicity, and improved indicators of liver function [28]. In humans, dietary calcium may help reduce fluoride absorption and mitigate fluoride toxicity [29,30,31]. However, studies investigating probiotics in fluoride-induced ileal injury remain limited.

Probiotics are colonies of live microorganisms that offer quantifiable health benefits. By regulating the gut microbiota, they could maintain intestinal barrier function, regulate immune function, and stimulate the generation of circulating metabolites that exert systemic effects [32]. Bifidobacterium constitutes one of the most significant bacterial taxa in the human intestine, with its characteristics and mechanisms of action having been extensively documented since 1950 [33,34,35]. Over the past two decades, multiple probiotic candidates from the Bifidobacterium and Lactobacillus genera have been shown in experimental and clinical studies to reduce manifestations associated with intestinal injury [36,37,38,39]. A recent study found that male mice with Bifidobacterium supplementation showed a reduction in the damage to the high-fat diet-driven impairment of the gut barrier [40]. A random trial involving 100 students with diarrhea showed that taking Bifidobacterium could effectively alleviate diarrhea symptoms and restore the disrupted intestinal flora caused by diarrhea [41]. Furthermore, our group’s previous research demonstrated that Lactobacillus johnsonii BS15 could restore ileal microbial dysbiosis, heightened intestinal permeability, and aberrant inflammatory responses resulting from fluoride exposure in drinking water, concurrently ameliorating hippocampal injury in mice [20,21]. This suggests that probiotic intervention in the ileum could offer a means of preventing and treating intestinal and extraintestinal diseases induced by fluoride. However, even when classified within a single species, distinct probiotic strains could display markedly divergent characteristics and may yield substantially different clinical outcomes. Consequently, further investigation is required to establish the effectiveness of a beneficial probiotic strain [42]. Discovering and studying the probiotic properties of more bacterial strains in specific application scenarios not only provides more bacterial strain resources, but also offers more possibilities for future disease prevention and treatment.

The potential beneficial impacts of B. animalis subsp. animalis GY007 on ileal dysfunction and ecological imbalance caused by fluoride were assessed in this study. Our research may offer a conceptual framework to support the use of probiotic interventions for managing the intestinal injury associated with excessive fluoride exposure.

2. Materials and Methods

2.1. Bacterial Strain and Culture

The strain GY007 was stored in the China center for type culture collection with preservation number CCTCC AB 2025172. GY007 was purified and inoculated into the prepared culture medium broth at 37 °C for 24 h and subcultured for three generations, to ensure activity. The third-generation bacterial suspension was centrifuged at 4000 g for 15 min and resuspended in normal saline, which was repeated three times for washing. The concentration of GY007 was adjusted to 1 × 10^9^ CFU mL^−1^ using the colony counting method for in vivo experiments.

2.2. Animal and Experiment Design

As shown in Figure 1A, the C57BL/6J mice (male, 3 weeks old) were procured from Si Pei Fu and maintained under standard caging conditions. Throughout the experiment, animals were housed under standardized housing parameters (20–22 °C) with a 12:12 h light–dark cycle, and six mice were accommodated in each cage. Mice were maintained with free access to standard chow and water (i.e., food and water were continuously available and not restricted) and monitored daily for health. Following a 7-day acclimation period, the animals were randomly allocated to three groups (n = 12 per group): a control group (Ctrl), a fluoride-exposed group (F), and a fluoride-exposed group supplemented with GY007 (F+Bi). All treatments were administered once daily via oral gavage, with the following treatment. The F+Bi group received sequential gavage of two 0.2 mL preparations: a fluoride solution (24 mg/kg, dissolved in normal saline) and a GY007 bacterial suspension (1 × 10^9^ CFU/mL, prepared in normal saline). The F group was gavaged with an identical volume of the fluoride solution (0.2 mL), followed by 0.2 mL of normal saline. The Ctrl group was administered an equivalent total volume of normal saline alone. All the experimental procedures lasted for eight weeks. The doses of fluoride solution and duration of the experiment were as given in previous studies [14,22].

On the final day of the 8-week experimental period, blood was collected by retro-orbital sampling from mice. The blood samples were centrifuged to isolate the serum, which was then stored at −80 °C for testing. All mice were culled by cervical dislocation and sampling was conducted immediately. The ileum was excised and opened; luminal contents were gently scraped into sterile tubes. Ileal contents were promptly frozen at −80 °C for subsequent 16S rRNA high-throughput sequencing and non-targeted metabolomics. The collected ileal tissue was flushed with ice-cold saline, cut into small pieces and immediately stored at −80 °C for testing. All procedures were performed on ice. The partial ileal tissue without intraluminal material was in 4% paraformaldehyde solution and partial ileal tissues containing luminal content were in methanol-carnoy for at least 24 h for fixation.

2.3. Metabolite Detection and Bioinformatics Analysis

A total of 24 ileal contents samples were performed to untargeted metabolite profiling, with 8 samples per group. Metabolite profiling and the associated bioinformatics workflows were carried out in accordance with procedures reported previously [43].

2.4. 16S rRNA Sequencing and Analysis

A total of 24 ileal contents samples were performed to 16S rRNA sequencing, with 8 samples per group, corresponding to the metabolite detection samples. Briefly, ileal content DNA was extracted using the E.Z.N.A. Stool DNA Kit. The V3–V4 region of the 16S rRNA gene was amplified with barcoded primers 338F/806R, and amplicons were purified, quantified, pooled in equimolar amounts, and sequenced on an Illumina MiSeq platform (2 × 250 bp, paired-end). Reads were processed in QIIME2 2024.5 [44] for demultiplexing, primer trimming, quality control, and denoising/merging/chimera removal with DADA2 [45]; ASVs were aligned with MAFFT [46] and a phylogeny was built with FastTree2 [47]. Taxonomy was assigned using a Naive Bayes classifier against Greengenes2 2022.10 [48,49]. Alpha diversity (Shannon, Simpson), beta diversity (Bray–Curtis; PCoA), and PERMANOVA were performed in QIIME2 and R V4.3.3 [50,51]. Random forest was used to discriminate groups, and Mantel and Spearman correlation analyses were applied to assess microbiota–metabolome and phenotype associations [52,53]. Detailed information is provided in the Supplementary File S1.

2.5. Real-Time Quantitative PCR (RT-qPCR) of mRNA Expression Levels

Five specimens from each experimental group were randomly chosen (15 in total) for RNA isolation in this part. Total RNA from ileal tissue was extracted using procedures that have been reported previously [22]. Primer information is provided in Supplementary Table S1.

2.6. Histology

The subsequent steps were the same as described in the previous research [43]. The anti-MUC2 (1:500), anti-4-HNE (1:100), anti-ZO-1 (1:100), anti-Claudin-1 (1:50), anti-Occludin (1:500), anti-IRF7 (1:1000), anti-TLR9 (1:100), anti-NF-κb (1:1000), anti-IRAK4 (1:1000), and anti-Myd88 (1:200) were used to assess the condition of the ileal tissue.

2.7. Biochemical Analysis

The serum samples and the ileal tissue were estimated. The blood samples were centrifuged at 3000× g for 15 min at 4 °C. Subsequently, serum diamine oxidase (DAO) and D-lactate were measured using mouse-specific ELISA kits (Jiangsu Enzyme Technology, Yancheng, Jiangsu, China). Antioxidant indices in the ileal tissue were measured using commercial kits (Jiancheng Bioengineering Institute, Nanjing, Jiangsu, China), including total antioxidant capacity (T-AOC; A015-3-1), catalase (CAT; A007-1-1), reduced glutathione (GSH; A006-2-1), malondialdehyde (MDA; A003-1-2), hydrogen peroxide (H_2_O_2_; A064-1-1), and superoxide dismutase (SOD; A001-3-2).

2.8. Statistical Analysis

Normal distributions were assessed using the Shapiro–Wilk normality test. If the data were not normally distributed, normality tests were repeated after log-transformation. Normally distributed data with homogeneity of variance were analyzed using t-tests (two independent samples) or one-way ANOVA with LSD post hoc multiple analysis (comparisons among multiple groups). Normally distributed data with heterogeneity of variance among multiple groups were analyzed using one-way ANOVA with Dunnett’s T3 post hoc analysis. Data that did not reach a normal distribution were tested using the Kruskal–Wallis test at p < 0.05, followed by the Wilcoxon rank-sum test. Data are presented as the mean ± standard deviation and were analyzed using IBM SPSS Statistics 27. * p < 0.05, ** p < 0.01, and *** p < 0.001 were considered significant. The mapping software used was GraphPad Software 8.0.

3. Results

3.1. GY007 Could Alleviate Intestinal Barrier Damage Induced by Fluoride

Consistent with the animal handling protocol outlined above, serum samples were obtained from the mice (Figure 1A). As illustrated in Figure 1B,C, the F group exhibited markedly (p < 0.001) higher levels of D-lactate and DAO than control mice. Mice receiving GY007 (F+Bi) displayed reduced circulating D-lactate and DAO concentrations, relative to the F group (p < 0.05). For both parameters, the Ctrl and F+Bi groups did not differ significantly (p > 0.05). Tight junction proteins were assessed in the ileal tissue of the mice. In the F group, both mRNA expression and protein levels of Claudin-1 (Figure 2A,B), Occludin (Figure 2D,E) and ZO-1 (Figure 2G,H) were markedly decreased in comparison to the remaining groups (p < 0.05). Treatment of GY007 significantly restored these markers (p < 0.05), with no significant difference observed between the Ctrl and F+Bi (p > 0.05). To evaluate the in situ expression patterns of these proteins, immunofluorescence staining was carried out thereafter. As depicted in Figure 2C,F,I, a significant decrease in these proteins was detected in the F group versus Ctrl, while GY007 administration restored their expression to levels comparable with those observed in controls. Overall, the results indicated that fluoride would cause enhanced intestinal permeability of the ileum and decreased expression of tight junction proteins, while GY007 could inhibit this negative impact.

3.2. GY007 Could Repair the Ileal Oxidative Damage Induced by Fluoride

Analyses were conducted to further evaluate the oxidative stress profile in the mouse ileum. Significant variations (p < 0.05) of SOD (Figure 3B) and CAT (Figure 3D) activities together with MDA (Figure 3A), T-AOC (Figure 3C), and H_2_O_2_ (Figure 3E) contents were observed in the F group. Administration of GY007 significantly modulated MDA, SOD, T-AOC, and CAT measurements (p < 0.01), restoring these indices to values that were not statistically different from those in the Ctrl group with no significant difference observed between the Ctrl and F+Bi (p > 0.05). Although GY007 treatment was able to significantly reverse (p < 0.05) the trend of fluoride-induced accumulation of H_2_O_2_ in the ileum, H_2_O_2_ levels remained substantially higher than the Ctrl group (p < 0.05). Meanwhile, across the groups, no statistically meaningful differences were observed, regardless of whether fluoride exposure or GY007 dosing was applied (p > 0.05, Figure 3F). Then, the mRNA expressions of Duox2 and Duoxa2 were detected. As presented in Figure 3G,H, fluoride exposure led to significantly upregulated (p < 0.05) ileal Duox2 and Duoxa2 mRNA levels relative to the Ctrl group, while GY007 could significantly reverse (p < 0.01) this trend. Further immunofluorescence results confirmed that the lipid peroxidation reaction in the F group was enhanced. The use of GY007 could reduce the lipid peroxidation reaction in the ileum of mice (Figure 3I). In aggregate, these data implied that fluoride causes oxidative stress damage to the ileum of mice, while the use of GY007 could partially alleviate these adverse effects.

3.3. The Influences of Fluoride Exposure and Probiotic GY007 on Function and Mucus Layer of the Ileum

Reg3a, b, g, MUC2 and Ramp1 are important indicators for characterizing intestinal homeostasis and function [54]. Therefore, the mRNA expression levels of them were further examined. After fluoride treatment, ileal mRNA levels of Reg3a, b, g and MUC2 were reduced, consistent with a loss of intestinal homeostasis (p < 0.01, Figure 4A–D). The application of GY007 could significantly reverse the transcriptional levels of them (p < 0.05) to values comparable with the Ctrl group (except for Reg3a). Immunofluorescence analyses suggested that the staining intensity of MUC2 and Ramp1 were diminished in the ileum of mice exposed to fluoride, while the use of GY007 could enhance the expression (Figure 4E,F). MUC2 serves as a key scaffolding element within the external mucous barrier layer [55]. More detailed examination of the defining features of mucus in the ileum was observed. Evaluating intestinal mucus characteristics in fixed tissue slices remains difficult, largely because mucus preservation in fixed specimens is inconsistent [56]. Fixed sections from the Ctrl, F, and F+Bi groups were subjected to Alcian blue–periodic acid–Schiff (AB/PAS) staining and inspected to assess whether fluoride treatment and GY007 administration produced changes in the intestinal mucus (Figure 4G). In comparison with the Ctrl, the surface of the ileal villi and the mucus in the inter-villus spaces in the F group decreased, while the F+Bi group showed a trend toward restoration. Glycogen, neutral mucoprotein, and various glycoproteins appear purple-red, while acidic mucoprotein, proteoglycan and hyaluronic acid appear blue in the AB/PAS staining. The mucin components in the F group were particularly different, more resembling acidic mucopolysaccharides, while the F+Bi group and the Ctrl group were more similar to neutral mucopolysaccharides. These results suggeste that fluoride exposure compromises the intestinal homeostasis, function and mucus properties of the ileum, while GY007 could produce a positive effect.

3.4. Impacts of Fluoride Exposure and GY007 on Toll-like Receptor (TLR) Signaling Pathway and Downstream Inflammatory-Related Factors

TLR-mediated signaling has been widely investigated as a key mechanism through which probiotics influence host regulatory processes. The interaction between them could trigger local immune responses in the intestine, which may affect the homeostasis and communication within the body [57]. This part focused on the levels of key components within the TLR signaling pathway of the ileum in mice (Figure S1). As shown in Figure 5A–F, GY007 significantly decreased the mRNA expression and protein content of TNF-α, IL-6, and IFN-α within the murine ileum (p < 0.01), restoring them to levels not significantly different from those of the control mice. We detected IL17, IL-10, and IFN-γ, additionally (Figure 5G–I). Fluoride administration markedly elevated (p < 0.01) the level of mRNA expression of IL-17 in the ileum. GY007 had a reversing trend but no statistical significance (p = 0.062). Across the three groups, IL-10 and IFN-γ levels did not differ significantly (p > 0.05). As shown in Figure 6, GY007 significantly attenuated the fluoride-induced upregulation of ileal IRAK4, IRF7, Myd88, NF-κb, and TLR9 mRNA levels (p < 0.05). Immunofluorescence assay showed a trend consistent with the mRNA expression levels. Notably, the mRNA levels of TLR2, TLR3, TLR5, TLR7, and TLR8 did not differ significantly across the experimental groups (p > 0.05, Figure S2A–E). In the F group, TLR4 mRNA expression was markedly elevated (p < 0.05), while GY007 showed a tendency towards a decrease that did not achieve statistical significance (Figure 2F). These results suggest that GY007 may modulate the changes in TNF-α, IL-6, and IFN-α levels caused by fluoride exposure through TLR9, although a direct causal relationship has not been functionally validated.

3.5. GY007 Could Restore Microbial Structure in High Fluoride Exposure-Induced Ileal Injury

To explore the function of probiotics from a microbiological perspective, the 16S rRNA high-throughput sequencing of the ileal content was conducted. Rarefaction analysis assessed the observed species across the collected samples. The findings indicated that, in each of the three cohorts, the accumulation of detected genes approached a plateau, suggesting that sequencing depth was sufficient to capture most of the measurable diversity (Figure S3A). As illustrated in Figure 7A, alpha diversity analysis showed that the F group had a significantly lower Simpson index than the Ctrl group (p < 0.05). In contrast, the Shannon diversity index exhibited a decreasing trend in the F group, although this change did not reach statistical significance (p > 0.05). GY007 could significantly reverse the reduction in diversity caused by fluoride (p < 0.05). As shown in Figure 7B, the microbial communities among the three groups each exhibited unique patterns (p < 0.001, R^2^ = 0.46). The intestinal microbial distribution structure of F+Bi was more similar to that of the control. Analysis at the phylum-level profiling indicated that the ileal microbiome mainly comprised Firmicutes_D, Actinobacteriota, Bacteroidota, Desulfobacterota_I, and Firmicutes_A (Figure 7C). It is worth noting that the abundances of Firmicutes_A and Actinobacteriota within the mouse ileum was significantly changed after fluoride exposure (p < 0.001), while GY007 could significantly reverse this trend (p < 0.05, Figure S3B). Genus-level composition in the top five genera predominantly comprised Lactobacillus, Limosilactobacillus, Dubosiella, Bifidobacterium, and Ligilactobacillus (Figure 7D). Both the LEfSe procedure and a random forest model from machine learning workflows provided additional evidence of pronounced differences in gut microbial composition across the three groups. Also, seven biomarkers were simultaneously identified by both methods (Figure 7E,F and Figure S3C). Within the set of marker genera identified, in comparison with the control mice, fluoride exposure significantly altered the abundance of the majority of these taxa (p < 0.01, Figure 7G). Although there was no statistical significance, the presence of GY007 tended to make the abundances of these markers closer to those of the Ctrl group, even in Dubosiella (Figure 7G). These observations indicated that fluoride exposure may disrupt the diversity and structure of the ileum microbiota, whereas GY007 could relieve this condition.

3.6. Effects of GY007 on Ileal Metabolic Profiles in Mice Subjected to Fluoride Exposure

To assess ileal metabolic alterations in fluoride-exposed mice following administration of GY007, the metabolomics analysis was conducted. In total, 2479 metabolites were detected, comprising 1224 acquired in positive ion (POS) and 1255 acquired in negative ion (NEG). Based on LC–MS-derived ileal profiles, the PCA score plots demonstrated distinct clustering of the Ctrl and F groups in both of the ion acquisition modes (Figure 8A,B). Following GY007 administration, samples from the F+Bi group generally shifted away from the F group in the multivariate space, although partial overlap persisted (Figure 8C,D). These findings suggested that GY007 administration partially mitigates fluoride-induced metabolic dysfunction and may confer a protective effect on the ileum. To identify metabolites exhibiting statistically significant differences across the three groups, variables meeting the thresholds of a VIP > 1 and a p < 0.05 were selected as differentially expressed metabolites (DEMs). A total of 295 DEMs were detected in the F versus F+Bi contrast; 346 DEMs were found when comparing Ctrl with F (Figure 8E, Tables S2 and S3). When evaluating the differences between the Ctrl and F groups, 169 were upregulated and 177 were downregulated in the F group (Figure 8E, Table S2). When evaluating differences between the F and F+Bi group, 175 were downregulated and 120 were upregulated in the F+Bi group (Figure 8F, Table S3). The Venn diagram in Figure 8G shows that the 111 common DEMs were detected in both comparisons (Figure 8E). The expression profiles of 27 upregulated and 21 downregulated metabolites in F were significantly restored in the F+Bi. We focused on the changes of nine specific DEMs. The accumulation of hippuric acid could lead to the excessive production of the total in vivo level of reactive oxygen species [58]. Isocytosine is the positional isomer of cytosine and is involved in the metabolism of bacteria [59]. MN6249 (hippuric acid, HIA) and MP3188 (isocytosine, ISO) were observed to increase in F and decrease in F+Bi (p < 0.01, Figure 8F,G). The concentration of bile acids in the mice exposed to the fluoride environment is worthy of attention [60]. As shown in Figure 8H–K, four bile acids, alcohols and their derivatives, including MN18188 (TRI), MN18722 (OIC-7α), MP21633 (OIC-3), and MP22898 (OIC-3α), showed a significant reduction in the F group (p < 0.05). GY007 could reverse this trend (p < 0.05). 18-Glycyrrhetinic acid [61], Saikosaponin B2 [62] and Sinapinic acid [63] could have anti-inflammatory and antioxidant stress effects. GY007 could markedly reverse (p < 0.05) the decrease in the abundance of MN19753 (18-Glycyrrhetinic acid, GLA), MP26064 (Saikosaponin B2, SAB), and MP9998 (sinapinic acid, SIA) caused by fluoride (Figure 8L–N). Based on the 13 markers for inflammatory bowel disease identified by the cohorts [64], seven of them were annotated and evaluated in this study. Except for MN6908 (L-Carnitine) and MN7724 (L-Tryptophan), the remaining metabolites showed no significant differences among the three groups (Figure S4). Bubble plots were employed to depict the key metabolic pathways most strongly enriched by the identified biomarkers in the two comparisons (Figure 8O,P). The metabolites of Ctrl vs. F were mainly involved in bile secretion, ABC transporters, and histidine metabolism (Figure 8O). There were four pathways enriched in F vs. F+Bi, namely purine metabolism, histidine metabolism, longevity regulating pathway, and arachidonic acid metabolism (Figure 8P).

3.7. Relevance Analysis of the Microbiota and Metabolism with Ileal Function Variables

To clarify how ileal molecules may mediate the probiotic-driven restoration of fluoride-induced intestinal injury, we integrated the datasets generated above—namely, key genera identified by 16S rRNA sequencing and representative DEMs from the metabolomics screen—and subsequently performed Pearson correlation and Mantel test analyses. As shown in Figure 9A, the obtained results confirmed that there was an interesting correlation between the metabolite groups, the bacterial genus groups and the key indicators. Myd88 and TNF-α-mRNA are the most explanatory variables in the case of probiotics reversing fluoride-induced intestinal damage (for Myd88, the genus: Mantel’s r = 0.63; the metabolites: Mantel’s r = 0.59; for TNF-α-mRNA, the genus: Mantel’s r = 0.60 and the metabolites: Mantel’s r = 0.58; all p < 0.01), with MDA and Occludin protein levels being the next most significant (p < 0.01). There was an expected correlation among the indicators of intestinal function. Furthermore, we conducted Spearman analysis to evaluate the correlations between each key factor in the metabolites, the genus and the intestinal function indicators, separately. As shown in Figure 9B, among the 30 indicators related to the function of the ileum, seven key bacterial genera were correlated with at least sixteen of the indicators, while nine key metabolites were correlated with at least three of the indicators. The Bifidobacterium, MP3188 (ISO), MN18722 (OIC-7α), and MP9998 (SIA) showed a closer correlation with more of the aforementioned experimental indicators. It is worth noting that Mantel and Spearman analyses indicate co-variation, not directionality.

4. Discussion

High fluoride-associated disorders constitute some of the most prevalent and severe endemic conditions worldwide, with particularly pronounced impacts on the gastrointestinal tract [14,65]. Consequently, strategies focused on mitigating fluoride-induced ileal injury have become a major area of research interest and significance. Probiotic-based interventions offer a promising approach to alleviating intestinal injury. The therapeutic effect of B. animalis subsp. animalis GY007 on high fluoride-induced intestinal damage in mice was evaluated. A review of the existing literature revealed the absence of a universally accepted standard for determining high fluoride exposure concentrations. The chosen dosage of 24 mg/kg body weight NaF and the duration were determined in accordance with our earlier investigations [14,22] and other published research [66,67,68,69]. According to the dose conversion formula of the U.S. Food and Drug Administration (FDA), the equivalent human dose is approximately 23.8 mg/L for an adult weighing 55 kg and with a daily water intake of 4.5 L [20,70]. In contrast, the World Health Organization (WHO) recommends a guideline value for fluoride in drinking water of 1.5 mg/L, and the U.S. Environmental Protection Agency (EPA) primary maximum contaminant level (MCL) is 4.0 mg/L [71,72]. In practice, the primary strategy to mitigate fluoride-related health risks is to reduce fluoride intake [73]. Where low fluoride water or foods are not readily achievable, our findings suggest that GY007 could be explored as an adjunct, diet-deliverable approach to support intestinal barrier function during high fluoride exposure, rather than as a substitute for source control. Bifidobacterium animalis strains have been used as probiotic ingredients in foods and dietary supplements under FDA “no questions” GRAS notices [74], and they have also been officially included in the “List of Microorganisms Approved for Use in Food,” published by the National Health Commission (NHC), supporting their feasibility as a food ingredient [75]. Future work should include dose optimization and human studies in fluoride-exposed populations, incorporating clinically relevant outcomes (e.g., gastrointestinal symptoms and stool microbiome profiles).

Increased intestinal epithelial permeability is a critical determinant in the pathogenesis of various gastrointestinal disorders [76]. Tight junctions form the core structure of the selective permeability barrier between intestinal epithelial cells, consisting of tight junction proteins, including transmembrane components such as Claudin-1 and Occludin, and peripheral membrane-associated scaffold proteins such as ZO-1. Tight junction proteins create highly organized intercellular contacts through tight adhesion [77,78]. The disruption or loss of them leads to increased intestinal mucosal permeability, allowing the passage of pathogenic microorganisms and harmful substances, thereby triggering and exacerbating intestinal inflammation [79]. The data from this study demonstrated that fluoride exposure significantly elevated the blood levels of D-LA and DAO, resulting in increased intestinal epithelial permeability in mice. The mRNA expression and protein levels of the three tight junction proteins further confirmed the impairment of the intestinal barrier function in mice. Exposure to fluoride-containing drinking water has been shown to impair intestinal development across multiple segments and to suppress tight junction protein expression, thereby increasing mucosal permeability. Among the segments of the intestine, the ileum appears to experience the greatest adverse effects when exposed to fluoride [20,21]. Administration of GY007 at 1 × 10^9^ CFU ameliorated the fluoride-induced increase in intestinal epithelial permeability, providing a robust basis for restoring intestinal barrier function.

Our recent study employed single-cell sequencing technology to investigate fluoride-induced damage to the ileum. The results showed that the expression of Duox2 and Duoxa2 in the ileal cells of fluoride-exposed mice was elevated, and the level of 4-HNE also increased [14]. Consistent with these observations, this study found that oxidative stress levels in the ileal tissues of mice exposed to a high fluoride exposure were elevated. Duox2 is an epithelial antimicrobial dual oxidase, with Duoxa2 serving as a key maturation factor for the enzyme. MacFie et al. [80] have reported that in individuals experiencing active ulcerative colitis, the mature Duox2–Duoxa2 complex is the main enzyme system that generates reactive oxygen species (specifically hydrogen peroxide). Administration of GY007 permits the targeted modulation of these indices. Fluoride treatment decreased intestinal SOD and CAT and increased MDA in mice, which has been observed in both rat and chicken models [81,82]. Notably, fluoride exposure did not elicit a significant alteration in ileal glutathione (GSH) levels in the present study. This finding suggested that fluoride may exert a distinct pattern of oxidative stress injury in the ileum; however, additional evidence is required to substantiate this conclusion. While GY007 exhibited a partial reduction in H_2_O_2_ levels, its ability to fully alleviate oxidative stress remains limited, and further optimization of the GY007 administration protocol may enhance its efficacy. These findings suggest that GY007 could partially alleviate oxidative stress damage in the ileum of mice caused by fluoride exposure, but further research is required to fully assess its potential.

Notably, some findings in this study differ from prior reports, particularly regarding Reg3b/Reg3g and ileal Bifidobacterium abundance. Previous work reported increased Reg3b/Reg3g expression and higher MUC2 and Ramp1 abundance in the ileum after chronic fluoride exposure, where a mouse model was exposed to drinking water fluoride at a concentration of 25/50 ppm for a duration of 56 weeks [14], whereas Reg3b/Reg3g showed an opposite trend in our gavage model (24 mg/kg NaF for 8 weeks). Fluoride exposure reduced ileal Bifidobacterium in the present study, whereas our earlier work in ICR mice using fluoride exposure via drinking water for 10 weeks showed an increase [20]. This discrepancy may be attributable to differences in the experimental animal, fluoride exposure paradigms, including the route of administration, and the duration of treatment. Reg3 proteins are thought to contribute to restoration and repair processes within the intestinal mucosa [80,83]. It exhibits antibacterial and anti-inflammatory activities, supports tissue regeneration, and contributes to the maintenance of intestinal homeostasis [84,85,86,87]. One study reported that mice lacking Reg3b exhibited exacerbated colitis manifestations relative to their wild-type counterparts [88]. The MUC2 gene encodes the predominant secreted mucin in the intestinal tract, constituting a major structural component of the mucus layer and acting as a critical contributor in protecting the ileal mucosa from infectious challenges [56]. When Ramp1 is lacking, animals exhibit heightened epithelial stress and become more vulnerable to infection by pathogenic bacteria [89,90]. Overall, our observations indicated that fluoride exposure disrupts intestinal mucosal homeostasis. Treatment with GY007 significantly reversed the changes in the indicators associated with ileal mucosal dysfunction induced by fluoride. The distinct Reg3a expression pattern may suggest a different regulatory response to GY007, which warrants further investigation. Future studies should investigate how GY007 modulates the regenerating gene family dynamics within the murine small intestine subjected to long-term fluoride exposure via drinking water. Elucidating whether the probiotic responses are tuned to specific physiological processes will provide a rationale for targeted investigations into probiotic engraftment and accumulation.

Increased intestinal permeability can activate pro-inflammatory signaling pathways, thereby contributing to the development of intestinal inflammation [54]. Consistent with this, fluoride exposure within this investigation increased ileal mRNA levels of TNF-α, IL-6, and IFN-α, whereas GY007 administration significantly downregulated the expression of pro-inflammatory mediators. Accumulating evidence indicates that probiotics can modulate inflammatory mediators. One study reported that a combined probiotic formulation comprising Clostridium butyricum and Bifidobacterium infantis ameliorated intestinal injury by normalizing IFN-γ, IL-10, and TNF-α levels [91]. Streptococcus thermophiles mitigated inflammatory responses by lowering TNF-α and enhancing IFN-γ and IL-10 [92]. Notably, fluoride exposure in the present study did not significantly alter IL-10 or IFN-γ levels. Although GY007 tended to attenuate the fluoride-associated increase in IL-17, this effect did not reach statistical significance (p = 0.062). TNF-α, IL-6, and IFN-α are downstream inflammatory mediators of Toll-like receptor (TLR) signaling (Figure S1). TLRs are pattern recognition receptors and are essential mediators of immune activity at mucosal surfaces [93]. Further detection of the key mediators on the TLR pathway revealed that fluoride significantly increased the levels of IRF7, IRAK4, Myd88, and NF-κb, while GY007 could reverse this trend. During the process of fluoride damage, only TLR9 and TLR4 showed significant changes, while GY007 only had an effect on TLR9. Situated within endosomal compartments, TLR9 acts as a specialized DNA sensor by detecting CpG dinucleotide motifs that are unmethylated or only minimally methylated, including those in immune complexes [94], and subsequently leads to the engagement of Myd88 [95]. As a pivotal adaptor in the TLR9 signaling cascade, Myd88 facilitates NF-κb phosphorylation and activation by recruiting members of the interleukin-1 receptor-associated kinase (IRAK) family. Upon activation, NF-κb translocates to the cell nucleus, where it induces the transcription of multiple pro-inflammatory cytokines and co-stimulatory molecules, culminating in tissue injury [96]. Overall, these data suggest that the key components of the Toll-like receptor signaling pathway may be involved in the modulation of immune homeostasis in the ileum of mice by the intervention with GY007. Future investigations should conduct further research in relevant areas to clarify the mechanism of the effects of probiotics.

The small intestine not only acts as a barrier but also provides a living environment for microorganisms [56]. On the basis of the AB/PAS staining results, we subsequently performed a more detailed analysis of the intestinal microbial community. The findings indicated that probiotic intervention restored fluoride-disrupted gut microbiota diversity, as shown by the rebound in the reduced Simpson and Shannon indices and the normalization of community composition and structure. We conducted a comparative analysis between groups to further assess the variation in Bifidobacterium levels across the different treatment cohorts. Notably, both LEfSe and machine learning analyses identified Bifidobacterium as a discriminative biomarker; moreover, it exhibited the highest feature importance score in the machine learning model, further underscoring the role of Bifidobacterium animalis subsp. animalis GY007 in modulating and maintaining gut microbial homeostasis. Administration of GY007 was associated with an enrichment of intestinal Bifidobacterium and Dubosiella, accompanied by a decrease in Paramuribaculum and Kineothrix et al. Bifidobacteria are well-established probiotics recognized for their beneficial effects on gut health. Compared with Akkermansia muciniphila, Dubosiella has been reported to exhibit stronger probiotic efficacy, providing greater protection against colonic mucosal barrier injury and attenuating intestinal inflammation [97]. Kineothrix is positively correlated with intestinal inflammation [98], while Paramuribaculum is regarded as a potential pathogenic bacterium associated with inflammatory bowel disease (IBD) [99]. GY007 may facilitate intestinal repair by modulating the microbiota environment, as reflected by the enrichment of advantageous taxa alongside a decline in potentially pathogenic bacterial groups.

The untargeted metabolomics analysis identified a total of 48 metabolites present in both comparisons. For example, the F group showed elevated levels of HIA and ISO, whereas both metabolites were markedly lower in the F+Bi group; conversely, GLA, SAB, and SIA were diminished under fluoride exposure but became significantly more abundant following administration with GY007. HIA is a benzoic acid-related compound and ranks among the predominant organic acids detected in mammalian urine, functioning as an important co-metabolite that links microbial activity in the gut with host metabolism [100,101]. Ni et al. demonstrated that the gut microbiota-mediated and host-dependent accumulation of hippuric acid results in an increase in reactive oxygen species (ROS) levels [58]. Evidence from both animal models and cell-based systems indicates that GLA contributes to preserving gut homeostasis while regulating the abundance of proteins involved in tight junction architecture, cell adhesion, and focal adhesion assemblies [102,103,104]. The selection of ISO (MP3188), OIC-7α (MN18722), and SIA (MP9998) was based on their observed correlations with multiple ileal injury indicators across the experimental groups. ISO, a nucleobase derivative, may reflect altered microbial nucleotide metabolism [105], while OIC-7α, a bile acid biosynthesis intermediate, is implicated in regulating intestinal barrier integrity and inflammation pathways through receptors like FXR [106,107]. SIA, a phenolic compound, is known for its antioxidant and anti-inflammatory effects, which contribute to reducing intestinal inflammation [63,108,109,110]. Furthermore, we performed KEGG pathway analysis on the metabolites showing differential abundance. The histidine metabolism signaling pathway was simultaneously observed in both the Ctrl vs. F comparison and the F vs. F+Bi comparison, and it holds significant importance. The metabolic process of histidine (HIS) is closely associated with the intestinal microbial community and appears to be an important determinant of the onset and worsening of chronic disorders [111]. Among human proteins, HIS contains the smallest proportion of essential amino acids. Variations in its abundance, as well as exogenous supplementation, have been linked to multiple potential health-promoting effects, notably antioxidant activity and attenuation of inflammatory responses [112,113]. However, excessive intake of HIS may elicit detrimental outcomes, including cell apoptosis and increased tissue damage [114]. This indicated the importance of maintaining a stable state rather than simply increasing or decreasing the levels of substances. Subsequently, we conducted a comprehensive analysis of the combined indicators of the microbial community, metabolic community, and intestinal function. There were correlations between the microbial community, metabolic community and the intestinal function indicators after the GY007 intervention. Further detailed analysis revealed that the Bifidobacterium was associated with each of these indicators. Among the several metabolites of concern, the prioritized metabolites, ISO, OIC 7α, and SIA, emphasized the integrative metabolic shifts that may explain the therapeutic effects of GY007 in alleviating fluoride-induced damage. The above results further demonstrate the significance of the Bifidobacterium animalis subsp. animalis GY007 and the substances that we need to pay attention to particularly.

Although this study has demonstrated the therapeutic efficacy of GY007 in alleviating fluoride-induced ileal injury, certain limitations remain. Firstly, the analysis was limited to the ileum, a specific segment of the intestine, and therefore the findings may not be generalizable to the entire gastrointestinal system. Additional studies are warranted to evaluate the impact of GY007 on other regions of the intestine. Secondly, although the 24 mg/kg body weight dosage of sodium fluoride over 8 weeks induced significant toxic effects on the murine ileum in the present study, future research should include preliminary toxicity assessments of sodium fluoride following established toxicity guidelines. Thirdly, additional investigation is required to clarify whether GY007 alleviates damage to the ileum through a primary mechanism acting on the tissue itself, or instead via secondary pathways that mediate the observed protective outcome. This could be explored through future experiments involving antibiotic-induced perturbation of the gut microbiota.

5. Conclusions

The investigation demonstrated that B. animalis subsp. animalis GY007 as a potential probiotic strain could mitigate the ileal injury and restore the microbiota–metabolite balance disrupted by high fluoride exposure. Its principal functions include maintaining the integrity of the ileal epithelial barrier and associated mucosal functions, modulating antioxidant capacity, and influencing inflammatory mediators potentially linked to the TLR9/NF-κb/IRF7 signaling pathway, as well as restoring the dysbiosis of both the microbiota and metabolome. Microbial analysis further highlighted the role of Bifidobacterium animalis in this process. Metabolic analysis identified ISO, OIC-7α, and SIA as key metabolites (Figure 10). However, the underlying mechanism remains incompletely understood, and further studies are needed to explore this aspect. Future work should employ causality-oriented approaches, such as targeted microbiota/metabolite depletion or modulation. In summary, this research identified a probiotic that could mitigate ileal injury caused by high fluoride exposure, providing valuable insights for developing therapeutic strategies to alleviate the biological harm associated with environmental fluoride pollution. Moreover, our findings offer an integrative perspective on microbiota–host interactions in the context of fluoride exposure.