Network toxicology of perfluorooctanoic acid-induced hepatointestinal injury and the protective effect of Lycium barbarum polysaccharides against perfluorooctanoic acid in broilers

Abstract

Given the extensive organ toxicity of perfluorooctanoic acid (PFOA), this study aimed to elucidate its mechanisms of hepatointestinal toxicity and to evaluate the protective efficacy of Lycium barbarum polysaccharides (LBP). In the initial phase, network toxicology and molecular docking were used to identify protein targets associated with PFOA-induced hepatointestinal toxicity. A total of 224 key protein targets were identified, which were mainly enriched in the PI3K-AKT signaling pathway. Based on these findings, an animal experiment was conducted. The study utilized 480 one-day-old male broilers which were randomly assigned to six groups, with each group containing eight replicates of ten chicks. The experimental period was from day 4 to day 42. The groups were as follows: Group C, blank control; Group P, 1.5 mg/L PFOA; Group D, 1.5 mg/L PFOA + 0.4% LBP; Group Z, 1.5 mg/L PFOA + 0.6% LBP; Group G, 1.5 mg/L PFOA + 0.8% LBP; Group L, 0.8% LBP only. The results showed that in comparison with group C, group P exhibited a significant decrease in ADG, ADFI levels, intestinal tight junction proteins expression, and activities of antioxidant enzymes. In contrast, pro-inflammatory cytokine levels were significantly increased. Supplementation with 0.8% LBP significantly ameliorated these adverse effects. Furthermore, PFOA exposure resulted in the inhibition of the PI3K-AKT pathway, while 0.8% LBP activated this pathway. To further understand the mechanisms of PFOA toxicity via the liver-intestine interactions, 16S rRNA sequencing was conducted, revealing that PFOA increased the abundances of Firmicutes and Campylobacterota while reducing Bacteroidota and Proteobacteria. The 0.8% LBP supplementation improved the microbiota composition by enhancing the diversity and abundance of beneficial bacteria, thereby conferring intestinal protection. These findings suggest that PFOA exposure can induce hepatointestinal toxicity, while LBP may serve as a feed additive to mitigate the adverse effects of PFOA. This study provides a novel approach for investigating PFOA toxicity and proposes innovative strategies for developing feed supplements to safeguard liver and intestinal health.

Graphical abstract

Introduction

Perfluorooctanoic acid (PFOA), a representative and extensively utilized perfluoroalkyl substance (PFAS). It finds application in numerous industrial and consumer goods, including non-stick coatings for cookware, waterproof textiles, food packaging materials, and firefighting foams, owing to its exceptional hydrophobic and oleophobic properties (Post et al., 2012). However, PFOA exhibits remarkable environmental persistence, bioaccumulation, and potential for long-range migration, rendering it a pervasive global contaminant (Brunn et al., 2023). It has been well-documented that PFOA is capable of causing multiorgan toxicity, with detrimental effects reported in the liver, kidneys, immune system, nervous system, reproductive organs, heart, and intestines (Chen et al., 2025; Li et al., 2025a; Liu et al., 2024; Manera et al., 2022; Nie et al., 2023; Zhang et al., 2024c). This widespread exposure has elicited significant concerns within the global public health community.

Given its role as the primary organ for material metabolism and PFOA accumulation, the liver has been the most prominent and widely investigated target of PFOA toxicity. A substantial correlation has been demonstrated between PFOA exposure and biomarkers indicating liver damage, including increased serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST). Additionally, PFOA exposure has been demonstrated to induce liver enlargement, inflammatory cell infiltration, steatosis and necrosis (Gao et al., 2025). PFOA has been verified as a potent agonist of PPARα, capable of substantially modifying liver fatty acid uptake and synthesis, thereby contributing to liver lipid metabolism disorders; this finding represents a pivotal advance in elucidating its hepatotoxicity (Yang et al., 2023). Advances in understanding the health risks of PFOA, together with the growing acceptance of the liver-intestine interactions concept in toxicology, have prompted investigators to examine the compound's effects on intestinal health and their interplay with liver injury (Han et al., 2025; Xu et al., 2022). The intestine constitutes the first line of defense against orally ingested PFOA, the gut microbiota profile and the integrity of the intestinal barrier are crucial determinants of host health (Mobley et al., 2025). Early research suggests that PFOA exposure may alters the gut microbiota, reducing beneficial microbes and promoting opportunistic pathogens. This dysbiosis may compromise the intestinal barrier by impairing tight junctions between epithelial cells, which in turn increases gut permeability and facilitates the entry of endotoxins and other harmful compounds into the portal vein (Chen et al., 2025). This sequence of events may trigger the immune-inflammatory response in the liver, worsen liver injury, and establish a detrimental cycle. Thus, while the hepatotoxicity of PFOA is relatively well understood, its impact on intestinal equilibrium and potential synergistic repercussions on overall health via the liver-intestine interactions represent a burgeoning and critical research avenue.

To address these challenges, network toxicology has developed into a new and powerful tool. It integrates bioinformatics with large databases to construct association network linking compounds, toxicity targets, and disease phenotypes for holistic toxicological evaluation (Hao et al., 2025). Advances in high-throughput biotechnologies have facilitated the growing use of integrated network toxicology-omics approaches for pinpointing toxicological targets of diverse chemical compounds, particularly environmental pollutants (Wang, 2025). Molecular docking provides predictions of ligand-target binding modes and interaction strengths (Pinzi and Rastelli, 2019). These integrated approaches enable researchers to delineate toxin-biomolecule interactions, thereby illuminating fundamental toxicological mechanisms and their effects on organisms For example, the combined use of network toxicology, molecular docking, and experimental validation has successfully elucidated the mechanistic basis of PFOA-induced reproductive toxicity and bone metabolism disruption (Shao and Fan, 2025; Wang et al., 2025a).

Accumulating evidence indicates that Lycium barbarum polysaccharides (LBP) can alleviate the adverse effects induced by exposure to environmental pollutants in humans. LBP has garnered significant interest for its potential as a prebiotic and its favorable impacts on the liver. LBP, a major active compound extracted from the medicinal fruit wolfberry, exhibits various activities including antioxidative, anti-tumor, immune-regulating, liver-protective, and neuroprotective properties (Tian et al., 2019). Previous studies have reported the neuroprotective role of LBP against the herbicide 2,4-dichlorophenoxyacetic acid (2,4-D) and its hepatoprotective effects against di (2-ethylhexyl) phthalate (DEHP) and cadmium-induced liver damage; additionally, LBP mitigates testicular damage in zebrafish exposed to nonylphenol, suggesting its potential protective impact against environmental pollutants (Liu et al., 2021; Tang et al., 2017; Varoni et al., 2017; Zhou et al., 2022). Thus, a proposed hypothesis is that LBP provides protection from PFOA-mediated organ toxicity in broilers.

We combined network toxicology, molecular docking, and in vivo experiments to systematically investigate the mechanisms of PFOA-induced hepatointestinal toxicity in broilers, and the protective effects of LBP. This study aimed to elucidate the hepatotoxic and enterotoxic effects of PFOA and to assess the efficacy of LBP in mitigating these effects, thereby establishing a scientific foundation for the use of LBP as a feed supplement.

Materials and methods

Potential targets of PFOA

The SDF file for PFOA, obtained from PubChem database, was imported into the PharmMapper database. The search was constrained to the human proteome dataset, with default settings retained for all other parameters. The identified targets were then standardized using the UniProt database to obtain their official gene symbols.

The potential targets of the disease

The keywords "liver lipid metabolism, liver injury, hepatotoxicity, intestinal toxicity, intestinal injury" were used for retrieval in GeneCards (https://www.genecards.org) and OMIM (http://www.omim.org/). The Venn plot of the intersection of the genes related to "Perfluorooctanoic acid" and "disease" was obtained through the Venny 2.1.0 software. The protein targets were imported into the STRING database to construct a protein-protein interaction (PPI) network. Visual analysis of the key protein targets, ranked by degree, was performed using Cytoscape 3.7.1. Then, the key protein targets were screened using the CentiScaPe 2.2 plugin in Cytoscape 3.7.1. software.

GO annotation and KEGG enrichment

The screened protein targets were analyzed using the DAVID database, with the results were visualized using an online bioinformatics platform (https://www.bioinformatics.com.cn/).

Molecular docking

The top six key protein targets from the PPI network were selected for molecular docking with PFOA. The source of the crystal structures was the PDB database (https://www.rcsb.org/). These structures were processed in AutoDockTools 1.5.7, where hydrogen atoms were added, and the files were exported in pdbqt format. Molecular docking was then performed. The docking results were visualized using PyMOL software, followed by analysis of non-covalent interactions using the PLIP tool (https://plip-tool.biotec.tu-dresden.de/plip-web/plip/index).

Animal ethics statement

This study was approved by the Experimental Animal Ethics Committee of Hebei Agricultural University, and was conducted in accordance with national regulations and animal welfare standards.The ethical approval number is 2024039.

Experimental materials

LBP and PFOA were obtained from Xi'an Shengqing Biotechnology Co., Ltd. (purity ≥ 80%) and Sigma-Aldrich (purity ≥ 95%), respectively.

Animal grouping

The study utilized 480 one-day-old male Arbor Acres broilers (49.8 ± 3.51 g) which were randomly assigned to six groups, with eight replicates per group and ten chicks per replicate. Group C, blank control; Group P, 1.5 mg/L PFOA in drinking water (based on our previous research); Group D, 1.5 mg/L PFOA + 0.4% LBP; Group Z, 1.5 mg/L PFOA + 0.6% LBP; Group G, 1.5 mg/L PFOA + 0.8% LBP; Group L, 0.8% LBP only (Long et al., 2020). The broilers were reared in three-layer cages with appropriate light and ad libitum access to feed. The specific composition and nutrient levels of the diets are presented in Table 1. The experimental period was from day 4 to day 42.

Table 1.

Nutrient levels of basic diet formulas for experimental animals(Dry matter MJ / kg %).

	Content
	1-21 days of age	22-42 days of age
Corn grain	55.58	58.1
Soybean meal	23.98	20.55
Fish meal	4	4
Cottonseed meal	4	3.12
Corn gluten meal	3	4
Soybean oil	3.4	4.41
Limestone	1	0.9
Lysine(Lys)	0.8	0.9
NaCl	0.32	0.32
CaHPO4	1.55	1.4
Methionine(Met)	0.27	0.2
Trace mineral premix	2	2
Choline-Cl	0.1	0.1
Total	100.00	100.00
Nutrient levels
Metabolic energy (ME)/(MJ/kg)	12.34	12.76
Crude protein(CP)	21.8	20.7
Calcium(Ca)	1.05	0.96
Phosphorus(P)	0.56	0.52
Lysine(Lys)	1.09	1

Note: Pre mixed feed for broilers can be supplied as full price compound feed per kilogram: Vitamin A 5000IU, Vitamin D3 1500IU, Vitamin E 10 IU, Vitamin K3 1.5 mg, Vitamin B1 3.2 mg, Vitamin B2 7.6 mg, Vitamin B6 3.2 mg, Vitamin B12 12 μg, Niacin 40 mg, Pantothenic acid 15 mg, Folic acid 1.5 mg, Biotin 0.18 mg, Cu 5 mg, Fe 82 mg, Mn 95.2 mg, I 1.14 mg, Se 0.20 mg, Zn 35 mg.

Sampling and measurements

Body weight was recorded at the end of each dietary phase to calculate average daily gain (ADG), average daily feed intake (ADFI), and feed conversion ratio (FCR). On day 42, one bird per replicate was fasted for 12 hours, and final body weight was recorded. Blood was then collected from the wing vein, followed by euthanasia and necropsy. The liver was weighed to calculate the organ index, expressed as g/100 g body weight. Following collection, all samples (liver, duodenum, and cecal contents) were snap-frozen and maintained at −80°C until further analysis. A portion of the liver and duodenum were thoroughly rinsed with normal saline and then fixed in 4% paraformaldehyde for paraffin sectioning. No experimental animals died throughout the study period.

Detection of serum indicators

Levels of ALT (C009-2-1), AST (C010-2-1) in serum were determined using commercial kits (Nanjing Jiancheng Bioengineering Institute, Nanjing, Jiangsu, China). Immunoglobulin G (IgG, ml042771) in serum was determined using corresponding commercial kits (Shanghai Enzyme-Linked Biotechnology Co., Ltd., Shanghai, China).

Liver and intestinal histopathology

For histological analysis, one sample per replicate was processed. Three separate areas from the liver and intestinal tissues were embedded, cut into 5 μm sections using a microtome (Leica SM2010 R), followed by hematoxylin and eosin (H&E) staining for histological examination under a light microscope.

Oxidative stress and protein analysis in liver and intestine

The levels of Catalase (CAT, A007-1-1), Malondialdehyde (MDA, A003-1-2), Total antioxidant capacity (T-AOC, A015-3-1), Glutathione peroxidase (GSH-Px, A005-1-2), Superoxide dismutase (SOD, A001-3-2) in the organs were measured using commercial assay kits (Nanjing Jiancheng Bioengineering Institute, Jiangsu, China) according to the manufacturers' protocols. Similarly, the concentrations of Zonula Occludens-1(ZO-1, ml060977), Occludin (ml060976), B-Cell Lymphoma 2(Bcl-2, ml037011), BCL2-Associated X Protein (BAX, ml059826), Interleukin-1β (IL-1β, ml002790), Tumor necrosis factor-α (TNF-α, ml059835), Interleukin-6(IL-6, ml059839) were determined with corresponding commercial kits (Shanghai Enzyme-Linked Biotechnology Co., Ltd., Shanghai, China).

Gene expression analysis in liver and intestine

Total RNA was isolated from organ samples with a commercial kit (LS2052, Shanghai Promega Biological Products Co., Ltd., China). Subsequently, cDNA was synthesized using the reverse transcription kit (LS2052, Shanghai Promega Biological Products Co., Ltd., China). For Quantitative real-time PCR (qPCR) analysis, amplification was performed on fluorescence quantitative PCR instrument (Gentier 48 R, Tianlong Technology Co., Ltd., Xi'an) to determine the mRNA expression of relevant genes. The primers were designed and supplied by Dalian Bao Biological Company, with the internal reference gene being β-actin (Table 2). Quantification of gene expression was determined by the 2^−△△Ctmethod.

Table 2.

Primer Sequence and information.

Gene	Primer sequences(5′→3′)	GenBank login number
AKT1	F: 5′-AAGAAACAGGAGGAAGAGATGATGG-3′	XM_046917866.1
	R: 5′-GAAACTTCCATTTCTTCAGCACCTG −3′
PI3K	F: 5′-CTTCTGGAGTCCTATTGTCG-3′	XM_046923916.1
	R: 5′-CACCTTCTGGGTCTCATCTT −3′
Bax	F: 5′-GAGCACCTGAAGAGGGTCTATGG-3′	NC_052565.1
	R: 5′-GAACAGATGGGTGACCACATTGAC-3′
Bcl-2	F: 5′-GTGGAATTGTACGGCAACAGTATG-3′	NC_052533.1
	R: 5′-ACCAGAACCAGGCTCAGGATG-3′
Caspase 3	F: 5′-TTTGTTTGTGTGTTGCTAAGCCATG-3′	NC_052535.1
	R: 5′-GACTTCTGCACTTGTCACCTCTG-3′
Claudin-1	F: 5′-AGATCCAGTGCAAGGTGTACG-3′	NM_001013611.2
	R: 5′-AAACACACCAACCAGACCCA-3′
Occludin	F: 5′-TGAATGCACCCACTGAGTGTT-3′	NM_205128.1
	R: 5′-CCAGAGGTGTGGGCCTTAC −3′
ZO-1	F: 5′-TTC AGG TGT TTC TCT TCC TCC TC-3′	XM_015278981.2
	R: 5′-CTG TGG TTT CAT GGC TGG ATC-3′
β-actin	F: 5′-TTGTTGACAATGGCTCCGGT-3′	NM_205518.2
	R: 5′-TCTGGGCTTCATCACCAACG-3′

AKT1=AKT Serine/Threonine Kinase 1; PI3K=Phosphatidylinositol 3-Kinase; Bax=BCL2-Associated X Protein; Bcl-2 = B-Cell Lymphoma 2; Caspase 3=Cysteine-Aspartic Acid Protease 3; ZO-1=Zonula Occludens-1.

Analysis of intestinal flora

For sequencing analysis, DNA was extracted from cecal contents of Group C (blank control), Group P (PFOA model), and Group G (0.8% LBP + PFOA). Following library construction with the TruSeq® DNA PCR-Free Sample Preparation Kit, library quality was evaluated via Qubit and qPCR. Qualified libraries were subsequently sequenced on the NovaSeq 6000 system.

Sequence clustering was performed using Uparse (v7.0.1001) with a 97% identity threshold to generate operational taxonomic units (OTUs). Taxonomic annotation of OTUs was conducted against the SILVA138 SSU rRNA database via the Mothur algorithm. Alpha diversity and UniFrac-based UPGMA trees were analyzed in QIIME. Visualization tools (rarefaction, rank abundance, and species accumulation curves; PCoA and NMDS ordination) were implemented in R (v2.15.3). LEfSe analysis was performed using LEfSe software, and Metastats analysis was conducted using R software. Spearman correlations (significance-tested via corr.test) between species and environmental factors were visualized as a heatmap using pheatmap. Functional prediction was conducted based on OTU trees and gene annotations in the Greengenes database.

It was completed in collaboration with Beijing Novogene Bioinformatics Technology Co., Ltd.

Statistical analysis

Data were analyzed using one-way analysis of variance (ANOVA) followed by Duncan's multiple range test for comparisons among multiple groups, using SPSS 25.0. Results are expressed as mean ± standard error of the mean (SEM). Statistical significance was set at P < 0.05. Bar charts were generated using GraphPad Prism 9.0.

Results

Potential targets of PFOA and diseases

The targets of PFOA screened from the PharmMapper database were standardized using the UniProt database. After deduplication, 285 potential targets were obtained. The screening and deduplication of disease-related genes obtained from GeneCards and OMIM resulted in a final set of 4,301 targets. Among these, targets retrieved from the GeneCards database using the keywords "liver lipid metabolism, liver injury, hepatotoxicity, intestinal toxicity, intestinal injury" were further screened with a Relevance score ≥ 10. Overlapping protein targets of the disease and PFOA were identified, resulting in 224 key targets, with the intersection visualized in a Venn diagram (Fig. 1A).

Fig. 1.

Acquisition of overlapping targets for PFOA and diseases. (A) Venn plot. (B) PPI Network Plot and Visualization analysis. (C) Core genes screened by the Centiscape 2.2 plugin.

Key protein targets were analyzed using the STRING database. The constructed PPI network consisted of 224 nodes and 2,737 edges, exhibiting an average node degree of 24.4 and a significant PPI enrichment p-value (< 1.0e-16). Subsequently, it was visualized in Cytoscape software for a degree-based analysis. The degree values of the target proteins ALB, AKT1, ESR1, HSP90AA1, EGFR, CASP3, PPARG, SRC, MMP9, and HSP90AB1 were among the highest (Fig. 1B). These targets have been previously associated with pathological processes relevant to PFOA-induced hepatointestinal toxicity.

PPI network data were screened for gene clusters and core targets using the CentiScaPe 2.2 plugin. The threshold parameters were as follows: Betweenness unDir > 257.7636364, Closeness unDir > 0.002134357, Degree unDir > 49.76363636. After screening, a total of 45 nodes and 988 edges were obtained. The identified core targets are shown in Fig. 1C. The screened core genes include 45 genes such as ALB, AKT1, ESR1, HSP90AA1, EGFR, CASP3, PPARG, MMP9, HSP90AB1, and SRC.

GO functional and KEGG pathway enrichment analysis

The 45 core proteins identified using the CentiScaPe 2.2 plugin were subsequently subjected to GO functional annotation and KEGG signaling pathway enrichment analysis using the DAVID database. The findings are presented in Fig.s 2A and 2B. The significant biological processes (BP) mainly encompassed terms such as the negative regulation of apoptosis and nuclear receptor-mediated steroid hormone signaling. The cellular components (CC) included mitochondria and cytosol, among others. The molecular functions (MF) included enzyme binding and protein binding, among others. The KEGG analysis revealed substantial enrichment in signaling pathways such as IL-17, PI3K-AKT, HIF-1, FoxO, and PPAR signaling pathways.

Fig. 2.

Enrichment analysis and molecular docking of core targets. (A) GO functional annotation; (B) KEGG signaling pathway enrichment analysis; (C) Molecular docking: a. PFOA and AKT1; b. PFOA and ALB; c. PFOA and CASP3; d. PFOA and HSP90AA1; e. PFOA and MMP9; f. PFOA and PPARG. The green structure represents amino acid residues of the protein, the yellow structure represents PFOA, the yellow dotted lines represent hydrogen bonds, and the numbers beside the lines represent bond distances.

Molecular docking

To validate the network pharmacology prediction results, the top six key protein targets were selected for molecular docking. The results are shown in Fig. 2C. PFOA exhibited binding interactions with these protein targets via hydrogen bonds and other interactions.

The influence of LBP on PFOA-induced in growth performance

According to Table 3, compared with Group C, Group P exhibited significantly lower ADG and ADFI, but a higher FCR . In contrast Group G showed significantly higher ADG and ADFI, along with a lower FCR compared to Group P. However, during the period from day 22 to day 42, no significant differences in ADFI were observed among the groups.

Table 3.

Evaluation of LBP's Effects on Broiler Growth Performance.

Item	Treatment groups						SEM	P-value
	C	P	L	D	Z	G
4 to 21 days of age
ADG, g	37.27a	33.72d	37.52a	35.23c	35.67c	36.5b	0.208	<0.001
ADFI, g	48.8a	46.04e	48.62ab	46.87d	47.13cd	47.87bc	0.223	<0.001
FCR	1.31b	1.37a	1.3b	1.33ab	1.32b	1.31b	0.001	0.04
22 to 42 days of age
ADG, g	81.36ab	77.4d	82.37a	78.99cd	79.56bc	80.49abc	0.334	<0.001
ADFI, g	137.21	135.55	137.08	136.01	136.28	137.04	0.252	0.314
FCR	1.69bc	1.75a	1.66c	1.72ab	1.71abc	1.7abc	0.001	0.033
4 to 42 days of age
ADG, g	81.36ab	77.4d	82.37a	78.99c	79.56c	80.49b	0.242	<0.001
ADFI, g	93.01a	90.8d	92.85a	91.44cd	91.7bc	92.45ab	0.195	<0.001
FCR	1.57cd	1.63a	1.55d	1.6b	1.59cb	1.58cb	0.001	<0.001

The presence of distinct superscript letters within a row indicates a statistically significant difference (n = 8, P < 0.05). SEM: standard error of the mean; ADG : average daily weight gain; ADFI: average daily feed intake; FCR : feed conversion ratio. C: blank control group; P: 1.5mg/L PFOA model group; D: 1.5mg/L PFOA+0.4%LBP; Z: 1.5mg/L PFOA+0.6%LBP; G: 1.5mg/L PFOA+0.8%LBP; L: 0.8%LBP control group.

The influence of LBP on liver and immune function following PFOA exposure

Compared with Group C, a significant increase in the liver index, ALT and AST activities was observed following PFOA exposure. Conversely, intervention with Group G significantly reduced these liver parameters relative to Group P. Furthermore, compared with Group C, serum IgG induced by PFOA was significantly reduced, while compared with group P, IgG in group G was significantly increased (P < 0.001) (Table 4).

Table 4.

The impact of LBP on liver function and immune function.

Item	Treatment groups						SEM	P-value
	C	P	L	D	Z	G
Liver
Liver Index,%	2.04b	2.17a	2.04b	2.15ab	2.11ab	2.07ab	0.012	0.047
ALT, U/L	6.23b	7.11a	6.05b	6.87ab	6.56ab	6.23b	0.118	0.056
AST, U/L	109.06b	127.17a	108.8b	122.14ab	118.23ab	111.12b	2.07	0.034
Immune function
IgG, g/L	35.01ab	30.38e	36.23a	32.42d	33.43cd	34.54bc	0.322	<0.001

The presence of distinct superscript letters within a row indicates a statistically significant difference (n = 8, P < 0.05). SEM: standard error of the mean; ALT : Alanine aminotransferase; AST : Aspartate aminotransferase; IgG :Immunoglobulin G. C: blank control group; P: 1.5mg/L PFOA model group; D: 1.5mg/L PFOA+0.4%LBP; Z: 1.5mg/L PFOA+0.6%LBP; G: 1.5mg/L PFOA+0.8%LBP; L: 0.8%LBP control group.

Protective effect of LBP against PFOA-induced histopathology in the liver and intestine

Compared with Group C, PFOA exposure caused significant damage to liver tissue, characterized by blurred liver cell structures, vacuolar degeneration, and pyknosis and karyolysis. Compared with Group P, LBP supplementation alleviated this damage, in particular, in Group G, the hepatocytes were neatly arranged and the nuclear structures were intact (Fig. 3A).

Fig. 3.

Effects of LBP on histomorphology of broilers induced by PFOA. (A) Liver histological morphology (→ indicates vacuolar degeneration; ▲ indicates nuclear pyknosis); (B) Intestinal histological morphology (→ indicates villi rupture; ▲ indicates villi shedding). C: blank control group; P: 1.5mg/L PFOA model group; D: 1.5mg/L PFOA+0.4%LBP; Z: 1.5mg/L PFOA+0.6%LBP; G: 1.5mg/L PFOA+0.8%LBP; L: 0.8%LBP control group.

Compared with Group C, PFOA exposure induced epithelial cell degeneration, necrosis, exfoliation, and villus rupture. Compared with Group P, LBP supplementation, particularly in Group G, notably restored mucosal integrity, resulting in neatly arranged villi (Fig. 3B).

LBP's effect on PFOA-induced changes in oxidative stress indicators

Compared with Group C, PFOA exposure was associated with significant decreases in antioxidant enzyme levels in the liver (SOD, GSH-Px, CAT) and intestine (SOD and T-AOC) (P < 0.001), as well as a marked increase in MDA content in both tissues. Compared with Group P, all the above indicators in Group G were markedly reversed, with the exception of liver GSH-Px activity. The results are shown in Table 5.

Table 5.

The effect of LBP on liver and intestinal oxidative stress indicators.

Item	Treatment groups						SEM	P-value
	C	P	L	D	Z	G
Liver
SOD, U/mgprot	120.45ab	102.81d	130.38a	105.7cd	110.06bcd	117.01bc	2.105	<0.001
MDA, nmol/mgprot	1.11bc	1.23a	1.09c	1.21ab	1.13abc	1.11bc	0.016	0.03
GSH-Px, U/mgprot	97.72ab	89.24c	98.99a	90.11bc	92.22abc	95.26abc	1.121	0.044
CAT, U/mgprot	7.46ab	6.38c	7.91a	6.75bc	7.06abc	7.38ab	0.135	0.01
Intestinal
SOD, U/mgprot	104.07a	85.26e	105.56a	88.15d	95.65c	101.06b	0.003	<0.001
MDA, nmol/mgprot	2.29c	3.02a	2.02d	2.75b	2.46c	2.38c	1.154	<0.001
T-AOC, mM	0.76a	0.72d	0.76a	0.73cd	0.74bc	0.75ab	0.057	<0.001

The presence of distinct superscript letters within a row indicates a statistically significant difference (n = 8, P < 0.05). SEM: standard error of the mean; SOD : Superoxide dismutase; GSH-Px : Glutathione peroxidase; CAT : Catalase; T-AOC:Total sntioxidant capacity; MDA:Malondialdehyde. C: blank control group; P: 1.5mg/L PFOA model group; D: 1.5mg/L PFOA+0.4%LBP; Z: 1.5mg/L PFOA+0.6%LBP; G: 1.5mg/L PFOA+0.8%LBP; L: 0.8%LBP control group.

LBP's effect on PFOA-induced changes in pro-inflammatory cytokine

Compared with Group C, PFOA exposure dramatically augmented the levels of IL-1β, IL-6, and TNF-α in the liver and intestine. Compared with Group P, the levels of these factors in Group G were substantially lowered (Table 6).

Table 6.

Evaluation of LBP's Impact on Liver and intestinal inflammatory.

Item	Treatment groups						SEM	P-value
	C	P	L	D	Z	G
Liver
TNF-α, pg/mL	26.22c	29.5a	26.18bc	28.95ab	27.62abc	26.8c	0.639	0.013
IL-1β, pg/mL	64.76d	71.26a	64.98d	70.23ab	67.89bc	65.65cd	0.513	<0.001
IL-6, pg/mL	26.22b	29.5a	26.18b	28.95ab	27.62ab	26.8b	0.323	0.003
Intestinal
TNF-α, pg/mL	52.51b	57.58a	51.88b	56.92a	54.45ab	52.99b	0.583	0.008
IL-1β, pg/mL	53.56c	61.89a	51.91c	60.46ab	58.12b	54.04c	0.735	<0.001
IL-6, pg/mL	18bc	20.41a	17.37c	19.41ab	18.84abc	18.02bc	0.254	0.002

The presence of distinct superscript letters within a row indicates a statistically significant difference (n = 8, P < 0.05). SEM: standard error of the mean; TNF-α : tumor necrosis factor-α; IL-1β: Iinterleukin - 1β; IL-6 : Interleukin-6. C: blank control group; P: 1.5mg/L PFOA model group; D: 1.5mg/L PFOA+0.4%LBP; Z: 1.5mg/L PFOA+0.6%LBP; G: 1.5mg/L PFOA+0.8%LBP; L: 0.8%LBP control group.

LBP's effect on PFOA-induced changes in intestinal tight junction proteins

Compared with Group C, PFOA exposure substantially lowered the protein abundance of Occludin and ZO-1, along with mRNA levels of Occludin, Claudin-1, and ZO-1. An upregulation was observed in Group G compared to Group P, manifested as higher protein levels of ZO-1 and Occludin and elevated mRNA levels of ZO-1, Occludin, and Claudin-1 (Table 7 and Fig. 4).

Table 7.

Modulation of Pathway-Associated and Tight Junction Proteins by LBP.

Item	Treatment groups						SEM	P-value
	C	P	L	D	Z	G
Liver
BAX, pg/mL	6.01c	7.89a	5.97c	7.37b	6.98b	6.17c	0.12	<0.001
Bcl-2, pg/mL	58.79a	48.32e	59.13a	50.91d	54.15c	55.97b	0.611	<0.001
Intestinal
BAX, pg/mL	4.76d	6.8a	4.45e	6.28b	5.46c	5.02d	0.128	<0.001
Bcl-2, pg/mL	54.02ab	42e	55.08a	45.12d	49.59c	53.03b	0.741	<0.001
Intestinal TJ protein
ZO-1, ng/mL	64.77ab	56.32d	65.16a	60.89c	61.75c	62.27bc	0.549	<0.001
Occludin, ng/mL	14.4ab	12.08d	14.64a	12.95cd	13.46bc	13.98ab	0.178	<0.001

The presence of distinct superscript letters within a row indicates a statistically significant difference (n = 8, P < 0.05). SEM: standard error of the mean; BAX: BCL2-Associated X; Bcl-2: B-cell lymphoma-2; ZO-1: zonula occludens-1. C: blank control group; P: 1.5mg/L PFOA model group; D: 1.5mg/L PFOA+0.4%LBP; Z: 1.5mg/L PFOA+0.6%LBP; G: 1.5mg/L PFOA+0.8%LBP; L: 0.8%LBP control group.

Fig. 4.

Effects of LBP on PFOA-induced intestinal tight junction proteins in broilers. The presence of distinct superscript letters indicates a statistically significant difference (n = 8, P < 0.05). ZO-1=Zonula Occludens-1. Statistical evaluation was done using one-way ANOVA. C: blank control group; P: 1.5mg/L PFOA model group; D: 1.5mg/L PFOA+0.4%LBP; Z: 1.5mg/L PFOA+0.6%LBP; G: 1.5mg/L PFOA+0.8%LBP; L: 0.8%LBP control group.

The influence of LBP on PFOA-mediated alterations in PI3K-AKT signal transduction

Compared with Group C, PFOA exposure sharply diminished the protein levels of Bcl-2 in the liver and intestine, as well as the mRNA levels of BCL-2, AKT1, and PI3K. Additionally, PFOA exposure significantly increased the protein levels of Bax and the mRNA levels of BAX and CASP3. Compared with Group P, all the above indicators in the liver and intestine of Group G were markedly reversed (Table 7 and Fig.s 5A-B).

Fig. 5.

Analysis of PI3K-AKT signaling pathway genes and correlation of liver and intestinal indicators. (A) Liver PI3K-AKT pathway genes; (B) intestinal PI3K-AKT pathway genes; (C) Liver-intestine correlation of indicators. The presence of distinct superscript letters indicates a statistically significant difference (n = 8, P < 0.05). AKT1=AKT Serine/Threonine Kinase 1; PI3K=Phosphatidylinositol 3-Kinase; Bax=BCL2-Associated X Protein; Bcl-2 = B-Cell Lymphoma 2; Caspase 3=Cysteine-Aspartic Acid Protease 3. Statistical evaluation was done using one-way ANOVA. *P < 0.05, ** P < 0.01. C: blank control group; P: 1.5mg/L PFOA model group; D: 1.5mg/L PFOA+0.4%LBP; Z: 1.5mg/L PFOA+0.6%LBP; G: 1.5mg/L PFOA+0.8%LBP; L: 0.8%LBP control group.

Correlation analysis of liver and intestinal indicators

As shown in Fig. 5C, markedly positive correlations were identified between intestinal and liver levels of both pro-inflammatory cytokine and MDA (P < 0.01). The activity of intestinal antioxidant enzymes was significantly positively correlated with that of liver antioxidant enzymes, while the activity of intestinal SOD showed no significant correlation with that liver GSH-Px. Gene expression levels of the intestinal and liver PI3K-AKT pathways were positively correlated.

The effect of LBP on PFOA-induced changes in intestinal flora

The data presented in Fig. 6A–C confirmed that the achieved sequencing depth supported a reliable representation of the cecal microbiota profile. Analysis of the samples yielded 1,119 operational taxonomic units (OTUs). Among them, Group C had 957 OTUs (138 unique OTUs), Group P had 881 OTUs (49 unique OTUs), and Group G had 907 OTUs (58 unique OTUs). PFOA exposure suppressed OTU richness, whereas LBP supplementation was associated with increased OTU richness (Fig.s 6D-E). This suggests that high-dose LBP may improve intestinal microbiota homeostasis disrupted by PFOA.

Fig. 6.

The effect of LBP on PFOA-induced changes in intestinal flora sample size and OTUs. (A) Dilution curve; (B) Species accumulation box plot; (C) Rank Abundance curve; (D) OTU Wayne diagram; (E)Total OTUs in each group. C: blank control group; C:blank control group; P: 1.5mg/L PFOA model group; G: 1.5mg/L PFOA+0.8%LBP.

Intervention of LBP against PFOA-induced dysbiosis in intestinal flora diversity

Alpha diversity analysis showed no statistically significant intergroup variations (Fig. 7A). NMDS and PCoA plots (Fig. 7B) were generated based on the binary Jaccard algorithm. These plots revealed certain differences among the three groups, suggesting that PFOA exposure substantially reshaped the structural profile of the cecal microbial community. LBP supplementation was associated with partial restoration of the intestinal microbiota imbalance induced by PFOA. The UPGMA clustering tree (Fig. 7C) showed high similarity between Group C and Group G, while similarity between these two groups and Group P was low. In line with above findings, the results suggest that LBP may mitigate gut microbiota dysbiosis caused by PFOA. In addition, the intestinal microbiota in this study primarily consisted of the Bacteroidota, Firmicutes, and Campylobacterota.

Fig. 7.

The effect of LBP on PFOA-induced changes in intestinal flora diversity. (A) Alpha diversity; (B) PCoA and NMDS analysis; (C) UPGMA cluster tree. C: blank control group; C:blank control group; P:1.5mg/L PFOA model group; G:1.5mg/L PFOA+0.8%LBP.

Effects of LBP on PFOA-induced changes in intestinal microbiota abundance

At the phylum level, PFOA exposure increased the Firmicutes, Campylobacterota, and Verrucomicrobiota and decreased the proportions of Bacteroidota and Proteobacteria. The addition of LBP reversed these changes, leading to a further elevation in Verrucomicrobiota abundance. At the genus level, PFOA exposureincreased the Faecalibacterium, Campylobacter, and Rikenella, and decreased the proportions of Alistipes and Barnesiella. However, the addition of LBP reversed these changes, leading to a further elevation in Rikenella abundance (Fig. 8A-B).

Fig. 8.

Effects of LBP on PFOA-induced changes in intestinal microbiota abundance. (A) Phylum level; (B) Genus level; (C) Linear discriminant analysis (LDA) score (LDA > 3); (D) Linear discriminant analysis effect size (LEfSe) analysis. C: blank control group; C:blank control group; P:1.5mg/L PFOA model group; G:1.5mg/L PFOA+0.8%LBP.

To further reveal the dominant species in the cecal microbiota of broilers, a species comparative analysis was performed, as shown in Fig. 8C-D. Group G was enriched in p__Proteobacteria, including c__Gammaproteobacteria, o__Burkholderiales, and f__Sutterellaceae. Group P was enriched in f__Acidaminococcaceae, o__Acidaminococcales, and p__unidentified bacteria, including c__Saccharimonadia, o__Saccharimonadales, f__Saccharimonadaceae, c__Bacteroidia, o__Bacteroidales, and f__Bacteroidaceae.

The MetaStat complex heatmap highlighted the species differences among the groups (Fig. 9A). Marked differences in the abundances of Colidextribacter, Parasutterella, Phascolarctobacterium, and Ruminococcus were observed between Group C and Group P. Marked differences in the relative abundances of Odoribacter and Parasutterella were observed between Group P and Group G; and Lachnoclostridium exhibited a significant difference between Group C and Group G.

Fig. 9.

Effects of LBP on PFOA-induced changes in intestinal microbiota: MetaStat complex heatmap, clustering heatmap, and ternary phase diagrams. (A) MetaStat complex heatmap; (B) Clustering heatmap; (C) Ternary diagram showing phylum-level abundances; (D) Ternary diagram showing genus-level abundances. C: blank control group; C:blank control group; P:1.5mg/L PFOA model group; G:1.5mg/L PFOA+0.8%LBP.

Clustering analysis of the top 10 species at the phylum level was performed. The heatmap revealed that the species composition of Group C and Group G was more similar. Deferribacteres, Verrucomicrobiota, Cyanobacteria, Proteobacteria, Actinobacteriota, and Actinobacteria were more abundant in Group G, while Campylobacterota and unidentified_Bacteria were more abundant in Group P (Fig. 9B).

Ternary phase diagrams indicated that Firmicutes and Bacteroidota exhibited relatively high relative abundances among the three groups, while Campylobacterota was mainly concentrated in Group P (Fig. 9C). At the genus level, Barnesiella, Parasutterella, and Ligilactobacillus were predominantly found in Group C and Group G, while UCG-005 and Campylobacter were mainly concentrated in Group P. These results corroborate the findings described above (Fig. 9D).

Correlation analysis

Fig. 10A displays the correlation analysis conducted for the top 35 bacterial genera. The levels of intestinal and liver inflammatory cytokine, liver MDA level, expression levels of BAX and CASP3 in the liver and intestine, and liver function indices were strongly inversely correlated with Phascolarctobacterium and strongly positively associated with Fusicatenibacter. The serum IgG level, liver and intestinal antioxidant enzymes, expression levels of BCL-2, AKT1, and PI3K in the liver and intestine, and expression levels of intestinal tight junction proteins were significantly correlated positively with Phascolarctobacterium but inversely with Fusicatenibacter. The liver TNF-α level was significantly negatively correlated with Parasutterella and Mucispirillum.

Fig. 10.

Correlation analysis and functional prediction analysis. (A) Correlation analysis; (B) KEGG level 1 heatmap; (C) KEGG level 3 heatmap. Statistical analysis for correlation was performed using Pearson correlation analysis. *P < 0.05, ** P < 0.01. C: blank control group; C:blank control group; P:1.5mg/L PFOA model group; G:1.5mg/L PFOA+0.8%LBP.

Functional prediction

As shown in Fig. 10B, at the KEGG level 1, the PFOA-treated Group P was over-represented in Environmental_Information_Processing and Cellular_Processes, while Group C and Group G showed the opposite pattern, with enrichment in Human Diseases, Organismal Systems, Metabolism, and Genetic Information Processing.

As shown in Fig. 10C, at the KEGG level 3, a pronounced difference was found among the three groups. Group C was markedly enriched in Genetic Information Processing and Metabolism, Group G was markedly enriched in Metabolism, and Group P was significantly enriched in Environmental Information Processing and Genetic Information Processing (Environmental Information Processing is an important category of biological metabolic pathways). In summary, functional prediction of the intestinal flora suggested that PFOA may exert toxic effects by affecting biological metabolism, while LBP exerts a protective effect by improving metabolism.

Discussion

We combined network toxicology, molecular docking, and in vivo experiments to systematically explore the mechanisms underlying PFOA-induced hepatointestinal toxicity. Potential toxicity targets in the liver and intestine were identified from multiple databases, and molecular docking simulations were used to predict PFOA-target binding affinities. Furthermore, experimental evidence from animal models demonstrated a correlation between PFOA-induced liver and intestinal toxicity and the PI3K-AKT pathway. We assessed the therapeutic potential of a Chinese herbal medicine against PFOA-induced organ toxicity. In vivo experiments indicated that LBP may alleviate the organ toxicity associated with PFOA. By employing a multi-dimensional approach, this study lays the groundwork for deciphering the toxicity mechanisms of environmental pollutants and paves the way for preventive and therapeutic interventions against PFOA-related liver and intestinal damage.

Network toxicology plays a central role in toxicity prediction and elucidating toxicological mechanisms (Ge et al., 2024). The combined application of network toxicology and molecular docking in recent investigations has revealed the pathological mechanisms of action for diverse pollutants. These methods, for instance, have successfully revealed the key targets and pathways implicated in aflatoxin B1-induced hepatointestinal toxicity (Ge et al., 2024). Recent research utilizing network toxicology and molecular docking has elucidated the molecular mechanisms underlying diisononyl cyclohexane-1,2-dicarboxylate (DINCH)-induced hepatotoxicity, particularly its interactions with core protein targets (Xin et al., 2025). The application of these methods in prior studies has revealed key molecular events in PFOA-induced disruption of spermatogenesis (Luo et al., 2025). To investigate PFOA's hepatointestinal toxicity, we first identified potential targets using network toxicology and then verified the interactions through molecular docking by assessing the binding affinity of PFOA with core proteins.

In this study, network toxicology analysis identified 224 overlapping targets. Among these, ALB and AKT1 exhibited relatively high degree rankings and were considered potential pathogenic targets of PFOA. A decline in ALB, a liver-derived protein critical for transport and homeostasis, can cause an accumulation of free PFOA in liver tissue, thereby exacerbating its direct hepatotoxic effects (Chi et al., 2018). Furthermore, an imbalance in ALB and disruption of the internal environment may cause liver tissue edema and damage to the intestinal barrier. The role of AKT1 in cellular regulation involves orchestrating critical processes like metabolism, proliferation, and apoptosis via phosphorylation of downstream targets (Ge et al., 2024). Previous studies have suggested that the reproductive toxicity of PFOA in mice is closely associated with the down-regulation of AKT1 expression (Wang et al., 2025a). To further elucidate the hub targets related to liver and intestinal toxicity of PFOA, 45 key genes were identified based on the PPI network, using the Centiscape algorithm. The top ten key genes included ALB, AKT1, ESR1, HSP90AA1, EGFR, CASP3, PPARG, MMP9, HSP90AB1, and SRC.

Albumin (ALB) is essential for cellular homeostasis, and AKT1 is a known regulator of intracellular signaling cascades. CASP3 orchestrates apoptotic cell death, while MMP9 participates in tissue repair and inflammatory responses. PPARG has been shown to regulate various physiological processes, including lipid metabolism, inflammatory responses and immunity. HSP90AB1 and HSP90AA1 are known to act as molecular chaperones for proteins such as AKT1. ESR1, EGFR, and SRC are established regulators of key cellular processes-including signal transduction, proliferation, and differentiation, with well-documented roles in oncogenesis. PFOA exposure is associated with carcinogenic effects and poses a significant risk for multiple cancer types, notably liver and kidney cancer, as evidenced by prior research (Lee et al., 2023). Furthermore, PFOA exposure has been linked to oxidative stress, apoptosis, and imbalances in intestinal homeostasis (Chen et al., 2025; Wang et al., 2025a). Consequently, it is hypothesized that the mechanisms underlying the hepatointestinal toxicity of PFOA not only contribute to the development of cancer cells but also associated with oxidative stress, inflammatory responses, apoptosis, and disruption of homeostasis.

To gain a comprehensive understanding of the hepatointestinal toxicity of PFOA, enrichment analysis was performed on the 45 key genes. GO functional enrichment analysis indicated that apoptosis was the primary biological process associated with the liver and intestinal toxicity of PFOA. Furthermore, the PI3K-AKT signaling pathway was highlighted as the predominant toxicity-related pathway by KEGG enrichment analysis. The PI3K-AKT pathway orchestrates a range of fundamental cellular processes, encompassing metabolism, cell survival, migration, and proliferation (Wang et al., 2025b). This pathway also regulates the FoxO signaling pathway, which in turn influences cellular apoptosis. Previous studies have reported that PFOA exposure is associated with downregulation of PI3K and AKT mRNA expression, which may contribute to reproductive toxicity (Wang et al., 2025a). Additionally, the IL-17 signaling pathway contributes to the host's acute immune defense by promoting the release of cytokines and chemokines, facilitating neutrophil recruitment, and regulating helper T cell differentiation, PFOA exposure has been associated with increased IL-17 levels, which may be linked to immunotoxicity (Sun et al., 2025). Moreover, signaling pathways such as HIF-1, PPAR, and MAPK warrant further investigation.

Molecular docking confirmed the direct binding of PFOA to its target proteins. The findings suggest that PFOA exhibits robust binding affinity with the six key target proteins, which may contribute to the long-term biological effects of this persistent pollutant (Luo et al., 2025). By binding to these critical proteins, PFOA may interfere with their normal functions, potentially leading to sustained disruption of their functions. Even with fluctuations in environmental PFOA concentrations, it has been demonstrated to exert long-term and persistent adverse effects on the liver and intestines. This insight is important for the risk assessment of health implications linked to chronic PFOA exposure.

In vivo experiments on broilers provide direct support for network toxicology and molecular docking. Abnormal liver indices, including elevated levels of AST and ALT, are indicative of liver damage, with the degree of elevation reflecting the severity of injury. The histological integrity of an organ serves as the basis for the realization of its functions. Following PFOA exposure, liver indices increased, accompanied by elevated AST and ALT levels. Histopathological analysis of tissue sections revealed a loss of histological integrity in the liver and intestines, suggesting a potential impact of PFOA on liver and duodenal function. The liver acts as a vital metabolic center, and the intestine is tasked with nutrient uptake. Therefore, PFOA exposure reduced average daily weight gain. The prolongation of PFOA exposure time has been demonstrated to result in the intensification of organ damage and the augmentation of inflammatory cytokine activity. According to previous research, PFOA can disrupt the balance between pro-inflammatory and anti-inflammatory cytokine and​induce organ damage by inflammatory responses (Li et al., 2025c). The activation of inflammatory cytokine may trigger oxidative stress, thereby creating a vicious inflammation-oxidation cycle (Hoesel and Schmid, 2013). In addition, inflammatory cytokine also affect the production of different immunoglobulins and influence the levels of immunoglobulins in the body. This previously reported conclusion supports the experimental data from this work. PFOA exposure was associated with increased the levels of pro-inflammatory cytokine in the liver and duodenum. Pro-inflammatory stimulation resulted in a marked decrease in the activities of SOD, T-AOC, GSH-Px, and CAT. Excessive free radicals triggered lipid peroxidation reactions, resulting in a significant increase in MDA.The results suggest oxidative damage in the broilers' liver and intestines. Moreover, the enhanced inflammatory response suppressed IgG production, thereby impairing immune function. The above indicators are consistent with the previous experimental results (Li et al., 2025c; Shi et al., 2024; Zhang et al., 2021). Previous research has suggested the potential of orally administered natural Chinese herbs as a therapeutic strategy to counteract PFOA-induced inflammation, oxidative stress, and apoptotic cell death (Liu et al., 2016; Zhang et al., 2024b). After 0.8% LBP intervention, the levels of AST and ALT, liver index, and pro-inflammatory factor levels decreased, and the histopathology recovered, the levels of antioxidant enzymes and IgG were upregulated, suggesting that 0.8% LBP may be a natural drug for inhibiting the liver and intestinal toxicity of environmental pollutant PFOA (Varoni et al., 2017). By regulating cytokine levels, LBP notably improved broiler growth performance and immune function, supporting its use as a potential feed additive (Long et al., 2020).

Based on the KEGG analysis of network toxicology, along with the hepatointestinal toxicity in broilers induced by PFOA exposure through inflammation and oxidative stress, the PI3K-AKT pathway was selected for further investigation. The findings demonstrated that PFOA exposure was associated with suppressed expression of PI3K and AKT1, which may inhibit the PI3K-AKT signaling pathway and reduce AKT1 phosphorylation levels (Zhang et al., 2024b). This disturbance may shift the intracellular equilibrium between anti-apoptotic and pro-apoptotic signals. The results of this study suggest that PFOA exposure was associated with altered mRNA expression of Bcl-2 and Bax, which may contribute to Caspase-3 activation and subsequent apoptosis. The addition of LBP activated this pathway and inhibited the occurrence of apoptosis, aligning with the results of this study. While the findings of this study suggest that PFOA-induced hepatointestinal toxicity involves the PI3K-AKT signaling pathway, the precise mechanisms underlying this effect require further validation.

With the continuous in-depth research of the liver-intestine interactions theory in toxicology, the hepatotoxicity of PFOA has become relatively well understood. Combined with this study, there is a wide and significant correlation between the duodenum and the liver in oxidative stress, inflammatory response, and the PI3K-AKT signaling pathway. These findings support the close interaction of the liver-intestine interactions in the processes of oxidation, inflammation, and apoptosis, and provide data to support further investigation into the mechanisms linking intestinal injury and liver injury (Zhang et al., 2024a). Therefore, this study subsequently carried out an investigation into the intestinal barrier of broilers. This study found that PFOA exposure was associated with impaired intestinal tissue integrity, which may contribute to down-regulation of intestinal tight junction protein expression and can increase host susceptibility to enteric pathogens. At the same time, intestinal barrier dysfunction is also accompanied by the occurrence of intestinal inflammation. Our experimental results confirmed the existence of intestinal inflammation. The results demonstrate an impairment of intestinal barrier integrity in broilers following PFOA exposure (Li et al., 2025b). As a promising prebiotic, LBP has been shown to regulate the expression of tight junction proteins and inflammatory responses, partially restore the intestinal morphological structure, reduce intestinal permeability, decrease the influx of pathogenic bacteria into the body, mitigate PFOA-induced intestinal injury and safeguard intestinal barrier integrity in broilers (Li et al., 2023).

The gut microbiota is essential for the maintenance of intestinal barrier integrity, overall gut homeostasis, and the promotion of growth in chickens (Liu et al., 2023). The findings indicate that PFOA exposure was associated with disruption of microbial homeostasis, with Firmicutes and Campylobacterota increasing and Bacteroidota and Proteobacteria decreasing. The pathogenic mechanisms of Campylobacterota, including immune activation, inflammation induction, direct damage to the intestinal epithelium, and increased permeability leading to bacterial translocation, show similarities to the intestinal damage associated with PFOA exposure in the present study (Kemper and Hensel, 2023). The association between Bacteroidota depletion and liver injury may be relevant to the hepatotoxic effects of PFOA observed in this study (Haraguchi et al., 2019). The significant shifts in Bacteroidota abundance noted in this study are consistent with its known critical function in oxidative stress homeostasis (Shi et al., 2021). The observation that Firmicutes abundance increased while Proteobacteria abundance decreased is consistent with prior research supporting the concept of inherent environmental adaptability in microorganisms (Costa et al., 2025). An increase was observed in beneficial bacteria such as Faecalibacterium, Fusicatenibacter, Rikenella, and Odoribacter following PFOA exposure. Metabolites derived from these bacteria have been implicated in the regulation of inflammation and maintenance of intestinal barrier function, and this response may serve to counteract the hyperinflammatory state induced by PFOA (Pang et al., 2021; Sokol et al., 2008; Tian et al., 2024). The anti-inflammatory effects mediated by acetate and propionate from Phascolarctobacterium are consistent with the possibility that correlation patterns observed at the genus level (Ahmad et al., 2025; Wu et al., 2017). LBP supplementation was associated with enrichment of beneficial bacterial species, increased their proportion, inhibition of harmful bacteria, and enhanced production of short-chain fatty acids (SCFAs). These microbial changes are consistent with the known functions of SCFAs in anti-inflammatory and metabolic regulation, supporting the potential role of LBP as a prebiotic (Zhu et al., 2020). Nevertheless, the precise mechanisms through which the gut microbiota modified by PFOA exposure contribute to its biological effects remain to be elucidated.

The results of functional predictions support the aforementioned conclusion. PFOA exposure was associated with disruption of dynamic equilibrium between beneficial and pathogenic bacteria, induction of intestinal microbiota disorders, and potential adverse effects mediated by microbiota metabolites. These changes may facilitate the entry of intestinal toxins into the liver through systemic circulation, potentially contributing to organ damage (Ma et al., 2021). Collectively, these findings suggest that prolonged PFOA exposure may adversely affect the liver, microbiota, and intestine in broilers. LBP supplementation was associated with enhanced energy levels, increased proportion of beneficial bacteria, enhanced production of SCFAs, and protection of the intestinal epithelium against damage. Consequently, a key priority for future studies is to decipher PFOA's impact on metabolic pathways and to define LBP's metaboloprotective actions.

Conclusion

In this study, network toxicology, molecular docking, and in vivo experiments in broilers were integrated to identify key targets and mechanisms associated with PFOA-induced hepatointestinal toxicity, as well as the involvement of the liver-intestine interactions in mediating the combined effects of PFOA on the liver and intestine. In addition, the protective effects of LBP against PFOA-induced toxicity in the liver and intestine were observed, providing a foundation for the development of intervention strategies to mitigate PFOA-associated toxicity.

Data availability

Data will be made available on request.

CRediT authorship contribution statement

Nana Gao: Writing – original draft, Project administration, Investigation, Formal analysis, Data curation. Fang Huang: Visualization, Software. Yang Li: Visualization, Validation, Software, Resources. Yujia Wu: Methodology, Conceptualization. Heping Bai: Formal analysis. Xu Liu: Software, Methodology. Xiaodan Wang: Writing – review & editing, Supervision, Methodology, Funding acquisition.

Disclosures

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.