Optimized Extraction of Soluble Dietary Fiber from Lyophyllum decastes and Its Effect on Hypolipidemic and Gut Microbiota in Mice

Abstract

Lyophyllum decastes soluble dietary fiber (LDSDF) is a polysaccharide-based active ingredient derived from the edible and medicinal fungus L. decastes. However, its extraction methods remain unoptimized, and its hypolipidemic and gut microbiota effects have yet to be thoroughly investigated in mice. In this study, response surface optimization of the LDSDF extraction method indicated an optimal extraction temperature of 99 °C, a solid/liquid ratio of 25:1 mL/g, and an extraction time of 1.9 h. The optimal ethanol precipitation parameters were a concentration ratio of 3.9, an ethanol concentration of 74.4%, and a precipitation time of 16.4 h. These conditions afforded an LDSDF yield of 15.83%. Following 6 weeks of oral gavage of LDSDF in obese mice, the results showed that LDSDF inhibited increases in body and organ weight; reduced serum levels of total cholesterol, triglycerides, and low-density lipoprotein cholesterol; increased serum levels of high-density lipoprotein cholesterol; decreased alanine aminotransferase and aspartate aminotransferase activities; and lowered systemic levels of pro-inflammatory cytokines (tumor necrosis factor-α, interleukin-6, and interleukin-1β). Concurrently, it elevated the hepatic activities of superoxide dismutase, catalase, and glutathione peroxidase; reduced malondialdehyde levels; and mitigated lesions in liver and epididymal fat cells. Meanwhile, 16S rRNA sequencing revealed that LDSDF significantly alleviated intestinal flora imbalances. Overall, this study established an optimized extraction process to obtain LDSDF with a high yield and confirmed the hypolipidemic and gut microbiota-modulating efficacy of this active ingredient, highlighting its potential for use as a functional food ingredient.

1. Introduction

The global prevalence of obesity continues to rise rapidly [1]. Along with smoking, physical inactivity, and a high-fat diet (HFD), obesity is a well-established risk factor for hyperlipidemia (HLP) [2,3]. HLP is a metabolic disorder characterized by elevated blood lipid levels and is a major contributor to the development of both atherosclerosis and coronary heart disease [4,5]. Due to its steadily increasing prevalence, the World Health Organization recognizes this disorder as a growing global public health challenge.

Statins and fibrates are the most commonly prescribed first-line pharmacotherapeutics for HLP. However, their long-term use can lead to various adverse effects, inspiring research into natural alternatives with lipid-lowering properties [6]. Dietary fiber (DF), which is classified as soluble dietary fiber (SDF) or insoluble dietary fiber, is attracting increasing attention in the health supplementation industry due to its beneficial physiological effects [7,8]. Numerous studies have demonstrated that SDF can substantially reduce HLP [9,10]. For example, in mice with HFD-induced obesity, supplementation with fermented wheat bran (i.e., SDF) reduced body and organ weight, ameliorated dyslipidemia, and significantly lowered the activities of serum liver enzymes [11]. These findings highlight the efficacy of SDF in alleviating HLP and its associated complications.

Current research on DF has predominantly focused on conventional plant sources such as fruits, vegetables, and grains [12]. Recently, edible mushrooms have emerged as novel microbial sources of DF and are attracting increasing research interest owing to their unique functional properties and therapeutic potential [13]. Lyophyllum decastes, commonly known as the lotus leaf mushroom, is both a medicinally and culinarily important fungus. Its fruiting bodies contain polysaccharides, polyphenols, proteins, and dietary fibers, which contribute to its nutritional and pharmacological value [14].

Previous research on the bioactive constituents of L. decastes has largely focused on its constituent polysaccharides, with limited investigation of its DF components. Research has demonstrated that L. decastes is a promising fungal source of SDF, with a notably higher content than that found in many commonly consumed edible mushrooms, including Agaricus bisporus, Pleurotus eryngii, Flammulina velutipes, and Lentinus edodes [15]. Common techniques for extracting DF, particularly SDF, include hot-water extraction [16], chemical extraction [17], biological fermentation [18], enzymatic extraction [19], and ultrasonic-assisted extraction [20]. These methods differ considerably in terms of extraction efficiency and the physicochemical properties of the resulting DF [21,22,23]. Among them, hot-water extraction is a simple and environmentally friendly process that achieves high yields of L. decastes SDF (LDSDF). However, the optimal extraction parameters remain poorly defined.

Although LDSDF exhibits significant hypolipidemic activity in vitro [15], its hypolipidemic efficacy in HFD animal models requires further investigation. Beyond its effects on diseases such as obesity and HLP, an HFD significantly affects the gut microbiota [24,25]. Specifically, changes in microbiota composition and activity are strongly linked to metabolic diseases induced by an HFD [26,27]. Research has shown that DF consumption modulates both the composition and function of the gut microbiota [28] while improving gut barrier function and host metabolism [29]. Additionally, DF supplementation has been shown to reduce weight gain and lower blood lipid levels in an HFD-induced obese mouse model [30,31]. Nevertheless, the impact of DF on the gut microbiota of obese mice remains unclear.

Given the underexplored potential of L. decastes, this study aims to establish an optimized extraction protocol for LDSDF to obtain a high-quality, bioactive product with maximal yield. Subsequently, the efficacy of LDSDF in alleviating HLP is evaluated by establishing an HFD-induced obese mouse model and implementing LDSDF intervention therapy. Furthermore, 16S rRNA technology is used to examine the diversity and species of the gut microbiota in obese mice following LDSDF intervention. Ultimately, this study aims to comprehensively evaluate the efficacy of LDSDF in preventing metabolic diseases to support the development of functional, hypolipidemic products from natural edible fungi.

2. Materials and Methods

2.1. Materials

The fruiting bodies of L. decastes were purchased from Jiangsu Gubentang Biotechnology Co., Ltd. (Taizhou, China). All chemicals and solvents were of analytical grade and were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Commercial assay kits were employed to analyze the serum lipids, hepatic enzymes, oxidative stress markers, and inflammatory cytokines according to the manufacturers’ instructions. DNA extraction and polymerase chain reaction (PCR) reagents were used for gut microbiota analysis. Detailed information, including vendor names and catalog numbers for all commercial kits and key reagents, is provided in Supplementary Table S1.

2.2. Preparation and Optimization of the Water-Extraction Process for LDSDF

2.2.1. Preparation of LDSDF

LDSDF was prepared via a two-step approach, namely hot-water extraction and ethanol precipitation. The sample powder was prepared by grinding dried L. decastes fruiting bodies, followed by crushing and sifting through a 60-mesh screen. The resulting powder (5 g) was extracted with distilled water at a liquid-to-solid ratio of 40:1 (mL/g) under magnetic stirring at 90 °C for 1 h. After centrifugation (8000× g, 20 min), the supernatant was collected, concentrated, and freeze-dried to obtain the aqueous extract, whose yield was calculated using Equation (1). The supernatant from the extraction step was then subjected to ethanol precipitation by mixing with an equal volume of 75% ethanol and standing for 12 h. The resultant precipitate was collected by centrifugation (8000× g, 20 min), redissolved in distilled water, concentrated to a ratio of 4:1, and finally freeze-dried to yield the crude LDSDF. The yield and extraction efficiency of LDSDF were calculated using Equations (2) and (3), respectively. The LDSDF content was determined according to the enzymatic hydrolysis method described in the GB 5009.88-2014 national standard, “Determination of Soluble Dietary Fiber in Foods.” [32]

$$YieldoftheaqueousextractofL.decastes\left(\%\right)={\displaystyle \frac{m_1}{m}}\times 100\%$$

(1)

$$YieldofcrudeLDSDF\left(\%\right)={\displaystyle \frac{m_2}{m}}\times 100\%$$

(2)

$$LDSDFextractionefficiency\left(\%\right)=YieldofcrudeLDSDF\left(\%\right)\times ContentofLDSDF\left(\%\right)\times 100\%$$

(3)

where m₁ is the weight of the aqueous extract of L. decastes (g), m₂ is the weight of the crude LDSDF (g), and m is the weight of the raw material (g).

2.2.2. Optimization of the Water-Extraction Process for LDSDF

Based on the extraction protocol established above, a single-factor experimental design was used to evaluate the effects of the extraction temperature (60, 70, 80, 90, and 100 °C), liquid-to-solid ratio (10:1, 20:1, 30:1, 40:1, and 50:1 mL/g), and extraction time (1, 2, 3, 4, and 5 h) on the yield of the L. decastes aqueous extract. The obtained results allowed the identification of the key parameters affecting the yield and defined their preliminary ranges.

Response surface methodology (RSM) experiments were carried out using a Box–Behnken design (BBD) implemented in Design-Expert software (version 8.0.6). The extraction temperature (A), liquid-to-solid ratio (B), and extraction time (C) were selected as independent variables. A three-factor, three-level experimental design was established, and the resulting data were subjected to regression analysis. The variables and their corresponding levels are presented in

Table 1. Box–Behnken design factors and coding levels for optimization of the water-extraction process.

Factor	Coding	Coding Levels
		−1	0	1
Temperature (°C)	A	80	90	100
Liquid-to-solid ratio (mL/g)	B	10:1	20:1	30:1
Time (h)	C	1	2	3

.

2.2.3. Optimization of the LDSDF Ethanol Precipitation Process

A single-factor experimental design was used to evaluate the effects of the concentration ratio (1:1, 2:1, 3:1, 4:1, and 5:1), ethanol concentration (30%, 45%, 60%, 75%, and 90%), and ethanol precipitation time (4, 8, 12, 16, and 20 h) on the yield, content, and extraction efficiency of LDSDF during the ethanol precipitation of L. decastes extracts. The independent variables considered for this purpose were the concentration ratio (A), ethanol concentration (B), and precipitation time (C). The conditions were optimized via RSM and are summarized in

Table 2. Box–Behnken design factors and coding levels for optimization of the ethanol precipitation process.

Factor	Coding	Coding Levels
		−1	0	1
Concentration ratio	A	3:1	4:1	5:1
Ethanol concentration (%)	B	60	75	90
Time (h)	C	12	16	20

.

2.3. Design of the Animal Experiments

In this study, 48 healthy male ICR mice (weight: 20 ± 2 g) of specific pathogen-free (SPF) status were sourced from Shanghai Zhongmo Biotechnology Co., Ltd. (Shanghai, China) under the experimental animal license number SCXK (Zhejiang) 2024-0004. High-fat and basic feed were purchased from Jiangsu Xietong Pharmaceutical Biotechnology Engineering Co., Ltd. (Nanjing, China). The animal experiments complied with ARRIVE guidelines and were approved by the related ethical regulations of Shanghai Academy of Agricultural Sciences (protocol number: SAASPZ0424110).

After a one-week acclimatization period under standard conditions, the 48 mice were randomly assigned to 6 groups (n = 8), namely the normal control (NC); HFD; simvastatin (SV); and low-, medium-, and high-dose LDSDF (SL, SM, SH) groups. The NC group received a conventional diet, whereas all other groups were fed an HFD. Mice in the SV group were administered simvastatin (3 mg/kg/day) via oral gavage. The SL, SM, and SH groups received LDSDF at doses of 100, 300, and 500 mg/kg/day, respectively, via oral gavage. Equal volumes of physiological saline were administered daily to the NC and HFD groups. All treatments were administered once daily for 6 weeks. Body weight and food intake were monitored regularly throughout the 6-week gavage period [33,34].

All animals were housed under SPF conditions, with the ambient temperature maintained at 25 ± 2 °C and relative humidity at 50 ± 5%, over a 12 h light/dark cycle. Food and water were available ad libitum for the study duration (Figure 1) [35].

Two days before terminating the experiment, fecal samples were collected in sterile centrifuge tubes. Immediately upon collection, the samples were flash-frozen in liquid nitrogen and preserved at −80 °C.

Prior to terminal blood collection, mice were fasted for 12 h with free access to water. Blood samples were collected (in sterile tubes) from the orbital sinus under anesthesia and centrifuged at 4 °C to isolate serum, which was aliquoted and preserved at −80 °C.

Following phlebotomy, mice were sacrificed via cervical dislocation. The liver and epididymal fat tissues were promptly excised and weighed. Two portions were processed for each tissue; one was fixed in 4% paraformaldehyde for histological analysis, while the other was snap-frozen in liquid nitrogen and preserved at −80 °C for subsequent detection of organ indices [36].

2.4. Analysis of Body Weight, Liver Index, and Epididymal Fat Index

The body weights of the mice were monitored weekly throughout the study, and daily food intake was recorded for each group. In the terminal week of the study, the final body weight was measured before blood collection, and the total change in body weight was calculated. Following dissection, the liver and epididymal fat tissues were excised and weighed to determine the organ-to-body weight ratios (organ coefficients).

2.5. Biochemical Analysis of Serum and Organ Tissues

Serum biochemical indicators, including the total cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C), were determined using commercially sourced assay kits. Additionally, the alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels were measured by kits to evaluate liver function. The concentrations of inflammatory cytokines—tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6), and interleukin-1β (IL-1β)—in the mouse serum were measured using enzyme-linked immunosorbent assay kits. All operations were conducted in accordance with the manufacturer’s guidelines.

Fixed liver and epididymal fat tissues were processed for histological analysis. Liver tissue was stained and analyzed using hematoxylin and eosin (H&E) and Oil Red O, while epididymal fat was analyzed by H&E staining. Histopathological changes in the stained tissue samples were observed and evaluated using an optical microscope.

To assess the antioxidant activity, a 10% (w/v) liver homogenate was prepared by homogenizing liver tissue (0.1 g) in physiological saline (0.9 mL) using a homogenizer in an ice bath. The homogenate was then centrifuged at 3500× g and 4 °C for 10 min. The superoxide dismutase (SOD), catalase (CAT), glutathione peroxidase (GSH-Px), and malondialdehyde (MDA) levels in the supernatant were determined using commercial kits.

2.6. Analysis of Gut Microbiota Composition

Total microbial DNA was extracted from mouse fecal samples using the DNAzol Genomic DNA Rapid Extraction Kit. Primers 341F and 805R were used to amplify the V3-V4 hypervariable region of the 16S rRNA gene, and sequencing libraries were created. Paired-end sequencing was performed on the Illumina MiSeq platform (www.majorbio.com). Raw sequencing data were processed using the DADA2/Deblur pipeline on the MajorBio Cloud Platform (www.majorbio.com) to generate a high-resolution table of amplicon sequence variants (ASVs). A Sobs index dilution curve was plotted to confirm sufficient sequencing depth. Significant variations in the α-diversity indices (Chao, Ace) between groups were evaluated using Kruskal–Wallis (KW) sum–rank tests. β-Diversity was evaluated based on Bray–Curtis distances using principal coordinate analysis (PCoA). The relationships between microbial community abundance and different environmental factors were examined using Spearman’s correlation. Signature microorganisms were identified in several groups using linear discriminant analysis (LDA).

2.7. Statistical Analysis

All experiments were performed in triplicate. Data are presented as the mean ± standard deviation (SD) or the mean ± standard error of the mean (SEM), as appropriate. The normality of the data distribution was assessed using the Kolmogorov–Smirnov test. For comparisons among more than two groups, a one-way analysis of variance (ANOVA) was performed, followed by Tukey’s post hoc test. Statistically significant differences between groups are indicated in the figures and tables by “*” or “^#”. All statistical analyses and data visualization were performed using GraphPad Prism (version 8.0.2) and SPSS (version 27.0).

3. Results and Discussion

3.1. Optimization of the LDSDF Extraction Method

3.1.1. Optimization of the LDSDF Water-Extraction Process

Single-factor experiments indicated that the optimal parameters for the hot-water extraction of L. decastes were an extraction temperature of 90 °C, a liquid-to-solid ratio of 20:1 mL/g, and an extraction time of 2 h (Figure 2A–C). The extraction temperature (A), liquid-to-solid ratio (B), and extraction time (C) were selected as independent variables because of their significant effects on the extract yield. For the RSM design, the yield of the aqueous extract (Y) served as the response variable.

Table 3. Response surface design and results for the hot-water extraction of L. decastes.

No.	Temperature (°C)	Liquid-to-Solid Ratio (mL/g)	Time (h)	Yield of Aqueous Extract (%)
				Actual Value	Predicted Value
1	0	0	0	67.36	67.75
2	0	0	0	68.40	67.75
3	1	1	0	68.20	67.98
4	0	1	−1	65.20	65.21
5	0	1	1	67.40	67.29
6	−1	−1	0	61.00	61.20
7	1	−1	0	60.80	60.48
8	−1	0	−1	66.20	65.88
9	0	−1	−1	63.00	63.11
10	0	0	0	68.20	67.75
11	−1	0	1	65.80	65.60
12	0	0	0	68.20	67.75
13	1	0	1	66.80	67.12
14	−1	1	0	63.10	63.42
15	0	0	0	66.60	67.75
16	1	0	−1	68.00	68.20
17	0	−1	1	59.70	59.67

presents the 17 experimental runs generated by the BBD in Design-Expert software, along with the corresponding results. A quadratic regression model (Equation (4)) was developed to relate the response variable to the independent factors.

$$\mathrm{Y}=67.75+0.96\mathrm{A}+2.43\mathrm{B}-0.34\mathrm{C}+1.32\mathrm{AB}-0.20\mathrm{AC}+1.38\mathrm{BC}-0.80{\mathrm{A}}^2-3.68{\mathrm{B}}^2-0.25{\mathrm{C}}^2$$

(4)

The developed model was highly significant (p < 0.0001,

Table 4. ANOVA for the regression model of the LDSDF water-extraction process.

Source	Sum of Squares	DF	Mean Square	F Value	p Value	Significance
Model	132.24	9	14.69	35.41	<0.0001	significant
A	7.41	1	7.41	17.86	0.0039	**
B	47.05	1	47.05	113.38	<0.0001	**
C	0.91	1	0.91	2.20	0.1819
AB	7.02	1	7.02	16.92	0.0045	**
AC	0.16	1	0.16	0.39	0.5543
BC	7.56	1	7.56	18.23	0.0037	**
A2	2.70	1	2.70	6.51	0.0380	*
B2	56.90	1	56.90	137.21	<0.0001	**
C2	0.27	1	0.27	0.64	0.4502
Residual	2.90	7	0.41
Lack of fit	0.60	3	0.20	0.35	0.7932	not significant
Pure error	2.30	4	0.58
Cor total	135.15	16
R2	0.9785
Adj R2	0.9509		C.V.%	0.98
Pred R	0.9021		Adeq Precision	17.230

* denotes a significant difference (p < 0.05), while ** represents an extremely significant difference (p < 0.01).

). Moreover, the lack-of-fit test was not significant (p = 0.7932), indicating that the model adequately describes the experimental data and is suitable for predicting aqueous extract yield. The coefficient of determination (R² = 0.9785) implied that 97.85% of the total variation was explained by the model. The adjusted R² value of 0.9509 further confirmed a strong agreement between the predicted and actual values, with minimal experimental error. A low coefficient of variation (C.V. = 0.98%) confirmed high model reproducibility, and an adequate precision value of 17.230 (>4) indicated a strong signal-to-noise ratio and good model discriminability.

The relative influence of the extraction parameters on the yield decreased according to the following order: liquid-to-solid ratio (B) > extraction temperature (A) > extraction time (C). The quadratic term for temperature (A²) was significant (p < 0.05), while AB, BC, and B² had p-values < 0.01, indicating that the solid-to-liquid ratio significantly affected the aqueous extract yield.

Contour line shapes can be used to interpret the interaction effects between response variables. Elliptical contours indicate significant interaction between two variables (p < 0.05), whereas circular contours suggest no significant interaction effect on the aqueous extract yield (p > 0.05). Additionally, a steeper response surface slope implies a stronger interaction influence of the corresponding factors on the output. Contour plot analysis revealed elliptical profiles for the interactions between A and B, as well as between B and C, suggesting significant interaction effects. By contrast, the interaction between A and C was represented by a circular contour, indicating negligible interaction. These results demonstrate that the extraction temperature (A), liquid-to-solid ratio (B), and extraction time (C) significantly influence the extraction yield (Figure 2D–I). Furthermore, the steeper response surfaces for the AB and BC interactions compared with that of the AC interaction confirm that the combined influence of temperature, liquid-to-solid ratio, and liquid-to-solid ratio over time has a substantial impact on the aqueous extract yield.

The optimal extraction parameters determined through RSM were an extraction temperature of 99.38 °C, a liquid-to-solid ratio of 25.84:1 mL/g, and an extraction time of 1.93 h, giving an aqueous extract yield of 68.82%. For practical application, the parameters were adjusted to 99 °C, 25:1 mL/g, and 1.9 h, respectively. The corresponding validation experiments performed in triplicate gave an average yield of 68.40%, confirming the reliability of the established extraction conditions. Similar to the hot-water extraction of SDF from Agrocybe cylindracea, higher temperatures and ratios enhanced SDF solubility, likely through increased cell wall disruption [37].

3.1.2. Optimization of the LDSDF Ethanol Precipitation Process

Single-factor experiments indicated that the optimal conditions for LDSDF ethanol precipitation were a concentration ratio of 4, an ethanol concentration of 75%, and a precipitation time of 16 h. According to the data presented in

Table 5. Response surface design and LDSDF extraction efficiency.

No.	Concentration Ratio	Ethanol Concentration (%)	Time (h)	Extraction Rate of LDSDF (%)
				Actual Value	Predicted Value
1	−1	−1	0	7.29	7.58
2	1	−1	0	8.16	8.08
3	−1	1	0	7.88	7.96
4	1	1	0	5.92	5.62
5	−1	0	−1	12.49	12.25
6	1	0	−1	10.04	10.17
7	−1	0	1	11.82	11.69
8	1	0	1	11.69	11.93
9	0	−1	−1	8.78	8.74
10	0	1	−1	7.92	8.08
11	0	−1	1	9.89	9.72
12	0	1	1	8.26	8.30
13	0	0	0	16.16	15.87
14	0	0	0	16.08	15.87
15	0	0	0	15.49	15.87
16	0	0	0	15.92	15.87
17	0	0	0	15.68	15.87

, the response of the LDSDF extraction yield (Y) to the three factors—concentration ratio (A), ethanol concentration (B), and precipitation time (C)—was fitted to a second-order polynomial equation (Equation (5)).

$$\mathrm{Y}=15.87-0.46\mathrm{A}-0.52\mathrm{B}+0.3\mathrm{C}-0.71\mathrm{AB}+0.58\mathrm{AC}-0.19\mathrm{BC}-2.88{\mathrm{A}}^2-5.68{\mathrm{B}}^2-1.48{\mathrm{C}}^2$$

(5)

As outlined in

Table 6. ANOVA for the regression model of the LDSDF ethanol precipitation process.

Source	Sum of Squares	DF	Mean Square	F Value	p Value	Significance
Model	202.45	9	22.49	223.53	<0.0001	significant
A	1.68	1	1.68	16.73	0.0046	**
B	2.14	1	2.14	21.29	0.0024	**
C	0.74	1	0.74	7.33	0.0303	*
AB	2.00	1	2.00	19.90	0.0029	**
AC	1.35	1	1.35	13.37	0.0081	**
BC	0.15	1	0.15	1.47	0.2642
A2	34.88	1	34.88	346.55	<0.0001	**
B2	135.63	1	135.63	1347.71	<0.0001	**
C2	9.20	1	9.20	91.40	<0.0001	**
Residual	0.70	7	0.10
Lack of fit	0.39	3	0.13	1.69	0.7932	not significant
Pure error	0.31	4	0.078
Cor total	203.16	16
R2	0.9965
Adj R2	0.9921		C.V.%	2.85
Pred R	0.9666		Adeq Precision	42.076

* denotes a significant difference (p < 0.05), while ** represents an extremely significant difference (p < 0.01).

the lack of fit was not significant (p > 0.05), whereas the regression model was extremely significant (p < 0.0001), indicating that the model correlates well with the experimental data and is suitable for predicting and optimizing the ethanol precipitation parameters to enhance the LDSDF extraction yield. The coefficient of determination (R²) was 0.9965, and the adjusted R² was 0.9921. The proximity of both values to 1 demonstrated that the model exhibited high accuracy and reliability. Among the linear terms, the concentration ratio (A), ethanol concentration (B), and precipitation time (C) had the most significant effects on the result. The quadratic terms (A², B², and C²) and interaction terms (AB and AC) were also significant. Consequently, the factors influencing the extraction yield were ranked as follows: ethanol concentration (B) > concentration ratio (A) > precipitation time (C).

Response surface and contour plot analyses revealed that the two-factor interactions (AB and AC) had a more pronounced effect on the LDSDF extraction yield than the other interactions (Figure 3). Optimization using the design software identified an optimal concentration ratio of 3.94, an ethanol concentration of 74.36%, and a precipitation time of 16.37 h, yielding a predicted extraction efficiency of 15.91%. To ensure operational feasibility, these conditions were adjusted to a concentration ratio of 3.9:1, an ethanol concentration of 74.4%, and a precipitation time of 16.4 h. Three independent validation experiments produced an average extraction efficiency of 15.83%, closely matching the predicted value. These results demonstrate that the response surface model can accurately predict the LDSDF extraction yield. Among the investigated factors, ethanol concentration was identified as the primary determinant of precipitation efficiency, consistent with previous findings for materials such as Flammulina velutipes root dietary fiber [38].

3.2. Effects of LDSDF on the Body Weight, Liver Index, and Epididymal Fat Index of High-Fat Mice

The change in body weight served as an indicator of HFD-induced obesity in the experimental mice [39]. After a 1-week adaptation period, the average initial body weights were comparable across all groups (p > 0.05). Throughout the 6-week HFD feeding period, the HFD group exhibited significantly greater body weight gain than all other groups (p < 0.05), whereas the SV and sample treatment groups (SL, SM, and SH) showed more gradual weight increases. Final body weight was significantly higher in the HFD group (44.56 ± 4.93 g) than in the NC group (29.88 ± 1.42 g) following the feeding period (p < 0.05). A significant decrease was observed in the final weights of the sample groups relative to the HFD group (p < 0.05), indicating that LDSDF administration partially inhibited HFD-induced weight gain (Figure 4A). Although all HFD-fed groups (HFD, SL, SM, SH, and SV) gained significantly more weight than the NC group during the study period (p < 0.05), the SL, SM, and SH groups exhibited significantly less weight gain than the HFD group, further supporting the inhibitory effect of LDSDF gavage on obesity development (Figure 4C). Moreover, neither HFD nor SDF intervention significantly affected food intake, with no differences observed between groups (p > 0.05; Figure 4D). The magnitude of weight reduction achieved with LDSDF is consistent with findings for Ganoderma DF, which curbed weight gain by 15–20% in HFD mice, a phenomenon attributed to modulated energy metabolism [40].

Extended HFD consumption in mice resulted in augmented epididymal fat tissue bulk and hepatic lipid deposition [41]. The HFD group exhibited significant increases in liver and epididymal fat weights and their corresponding coefficients relative to the NC group (p < 0.05,

Table 7. Effects of LDSDF on organ weight and coefficients in high-fat mice.

Group	Liver Weight (g)	Liver Coefficient (%)	Epididymal Fat Weight (g)	Epididymal Fat Coefficient (%)
NC	0.87 ± 0.153 d	3.08 ± 0.477 b	0.57 ± 0.058 d	2.02 ± 0.154 e
HFD	1.63 ± 0.153 a	4.05 ± 0.068 a	1.80 ± 0.100 a	4.48 ± 0.229 a
SL	1.10 ± 0.100 bc	3.20 ± 0.114 b	1.00 ± 0.100 b	2.91 ± 0.104 b
SM	1.33 ± 0.153 b	3.53 ± 0.234 b	1.37 ± 0.115 b	3.63 ± 0.299 bc
SH	1.37 ± 0.115 cd	3.52 ± 0.216 b	1.33 ± 0.058 c	3.44 ± 0.123 cd
SV	1.10 ± 0.100 cd	3.15 ± 0.093 b	1.10 ± 0.100 c	3.15 ± 0.093 d

Different lowercase letters indicate significant differences among groups (p < 0.05).

). In contrast, LDSDF supplementation significantly reduced these coefficients in HFD-fed mice (p < 0.05), indicating that LDSDF intervention attenuated diet-induced increases in organ weights and restored values to levels observed in normal animals (Figure 4E,F). These findings are consistent with previous reports showing that mushroom DF alleviates visceral fat hypertrophy [42].

3.3. Effect of LDSDF on the Blood Biochemical Parameters of High-Fat Mice

3.3.1. Effect of LDSDF on the Blood Lipid Level

After 6 weeks of HFD feeding, the mice exhibited significant dyslipidemia, characterized by elevated serum levels of TC, TG, and LDL-C (1.51-, 3.21-, and 3.05-fold increases, respectively), coupled with a reduction in HDL-C (0.53-fold decrease) relative to the NC group (Figure 5A–D).

LDSDF intervention effectively ameliorated this condition in a dose-dependent manner. Specifically, the SL, SM, and SH treatment groups exhibited progressively reduced atherogenic lipid levels (TC, TG, and LDL-C) and increased HDL-C compared with the HFD group. The SH group showed the most substantial improvement among the treatment groups, with 0.65-, 0.44-, and 0.37-fold reductions in TC, TG, and LDL-C, respectively, along with a 1.28-fold increase in HDL-C. The SV (positive control) group also demonstrated a strong lipid-lowering effect with a notable 1.36-fold increase in HDL-C (Figure 5A–D). In line with previous findings on bioactive fungal polysaccharides, the present study further demonstrates that LDSDF intervention effectively ameliorated diet-induced dyslipidemia in a dose-dependent manner, with higher doses yielding more pronounced lipid-lowering effects [40].

3.3.2. Effect of LDSDF on Liver Function in High-Fat Mice

AST and ALT are widely used biomarkers for assessing hepatic injury, with elevated levels of these enzymes indicating impaired liver function [43]. A significant increase in ALT and AST levels was observed in the HFD group compared with the NC (p < 0.05), indicating substantial diet-induced liver damage. In contrast, ALT and AST levels were significantly lowered in the SV and LDSDF treatment groups, with the greatest reduction being observed in the SH group (Figure 5E,F). These findings demonstrate that LDSDF effectively attenuates hepatic injury in obese mice by reducing serum transaminase levels in a dose-dependent manner. This protective effect is consistent with reports of Ganoderma DF improving hepatic function [40].

Obesity is frequently associated with systemic inflammation, as demonstrated in previous studies [44,45]. To evaluate this in greater detail, the effects of LDSDF on serum levels of pro-inflammatory cytokines (TNF-α, IL-6, and IL-1β) were examined in HFD-fed mice. Relative to the NC group, HFD induced a pronounced systemic inflammatory response, characterized by significant elevations in the serum levels. However, the administration of LDSDF at varying doses markedly reduced the levels of these inflammatory markers in HFD mice. The SH group demonstrated the most substantial reductions, namely 0.35-, 0.18-, and 0.50-fold for TNF-α, IL-6, and IL-1β, respectively (Figure 5G–I). Consistent with the finding that DF from other edible fungi reduces inflammatory markers in obese mice, these results further confirm that LDSDF effectively attenuates hepatic injury in obese mice [46].

3.4. Effect of LDSDF on the Organ Tissue Morphology and Biochemical Indexes of High-Fat Mice

Obesity is commonly associated with excessive lipid accumulation and aberrant hepatic lipid metabolism, often leading to fatty liver disease [47]. In the NC group, hepatocytes were compact, uniformly stained, and devoid of lipid vacuoles or inflammatory infiltrates. In contrast, prominent macrovesicular steatosis, characterized by numerous lipid vacuoles within the cytoplasm, was observed in the HFD group. LDSDF treatment markedly attenuated hepatic steatosis in the obese mice, restoring cellular architecture and reducing lipid droplet accumulation, with the SH group showing the most pronounced improvement (Figure 6A).

The progression of obesity is accompanied by the expansion of white adipose tissue. Adipocytes in the HFD group were markedly hypertrophied relative to those in the NC group, with fewer cells per field of view, indicating cellular enlargement. LDSDF treatment reduced adipocyte size across all treatment groups in a dose-dependent manner, with the SH group exhibiting cell sizes closest to those of the NC group (Figure 6B).

Substantial lipid accumulation and aggregation were observed in the HFD group compared with the NC group, confirming HFD-induced hepatic steatosis [48]. LDSDF administration reduced hepatic lipid deposition in a dose-dependent manner (Figure 6C), indicating that LDSDF supplementation effectively alleviates hepatic steatosis and decreases intracellular lipid accumulation.

Hyperlipidemia is often accompanied by hepatocyte damage resulting from lipid peroxidation, which induces inflammation and compromises liver function [49]. The HFD significantly induced hepatic oxidative stress, as evidenced by significantly increased MDA levels and decreased activities of the antioxidant enzymes SOD, CAT, and GSH-Px relative to the NC group (p < 0.05). LDSDF administration markedly attenuated HFD-induced oxidative damage, as evidenced by increased enzyme activities and decreased MDA levels compared with the HFD group (p < 0.05), with the most pronounced effects observed in the SH group (Figure 6D–G). Hyperlipidemia is a major contributor to hepatic lesions and lipid accretion in mice, whereas SDF from edible fungi mitigates dyslipidemia and its associated pathologies. The present study provides further evidence supporting this beneficial role of edible fungal SDF [46,50].

3.5. Influence of LDSDF on the Gut Microbiota Composition of High-Fat Mice

According to previous studies, 16S rRNA sequencing is crucial for investigating the gut microbiota, as it enables the characterization of microbial diversity, composition, and functional potential [31,51,52]. As shown in Figure 7A, after defining an optimal sequencing depth, the curve stabilized, indicating that the current depth sufficiently captured species diversity within the samples, validating the sequencing strategy. α-Diversity, which reflects the species richness and evenness in individual samples, was used to evaluate the abundance and homogeneity of microbial communities. Considering the Ace and Chao indices, the HFD group exhibited a marked reduction in microbial richness compared with the NC group after the feeding period, indicating a significant HFD-induced decline in overall gut microbiota diversity (p < 0.05). In contrast, the SL, SM, SH, and SV groups showed increased microbial richness and evenness compared with the HFD group, with more pronounced intervention effects observed in the SM and SH groups (p < 0.05) (Figure 7B,C).

β-Diversity reflects compositional differences among microbial communities across different samples. PCoA revealed that principal coordinate 1 (PC1) and principal coordinate 2 (PC2) explained 22.99% and 17.32% of the variation, respectively, with a cumulative contribution of 40.31%, indicating that these two axes capture a substantial portion of the total variance. Moreover, distinct clustering of the NC and HFD groups indicated marked differences in gut microbial composition. After treatment with LDSDF and simvastatin, the SL, SM, SH, and SV groups exhibited partial separation from the HFD group, implying that these interventions partially altered the gut microbiota in obese mice (Figure 7D).

Analysis of the 16S rRNA sequencing data revealed the phylum-level gut microbiota composition [53], wherein the dominant phyla were Campylobacteriota, Thermodesulfobacteriota, Bacteroidota, and Firmicutes. Under HFD conditions, the relative abundance of Bacteroidota decreased, while that of Firmicutes and Thermodesulfobacteriota increased. Notably, elevated levels of Firmicutes and reduced levels of Bacteroidota have been consistently associated with obesity [54,55]. These shifts may have functional implications; a higher abundance of Firmicutes may promote fat absorption, whereas a reduction in Bacteroidota may compromise carbohydrate metabolism [56]. Accordingly, HFD-induced dysbiosis has been reflected by an elevated Firmicutes/Bacteroidota (F/B) ratio (Figure 8A) [57]. Consistent with these reports, the F/B ratio was significantly higher (p < 0.05) in the HFD group than in the NC group. Treatment of HFD mice with LDSDF or SV significantly counteracted this change, resulting in considerably lower F/B ratios in the SL, SM, SH, and SV groups (p < 0.05) (Figure 8C).

The predominant bacterial families identified in the gut microbiota of these mice were Lachnospiraceae, Erysipelotrichaceae, Desulfovibrionaceae, norank_p_Bacteroidota, Lactobacillaceae, Peptostreptococcaceae, and Bacteroidaceae. Compared with the NC group, the HFD group exhibited a higher abundance of Lachnospiraceae, Erysipelotrichaceae, and Lactobacillaceae, while the abundance of norank_p_Bacteroidota and Bacteroidaceae was reduced. Previous studies have shown that HFD feeding induces advanced hepatic steatosis, liver inflammation, and lipid accumulation in mice. These pathogenic alterations have been linked to higher abundances of Lactobacillaceae and Lachnospiraceae, leading to the synthesis of secondary bile acids [58].

Bacteroidaceae, which are recognized as beneficial gut microbes, play a key role in degrading complex carbohydrates into short-chain fatty acids (SCFAs). Importantly, SCFAs supply essential energy for intestinal cells and help maintain intestinal barrier integrity [59]. Supplementation with LDSDF significantly altered the gut microbiota in high-fat mice, reducing Erysipelotrichaceae and Lactobacillaceae levels while increasing norank_p_Bacteroidota and Bacteroidaceae (Figure 8B). Furthermore, Oscillospiraceae, whose abundance is inversely correlated with obesity and type 2 diabetes mellitus [60], were markedly elevated by LDSDF treatment. Overall, HFD induced gut microbiota dysbiosis in mice, characterized by elevated levels of harmful bacteria and reduced levels of beneficial bacteria at both the phylum and family levels. LDSDF administration effectively alleviated this diet-induced dysbiosis.

LDA is a commonly used statistical technique for classification and feature selection. In the study of mouse gut microbiota, LDA effect size (LEfSe) analysis is commonly applied to characterize the microbial taxa that differ significantly across sample groups [61,62]. In fecal samples from the NC group, several taxa showed significant variations, including Bacteroidota, Bacteroidia, Bacteroidales, Bacteroidaceae, and Bacteroides. Other differentially abundant taxa included Muribaculaceae, Pseudomonadota, Parasutterella, Sutterellaceae, and Burkholderiales. HFD feeding resulted in distinct microbial compositions characterized by higher abundances of Firmicutes, Ligilactobacillus, Actinomycetota, Coriobacteriia, Coriobacteriales, Leptogranulimonas, Atopobiaceae, Odoribacteraceae, Actinomycetes, and Bifidobacteriaceae. This represents a classic manifestation of HFD-induced dysbiosis. While the NC group was rich in several health-associated Bacteroidetes-related phyla, the HFD group was characterized by a significant overall increase in Firmicutes abundance, leading to an elevated F/B ratio, a core feature of obesity-associated dysbiosis. Following LDSDF intervention, the SL group showed differential abundances of Firmicutes and Massiliimalia, whereas the SM group was enriched in Allobaculum and Coprobacillaceae. However, no significant effects were observed for these two groups; the SM group contained only two beneficial genera (Allobaculum and Coprobacillaceae), and the SL group exhibited a comparable Firmicutes-dominant profile to the HFD group. In contrast, the SH group displayed a distinct fecal microbiota profile dominated by Faecalibacterium, Bacteroidaceae, Bacteroides, and Romboutsia. The community structure in the SH group was distinct from the HFD group, changing from Firmicutes to Bacteroidaceae and Bacteroides. Furthermore, beneficial butyrate-producing bacteria, including Romboutsia and Faecalibacterium, were observed in the SH group. These changes indicate that high-dose LDSDF successfully counteracted the negative effects of an HFD, restoring a more complete microbial community structure (Figure 9).

Previous reports have demonstrated that the phylum Firmicutes is implicated in both health and disease, with an increased F/B ratio being correlated with human obesity [63]. Additionally, a strong negative correlation has been observed between Muribaculaceae abundance and obesity risk, indicating that high-fiber diets may alleviate obesity by increasing its prevalence [64]. Furthermore, supplementation with Romboutsia has been demonstrated to enhance endothelial function in obese mice by regulating the gut microbiota and lipid metabolism [65]. The degradation of dietary polysaccharides by gut microbiota, including Bacteroidota, Muribaculaceae, Faecalibaculum, and Romboutsia, generates SCFAs [66], which support intestinal function and promote gut health. Specifically, butyrate from Faecalibaculum activates GPR41/43 receptors, suppressing hepatic SREBP-1c expression and cholesterol synthesis, while propionate from Bacteroidota inhibits ACC activity, reducing fatty acid synthesis [67].

Consistent with these findings, an HFD disturbed the gut microbiota equilibrium in mice. Dietary intervention with LDSDF partially reversed the HFD-induced gut microbial dysbiosis in both experimental groups; however, microbiota modulation was dependent on the LDSDF dosage. Specifically, a significant increase in the abundance of beneficial bacteria (Allobaculum and Coprobacillaceae) was observed in the SM group, while the SH group demonstrated markedly greater abundance of Faecalibaculum, Bacteroidota, and Romboutsia. These microbiota alterations likely regulate lipid metabolism through microbial metabolites, consistent with dietary fiber studies where SCFA-producing Bacteroidota/Faecalibaculum enrichment inhibits hepatic lipogenesis via PPARγ/AMPK activation and reduces energy harvest from Firmicutes [58,68]. Reduced F/B ratio and secondary bile acid-producing families (Lachnospiraceae) may further limit intestinal lipid absorption and cholesterol reabsorption [69]. Thus, LDSDF modulates the microbiota-metabolite axis to alleviate HFD-induced hyperlipidemia.

4. Conclusions

In the study, single-factor experiments and RSM were used to establish a robust and efficient protocol for the extraction and purification of SDF from L. decastes. The close agreement between experimental yields and model predictions under optimized conditions validated the reliability of the established methodology. This optimized protocol provided a solid foundation for the production of high-quality LDSDF for subsequent biological evaluations.

The biological efficacy of LDSDF was assessed in an HFD-induced obese mouse model. LDSDF supplementation alleviated HFD-induced metabolic disturbances by improving lipid profiles, reducing adiposity, attenuating inflammation and oxidative stress, and mitigating liver injury. These systemic benefits were associated with significant restructuring of the gut microbiota, characterized by increased diversity and enrichment of beneficial bacteria.

Overall, the findings highlight the significant potential of LDSDF as a novel prebiotic dietary fiber. The combination of high extraction yield and confirmed lipid-lowering effects highlights its practical and economic potential for use in functional foods or nutraceuticals targeting obesity and related metabolic disorders.

While the association between the lipid-lowering efficacy of LDSDF and the restructuring of gut microbiota is clear, the precise molecular mechanisms underlying this modulation, as well as the resulting changes in key functional metabolites (e.g., SCFAs, bile acids), require further investigation.