Feed formulas for the experimental mice included the normal control diet (containing 23.07% of protein,11.85% of fat and 65.08% carbohydrates,3.40 kcal/g), purchased from Beijing Keao Xieli Feed Co., Ltd., Beijing, China, and the high-fat diet (containing 23.25% of protein, 34.55% of fat and 27.20% of carbohydrates, 5.13 kcal/g), purchased from Jiangsu Pharmaceutical & Bioengineering Co., Ltd., Jiangsu, China. SPF-grade poplar wood shavings as mouse bedding purchased from Beijing Keao Xieli Feed Co., Ltd., Beijing, China. All other chemical reagents are at least of analytical grade.

Tremella fuciformis was collected from Sichuan Yudeyuan Ecological Agriculture Technology Co.Ltd., TongJiang, Sichuan, China. The fruit bodies were defatted with anhydrous ethanol and extracted water-soluble polysaccharides with 100 times pure water at 100°C for 2 h, Repeated the extraction twice; the mixed extract was filtered, concentrated and centrifuged to obtain the supernatant. Four times the amount of anhydrous ethanol was added to the supernatant, the polysaccharide was precipitated overnight and centrifuged to obtain TP crude polysaccharide, and the impurities were washed with anhydrous ethanol. Evaporate the ethanol, re-dissolve in pure water and use the TCA method to remove the protein. Dialysis (retention capacity of 8,000 Da) for 24 h, vacuum lyophilisation to obtain TP (Wu et al., 2019; Yuan et al., 2022). The phenol-sulphuric acid method determined the total sugar content using mannose as a standard (Wang et al., 2017).

Impurities in TP were analysed by the UV-2700i system of the Shimadzu UV–Vis spectrophotometer, and the wavelength was from 200 to 700 nm. The functional groups and glycosidic bonds of TP were determined by the Fourier transform-infrared (FT-IR) spectra (Thermo, American) with a resolution of 4 cm⁻¹ in the 4,000–400 cm⁻¹ region. TP (~0.5 mg) was mixed with KBr and pressed into thin slices with abrasive tools at room temperature (Yukun et al., 2018).

HPLC determined monosaccharide composition (Shu et al., 2019). In brief, the TP (typically 2 mg) was hydrolysed with 0.5 M sulphuric acid at 120°C under nitrogen for 4 h. The TP hydrolysate was neutralised with sodium hydroxide and then derivatised with 450 μl 1-phenyl-3-methyl-5-pyrazolone (PMP) solution (0.5 M, in methanol) and 450 μl of 0.3 M NaOH at 70°C for 30 min. The reaction was stopped by neutralisation with 450 μl of 0.3 M HCl and extraction with Chloroform. HPLC analyses were performed on a Diamonsil C18 (5 μm, 4.6 mm × 150 mm) at 25°C with detection at UV 245 nm. The mobile phase was aqueous 0.1 M potassium dihydrogen phosphate (solvent A) and acetonitrile (solvent B). A gradient of B from 45% to 100% in 40 min was used, and the injection volume was 10 μl. The standard monosaccharides (Man, Rha, GlcA, GalA, Glc, Gal, Xyl) were weighed to prepare a series of concentration solutions. The derivatization of monosaccharides and the chromatographic analysis were carried out according to the above method. The molar ratio of each monosaccharide in TP was calculated by the standard external method.

The molecular weight of the TP was confirmed by liquid chromatography on a Hitachi L-2000 system (Hitachi, Japan) with a YMC PACK-Diol 200 Å (300 mm × 8.0 mm, ID 5 μm,20 nm)column by elution with ultrapure water, column temperature 30°C, the flow rate of 0.8 ml/min, and the injection volume was 20 μl and detection by the refractive index (Meng et al., 2016).

Fifty male C57BL/6J mice (8-week-old, 20 ± 2 g) were purchased from SiPeiFu Biotechnology Co., Ltd., Beijing, China. Mice were housed at constant temperature and humidity (22 ± 2°C; 55 ± 5% humidity), with free access to food and water and a 12-h light/dark cycle. All animal procedures followed the Guidelines for Care and Use of Laboratory Animals of the National Institutes of Health. The Ethical Committee approved all animal experiments for the Protection of Laboratory Animals, Sichuan Industrial Institute of Antibiotics, Chengdu University (Chengdu, China; Approval Number: SIIA 20210702). After a week of acclimatisation, the mice were randomly divided into five groups (n = 10 per group): (1) NCD group (fed normal chow and gavaged with 0.9% saline daily); (2) NCD_TP_H group (fed normal chow and gavaged with 200 mg/kg of TP daily); (3) HFD group (fed high-fat chow and gavaged with 0.9% saline daily); (4) HFD_TP_L group (fed high-fat feed and gavage of 100 mg/kg of TP daily); (5) HFD_TP_H group (fed high-fat feed and gavage of 200 mg/kg of TP daily). The experiment lasted for 12 weeks, and the body weight and food intake of each mouse were recorded weekly. In the 12th week, all mice were fasted overnight, anaesthetized with sodium pentobarbital and then subjected to cervical dislocation for execution. Serum was collected, and tissues such as epididymal fat, perirenal fat, liver and brain were excised, weighed, and tissues were fixed in 4% formaldehyde, whilst the rest were rapidly frozen in liquid nitrogen for further analysis.

Eight-week-old male C57BL/6J mice were randomly divided into four groups (5 mice in each group) after acclimatisation, and all were fed with high-fat chow (Chang et al., 2015; Zhang et al., 2020). The faecal samples of the NCD group, NCD_TP_H group, HFD group and HFD_TP_H mice were collected. 100 mg of faecal samples were taken in 1 ml of 0.9% sterile saline, vortexed and mixed for 10 s, then centrifuged at 800g for 3 min, and the supernatant was collected as the grafted sample. Fresh samples were prepared within 10 min before gavage, and fresh samples (100 μl each) were gavaged daily for 2 months before mice were executed, and serum and fat were collected for the following step analysis.

The levels of TC, TG, LDL-C, and HDL-C in serum were measured according to the instructions of the kit (Jiancheng, Inc., Nanjing, China), and the levels of GLP-1, GLP-2, IL-6, LPS, and TNF-α in serum were measured with the Elisa kit (Ruikesi Biotechnology Co., Ltd., Chengdu, China).

In the 11th week of the experiment, mice were fasted overnight (12 h; Zhang et al., 2020). A glucose solution of 1 g/kg body weight was gavaged, and blood glucose was measured after 0, 15, 30, 60, 90, and 120 min.

Total RNA was extracted from the tissue using a Trizol reagent according to the kit instructions. Equal amounts of total RNA were used to synthesise cDNA with the EX RT kit (gDNA remover; Zoman Biotechnology Co., Ltd., Beijing, China). cDNA synthesised was evaluated using a real-time quantitative PCR system (StepOnePlus™ Real-Time PCR Detection System, Applied Biosystems, United States). The PCR was performed in duplicate at 95°C for 3 min and subjected to 40 cycles of 95°C for 120 s, 95°C for 5 s, and 60°C for 10 s. GAPDH was used as the reference gene. Data were analysed using the 2^−ΔΔCt method. The primers for the genes targeted in this study are listed in the Supplementary material (Supplementary Table 1).

Total protein was obtained by lysing the tissue in ice-cold RIPA buffer containing phosphatase and protease inhibitors. The protein concentration was determined using the BCA protein assay kit, diluting each group of samples to the same protein concentration. Total protein was separated on a 12% SDS-PAGE gel and then transferred to polyvinylidene difluoride membranes (Immobilon TM-P; Millipore, United States). After being blocked with 5% non-fat milk in TBST for 1 h, the membrane was probed with primary antibody overnight at 4°C. The membranes were then probed with secondary antibodies for 1 h at room temperature and exposed using an ECL chemical enhancer. Quantitative analysis of optical density was performed using ImageJ software.

Mice epididymal fat was fixed overnight in 4% formaldehyde solution, dehydrated, embedded and sectioned, then stained with haematoxylin and eosin, and the sections were imaged using a digital trinocular camera (BA210 Digital, Motic Group Co., Ltd.) microscope system (Chen C. et al., 2018).

It quantified short-chain fatty acids in faeces using a Perkin Elmer gas chromatograph and mass spectrometer (Perkin Elmer Technologies, United States). Short-chain fatty acids were measured in faeces concerning the previously described method (Zhang et al., 2020). Briefly, weigh 200 mg of faeces accurately, add 0.5 ml of 25% methanol solution, homogenise and vortex vigorously for 5 min; centrifuge at 4°C for 20 min at 10,000 rpm; remove the supernatant and centrifuge (10,000g, 10 min) at 4°C; add 10 times the volume of 0.04 M HCl solution to the supernatant and leave overnight at 4°C in an airtight container. The samples were passed through a 0.22 μm filter membrane into a clean micro gas-phase vial. The samples were filtered using a PerkinElmer Gas Chromatography-Mass Spectrometer (Shelton, Connecticut 0648 United States) with a polar DB-FFAP capillary column (30 m × 0.25 mm i.d., a film thickness of 0.25 μm) before analysis. The carrier gas was high-purity helium (99.999%) at a flow rate of 1 ml/min, and the auxiliary carrier gas was high-purity hydrogen (99.999%). The detector temperature and inlet temperature were both 250°C. The initial temperature is 100°C, ramped up to 180°C at a rate of 5°C per minute and held for 2 min, then ramped up to 250°C at a rate of 5°C per minute and held for 5 min, and the shunt ratio is 10:1.

Bacterial DNA was isolated from the faecal using a MagPure Soil DNA LQ Kit (Magen, Guangdong, China) following the manufacturer’s instructions. DNA concentration and integrity were measured using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, United States) and agarose gel electrophoresis. PCR amplification of the V3-V4 hypervariable regions of the bacterial 16S rRNA gene was carried out in a 25 μl reaction using universal primer pairs (343F: 5′-TACGGRAGGCAGCAG-3′; 798R: 5′-AGGGTATCTAATCCT-3′). The PCR products were purified with Agencourt AMPure XP beads (Beckman Coulter Co., United States) and quantified using a Qubit dsDNA assay kit. The concentrations were then adjusted for sequencing. Sequencing was performed on an Illumina NovaSeq6000 with two paired-end read cycles of 250 bases each (Illumina Inc., San Diego, CA; OE Biotech Company; Shanghai, China). Raw sequencing data were in FASTQ format. Paired-end reads were then preprocessed using cutadapt software to detect and cut off the adapter. After trimming, paired-end reads were filtered low quality sequences, denoised, merged and detect and cut off the chimaera reads using DADA2 (Chao and Bunge, 2002) with the default parameters of QIIME2 (Hill et al., 2003). At last, the software output the representative reads and the ASV abundance table. The representative read of each ASV was selected using QIIME 2 package. All representative reads were annotated and blasted against Silva database Version 138 (or Unite; 16 s/18 s/ITS rDNA) using q2-feature-classifier with the default parameters. The microbial diversity in faecal samples was estimated using the alpha diversity that includes the Shannon index and Simpson index. The Unifrac distance matrix performed by QIIME2 software was used for unweighted Unifrac Principal coordinates analysis (PCoA) and phylogenetic tree construction. The 16S rRNA gene amplicon sequencing and analysis were conducted by OE Biotech Co., Ltd. (Shanghai, China).

One-way analysis of variance and Tukey’s test were used to determine significant differences. All data were presented as the mean ± standard deviation, and a p-value <0.05 was considered statistically significant. Statistical analysis was performed using GraphPad Prism version 9.0 (San Diego, California, United States).

The yield of TP from Tremella fuciformis fruit body by hot water extraction was about 9.8%. There were no prominent absorption peaks at 254 nm and 280 nm, indicating that the impurities in TP were entirely removed and TP had been refined (Figure 1A). The molecular weight distribution of TP ranges from 10.07 to 66.29 kDa, and the main molecular weight was 37.28 kDa (Figure 1B).TP was composed of Man, Rha, GlcA, Glc, Gal, and Xyl in a molar ratio of 15.95:0.31:2.47:1.20:3.18:1.72 (Figure 1C). The infrared spectrum of TP belongs to typical polysaccharide characteristics (Figure 1D); the absorption peaks around 3,421 and 2,934 cm⁻¹ represent the stretching vibration of O-H and C-H, respectively. The absorption peaks around 1,732 and 1,615 cm⁻¹ were the stretching vibrations of a carboxyl group (COO-), and the absorption peak around 1,078 cm⁻¹ was owing to the stretching vibration of C-O-C and C-O-H bending vibration of pyranose ring (Wu et al., 2022; Yuan et al., 2022). The results showed that TP was a mannose-based heterosaccharide, and its monosaccharide composition was similar to the previously reported Tremella polysaccharide, but the composition ratio was inconsistent (Wu et al., 2019; Gao et al., 2022).

After 12 weeks of feeding, we observed that feeding a high-fat diet caused a significant increase in body weight, fat weight, and liver weight in the mice, which was reversed with the TP intervention. In HFD-fed mice, TP inhibited weight gain and reduced fat accumulation in a dose-dependent manner (Figures 2A–C). Morphological observations showed that TP significantly inhibited high-fat diet-induced fat accumulation (Figure 2D) and adipocyte expansion (Figures 2E–F) in mice. These results revealed that TP reduced weight gain and fat accumulation in HFD-fed mice. TP ameliorated obesity in association with the modulation of lipid metabolism.

At week 11 of the experiment, a glucose tolerance test was performed, and glucose tolerance was quantified as the area under the curve (AUC) from 0 to 120 min (Figures 2H–I); glucose levels at 15–120 min were higher in HFD-fed mice than in NCD-fed mice. Mice in the HFD_TP_L and HFD_TP_H groups had smaller glucose AUC and showed better glucose regulation compared to the HFD group. Table 1 showed the effect of TP on blood biochemical parameters in mice. We observed that HFD significantly increased fasting blood glucose (Figure 2G) and serum TG, TC, and LDL-C levels in mice, a trend that was reversed by feeding TP. Similarly, the feeding of TP significantly increased the serum levels of HDL-C, GLP-1, and GLP-2 compared to the HFD group, indicating a better regulation of blood glucose and lipids by TP.

Gut microbiota plays a causal role in the development of metabolic syndrome caused by obesity and obesity-related metabolic disorders, and its diversity and changes are one of the important biological characteristics of the development of obesity (Federico et al., 2017; Fan and Pedersen, 2021). Therefore, we analysed the effect of TP on the composition of the gut microbiota by pyrophosphate sequencing-based analysis of bacterial 16S rRNA gene (regions V3-V4) in mice faeces. PCoA analysis showed an apparent clustering of the microbiota composition in each group (Figure 3A), with PCoA1 showing a 46.31% difference, mainly reflecting the effect of a normal and high-fat diet on changes in the structure of the gut microbiota; the vertical coordinate PCoA2 showed a 12.28% difference, which mainly reflects the effect of TP on the gut microbiota. NMDS plots showed that the high-fat diet significantly alters the structure of gut microbes in mice (Figure 3B). Alpha diversity analysis showed that the HFD group had the smallest chao1 values (Figure 3C) and obserced_species (Figure 3D) compared to the other four groups, and these two values were significantly higher after feeding TP. As shown in Figures 3E,F, the shannon and simpson indices were significantly higher after the TP intervention. These results suggested that the HFD reduced species’ diversity in mice’s gut microbiota. In contrast, feeding TP significantly improved the gut microbiota, mainly in increased diversity.

Furthermore, to investigate the regulatory effects of TP on the gut microbiota, we compared the relative abundance of the major taxa in the faecal microbiota of mice from five dietary groups. The gut microbiota composition has significant differences at phylum, family, and genus levels. At the phylum level, the histogram of the relative abundance of gut microbiome showed (Figure 4A) that all experimental groups included Firmicutes, Bacteroidota, Desulfobacterota, Deferribacterota, and Proteobacteria, with Firmicutes and Bacteroidota being the major phylum, accounting for over 80% of the total. Compared to the NCD group, The HFD group significantly altered the gut microbiota composition, characterised by a significant increase in the relative abundance of Firmicutes (Figure 4B) and a significant decrease in Bacteroidota (Figure 4C), resulting in a significant increase in the Firmicutes/Bacteroidota ratio. Interestingly, the F/B ratio also significantly decreased in the NCD_TP_H group compared to the NCD group (p < 0.05), and the Firmicutes/Bacteroidota ratios were significantly (p < 0.05) reduced in the HFD_TP_L and HFD_TP_H groups compared with the HFD group (Figure 4D) in a dose-dependent manner and tended to restore to the level of the NCD group.

At the family level, the most abundant microorganisms were mainly Muribaculaceae, Lachnospiraceae, Oscillospiraceae, Rikenellaceae, Clostridia_UCG-014, Prevotellaceae, Erysipelotrichaceae, Ruminococcaceae, and the high-fat diet significantly altered the relative abundance of the gut microbiota (Figure 4E). Compared with the NCD and TP intervention groups, the relative abundance of Desulfovibrionaceae (Figure 4G), Clostridia_UCG-014, Clostridiaceae (Figure 4H), Lachnospiraceae and Bacteroidaceae was higher in the HFD group. Whilst Muribaculaceae (Figure 4F), Oscillospiraceae, Lactobacillaceae and Prevotellaceae had lower relative abundance. The abundance of gut microbiota in HFD and TP treatment groups was significantly altered at the genus level (Figure 4I). Compared with the NCD group and the TP intervention group, the abundance of beneficial microbiota was lower in the HFD group, such as Muribaculaceae, Lachnospiraceae_NK4A136_group, Alistipes (Figure 4J), Colidextribacter, Lactobacillus, GCA-900066575. Whilst the relative abundance of Desulfovibrio (Figure 4L), Bilophila (Figure 4K), Blautia, and Intestinimonas was higher in the HFD group. These results suggested that a high-fat diet altered the gut microbiota in mice, and TP intervention reversed the HFD-induced gut microbiota alterations, especially obesity-related microbes such as Muribaculaceae, Lachnospiraceae, Alistipes, Bilophila, and Desulfovibrio.

The effect of TP on the content of SCFAs and BCFAs in the faeces of mice was shown in Table 2. The content of SCFAs in the faeces of mice in the HFD group was significantly lower than in the other four groups. In contrast, the levels of acetic, propionic and butyric acids were significantly higher after feeding TP (p < 0.05), There was no significant change in the content of valeric acid. In addition, the TP intervention also influenced the changes in the content of BCFAs, mainly in the form of a significant increase in the content of isobutyric acid (p < 0.05), but no significant change in the content of isovaleric acid.

Gut hormones are essential messengers in the microbe-gut-brain axis. qPCR was performed to detect the expression of FFAR2, PYY, and GLP-1 in the gut, NPY, AgRP, POMC, and CART in the hypothalamus, and Western Blot to detect the protein expression of GLP-1R, POMC, NPY2R, and AgRP. We observed that the expression of FFAR2 was significantly lower in the HFD group compared to the NCD group. Compared with the HFD group, the relative expressions of FFAR2, PYY, and GLP-1 increased in the three TP-fed groups, and the expression of PYY and GLP-1 in the HFD_TP_L and HFD_TP_H groups was dose-dependent (Figures 5A–C). It indicated that TP caused an increase in the production of short-chain fatty acids such as acetic acid and propionic acid in the intestine of mice (Table 2), which activated the expression of its receptor FFAR2. NPY, AgRP, POMC, and CART were critical hypothalamus regulators controlling dietary intake. NPY and AgRP mainly act as food intake inhibition, whilst CART and POMC act as facilitators of feeding. Compared to the NCD group, the expression of NPY and AgRP was significantly increased in the HFD group and decreased in the NCD_TP_H groups, and HFD_TP_L and HFD_TP_H groups were significantly decreased compared with the HFD group (Figures 5D–E). CART and POMC were up-regulated in TP-fed groups (Figures 5F,G). Compared to the HFD group, the Western blot showed that both doses of TP caused a significant increase in the expression of GLP-1R, POMC and NPY2R and decreased the expression of AgRP (Figures 5H,I). These results suggested that TP regulated the release of intestinal hormones, which ultimately regulated the expression of regulators controlling diet in the hypothalamus to suppress obesity in mice.

To investigate the effects of gut microbes on obesity in mice fed TP, we transplanted faeces from four groups (NCD, NCD_TP_H, HFD, HFD_TP_H) of donor mice to HFD-fed mice and then monitored the obesity-related traits as well as metabolic changes in blood glucose and blood lipids. The results showed that transplantation of faeces from NCD, NCD_TP_H, and HFD_TP_H mice reduced body weight (Figure 6A) and organ weight (Figure 6C), especially liver weight, and reduced fat accumulation in the recipient mice compared to the HFD_HFD group (Figures 6B–D). The morphological observation was that FMT inhibited the expansion of adipocytes (Figures 6E,F). Table 3 showed the changes in blood biochemical parameters in mice, similar to the results in Table 1. Compared to the HFD_HFD group, the other three groups had significantly lower fasting glucose, TC, TG, and LDL-C and higher levels of HDL-C, GLP-1, and GLP-2. These results suggested that the weight reduction effect of TP on HFD-fed mice may be due to the modulation of the gut microbiota.

To further explore the regulation of gut microbes by TP, we analysed the gut microbial composition of four groups of FMT mice. A beta diversity analysis included PCoA, NMDS and UPGMA sample-level clustering analysis. The differences in the composition of the intestinal microbiota of the four groups of mice were observed by PCoA (Figure 7A); 21.07 and 14.55% for PCoA1 and PCoA2, respectively, NCD_H_HFD and HFD_H_HFD groups were mainly distributed in the right quadrant, whilst the other two groups were distributed in the left quadrant, indicating significant differences in the composition of the intestinal flora of the four groups; NMDS analysis also showed similar results (Figure 7B); the results of UPGMA sample hierarchical clustering analysis showed that each of the four groups clustered into one cluster except for HFD_HFD_6 sample (Figure 7C). In addition to this, there were significant differences in alpha-diversity between the four groups, the ACE value (Figure 7D), chao 1 value (Figure 7E) and observed_species (Figure 7F) of the HFD_HFD group were lower than those of the other three groups, and all three values increased after transplantation of faeces from mice fed TP. By comparing the shannon (Figure 7G) and simpson (Figure 7H) values between the four groups, we can see that the HFD_HFD group had lower microbial diversity than the other three groups. These results suggested that a high-fat diet altered the gut microbial structure of mice. However, transplantation of faeces from mice fed TP increased gut microbial abundance, suggesting a beneficial effect of TP on the regulation of gut microbes.

To further investigate specific changes in the gut microbiota, we compared the microbial composition of the four groups at different taxonomic levels. At the phylum level, similar to previous results, the HFD_HFD group significantly altered the gut microbiota composition by increasing the Firmicutes/Bacteroidota ratio, which was reduced after transplantation of TP-fed mice faeces (Figures 8A–D). In addition, at the family level, the microbial composition of the four groups also differed significantly (Figure 8E). The relative abundance of Actinobacteriota was lower in the HFD_HFD group, whilst the relative abundance of Proteobacteria was higher. In the HFD_HFD group, the relative population abundance of Muribaculaceae (Figure 8F), Deferribacteraceae, Marinifilaceae, Bacteroidaceae, Peptostreptococcaceae, Erysipelotrichaceae, Peptococcaceae (Figure 8G) and Tannerellaceae was lower, and was higher abundances of Clostridia_UCG-014 (Figure 8H) compared to the other three groups; at the genus level, transplanted TP-fed mice faeces significantly increased the relative population abundance of Blautia (Figure 8L), Lachnospiraceae_NK4A136_group (Figure 8J), Colidextribacter (Figure 8K) and Faecalibaculum and decreased the relative population abundance of Odoribacter, Tuzzerella, and Anaerotruncus compared to the HFD_HFD group (Figure 8I). These results suggested that HFD induced changes in the gut microbiota, yet the ratio of gut microbiota was closer to that of the NCD group after transplanting faeces from TP-fed mice.