Gut microbiota dysbiosis exacerbates polystyrene microplastics-induced liver inflammation via activating LPS/TLR4 signaling pathway in ducks

Abstract

Ubiquitous microplastics can bioaccumulate in organisms, resulting in detrimental health impacts, such as liver inflammation. Nonetheless, the exact mechanism by which polystyrene microplastics (PS-MPs) trigger liver inflammation via the gut-liver axis in ducks remains unclear. The purpose of this study was to clarify the impact of PS-MPs exposure to liver inflammation through the gut-liver axis in ducks. Our investigation indicated that exposure to PS-MPs markedly upregulated the levels of MDA and ROS in the liver tissue and enhanced the release of pro-inflammatory cytokines (TNF-α, IL-6, and IL-1β). Additionally, PS-MPs exposure increased the LPS level, which ultimately triggered the TLR4/NF-κB signaling pathway. Notably, exposure to PS-MPs resulted in a marked change in the gut microbiota composition, primarily indicated by an increase in the relative abundance of Brachyspiraceae and a reduction in that of CAG-74 and Oscillospiraceae. Metabolome analysis further revealed that different expressed metabolites (DEMs) in the positive and negative mode were identified between the control and HMPs groups, including 1-methylhistamine, DL-Methionine sulfoxide, Guanidinoethyl sulfonate, l-Cysteic acid, Deoxyinosine, Camp. Both metagenomic and metabolome analyses showed enrichment in the lysosomal pathway. Correlation analysis suggested association among representative gut microbiota, serum LPS, oxidative stress factors, liver DEMs and key liver inflammatory indicators. Our study sheds light on the mechanism by which PS-MPs exposure induced liver inflammation in ducks via the modulation of the gut-liver axis. These findings improved our understanding of the underlying mechanisms that contribute to PS-MPs-induced hepatotoxicity in avian species.

Introduction

Microplastics (MPs, typically with a diameter smaller than 5 mm) are small plastic particles originating either from industrial production or environmental degradation via physical and chemical processes (Lambert and Wagner, 2016). These particles have become pervasive contaminants, detected in terrestrial ecosystems, potable water supplies, and even atmospheric particulate matter (Khan and Jia, 2023). In addition, a significant proportion of MPs undergo atmospheric transport and subsequent deposition into lentic and lotic systems, thereby contributing to the contamination of urban water resources (Sun et al., 2022). Among the diverse array of MPs, polystyrene microplastics (PS-MPs) are of particular concern due to their small particle size and recalcitrance to biodegradation, facilitating their bioaccumulation in both flora and fauna (Khalid, et al., 2020; Li, et al., 2020; Liu, et al., 2021). Beyond their intrinsic toxicity-derived from monomers and additives-MPs also serve as vectors for hazardous pollutants, antimicrobial resistance genes, and pathogenic microorganisms (Sheng, et al., 2021), collectively posing a substantial ecotoxicological risk to aquatic biota.

Accumulating empirical evidence demonstrated that MPs exposure can induce oxidative stress-mediated damage, leading to hepatotoxicity, nephrotoxicity, neurotoxicity, and so on (Prokić, et al., 2019; Prata, et al., 2020). Numerous investigations have demonstrated that MPs damage the liver by causing oxidative stress, triggering inflammation, and leading to mitochondrial injury (Capó, et al., 2021; Félix, et al., 2023; Yin, et al., 2023). Following a 30-day exposure to MPs, the liver of the mouse exhibited significant vacuolar degeneration, swelling of hepatocytes, and inflammatory cell infiltration (Li, et al., 2021). Ducks, being obligate aquatic foragers, are highly susceptible to PS-MP exposure through environmental matrices and contaminated feed. This ingestion pathway poses a potential risk for PS-MP residue transfer into duck products intended for human consumption.

The gut microbiota comprises diverse bacterial, viral, and fungal communities residing in the intestinal tract (Cryan, et al., 2019). Extensive research has established the critical role of the gut microbiota in immune modulation (Gao, et al., 2018; Chénard, et al., 2020), modulating the metabolism of lipids (Sheng, et al., 2018; Schoeler and Caesar, 2019) and bile acids (Molinaro, et al., 2018; Rowland, et al., 2018; Zeng, et al., 2019) as well as neuromodulation (Filosa, et al., 2018). In particular, the microbiota significantly influences the intestinal barrier function of the organism. As an important part of the gut, intestinal flora not only forms a "microbial barrier" with the intestinal mucosa to prevent the invasion of pathogenic bacteria, but also participates in the intestinal immune process, and constructs the "gut-liver axis" (Hamoud, et al., 2018), "gut-lung axis" (Bingula, et al., 2017), "gut-kidney axis" (Rossi, et al., 2015) and "gut-brain axis"(Keightley, et al., 2015) pathways with multiple organs outside the immune system(Zeng, et al., 2025). Under normal circumstances, there is a complex and delicate dynamic balance between intestinal flora and the host. Even brief contact with microplastics may disrupt gut microbiota composition, leading to measurable changes in microbial diversity within a short period (Xie, et al., 2022).

The equilibrium of the gut-liver axis is closely tied to gut microbiota composition, intestinal barrier health, and bile acid metabolism, all of which interact interdependently (Buzzetti, et al., 2016). Once the composition and function of intestinal microecology are changed, the barrier function of intestinal mucosa is damaged, and its permeability is changed. Microbial metabolites in the intestine may reach the liver through portal circulation, interfering with bile acid metabolism and provoking innate immune activation. The subsequent release of inflammatory and vasoactive factors induces oxidative and ER stress, ultimately resulting in hepatocyte degeneration, necrosis, and apoptosis. Enterohepatic recirculation mediates continuous exchange between intestinal absorption (portal system) and biliary excretion, potentially causing cumulative hepatocyte exposure to MPs/NPs with hepatic implications (Chiang, et al., 2024). Notably, the inflammatory mechanisms of PS-MPs in duck liver models remain uncharacterized.

This study aims to investigate how intestinal microbiota interacts with liver inflammation to enhance our comprehension of the hepatotoxic effects of PS-MPs in ducks. The ducks were exposed to PS-MPs (1 mg/L, 100 mg/L) for four weeks. To assess PS-MPs-induced hepatotoxicity, we conducted hepatic histopathological examinations and quantified inflammatory protein expression. Concurrently, we analyzed alterations in gut microbial composition and performed non-targeted hepatic metabolomic profiling following PS-MPs exposure. Our study provides a structural framework by which PS-MPs induced liver inflammation via gut-liver axis.

Materials and methods

Ethics statement and consent to participate

Animal experiments were approved by Institutional Animal Care and Use Committee of Yangzhou University (Jiangsu, China) and conducted in accordance with Chinese ethical guidelines (Jiangsu Administration Rules for Laboratory Animal Use, China) and GB14925-2010 standards for laboratory animal housing.

Experimental animals and sample collection

One-day-old Shaoxing ducks (Hanchao Poultry Company, Hangzhou, China) were acclimatized for seven days before randomization into three groups (n = 20/group): (1) control (ultrapure water, CK), (2) low-dose PS-MPs (1 mg/L, LMPs), and (3) high-dose PS-MPs (100 mg/L, HMPs), (5 μm in size, BaseLine Scientific Co., Ltd, Tianjin, China). The exposure doses of PS-MPs used in this study were selected with reference to previous toxicological research conducted in mice and poultry (Shengchen, et al., 2021; Zhang, et al., 2022). In these earlier studies, concentrations of 1 mg/mL, 10 mg/mL, and 100 mg/mL were administered to chickens (Zhang, et al., 2022). Additionally, since 1 mg/L has been reported as an environmentally relevant concentration of PS-MPs (Sun, et al., 2021a, 2021b), we chose two moderate doses—1 mg/mL and 100 mg/mL—for the current experiment. Fresh PS-MPs suspensions were prepared daily in ultrapure water and administered once per day. All ducks were maintained under controlled conditions (standard diet, 17L:7D photoperiod, constant temperature) for four weeks. Following weighing and euthanasia, liver tissue and cecal content samples were collected from all ducks for further analysis.

Hematoxylin and eosin

Tissue samples were fixed in 4 % paraformaldehyde and processed into 5-μm-thick sections. For histological examination, sections were stained with hematoxylin and eosin (H&E), then visualized and imaged using an Olympus microscope (Tokyo, Japan).

Inflammatory indices detection

Hepatic concentrations of pro-inflammatory cytokines were quantified using commercially available ELISA kits (Sino-uk Institute of Biological Technology, Beijing, China) following the manufacturer's protocols, including the interleukin (IL)-1β, IL-6, and tumor necrosis factor (TNF)-α.

Antioxidant indexes detection

Serum oxidative stress markers, including Reactive Oxygen Species (ROS), Malondialdehyde (MDA), Catalase (CAT) and Superoxide dismutase (SOD), were analyzed using commercial assay kits (Beijing Sino-uk Institute of Biological Technology) according to the manufacturer's instructions.

Western blotting

Liver tissue homogenization was performed using ice-cold RIPA lysis buffer (Beyotime). After centrifugation, protein concentrations were normalized and separated by 12 % SDS-PAGE (Genscript), followed by transfer to PVDF membranes (Millipore). Membranes were blocked with 5 % non-fat milk (1-2 h) before incubation with primary antibodies (4°C overnight; see Table S1) and corresponding secondary antibodies (1 h). Protein signals were visualized using a ChemiDoc imaging system (Bio-Rad) and quantified with ImageJ 1.54d software (National Institutes of Health, Bethesda, MD, USA).

Metabolome analysis of liver tissue

Liver tissue samples (80 mg) were immediately frozen in liquid nitrogen, homogenized in water with ceramic beads, and extracted using methanol/acetonitrile (1:1, v/v). After being centrifuged at 14,000 g for 20 minutes at 4°C, the supernatants were dried under vacuum and reconstituted in a mixture of acetonitrile and water (1:1) for LC-MS analysis. Chromatographic separation utilized a Waters ACQUITY UPLC HSS T3 C18 column (1.8 µm, 2.1 × 100 mm). Raw data were processed using ProteoWizard, and multivariate statistical analysis (OPLS-DA) identified differentially expressed metabolites (DEMs) (VIP ≥ 1.0, p < 0.05) which were subsequently annotated against the KEGG database (n = 10 biological replicates).

Non-targeted metabolomics analysis of gut microbiota

The OMEGA Mag-Bind Soil DNA Kit (Omega Bio-Tek, USA) was used to extract total microbial genomic DNA, adhering to the manufacturer's instructions. Metagenomic libraries (∼400 bp inserts) were prepared with the Illumina TruSeq Nano DNA LT Library Prep Kit and sequenced (PE150) on an Illumina NovaSeq platform (Personal Biotechnology Co., China). Raw reads were quality-filtered using Cutadapt (v1.2.1), fastp (v0.23.2), and Minimap2 (v2.24). Taxonomic classification was performed with Kraken2 (v2.0.8-beta) against a GTDB-derived database. Reads were assembled into contigs using Megahit (v1.1.2) (meta-large preset). Contigs (>300 bp) were clustered (MMseqs2) and taxonomically annotated via alignment against NCBI-nt (MMseqs2 taxonomy mode). Genes were predicted using Prodigal (v2.6.3), and CDSs were clustered (MMseqs2 easy-cluster). High-quality reads were mapped to predicted genes (Minimap2, -ax sr –sam-hit-only), and abundances were quantified using feature Counts. Non-redundant genes were functionally annotated against the KEGG database (MMseqs2 search mode). LEfSe identified differentially abundant taxa/functions across groups. Beta diversity (Bray-Curtis distances) was assessed via PCoA, NMDS, and UPGMA clustering.

Statistical analysis and data available

The data analysis was conducted using one-way ANOVA. The LSD method was utilized to examine differences among groups with IBM SPSS Statistics 25 software (Armonk, NY, USA). Data are expressed as mean ± standard deviation (SD), with statistical significance set at a p-value of ≤ 0.05. The Metagenomics sequencing read data were generated in this study has been uploaded to the Genome Sequence Archive database with the accession number CRA026210.

Results

PS-MPs exposure aggravated liver injury in ducks

Anatomical observations were carried out to evaluate the impact of different PS-MPs doses on Shaoxing ducks. The results indicated that the body weight and liver weight of ducks in HMPs group were observably decreased compared with CK group (Fig. 1A and 1B). H&E staining further indicated that LMPs ducks showed marked vacuoles in the liver, while HMPs ducks showed significant inflammatory cell infiltration (Fig. 1C).

Fig. 1.

PS-MPs-induced hepatic injury in Shaoxing ducks. (A-B) Body and liver weight. (C) H&E staining of liver sections (scale bars: 200 μm and 50 μm). Red arrows represent inflammatory cell infiltration. Blue arrows represent vacuoles. The data analysis was conducted using one-way ANOVA.The error bars represent the means± SD of three replicates. Different lowercase letters indicate significant differences (P < 0.05).

PS-MPs exposure induced liver inflammation

We investigated the inflammatory cytokines in ducks to determine the effect of PS-MPs on the liver inflammation. We observed a significant increase in liver TNF-α and IL-6 in the HMPs group, while IL-1β showed no significance compared to the control group (Fig. 2A). The protein expression of iNOS and COX2 in liver were significantly upregulated in LMPs and HMPs groups (Fig. 2B). Notably, the level of LPS in serum was obviously increased in HMPs group (Fig. 2C). To further investigate whether the change of LPS would affect TLR4 signaling pathway, we found that PS-MPs exposure upregulated the protein relative expressions of TLR4, MyD88, phosphorylated IκBα (p-IκBα) and phosphorylated NF-κB (p-NF-κB) in the liver (Fig. 2D).

Fig. 2.

PS-MPs exposure induced LPS/TLR4 signaling pathway in the liver. (A) The levels of liver inflammation cytokines. (B) The relative protein expression levels of iNOS and COX. (C) The level of serum LPS. (D) The relative protein expression levels of TLR4, MyD88, IκBα, p-IκBα, NF-κB and p-NF-κB. The data analysis was conducted using one-way ANOVA. The error bars represent the means± SD of ten replicates. Different lowercase letters indicate significant differences (P < 0.05).

PS-MPs exposure caused oxidative stress

To explore whether liver inflammation affect the status of oxidative stress, the indexes of redox biology were detected. The serum levels of MDA and ROS in the HMPs group were significantly upregulated, while SOD and CAT were remarkably downregulated (Fig. 3).

Fig. 3.

PS-MPs exposure activated oxidative stress. The ROS (A), SOD (B), MDA (C) and CAT (D) levels of serum in Shaoxing ducks. The data analysis was conducted using one-way ANOVA. The error bars represent the means± SD of ten replicates. Different lowercase letters indicate significant differences (P < 0.05).

PS-MPs exposure affects liver metabolites

Liver metabolic profiles following PS-MPs exposure were analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS). Partial least squares discriminant analysis (PLS-DA) was subsequently employed to identify differentially expressed metabolites in both positive and negative ionization modes. There was a clear differentiation between the control and HMPs groups as shown by the PLS-DA score plots. (Fig. 4A-4D). Compared with the control group, 89 up-regulated metabolites and 123 down-regulated metabolites were identified in the positive mode, while 60 up-regulated metabolites and 43 down-regulated metabolites in the negative mode (Fig. 4E and 4F). Besides, the main top 20 important metabolites are showed (Fig. 4G and 4H). And most distinguishing unique metabolites were observably upregulated in the HMPs group (Table S2 and S3). KEGG enrichment analysis indicated that DEMs in positive mode mainly enriched in Cysteine and methionine metabolism, Histidine metabolism, Tyrosine metabolism and Purine metabolism, while DEMs in negative mode mainly enriched in Taurine and hypotaurine metabolism, cAMP signaling pathway, cGMP-PKG signaling pathway, Lysosome and Purine metabolism (Fig. 4E and 4F). Metabolites involved in these pathways are mainly 1-methylhistamine, DL-Methionine sulfoxide, Guanidinoethyl sulfonate, l-Cysteic acid, Deoxyinosine, Camp, etc. Correlation analyses of differential metabolites in the positive mode and negative mode are listed in Fig. 4G and 4H. In the positive mode, there are significant associations among pesticides (Cyproconazole, Precymidone), markers of oxidative stress (DL-Methionine sulfoxide), and lipid metabolites (Palmitoylcarnitine). In the negative mode, there are significant correlations between Chlorophenol pollutants (2,4-dichlorophenol, 4-chlorophenol), pesticides (Imidacloprid) and oxidative stress markers (L-Cysteic acid), energy metabolites (Phosphocreatine).

Fig. 4.

PS-MPs exposure affects liver metabolites. (A-B) PLS-DA score plots. (C-D) Cluster analysis of differential metabolites. (E-F) Differential material screening map. (G-H) Importance analysis of differential metabolites. (I-J) KEGG enrichment pathways. (K-L) Correlation analysis of differential metabolites.

PS-MPs exposure disturbed the intestinal microbiota composition

Gut microbiota alterations were assessed via metagenomic sequencing comparing HMPs and CK ducks. Metagenomic sequencing of cecal samples yielded an average of 75,545,100 reads per sample (Table S4). Clustering at 97 % sequence similarity identified 14,922 operational taxonomic units (OTUs), comprising 1,545 unique to CK group, 6,864 unique to HMPs group, and 12,691 shared between groups (Fig. 5A). β-diversity analysis revealed distinct clustering patterns between groups (Fig. 5B, D), while α-diversity metrics showed significant intergroup differences (Fig. S1). The microbial composition was dominated by Firmicutes-A, Bacteroidota, and Firmicutes at the phylum level. Notably, HMPs ducks exhibited significantly elevated abundances of Spirochaetota, Cyanobacteria, and Campylobacterota compared to controls (Fig. 5C). Oscillospiraceae, Lachnospiraceae, Ruminococcaceae, and Acutalibacteraceae were the dominant family between HMPs and CK groups at the family level. Furthermore, the relative abundance of Brachyspiraceae, CAG-74, and Oscillospiraceae were markedly elevated in HMPs group compared to CK group (Fig. 5E). A total of top 20 differential gut microbiota is showed in the Fig. 5F, including Brachyspira, Coproplasm, CAG-303 and Limisoma. Linear discriminant analysis (LDA) distribution diagrams (LDA score>3) and the LEfSe algorithm was used to identify specific taxa with different distributions between the two groups. Three taxa were over‐represented (including the family Brachyspiraceae) and under‐represented (including the family Oscillospiraceae and the order Propionibacteriales) in the HMPs group (Fig. 5G and 5H). KEGG enrichment analysis indicated that different expressed microbiota in gut mainly enriched in Phosphotransferase system (PTS), Lysosome, Porphyrin metabolism and Starch and sucrose metabolism (Fig. 5I).

Fig. 5.

PS-MPs exposure alters gut microbiota composition. (A) OTUs Venn diagram. (B) PCoA analysis. (C) The 10 most dominant microbial phylum. (D) NMDS analysis. (E) The 10 most dominant microbial family. (F) Importance analysis of top 20 microbiota. (G) LDA scores analysis (LDA>3). (H) LEFse analysis of inter-group differences in the microbial community at the genus level (LDA > 2). (I) KEGG pathway of differential microbiota.

Correlations analysis of key variation factors

Pearson pairwise correlation analysis was performed to assess the relationship of serum LPS, representative gut microbiota, oxidative stress factors, liver DEMs and key liver inflammatory indicators (Fig. 6). Campylobacterota and Cyanobacteria abundance showed positive correlations with LPS concentration. The level of LPS was positively correlated with MDA and ROS, and negatively correlated with SOD and CAT. Furthermore, the oxidative metabolites DL-methionine sulfoxide and l-cysteic acid demonstrated positive relationships with both ROS/MDA and pro-inflammatory cytokines (TNF-α, IL-1β, IL-6), While exhibiting negative correlations with SOD/CAT activities.

Fig. 6.

Correlations analysis. (A-C) Pairwise Correlations between representative gut microbiota, serum LPS and oxidative stress factors, liver DEMs and key liver inflammatory indicators.

Discussion

MPs have emerged as a novel class of environmental contaminants with demonstrated biological toxicity. Studies indicate that MPs can translocate across biological barriers, accumulating in various tissues through systemic circulation. Substantial evidence suggests that MPs exposure induces multiple pathological effects, including oxidative damage, inflammatory responses, cellular dysfunction, and metabolic disturbances (Prata, da Costa, Lopes, Duarte and Rocha-Santos, 2020; Vethaak and Legler, 2021). In the previous research, we demonstrated that the PS-MPs exposure (1 mg/L and 100 mg/L) induced liver inflammation via the LPS-TLR4 signaling pathway in ducks. And H&E staining showed the marked vacuoles and significant infiltration of inflammatory cells in livers of ducks treated with PS-MPs. Exposure to PS-MPs induces lung pathology and ultrastructural alterations in chickens, including endoplasmic reticulum (ER) swelling, inflammatory cell infiltration, chromatin margination, and plasma membrane rupture(Lu, et al., 2024a). In a previous study, microplastics exposure triggered inflammation and lipid accumulation in fish liver, leading to changes in metabolic profiles and interference with lipid metabolism (Lu, et al., 2016). Thus, further investigation is warranted to elucidate the chronic hepatotoxic effects of PS-MPs and their molecular mechanisms.

PS-MPs exposure enhanced TLR4/NF-κB activation in intestinal epithelium, inducing inflammatory infiltration and cytokine production (Yao, et al., 2019). Elevated proinflammatory cytokines (IL-1β, IL-6) disrupted intestinal homeostasis (Opal and DePalo, 2000; Yao, Dong, Dai and Wu, 2019), while TNF-α exacerbated inflammation via NF-κB/MAPK activation (Sonis, 2010). Consistent with these findings, hepatic PS-MPs accumulation in ducks significantly increased IL-1β, IL-6 and TNF-α levels, accompanied by upregulated iNOS and COX-2 expression, suggesting their contribution in liver injury.

The NO and NF-κB signaling pathways are closely interconnected, with NO exerting regulatory effects at multiple stages of the NF-κB activation cascade (Hierholzer, et al., 1998). NF-κB activation plays a crucial role in initiating iNOS gene transcription (Mollace, et al., 2005). As crucial inflammatory mediators, cytokines demonstrate significant dysregulation during intestinal inflammation. NF-κB orchestrates the expression of pivotal inflammatory genes including iNOS, TNF-α, and COX-2, all of which play central roles in inflammatory pathogenesis (Heller and Krönke, 1994; Hotz-Behofsits, et al., 2010). When stimulated by ROS/RNS, this transcription factor triggers the expression of proinflammatory mediators including COX-2, TNF-α, iNOS, IL-1β, and IL-6, ultimately resulting in cellular damage and apoptotic pathways (Yamamoto and Gaynor, 2004). The cytokine-mediated reactivation of NF-κB forms a positive feedback mechanism that augments proinflammatory signaling and accelerates tissue deterioration (Neurath, et al., 1998).

Our findings indicate that PS-MPs exposure significantly elevates serum LPS levels. Existing evidence suggested that gut dysbiosis promotes LPS overproduction, which subsequently reach the liver via portal circulation (Arab, et al., 2018). As a potent activator of innate immunity, LPS triggers TLR4 signaling to stimulate proinflammatory cytokine release (Lu, et al., 2008; Krishnan, et al., 2021), highlighting the pivotal role of the LPS/TLR4 axis in gut-liver axis.

As a core component of the body’s inflammation system, the TLR/MyD88/NF-κB axis is extensively distributed throughout biological systems and critically involved in the development and regulation of multiple pathological conditions (Liu, et al., 2025). Toll-like receptors (TLRs) initiate pro-inflammatory immune responses (Ma, et al., 2021), and All TLR subtypes engage the MyD88-dependent signaling cascade, resulting in NF-κB activation and subsequent production of inflammatory cytokines and mediators (Zhang, et al., 2023b). The Nuclear factor kappaB (NF-κB) inhibitor protein (IκB) is an important member of the NF-κB signaling pathway. The higher the phosphorylation of IκBα, the lower the inhibitory effect on NF-κB and the stronger the degree of inflammatory response. Activation of the NF-κB pathway exacerbates hepatic inflammatory responses. We further detected the related proteins of TLR4 signaling pathway, and found that PS-MPs exposure upregulated the protein levels of TLR4, p-IκBα, p-NF-κB, and MyD88 in the liver, which showed increased inflammation of the liver.

ROS induction by MPs occurs via: (i) exogenous mechanisms involving environmental weathering factors (Pannetier, et al., 2019), and (ii) endogenous mechanisms through mitochondrial impairment following cellular internalization (Khan and Jia, 2023). Recent experimental models have established that MPs induce hepatic injury via ROS-mediated mechanisms, contributing directly or indirectly to the pathogenesis of metabolic dysfunction-associated fatty liver disease (MAFLD). Lu, et al. demonstrated that PS-MPs induce oxidative stress in Carp Intestinal Epithelial Cells. This oxidative stress, mediated by ROS, leads to cell cycle arrest, apoptosis, and autophagy(Lu, et al., 2024b). PS-MPs trigger oxidative stress, myocardial pyroptosis, inflammation, and mitochondrial/energy metabolism dysfunction by driving ROS overload, which leads to alterations in the NF-κB-NLRP3-GSDMD and AMPK-PGC-1α pathways (Zhang, et al., 2022).The significant changes of oxidate stress-related factors between HMPs and control group showed that oxidative damage was obviously aggravated in the HMPs group. Consistent with studies in zebrafish (Boopathi, et al., 2023), our data indicate that PS-MPs synergize with metabolic stressors to exacerbate oxidative damage and inflammation. MP/NP phagocytosis by Kupffer cells triggers metabolic dysregulation, promoting free fatty acid oxidation and consequent ROS overproduction that culminates in hepatic injury (Diehl, et al., 2020; Rudolph, et al., 2021; Prata, 2023). Our findings suggest that PS-MPs exposure triggers hepatic inflammation through activation of the LPS-TLR4-NF-κB signaling axis.

MPs disrupt intestinal homeostasis through three primary mechanisms: gut microbiota dysbiosis, induction of colonic inflammatory responses, and compromise of intestinal barrier integrity (Zhang, et al., 2023a). A 4-week PS-MPs exposure significantly altered gut microbiota composition, characterized by reduced beneficial taxa (CAG-74 and Oscillospiraceae) and enriched pathogenic bacteria (Brachyspiraceae). Brachyspira represents a monotypic genus (Brachyspiraceae) within the order Spirochaetales of the phylum Spirochaetes (Christodoulides, et al., 2022). Within the Brachyspiraceae family, three Brachyspira species have been established as poultry pathogens, demonstrating the capacity to induce avian intestinal spirochaetosis (AIS) in experimental chicken models.: Brachyspira alvinipulli (Swayne, et al., 1995; Stanton, et al., 1998), Brachyspira intermedia (Hampson and McLaren, 1999) and Brachyspira pilosicoli (Stephens and Hampson, 2002). Oscillospira is negatively correlated with inflammation-related diseases (Xia, et al., 2021; Li, et al., 2022, 2024), which belongs to family Oscillospiraceae.

Altered hepatic metabolite profiles were closely associated with liver injury markers, inflammatory responses, and intestinal barrier integrity (Wang, et al., 2022). In the present study, 24 DEMs in the positive mode and 14 DEMs in the negative mode are identified between the control and HMPs group. 1-methylhistamine, a histamine metabolite, was upregulated in the HMPs group, reflecting increased histamine release. DL-Methionine sulfoxide and l-Cysteic acid, which are oxidation products, were significantly up-regulated in the HMPs group, indicating an increase in oxidative stress levels in the HMPs group. The up-regulation of Deoxyinosine may be due to DNA damage leading to abnormal purine metabolism. Comprehensive KEGG enrichment analysis of metagenome sequencing and metabolome sequencing showed that lysosomes were overlapped pathway. Certain TLRs localized in lysosomes function as pattern recognition receptors, detecting pathogen-associated molecular patterns (PAMPs) and triggering inflammatory signaling cascades (Zhang, et al., 2021).

Campylobacterota and Spirochactota were Gram-negative phyla, and their abundance was positively correlated with LPS level. They activated the immune system by releasing LPS. LPS was positively correlated with ROS and MDA, indicating that LPS increased oxidative damage by inducing an inflammatory response. LPS was negatively correlated with SOD and CAT, because antioxidant enzymes are consumed in large amounts to fight oxidative stress. The abundance of Brachyspiracae and Osellospiraceae were positively correlated with ROS and MDA, suggesting that these bacteria aggravate oxidative damage through LPS or other mechanisms. LPS was positively correlated with 1-methylhistamine, reflecting the increased release of histamine in inflammatory response. LPS was positively correlated with DL-Methionine sulfoxide and l-Cysteic acid, indicating that LPS promoted amino acid oxidation through oxidative stress. cAMP was negatively related to the regulation of antioxidant enzymes (such as SOD and CAT) activity. Oxidation products such as DL-Methionine sulfoxide and l-Cysteic acid are positively correlated with inflammatory factors such as TNF-α and IL-6, suggesting that oxidative stress may drive inflammatory responses. In conclusion, Gram-negative bacteria (Campylobacterota and Spirochaetota) stimulate TLR4 signaling via LPS, subsequently activating the NF-κB pathway and promoting the secretion of pro-inflammatory cytokines (TNF-α, IL-1β, IL-6). This inflammatory cascade was accompanied by elevated oxidative stress markers and corresponding oxidation products.

Conclusion

In summary, excessive PS-MPs exposure disrupted gut microbiota equilibrium, exacerbating hepatic metabolic dysfunction and inflammatory responses via LPS/TLR4/NF-κB pathway activation. These findings provide novel insights into waterfowl health risks and elucidate gut-liver axis mechanisms in PS-MPs-induced hepatotoxicity.

CRediT authorship contribution statement

Ning Zhou: Writing – original draft, Investigation, Data curation. Tiantian Gu: Writing – review & editing, Investigation. Mingcai Duan: Data curation. Yong Tian: Formal analysis. Li Chen: Formal analysis. Tao Zeng: Methodology. Xuan Hou: Methodology. Xiaoyan Wang: Software. Qi Xu: Validation. Yu Zhang: Supervision, Resources. Lizhi Lu: Project administration, 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.