Abstract Recently, the nanozyme Pd@Pt has garnered attention due to its notable specific surface area and superior enzyme-like catalytic activity, leading to extensive examination and application in previous studies. However, the comprehensive impact of Pd@Pt nanozyme on treating metabolic disorders, such as diabetes and its associated conditions, remains largely unexplored. This research aimed to clarify how Pd@Pt influences metabolic balance at both the transcriptome and microbiome levels and to explore the interactions between microbiota and genes. We conducted an examination of mice subjected to a high-fat diet (HFD) following treatment with Pd@Pt. Transcriptome analysis was performed to identify differentially expressed genes (DEGs), and microbiome analysis was conducted to identify significant bacterial correlations associated with Pd@Pt exposure. The results indicated enhancements in glucose metabolism dysfunctions in the treated mice. Transcriptome analysis revealed that DEGs after Pd@Pt administration were enriched in the PI3K-Akt, NF-κB, and MAPK signaling pathways in the liver. Microbiome analysis identified four significant bacteria that exhibited a strong negative correlation with Pd@Pt exposure, while ten bacteria showed a positive correlation. Furthermore, a correlation network established among the gut microbiota, metabolites, and DEGs demonstrated a robust association. This research enhances our understanding of the mechanisms by which Pd@Pt affects the regulation of metabolic diseases in HFD-exposed environments and proposes a novel strategy for utilizing nanozymes in human health management. Keywords: Pd@Pt nanozyme, Metabolic homeostasis, Transcriptome, Microbiome, Metabolome Graphical abstract Image 1 [41]Open in a new tab Highlights * • Pd@Pt mitigates hyperglycemia induced by HFD. * • Pd@Pt significantly affected the structure and composition of gut microbiota. * • Pd@Pt significantly affected liver PI3K-Akt, NF-κB, and MAPK signaling pathways. * • Pd@Pt significantly affected the content of intestinal metabolites. 1. Introduction Nanozymes, with exceptional enzyme-mimicking properties, are emerging as promising alternatives to natural enzymes due to their superior attributes, such as convenient storage, adjustable catalytic activities, high stability, and scalability for large-scale production [[42][1], [43][2], [44][3], [45][4]]. Their potent ability to regulate oxidative stress is particularly significant in addressing metabolic diseases, where reactive oxygen species (ROS) play a crucial role, offering a unique therapeutic perspective through multifunctional activities [[46][5], [47][6], [48][7], [49][8], [50][9]]. By scavenging excess ROS or modulating pathologically related molecules, nanozymes demonstrate potential for effective treatment of metabolic disorders [[51][10], [52][11], [53][12]]. In previous studies, Pd@Pt nanozyme have been shown to exhibit enhanced peroxidase-, catalase-, and superoxide dismutase-like activities, with important implications for promoting wound healing and inhibiting bacterial infections [[54][13], [55][14], [56][15]]. Pd@Pt offers a promising and efficient approach for maintaining homeostasis through a universally applicable platform, holding significant potential for clinical diagnosis and therapy. The liver is a vital organ in mammalian physiology, fundamental to the maintenance of metabolic homeostasis and participating in numerous critical metabolic processes that contribute to overall health and wellness [[57]16,[58]17]. Impairment of hepatic metabolic functions can result in various metabolic disorders, including diabetes, obesity, and non-alcoholic fatty liver disease, underscoring the significance of elucidating the liver's role in energy metabolism [[59][18], [60][19], [61][20]]. However, the effects of Pd@Pt on the regulation of the liver associated with metabolic homeostasis are poorly understood. The intricate interplay between the gut microbiota and human health has spurred intensive investigation into strategies for manipulating the microbial ecosystem to mitigate metabolic disorders [[62]21,[63]22]. Increasing evidence highlights the critical contribution of gut dysbiosis to the development of several metabolic disorders [[64]23], diabetes [[65]24], nonalcoholic steatohepatitis [[66]25], and diabetic wound healing [[67]26]. Recognizing the potential of modulating the gut microbiota as a therapeutic avenue, recent research has turned to innovative approaches such as nanozymes. Nanozymes present a promising platform for precisely targeting and influencing specific microbial communities within the gut [[68]27,[69]28]. By leveraging the unique catalytic properties of nanozymes, it becomes feasible to selectively inhibit harmful bacteria while fostering the growth of beneficial species [[70]29]. This targeted modulation of the gut microbiota and metabolites holds significant implications for restoring balance and enhancing overall gut health in the context of metabolic diseases. Emerging evidence emphasizes the critical role of the gut-liver axis in metabolic homeostasis, where bidirectional communication between the gut microbiota and the liver plays a key role in regulating metabolic processes [[71]30]. The gut microbiota influences liver function through the production of metabolites such as short-chain fatty acids, bile acids, and other bioactive compounds, which modulate hepatic metabolism, inflammation, and insulin sensitivity [[72]31]. Conversely, the liver influences the composition and function of the gut microbiota through bile secretion and systemic immune responses [[73]32]. This intricate interplay highlights the importance of the gut-liver axis in metabolic diseases such as obesity and type 2 diabetes [[74]33,[75]34]. Understanding this axis provides a framework for exploring how gut microbiota regulation affects metabolic regulation, which is at the heart of this study. In this study, the regulatory effects of Pd@Pt on metabolic homeostasis in high-fat diet (HFD)-fed mice were investigated. Subsequently, transcriptome and intestinal microbiome analyses were carried out in HFD-fed mice to identify relevant alterations at the genetic and bacterial levels. Thereafter, metabolomics was used to further analyze the possible changes in metabolites. The integration of microbiome and transcriptome analyses allowed us to clarify the interaction between bacteria and genes for further exploration. Overall, our findings reveal the impact of Pd@Pt on metabolic homeostasis regulation and the underlying molecular mechanisms. We delve into the intricate mechanisms by which the mammalian liver and gut microbiota contribute to metabolic homeostasis and explore the implications for human health and disease. 2. Methods and materials 2.1. Main materials K[2]PtCl[4] and NaPdCl[4] were purchased from Aladdin (Beijing, China). L (+)-Ascorbic Acid was obtained from Adamas (Shanghai, China). Glucose was purchased from Sigma-Aldrich (St. Louis, MO, USA). 2.2. Animals This study was carried out in line with the guidelines of China Agricultural University Animal Ethics Committee. The animal experiments were conducted in the specific pathogen-free facility. Six-week-old male C57BL/6J mice were obtained from Beijing Vitality River Laboratory Animal Technology Co. After a one-week adaptation period, the mice were divided into groups fed with a standard chow diet (HFK Bioscience Co. Ltd.), a HFD (Research Diet, D12492), and a Pd@Pt group for four weeks. The mice in the Pd@Pt group were orally administered a 10 mg/kg solution of Pd@Pt nanozyme for the duration of the study. 2.3. Oral glucose tolerance test (OGTT) After the four-week feeding period, the mice were fasted for 6 h (from 8:00 a.m. to 2:00 p.m.) and then underwent an OGTT. Each mouse was orally given glucose of 2 g/kg of body weight. Subsequently, blood glucose levels were measured using a glucometer. Blood samples were collected from the tip of the tail vein at 0, 15, 30, 60, 90, and 120 min after glucose administration. 2.4. Hematoxylin-eosin (H&E) staining The colon, epididymis, and liver were immediately fixed in a 4 % formaldehyde solution. Subsequently, they were embedded in paraffin, sectioned, and stained with H&E. For further analysis, the stained tissue sections were examined under an optical microscope (Olympus, Tokyo, Japan). 2.5. Immunohistochemistry (IHC) staining Colon sections were first deparaffinized in xylene, and then hydrated in ethanol with concentrations of 100 %, 85 %, and 75 % respectively for 5 min each, followed by washing in distilled water. IHC staining of colon sections was carried out using mouse anti - antibody overnight. Subsequently, the sections were incubated with HRP-labeled goat anti-mouse IgG for 50 min at room temperature. For analysis, the slide was viewed on CaseViewer. 2.6. Immunofluorescent (IF) staining Colon sections were first deparaffinized in xylene. Subsequently, they were hydrated in ethanol at concentrations of 100 %, 85 %, and 75 % respectively, with each step lasting for 5 min, and then washed in distilled water. For IF staining, the colon sections were treated with mouse Rabbit anti-mouse Muc2 antibody overnight. Next, the sections were incubated with CY3-labeled goat anti-mouse IgG for 50 min at room temperature. After this, the colon sections were stained with DAPI for 10 min at room temperature. Finally, for analysis, the slide was viewed on CaseViewer. 2.7. Blood biochemistry analysis Blood biochemistry was used to quantify levels of aspartate aminotransferase (AST), high-density lipoprotein (HDL), low-density lipoprotein (LDL), cholesterol (CHO), triglycerides (TG), urea, lactic dehydrogenase (LDH), and alanine aminotransferase (ALT). 2.8. Hemolysis assay The compatibility of Pd@Pt with blood was evaluated using a hemolysis test. Various concentrations (0–800 μg/mL) of Pd@Pt were mixed with 2% rabbit red blood cells and incubated at 37°C for 2 h. After incubation, the mixture was centrifuged at 3000 rpm for 5 min, and the absorbance of the supernatant was measured at 560 nm using a microplate reader. 0.9% NaCl and ultrapure water served as the negative (Neg) and positive controls (Pos) for hemolysis, respectively. The hemolysis rate was determined using the measured absorbance values. The formula for calculating the hemolysis rate is: Hemolysis rate = [(ODsa - ODneg)/(ODpos - ODneg)] × 100 %, where ODsa represents the absorbance of the sample, ODneg is the absorbance of the negative control, and ODpos is the absorbance of the positive control. 2.9. Cell culture The HepG2, AML12, and Vero cells were obtained from the Stem Cell Bank of the Chinese Academy of Sciences, were cultured in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10 % fetal bovine serum (FBS) sourced from PAN Biotech. The cell cultures were then incubated under standard conditions of 37 °C in a 5 % CO[2] atmosphere. 2.10. Cytotoxicity of cells For CCK8 test: HepG2, AML12, and Vero cells were treated with the Pd@Pt nanozyme for 24 h. Then the cytotoxicity of different cells was evaluated by CCK8 kit (Abbkine, Wuhan, China) according to the manufacturer's protocol. For mitochondrial injury and endoplasmic reticulum injury analyses: HepG2 cells were plated in a 24-well plate and cultured for 24 h under standard cell culture conditions and then treated with Pd@Pt for an additional 24 h. To assess mitochondrial injury, MitoTracker Red CMXRos (Beyotime, Shanghai, China) was utilized, while for the evaluation of endoplasmic reticulum injury, ER-Tracker Green (Beyotime, Shanghai, China) was employed. The cells were cultured with Hoechst 33342 to stain the nuclei following the manufacturer's protocol. Finally, fluorescence microscope was used to capture images for further analysis. For cell apoptosis analysis: HepG2 cells were plated in a 24-well plate and cultured for 24 h under standard cell culture conditions and then treated with Pd@Pt for an additional 24 h. To assess the cell apoptosis, One Step TUNEL Apoptosis Assay Kit was used to evaluated according to the manufacturer's protocol. Finally, fluorescence microscope was used to capture images for further analysis. 2.11. The DNA injury analysis HepG2 cells were plated in a 12-well plate and cultured for 24 h under standard cell culture conditions and then treated with Pd@Pt for an additional 24 h. Immunofluorescence staining for γ-H2AX was then performed using the DNA damage assay kit for γ-H2AX immunofluorescence (Beyotime, Shanghai, China) following the manufacturer's protocol. Fluorescence microscopy was utilized to capture the images for further analysis. 2.12. Transcriptome analysis Liver transcriptome analyses, which encompassed RNA purification, reverse transcription, library preparation, and sequencing, were performed at Shanghai Majorbio Bio-pharm Biotechnology Co., Ltd. (Shanghai, China). Differential expression analysis was carried out using either DESeq2 or DEGseq, with a threshold of |log2FC| ≥1 and false discovery rate (FDR) criteria set at ≤ 0.05 for DESeq2 or ≤ 0.001 for DEGseq to identify significantly differentially expressed genes (DEGs). 2.13. Weighted gene correlation network analysis (WGCNA) analysis WGCNA was applied to investigate the scale-free network structure connecting genes and other factors with phenotypes. DEGs were input into the WGCNA using a specialized package in the R software environment. The relationships between genes and bacterial genera were extracted from the WGCNA results to establish the interaction network. 2.14. Gut microbiota analysis The E.Z.N.A. Fecal DNA Kit (Omega Bio-tek, Inc.) was utilized to extract genomic DNA from fecal samples. Amplification of the V3-V4 hypervariable region of the bacterial 16S rRNA gene was carried out using universal primers 338F (5′-ACTCCTACGGAGGCAGCAG-3′) and 806R (5′-GGACTACNNGGGGTATCTAAT-3′). Following amplification, deep sequencing was conducted on the Illumina Miseq/Novaseq platform (Illumina, Inc., USA) at Beijing Allwegene Technology Co., Ltd. The obtained data were analyzed using Linear Discriminant Analysis Effect Size (LEfSe) to identify bacterial taxa with significant differences, with statistical significance set at p < 0.05 and a Linear Discriminant Analysis (LDA) score greater than 4. Additionally, predicted functional profiles of the microbial communities were created using PICRUSt2, followed by the evaluation of statistically significant differences using STAMP. 2.15. Statistical analysis Data in this study are expressed as mean ± standard error of the mean. Statistical analyses were performed using Student's t-test to assess differences between two groups and one-way ANOVA to compare means among three groups. All statistical analyses were utilized by GraphPad Prism, with significance indicated as ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. 3. Results 3.1. Oral intake of Pd@Pt alleviates HFD-induced blood glucose disorder Pd@Pt nanozyme was synthesized according to the method previously established in our previous study ([76]Fig. 1A) [[77]35,[78]36]. As illustrated in the SEM image in [79]Fig. 1B, Pd@Pt exhibit a dendritic morphology, with a diameter ranging from 40 to 50 nm. To investigate the spatial distribution of the two metal components within the Pd@Pt structure, EDS elemental mapping analysis was conducted. The analysis revealed that pink particles representing Pd are predominantly concentrated in the core, while yellow particles indicating Pt are primarily located in the outer shell, as shown in [80]Fig. 1C. This indicates that Pd@Pt forms a bimetallic nanoparticle characterized by a core-shell architecture with a high degree of mesoporosity. Fig. 1. [81]Fig. 1 [82]Open in a new tab The impact of Pd@Pt on glucose tolerance and lipid metabolism. (A) Schematic representation of the Pd@Pt synthesis process; (B) Scanning electron microscopy (SEM) image of Pd@Pt; (C) Energy-dispersive X-ray spectroscopy (EDS) mapping of Pd@Pt; (D) Fasting blood glucose levels; (E) OGTT; (F) AUC during the OGTT; (G) Representative H&E and Masson staining of liver sections, scalar bar = 100 μm. To explore the regulatory effect of Pd@Pt on lipid and glucose metabolism in the body, we used continuous oral intake of HFD and accompanying Pd@Pt for four weeks. As shown in [83]Figs. S1A and S1B, an HFD significantly increased body weight and EP weight in mice, Pd@Pt although there is no significant difference, there is a trend of weight loss. To explore whether Pd@Pt administration affects glucose tolerance, an OGTT was performed. Pd@Pt tended to lower fasting blood glucose although there was no significant difference (p = 0.06) ([84]Fig. 1D). Notably, glucose tolerance ([85]Fig. 1E and F) was reduced in the HFD group and improved in the Pd@Pt group. In addition, we found that Pd@Pt had no significant effect on the histopathological characteristics and weight of the liver ([86]Fig. 1G and [87]Fig. S1C) and adipose ([88]Fig. S1J). Similarly, we found that Pd@Pt had no significant effect on serum biochemical parameters, including AST, ALT, CHO, TG, HDL, and LDL ([89]Figs. S1D–S1I). Taken together, these results indicated that Pd@Pt has impressive glucose-lowering effects. 3.2. Pd@Pt ameliorates the gut microenvironment Next, we will focus on analyzing how Pd@Pt enhances glucose metabolism from the perspective of the gut-liver axis. Histological examination of HE-stained tissue sections revealed that Pd@Pt had no significant impact on the histopathological characteristics of the gut ([90]Fig. 2A). HFD leads to the development of low-grade chronic inflammation and contributes to insulin resistance [[91]37,[92]38]. F4/80 is a well-established marker of inflammation, and mice fed an HFD for 4 weeks exhibited a significantly larger F4/80-positive subpopulation compared to those on a normal diet ([93]Fig. 2B). Notably, the oral intake of Pd@Pt significantly reduced F4/80 expression, indicating that Pd@Pt may mitigate the low-grade inflammation induced by the HFD. Prolonged exposure to a HFD has been shown to elevate the profile of intestinal inflammatory cytokines and impair mucosal barrier integrity. This results in diminished differentiation of goblet cells, a reduction in Muc2 production, a decrease in the tight junction protein claudin-1, and an increase in serum endotoxin levels [[94]39]. In the present study, we found that mice fed an HFD for 4 weeks exhibited significantly reduced Muc2 fluorescence intensity compared to those on a normal diet ([95]Fig. 2C). Notably, the oral intake of Pd@Pt significantly increased Muc2 expression, suggesting that Pd@Pt may enhance mucosal barrier integrity. Fig. 2. [96]Fig. 2 [97]Open in a new tab The influence of Pd@Pt on the gut structure. (A) The representative H&E staining of colon section; (B) The F4/80 staining of colon section; (C) The Muc2 staining of colon section. The representative staining of colon section, scalar bar = 100 μm. 3.3. The modulation of gut microbiota by Pd@Pt An increasing body of research highlights the substantial influence of gut microbiota dysbiosis and its metabolites on the onset and development of metabolic disorders [[98][40], [99][41], [100][42]]. To investigate the regulatory effects of Pd@Pt on gut microbiota, a high-throughput analysis of the V3-V4 region of the 16S rRNA gene was performed to assess changes in microbial community composition. The analysis of α-diversity indicated no significant changes in diversity (as measured by Chao and observed species) or richness (assessed through observed ASVs) among mice subjected to a HFD compared to the HFD control group ([101]Fig. 3A). In contrast, β-diversity analysis demonstrated distinct differences in the structure of bacterial communities between Pd@Pt-treated mice and those in the HFD group ([102]Fig. 3B). The alterations in the gut microbiota structure among HFD and Pd@Pt mice correlated with changes in relative abundance patterns observed at both the phylum and genus levels ([103]Fig. 3C and F). Additionally, we observed that treatment with Pd@Pt resulted in an increased Firmicutes to Bacteroidetes (F/B) ratio compared to the HFD group ([104]Fig. 3D). Also, we observed a notable increase in the relative abundance of several genera, including Parvibacter, Muribaculaceae, Desulfovibrio, and Candidatus_Saccharimonas, in mice fed an HFD. In contrast, administration of Pd@Pt significantly reduced the relative abundances of these genera ([105]Figs. S2A–S2D). Additionally, oral supplementation with Pd@Pt led to a significant increase in the relative abundances of genera such as Coriobacteriaceae_UCG-002, Streptococcus, Staphylococcus, Christensenellaceae_R-7_group, Monoglobus, Lachnospiraceae_NK4A136_group, Lactobacillus, Enterorhabdus, Clostridium_sensu_stricto_1, and Ileibacterium compared to the HFD group ([106]Figs. S2E–S2N). Furthermore, the analysis of the heatmap depicting dominant genera indicated an increasing trend in the number of dominant genera following Pd@Pt treatment ([107]Fig. 3F). Fig. 3. [108]Fig. 3 [109]Open in a new tab Structural and compositional analysis of the gut microbiota. (A) α-Diversity represented by observed ASV levels; (B) PLS-DA plot illustrating β-diversity of the gut microbiota; Relative abundance of the gut microbiota at the (C) phylum and (E) genus levels; (D) Comparison of the Firmicutes to Bacteroidetes (F/B) ratio between the Pd@Pt and HFD groups. (F) Comparison of the principal different genera in the HFD and Pd@Pt groups using heat maps. To assess the impact of Pd@Pt on the specific taxa of microorganisms in mice subjected to a HFD, we conducted LEfSe analyses spanning from the phylum to the species level, with significance established at p < 0.05 and LDA scores exceeding 3 ([110]Fig. 4A). The analysis revealed a predominance of Dubosiella in both the HFD and Pd@Pt groups, while Ileibacterium and Ileibacterium_valens were primarily concentrated in the HFD group. Furthermore, we conducted a comprehensive assessment of the alterations in gut microbiota in HFD-fed mice treated with Pd@Pt, utilizing Sankey diagrams to illustrate these changes from both phylum and species perspectives ([111]Fig. 4B). The PICRUSt2 analysis, integrating 16S rRNA gene sequencing data from the gut microbiota with widely available databases, enabled the prediction of the functional profile of the microbiota, facilitating the creation of a functional “map” [[112]36,[113]37]. As illustrated in [114]Fig. 4C, significant metabolic pathway differences related to glucose and lipid metabolism were observed in samples from the HFD group, including caprolactam degradation, carbon fixation pathways in prokaryotes, toxoplasmosis, inositol phosphate metabolism, and the citrate cycle (TCA cycle) (p < 0.05). Compared to HFD mice, photosynthesis, toluene degradation, bisphenol degradation, chloroalkane and chloroalkene degradation, fluorobenzoate degradation, dioxin degradation, ascorbate and aldarate metabolism, nicotinate and nicotinamide metabolism, and flavonoid biosynthesis (p < 0.05) ([115]Fig. 4C). Fig. 4. [116]Fig. 4 [117]Open in a new tab The comparative analysis of gut microbiota at the genus level. (A) The bacterial taxa with differing abundances between the HFD and Pd@Pt groups; (B) The relative abundance of gut microflora at the phylum level (left) and genus level (right) as depicted in Sankey diagrams for HFD and Pd@Pt samples; (C) The functional predictions of fecal microbiota in both the HFD and Pd@Pt groups as per the KEGG pathway analysis. PP: Pd@Pt. 3.4. The changes in fecal metabolomic features induced by Pd@Pt To define the functional alterations of Pd@Pt regulate-related gut microbiota and their relationship with the host, we profiled the metabolome in cecal contents using untargeted LC–MS/MS. The Human Metabolome Database (HMDB; [118]https://hmdb.ca) represents the largest and most comprehensive repository of biologically relevant metabolites specific to humans [[119]43,[120]44]. This database provides extensive information concerning human metabolites, encompassing their biological functions, physiological concentrations, associations with diseases, metabolic pathways, chemical reactions, and reference spectra. As shown in [121]Fig. S3A, the most HMDB compound classification was lipids and lipid-like molecules (818, 30.44%), followed by organic acids and derivatives (637, 23.71%) and organoheterocyclic compounds (405, 15.07%). In addition, metabolites can be classified into lipid, peptide, and other categories by KEGG compound classification. In the present study, we found that metabolism mainly focuses on [122]compounds with biological roles, phytochemical compounds, and lipids ([123]Figs. S4A–S4C). We further mapped the metabolites altered by Pd@Pt to their respective biochemical pathways through metabolic enrichment and pathway analysis based on KEGG annotations. The result displayed that the majority of terms in the fields of metabolism were mainly associated with metabolism of cofactors and vitamins, lipid metabolism, amino acid metabolism, and carbohydrate metabolism ([124]Fig. S3B). Notably, we identified a total of 3704 annotated differential metabolites between the HFD group and the Pd@Pt group, with 326 exhibiting increased expression and 180 showing decreased expression (log2|fold change| >1 and p < 0.05) ([125]Fig. S3C). PLS-DA demonstrated distinct metabolic profiles in both positive and negative ion modes between the two groups ([126]Fig. S3D). Similarly, the heatmap depicts these differences between the HFD and Pd@Pt groups ([127]Fig. S3E). To further analyze the 506 differential metabolites between the Pd@Pt and HFD groups, we first analyzed them by HMDB classification, and the results are shown these metabolites mainly involved in organoheterocyclic compounds (55, 14.32 %), organic acids and derivatives (80, 20.83 %), lipids and lipid-like molecules (124, 32.29 %), and others ([128]Fig. S5A). To facilitate a more intuitive analysis of the differential metabolites’ changing trends across various groups, we performed clustering of the differentially expressed metabolites. As shown in [129]Fig. S5B, the top 20 metabolites in abundance were observed to be significantly up-regulated and down-regulated. The OPLS-DA/PLS-DA model was employed as a supervised approach, utilizing a 7-fold cross-validation method to evaluate the predictive performance across different paired samples. Variable importance in projection (VIP) analysis of the first principal component was conducted to identify key metabolites that contribute significantly to classification outcomes [[130]45,[131]46]. Subsequently, we generated clustering heat maps and VIP bar charts to visualize the expression patterns of metabolites across samples within each differential group, as well as the p values associated with metabolites based on both VIP and univariate statistical analyses. We found that Octa-3,6-Dienedioylcarnitine, Panaxytriol, N-[2-Hydroxy-2-(4-Nitrophenyl)Ethyl]Acetamide, Oxaceprol, Oxyphencyclimine, Gentian Violet Cation, Methyl Violet, Dihydroethidium, Sisomicin Sulfate, 20-Hydroxy-Leukotriene E47-(7-Hydroxy-3,7-Dimethylocta-2,5-Dienoxy)Chromen-2.one, (3B,4B,11B,14B)-11-Ethoxy-3,4-Epoxy-14-Hydroxy-12-Cyathen-15-A1 14-Xyloside, Hovenine A, and Gamma-L-Glutamyl-L-Pipecolic Acid were positively correlated with the role of Pd@Pt. While, the contents of Pc (20:410:0), Nepetariaside, (E)-N-(2-Amino-4-Fluorophenyl)-3-(1-Cinnamy1-1H-Pyrazo1-4-Yl) Acrylamide, 2-Amino-4-Nitrophenol, Arthrobactin, and sdz-205,557 Hydrochloride were negatively related to the administration of Pd@Pt ([132]Fig. S6A). The provided visual depiction facilitates a discernible evaluation of the prominence and expression tendencies of differential metabolites. The KEGG pathway database represents a comprehensive collection of manually curated metabolic pathways that elucidate molecular interactions, physiological and biochemical reactions, as well as the intricate relationships among gene products. The histogram of KEGG revealed that the majority of terms in the fields of environment information processing (EIP) and human disease (HD) are linked to the adenosine 3′, 5′-cyclic monophosphate (cAMP) signaling pathway, nuclear factor kappa-B (NF-κB) signaling pathway, mitogen-activated protein kinase (MAPK) signal pathway, and insulin resistance, which are associated with the metabolism of glucose homeostasis ([133]Fig. S6B). KEGG pathway enrichment analysis involves assessing the enrichment of selected metabolites within a defined metabolic set. This process employs a hypergeometric distribution algorithm to identify pathways that exhibit significant enrichment of metabolites within the specified set. Next, we performed a KEGG enrichment analysis on the differentially expressed metabolites. This analysis revealed that the 15 most significantly enriched KEGG pathways associated with metabolism were identified and prioritized based on their enrichment scores (p < 0.05), as depicted in [134]Fig. 5A. Of particular interest, it was observed that the NF-κB and cAMP signaling pathway, which have been previously implicated in glucose metabolism [[135][47], [136][48], [137][49], [138][50]]. Notably, we also found that the differential metabolites also enrichment in the NF-κB and cAMP signaling pathway. These differential metabolites are also associated with insulin resistance, which is coupled with the KEGG pathway in terms of human disease ([139]Fig. S6B). Furthermore, the KEGG topology analysis focused on cutin, suberine and wax biosynthesis, nucleotide metabolism purine metabolism, pyrimidine metabolism, linoleic acid metabolism, arginine and proline metabolism, and cysteine and methionine metabolism ([140]Fig. 5B). Taken together, these observations define a mechanism whereby Pd@Pt regulates metabolic homeostasis, showing that Pd@Pt could be a potential nanomedicine to treat metabolic disease. The results suggest that the regulation of metabolites by Pd@Pt primarily targets energy metabolism. This finding indicates that Pd@Pt may ameliorate the alterations in energy metabolism induced by a HFD through the modulation of intestinal metabolites. Fig. 5. [141]Fig. 5 [142]Open in a new tab KEGG enrichment analysis. (A) KEGG Enrichment analysis; (B) The KEGG topology analysis. 3.5. Transcriptional profiling of the liver in mice treated with Pd@Pt To further explore the molecular mechanisms involved in blood glucose regulation by Pd@Pt, we performed transcriptome analysis of mouse liver. The results, illustrated in [143]Figs. S7A–S7C, unveiled a total of 505 DEGs when comparing the HFD group with the Pd@Pt-treated group. Among these DEGs, 107 genes exhibited upregulation, whereas 398 genes demonstrated downregulation (log2|fold change| >1 and p < 0.05). The heatmap depicting these DEGs between HFD and Pd@Pt group ([144]Fig. S7D). Subsequently, we subjected the DEGs to Reactome pathway annotation to pinpoint the most relevant genes for in-depth investigation. As shown in [145]Fig. S7E, we found that the top 2 abundant regulated Reactome pathways were signal transduction and metabolism. The enrichment analysis of DEGs based on the KEGG annotation revealed that the majority of terms in the fields of environment information processing (EIP) and organismal systems (OS) are linked to signaling transduction and the immune system ([146]Fig. S8). Collectively, these observations suggest that the predominant effects of Pd@Pt on the liver's transcriptional landscape are centered on metabolic and signal transduction pathways. KEGG analyses were performed to further delineate the principal signaling pathways modulated by the DEGs. Initially, the 17 most significant KEGGs associated with metabolic processes were identified and prioritized based on their enrichment scores (p < 0.05) ([147]Fig. 6A). Literature has previously implicated the MAPK, NF-κB, and phosphoinositide-3-kinases (PI3K)- protein kinase B (Akt) signaling pathway in glucose and lipid metabolism [[148]47,[149]48,[150]51]. Notably, we also found that the DEGs also enrichment in the MAPK, NF-κB, and PI3K-Akt signaling pathway. In addition, the enriched chordogram shows differential genes for different pathways ([151]Fig. S9). Together, these observations define a mechanism whereby Pd@Pt regulates metabolic homeostasis, showing that Pd@Pt could be a potential nanomedicine to treat metabolic disease. Fig. 6. [152]Fig. 6 [153]Open in a new tab The analysis of DEGs. (A) KEGG enrichment analysis; (B) GSEA analysis. Gene Set Enrichment Analysis (GSEA) represents a frequently employed analysis approach within various scientific domains, particularly in medical research [[154]52]. Unlike traditional enrichment analysis methods that rely on pre-filtering genes using fixed thresholds, GSEA mitigates the issue of inadequate information on potentially significant genes at the micro level. The distinct advantage of GSEA lies in its ability to bypass differential analysis and directly analyze the expression of all genes, thereby detecting gene sets with subtle but consistent differential expression patterns [[155]52]. In the context of GSEA, we identified 6 critical signaling pathways that were positively associated with the Pd@Pt treatment group and negatively associated with the HFD group, predominantly including ribosome biogenesis, oxidative phosphorylation, glycolysis and gluconeogenesis, glutathione metabolism, pyruvate metabolism, and phenylalanine metabolism ([156]Fig. 6B). These findings imply that Pd@Pt nanoparticles have the capacity to alleviate metabolic dysregulations by counteracting the signaling pathways perturbed by a HFD. To investigate the functional roles of the key genes modulated by Pd@Pt in maintaining metabolic equilibrium within the organism, we selected 17 genes from the aforementioned three pivotal signaling pathways for protein-protein interaction (PPI) and transcription factor (TF) analyses. Utilizing the STRING 12.0 database, we conducted PPI network analysis to pinpoint the principal gene targets within the pathways influenced by Pd@Pt. The analysis revealed that the genes with the highest interaction scores were CD14 and Tlr4, as well as Itgb4 and Itga4 ([157]Fig. 7A). Moreover, the co-expression patterns of proteins encoded by genes that display correlated expression across numerous experiments were investigated. The co-expression network of the 17 central genes is depicted in [158]Fig. 7B, with the genomic occurrence of gene families exhibiting similar expression patterns highlighted in [159]Fig. 7C. To identify potential upstream TFs that may be implicated in the observed variations, TF enrichment analysis was conducted on the DEGs within the critical pathways using the ChEA3 database, which draws upon information from ChIP-seq experiments ([160]Table S1). Fig. 7. [161]Fig. 7 [162]Open in a new tab The analysis of key genes. (A) STRING database-generated network visualization for the 17 DEGs; (B) Co-expression patterns observed in Mus musculus; (C) Co-occurrence of key genes; (D) The Sankey-bubble plots. The top 10 TFs identified from the literature ChIP-seq database included STAT3, CEBPD, EGR1, OLIG2, GATA2, ELK3, ERG, SALL4, TP53, and TP63 ([163]Fig. S10). It has been recently reported that the regulation of metabolic disorders is linked to TFs such as STAT3 and CEBPD [[164]53,[165]54]. To further explore the three-star pathways, key differentially expressed genes in these pathways were identified by Sankey-bubble plots ([166]Fig. 7D). The results indicated that the PI3K-Akt signaling pathway includes 5 hub genes, including Tlr4, Kitl, Nr4a1, and Erbb4. Similarly, the NF-κB signaling pathway included Tlr4, Cd14, and Gadd45a. Kitl, Nr4a1, Cd14, Klk1b4, Gadd45a, and Erbb4 were enriched in the MAPK signaling pathway ([167]Fig. 7D). Extensive research has implicated TLR4-mediated inflammation in the pathogenesis of intestinal and hepatic inflammation, which may be a critical factor contributing to liver-related pathologies [[168]55,[169]56]. Collectively, these findings suggest that the regulatory effects of Pd@Pt on glucose metabolism may be intricately linked to the modulation of inflammatory processes. The WGCNA algorithm is a frequently employed method for constructing gene co-expression networks. In such networks, genes that exhibit consistent expression patterns across various samples are grouped together, with the strength of their co-expression typically quantified by their correlation coefficients [[170]57]. Generally speaking, genes within the same co-expression network have similar expression forms, while genes in different gene co-expression networks have large differences in expression forms. Upon identification of the co-expression gene modules, these modules were correlated with relevant phenotypic data to investigate the association between the gene network and the phenotype, as well as to identify the pivotal genes within the network. To elucidate the interactions between genes and phenotypes subsequent to Pd@Pt treatment, WGCNA was employed. A total of 161 differentially expressed genes (DEGs) were integrated into the analysis and grouped into three distinct modules. The module-trait association analysis revealed that the MEblue module exhibited a positive correlation with the phenotype in the Pd@Pt-treated group, whereas it demonstrated a negative correlation with the phenotype in the HFD group ([171]Figs. S11A and S11B). Subsequently, the DEGs in the MEblue module were displayed as a hot map as shown in [172]Fig. S11C. 3.6. The correlation analysis between gut microbiota, metabolites, and DEGs Given that gut microbiota may regulate physiological functions through the action of metabolites [[173]40,[174]58], we conducted a correlation analysis among differential gut bacteria, metabolites, and differential genes. As shown in [175]Fig. 8A, Muribaculaceae, Desulfovibrio, Parvibacter, and Candidatus_Saccharimonas were significantly negative correlation with 3-Hydroxy-12-0xocholan-24-oic Acid, 17-(Acetyloxy)-3-Hydroxy-6-Methyl-(3B,5B,6A)-Pregnan-20-one, 7-Ketolithocholic Acid, Inecalcitol, 7-Dehydroxychol-8(14)-Enic Acid, Dioctyl Phthalate, 2-Hydroxyadenine, Ethyl 2-Amino-3-Sulfanylpropanoate, 5,8,11-Trihydroxyoctadec-9-Enoic Acid, 4-((6-Methoxyquinolin-8-Yl)Amino)Pentanoic Acid, 2′-Deoxyuridine, and Deoxyinosine. In contrary, these metabolites were significiantly positively with Streptococcus, Lachnospiraceae_NK4A136_group, Lactobacillus, Christensenellaceae_R-7_group, Coriobacteriaceae_UCG-002, lleibacterium, Staphylococcus, Clostridium_sensu_stricto_1, g_Monoglobus, and g_Enterorhabdus. Additionally, we observed significant negative correlations between the DEGs Ccne2, Erbb4, Il7, Rps6ka3, Eda2r, Itga4, Tir4, and Kith, and the metabolites including 2′-Deoxyuridine, Deoxyinosine, 3-Hydroxy-12-0xocholan-24-oic Acid, 17-(Acetyloxy)-3-Hydroxy-6-Methyl-(3B,5B,6A)-Pregnan-20-one, 7-Ketolithocholic Acid, 7-Dehydroxychol-8(14)-Enic Acid, Inecalcitol, Dioctyl Phthalate, 5,8,11-Trihydroxyoctadec-9-Enoic Acid, 4-((6-Methoxyquinolin-8-Yl)Amino) Pentanoic Acid, 2-Hydroxyadenine, and Ethyl 2-Amino-3-sulfanylpropanoate. While, Gadd45a, Kik1b4, Cd40, Mapk13, Hspb1, Itgb4, Dusp8, Cd14, and Nr4a1 were significantly positive with these metabolites ([176]Fig. 8B). [177]Fig. 8C depicts a heatmap of the correlation network, indicating a strong association between the differential genes and the gut bacterial composition ([178]Table S2). Collectively, these observations suggest that gut microbiota and their metabolites are critical targets for comprehending the metabolic balance modulated by Pd@Pt, thereby providing a robust theoretical foundation for advancing the clinical application of Pd@Pt through the gut-liver axis. The potential of nanozymes to modulate gut microbiota offers a novel avenue for developing strategies to enhance gut health and manage gut-associated disorders. Further research is necessary to fully harness the therapeutic potential of nanozymes in this context and to establish safe and efficacious nanozyme-mediated interventions for gut microbiota and metabolite regulation. Fig. 8. [179]Fig. 8 [180]Open in a new tab The correlation analysis network of gut microbiota, metabolites, and DEGs. (A) The correlation analysis of differential gut microbiota and differential metabolites between the Pd@Pt and HFD groups; (B) The correlation analysis of DEGs and differential metabolites between the Pd@Pt and HFD groups; (C) The correlation analysis of differential gut microbiota and DEGs between the Pd@Pt and HFD groups. 3.7. The biosafety analysis of Pd@Pt Considering the unique small particle size of nanozymes, evaluating their biosafety is imperative for their potential therapeutic applications. The diminutive dimensions can enhance cellular uptake and reactivity, yet they may also pose risks of toxicity and unintended biological effects. Firstly, after 24 h of incubation with Pd@Pt, AML12 and HepG2 cells retained >80 % viability across Pd@Pt at concentrations ranging from 0 to 300 μg/mL ([181]Fig. 9A; S12A). Biocompatibility is essential for the translation of nanozymes into clinical applications [[182][59], [183][60], [184][61]]. Then, we incubated rabbit red blood cells with varying Pd@Pt concentrations (0–800 μg/mL) at 37 °C for 2 h to evaluate the biocompatibility of the Pd@Pt in the blood circulation. As shown in [185]Fig. 9B, no significant hemolysis (less than 2 %) was observed at concentrations from 0 to 200 μg/mL after incubation of Pd@Pt with red blood cell suspension, indicating that the Pd@Pt had good biological safety. γ-H2AX, which is formed by the phosphorylation of the Ser-139 residue of the histone variant H2AX, serves as a well-established biomarker for double-strand DNA breaks. To quantitatively assess DNA damage, HepG2 cells were treated with Pd@Pt for 24 h, and the expression level of γ-H2AX was evaluated through immunofluorescence staining. As illustrated in [186]Fig. 9C, both the control group and the Pd@Pt group exhibited a weak green fluorescence in comparison to the positive group, suggesting that Pd@Pt has minimal potential for inducing DNA damage. The TUNEL assay revealed that only slight apoptosis of hepatocytes was observed under Pd@Pt exposure, indicating that Pd@Pt did not trigger significant cell apoptosis ([187]Fig. 9D). Subsequently, we evaluated the genotoxicity of Pd@Pt materials, and the results of comet assay showed that Pd@Pt did not cause obvious DNA damage tailing ([188]Fig. 9E). In addition, since orally administered drugs are predominantly metabolized and excreted through the liver and kidneys, it is essential to evaluate the effects of Pd@Pt on both liver and kidney function to comprehensively assess its biosafety. The results demonstrate that in vitro kidney and liver (Vero and HepG2 cells) retained over 80 % viability across Pd@Pt concentrations ranging from 0 to 300 μg/mL ([189]Figs. S12A–S12B), and histopathological analysis of kidney tissues revealed no structural abnormalities or pathological damage ([190]Fig. S12C). These findings collectively indicate that Pd@Pt does not induce hepatotoxicity or nephrotoxicity at the tested doses, further supporting its favorable biosafety profile for potential therapeutic applications. Understanding the mechanisms of interaction between nanozymes and biological systems will facilitate the design of safer, more effective therapeutic agents. Future studies should prioritize the rigorous evaluation of these safety profiles to advance the translational potential of nanozymes in clinical applications. Fig. 9. [191]Fig. 9 [192]Open in a new tab The biosafety analysis of Pd@Pt. (A) The CCK8 evaluated in AML12 cell; (B) Hemolysis assay of Pd@Pt in vitro; (C) The γ-H2AX staining; (D) The TUNEL assay.; (E) The comet assay. 4. Discussion ROS, immune dysregulation-induced inflammatory outbreaks and microbial imbalance play critical roles in the development of metabolic disorders [[193]5]. Recently, the regulation of nanozyme in modulating the composition and function of gut microbiota have attracted the attention of researchers. For example, Zhu et al. reported that metal-phenolic nanozyme influences the gut microbiome towards a beneficial state by enhancing bacterial diversity and shifting the compositional structure toward an anti-inflammatory phenotype [[194]62]. The 16S rRNA gene sequencing results indicate that post-oral of ZnPBA@YCW nanozyme can effectively regulate gut microbiota by enhancing the bacterial richness and diversity, significantly increasing the abundance of probiotics with anti-inflammatory phenotype while downgrading pathogenic E. coli to the same level as normal mice [[195]28]. In present study, structural perturbations in the composition of the intestinal bacterial community further lead to microbiota dysfunction, thereby exacerbating the intestinal inflammatory response. Pd@Pt can reverse the decrease in beneficial bacteria and increase in pathogenic bacteria caused by HFD and improve the overall composition of the gut microbiota. Additionally, it can also regulate the levels of gut microbial metabolites and increase the levels of beneficial metabolites. Therefore, the gut microbiota may serve as a critical target for the therapeutic effects of Pd@Pt in improving metabolic diseases, highlighting new avenues for research and treatment strategies. An increased F/B ratio has been associated with obesity in some studies [[196]63,[197]64], while, the higher F/B ratio in the Pd@Pt group compared to the HFD group was observed in our present study. At the same time, many studies have found no change in F/B or even a decrease in the ratio in obese animals and humans [[198]65,[199]66]. Conflicting results regarding F/B ratios in obesity studies can be attributed to several factors. Methodological differences, such as differences in DNA extraction, primer selection, and sequencing platforms, can lead to biases in microbial detection and quantification. Problems with subject recruitment and characterization, including uncontrolled lifestyle factors such as diet, antibiotic use, physical activity, and exposure to environmental pollutants, may confound the association between F/B ratios and obesity. Geographic and environmental factors (e.g., dietary habits and pollutant exposure in different regions) can also contribute to variations in microbiota composition. In addition, the high inter-individual variability of the gut microbiome and the possible presence of multiple categorical features associated with obesity, rather than just F/B ratios, further complicate the interpretation of results. Finally, the role of short-chain fatty acid production and metabolic endotoxemia in obesity may not be fully reflected by F/B ratios, as variation in these factors may be independent of the relative abundance of Firmicutes to Bacteroidetes-like bacteria [[200]67]. In previous studies, it has been reported that primary bile acids, synthesized in hepatocytes, are subsequently transformed into secondary bile acids by intestinal microbiota. These bile acids serve as pivotal nutrient sensors and metabolic regulators of lipid, glucose, and energy metabolism [[201]68,[202]69]. Disruptions in bile acid homeostasis, along with alterations in gut microbiota (dysbiosis), have been associated with the development of metabolic disorders such as diabetes and obesity [[203]70]. Notably, 7-ketolithocholic acid, a primary bile acid, effectively alters biliary bile acid composition and reduces cholesterol saturation, indicating its potential utility in gallstone dissolution by suppressing endogenous bile acid production and biliary cholesterol secretion [[204]71]. Besides, Li et al. reported that 7-ketolithocholic acid, a metabolite produced by the intestinal microbe Parabacteroides goldsteinii and suppressed by aspirin, functions as an FXR antagonist to promote intestinal epithelial repair, enhances Wnt signaling, and supports the self-renewal of intestinal stem cells, thereby exerting protective effects on gut health, particularly in the context of aspirin use [[205]72]. The findings imply that 7-acid may be a key regulator of gut microbiota composition and hepatic physiological status. Subsequently, we explored the relationship between DEGs and gut microbiota linked to 7-ketolithocholic acid to uncover the molecular underpinnings of metabolic homeostasis regulation by Pd@Pt nanozyme, leveraging the correlative insights from transcriptomic and microbiomic analyses. In this study, we observed the significant role of Pd@Pt nanomaterials in blood glucose regulation and explored their effects on intestinal flora and related metabolites. This finding not only provides a new direction for clinical diabetes treatment and prevention strategies, but also lays the foundation for our understanding of the potential mechanisms of Pd@Pt in the regulation of body metabolism. By improving the metabolic status of diabetic patients, Pd@Pt may bring novel interventions to the clinic, especially in drug-resistant and re-emerging cases, showing a bright application prospect. However, there are some limitations in this study. First, the limited sample size may affect the generalizability of the results. Therefore, future studies should expand the sample size to improve statistical efficacy. In addition, replacing the primate model may provide us with experimental data closer to the human physiological state, thus further validating the therapeutic effect of Pd@Pt and its mechanism. The more complex physiological properties of primates will help to reveal the potential multiple roles of Pd@Pt in microbiome regulation and systemic metabolism. Finally, future studies should emphasize the interactions between Pd@Pt and gut microbiota, especially targeting the effects of specific strains on systemic metabolism. These studies will contribute to an in-depth understanding of the mechanism of Pd@Pt in the treatment of diabetes and provide an important basis for the development of personalized medical protocols. Therefore, we expect that future studies will further advance the exploration of Pd@Pt in clinical applications for effective intervention strategies. 5. Conclusion Considering the pathological microenvironment characterized by hyperglycemia and hyperlipidemia in glucose and lipid metabolism disorders, the present study developed Pd@Pt nanozyme with a relatively simple composition, high activity, and controllable biosafety. Our findings demonstrated that Pd@Pt administration not only improved glucose tolerance but also elicited pronounced changes in the expression of genes critical to glycolipid metabolic signaling, such as those within the MAPK, NF-κB, and PI3K-Akt pathways. Moreover, analyses of 16S rRNA profiling and metabolites of fecal samples from mice treated with Pd@Pt revealed significant changes in the abundance of certain beneficial gut microbiota, which are strongly correlated with alterations in DEGs and metabolites. In summary, the current investigation underscores the Pd@Pt nanozyme's potential as a novel nanomaterial with the capacity to modulate metabolic pathways, thereby contributing significant insights into the utilization of integrated multi-omics analytical approaches for the exploration of nanozyme-mediated therapeutic strategies within the domain of metabolic disorders. CRediT authorship contribution statement Yanan Wang: Writing – original draft, Methodology, Investigation, Conceptualization. Nan Cheng: Supervision, Project administration, Conceptualization. Qi Zhang: Methodology, Investigation. Fei Chang: Methodology, Investigation. Teng Wang: Methodology, Investigation. Minrui Kan: Methodology, Investigation. Yutong Han: Methodology, Investigation. Baiqiang Zhai: Investigation, Funding acquisition. Kunlun Huang: Supervision, Project administration, Funding acquisition. Xiaoyun He: Supervision, Project administration, Conceptualization. Ethics approval and consent to participate The animal study was approved by the Animal Ethics Committee of China Agricultural University (approval number: AW62113202-4-1). Consent for publication All authors declare full consent for publication. Availability of data and materials Not applicable. Funding This work was supported by the major special project on agrobiotechnology breeding (2023ZD0406304), the 2115 Talent Development Program of China Agricultural University, and Henan Province Science and Technology Tackling Key Problems Project (242102311178). Declaration of competing interest The authors declare no competing interests. Footnotes ^Appendix A Supplementary data to this article can be found online at [206]https://doi.org/10.1016/j.mtbio.2025.101685. Appendix A. Supplementary data The following is the Supplementary data to this article: Multimedia component 1 [207]mmc1.docx^ (5.7MB, docx) Data availability Data will be made available on request. References