Research on the ultrasonic-assisted extraction, purification, structural characterization, and anti- glycolipid metabolic disorder (anti-GLMD) activity and mechanism of polysaccharide from Dryopteridis crassirhizomatis Rhizoma

Graphical abstract

Abstract

Dryopteridis crassirhizomatis Rhizoma, the traditional Chinese medicinal herb commonly known as “Mianma Guanzhong”, botanically originating from the dried rhizome of Dryopteris crassirhizoma Nakai, exhibits extensive pharmacological activities. To further elaborate on its medicinal potential, a homogeneous polysaccharide, designated as DCP-1-2, was ultrasonic-assisted extracted and purified from the dried rhizomes of Dryopteris crassirhizoma. Subsequently, comprehensive structural elucidation of DCP-1-2 was conducted using a combination of advanced analytical techniques, including HPLC, SERS, methylation analysis, and NMR spectroscopy. Notably, systematic in vitro assessments were performed to evaluate two key biological activities of DCP-1-2: its antioxidant capacity and its inhibitory effects on key enzymes implicated in glucose and lipid metabolism. To this end, as a critical crosslink between glucose and lipid metabolic pathways, AGEs were employed to establish a high-fat cell model. This cell model was then utilized to investigate the anti-GLMD activity of DCP-1-2 and to decipher the underlying molecular mechanisms. Collectively, these findings lay a solid experimental foundation for the potential application of Dryopteris crassirhizoma-derived polysaccharides in the prevention and management of GLMD.

1. Introduction

GLMD (A full list of abbreviations used throughout this manuscript is provided in the Supplementary Materials for reference.) is characterized by abnormalities in glucose and lipid metabolism [[1], [2]]. The resulting oxidative stress and inflammatory responses can exacerbate pancreatic β-cell dysfunction and promote the accumulation of AGEs[3]. In turn, AGEs further aggravate GLMD, creating a vicious cycle that triggers complications such as obesity, T2DM and CVD [[4], [5]]. Long-time use of commonly prescribed clinical drugs such as biguanides and statins often induces side effects, including liver and kidney dysfunction or muscle injury, highlighting the urgent need for safe natural therapeutic agents[6]. Thus, clarifying the relationship between GLMD and key regulatory pathways remains critical. The insulin receptor-mediated IRS1-PI3k-Akt pathway plays a central role in ameliorating GLMD. Insulin binding to the insulin receptor activates this pathway, promoting cellular glucose uptake and glycogen synthesis, while also regulating lipolysis and lipogenesis, directly improving insulin resistance-related GLMD [[7], [8]]. The SREBP-1c-PPAR-Fas pathway, activated by ligands of the peroxisome proliferator-activated receptor family binding to PPAR receptors, ameliorates dyslipidemia and glucose dysregulation in GLMD by inhibiting fatty acid β-oxidation, enhancing insulin sensitivity, and promoting energy metabolism[[9], [10]]. Conversely, activation of the NF-κB-JUN pathway could inhibit the IRS1-PI3k-Akt pathway through inflammatory cytokines, disrupts lipid metabolism gene expression, and exacerbates chronic inflammation and the vicious cycle of GLMD, becoming an important driver of GLMD progression[[11], [12]].

Research has shown that natural polysaccharides, as a class of bioactive macromolecules with wide-ranging activities, possess significant advantages such as low toxicity, multi-target effects, and good biocompatibility [[13], [14]]. These properties allow them to improve insulin resistance, inhibit fat synthesis, and enhance antioxidant capacity, making them a research hotspot in the field of anti-glycolipid metabolic disorder [[15], [16]]. Dryopteridis crassirhizomatis Rhizoma (Mianma Guanzhong), a traditional Chinese medicinal herb from the Dryopteridaceae family listed in the Chinese Pharmacopoeia, has been previously reported to exhibit various pharmacological activities, including antioxidant, anti-inflammatory, and antitumor effects[17]. Currently, both domestic and international research on its constituents has primarily focused on resorcinol derivatives, terpenes, and flavonoids, whereas polysaccharides remain relatively understudied[18]. Notably, there are no studies reporting the anti-GLMD activity and underlying mechanisms of its purified polysaccharides.

In this work, the optimal ultrasonic-assisted extraction conditions for water-soluble neutral polysaccharides from Dryopteridis crassirhizomatis Rhizoma were determined via single-factor, orthogonal, and response surface methodology experiments, and the purified polysaccharide fraction was characterized by HPLC, SERS, methylation analysis, and 1D/2D NMR spectroscopy, which revealed an arabinose-glucan backbone. The effects of DCP-1-2 on oxidative stress were examined in vitro, and its impacts on key enzymes mediating glycolipid metabolism were also evaluated. Subsequently, a high-fat HepG2 cell model was constructed using AGEs, and lipid contents (TC, TG, HDL-C, and LDL-C) were determined to verify the in vitro efficacy of DCP-1-2.  Its regulatory impact on oxidative stress levels was further assessed by measuring fluctuations in MDA and SOD contents. Finally, RT-qPCR was conducted to explore the anti-GLMD mechanisms of DCP-1-2.

2. Materials and methods

2.1. Materials and reagents

Dryopteridis crassirhizomatis Rhizoma (Batch Number: 220701) was purchased from Tongrentang (Beijing, China).DEAE-52 cellulose filler was obtained from Solarbio Biochemical Co., Ltd (Beijing, China), while Amberlite FPA 90Cl and Amberlite FPC 3500H were purchased from Macklin (Shanghai, China).The monosaccharides reference standards (L-arabinose, D-mannose, L-rhamnose, D-glucose, D-galactose, D-glucuronic acid, and D-galacturonic acid) were purchased from Aladdin Biochemical Co., Ltd. (Shanghai, China). High-glucose DMEM was purchased from Gibco (Waltham, MA, USA). Additionally, assay kits for TC, TG, HDL-C, LDL-C, SOD, and MDA were procured from Nanjing Institute of Bioengineering (Nanjing, Jiangsu, China). TRIzol reagent was purchased from Thermo Fisher Scientific (Waltham, MA, USA), the reverse transcription kit was purchased from ABclonal Technology (Wuhan, Hubei, China), and the Oil Red O Staining Kit was purchased from Solarbio Biochemical Co., Ltd. (Beijing, China). All other chemical reagents were of analytical grade or chromatographic grade.

2.2. Preparation and process optimization of DCP

2.2.1. Single-factor experimental single esign for DCP extraction process

The effects of ultrasonic extraction temperature (50, 60, 70, 80 °C), ultrasonic power (160, 180, 200, 220 W) and ultrasonic extraction time (30, 40, 50, 60 min) on the extraction rate of polysaccharides were studied.

2.2.2. RSM-BBD for optimizing DCP extraction process

Based on the results of single-factor experiments, the optimal extraction parameters were preliminarily identified. Subsequently, Response Surface Methodology coupled with Box-Behnken Design was adopted to further refine the extraction process, using the yield of DCP as the response variable. A three-factor, three-level experimental model was constructed, which comprised 17 experimental runs in total, including five replicates at the central point for evaluating the model’s goodness-of-fit within the central region. The coded values and corresponding levels of the independent variables are listed in Table 1.

Table 1.

Factors and levels of response surface method.

X1 Ultrasonic temperature (℃)	X2 Ultrasonic power (W)	X3 Ultrasonic time (min)
50	180	40
60	200	50
70	220	60

2.2.3. Extraction, separation and purification of DCP

After determining the optimal parameters for ultrasonic-assisted extraction, crude polysaccharide (abbreviated as cDCP) was isolated from dried Dryopteridis crassirhizomatis Rhizoma powder through extraction with deionized water, followed by sequential purification steps: concentration, Sevag method-based deproteinization, ethanol precipitation, and freeze-drying. Subsequently, cDCP was eluted through serially connected Amberlite FPC 3500H and Amberlite FPA 90Cl macroporous resin columns. The main component DCP-1 was further purified using DEAE-52 column chromatography, yielding the primary fraction DCP-1-2.

2.3. Structural and compositional analysis of DCP-1-2

2.3.1. Purity and molecular weight determination

The purity and molecular weight of DCP-1-2 were determined by HPGPC[19]. Accurately weighed amounts of DCP-1-2 and dextran standards were dissolved to prepare 1 mg/mL solutions. The analysis and separation were performed on a Waters 2695 instrument (Waters, USA) equipped with a 2414 refractive index detector and an Ultra hydrogel Linear 2000 column (12 μm, 7.8 × 300 mm). Chromatography was performed using ultrapure water as the mobile phase at a flow rate of 1 mL/min and temperature of 25 °C. The homogeneity of DCP-1-2 as a polysaccharide was confirmed by the presence of a single, symmetrical chromatographic peak.

2.3.2. UV–Vis and FT-IR spectroscopic detection

The UV–Vis spectrum (200–800 nm) and FT-IR spectrum (4000–400 cm⁻¹) of DCP-1-2 were obtained using a UV-2550 spectrophotometer and an FTIR-650 spectrometer (Shimadzu, Japan), respectively.

2.3.3. SERS analysis

Surface-enhanced Raman scattering (SERS) spectra of solid DCP-1-2 were recorded with a WITec Alpha 300 R confocal Raman microscope (WITec, Germany) using a silver plate as SERS-active substrate (enhancement substrate). Measurements were then performed with a 633 nm laser excitation source at a power of 20 mW and an integration time of 15 s.

2.3.4. Monosaccharide composition analysis

The monosaccharide composition of DCP-1-2 was determined according to a previously described method with modifications[20]. Briefly, polysaccharide samples were acid-hydrolyzed to cleave into their constituent monosaccharides, which were subsequently subjected to pre-column derivatization. The derivatives were analyzed using an Agilent 1260 Infinity II HPLC system (Agilent, USA) equipped with an Agilent Infinity Lab Poroshell 120 EC-C18 column. The separation was performed under isocratic elution with a mobile phase of acetonitrile and phosphate buffer (17.5:82.5, v/v) at a flow rate of 1.0 mL/min. Throughout the analysis, the column temperature was maintained at 25 °C, and the injection volume was fixed at 5 μL.

2.3.5. Methylation and GC–MS analysis

The glycosyl linkage analysis of DCP-1-2 was performed using the methylation method as previously described[21]. Briefly, the polysaccharide sample was first methylated. The methylation process typically involved protecting the glycosidic bonds by replacing the hydroxyl groups in the sugar molecules with methyl groups. Then, the methylated sample was subjected to acid hydrolysis, followed by reduction with NaBD₄ and acetylation with acetic anhydride to generate partially methylated alditol acetates (PMAAs). The resulting PMAAs were analyzed by an Agilent 7890B-7000D GC–MS (Agilent, USA) equipped with a capillary column (HP-5MS, 30 m × 0.25 mm). The analytical conditions were set as follows: Both the injection port and ion source were maintained at 250 °C. The column temperature program started at 160 °C, increased to 240 °C at a rate of 2 °C/min, held for 20 min, and then rose to 250 °C at a rate of 10 °C/min. Helium served as the carrier gas at a flow rate of 1.2 mL/min.

2.3.6. NMR characterization

For NMR characterization of DCP-1-2, the sample was first prepared by dissolving 50 mg of the polysaccharide in D₂O. The NMR measurements were then performed with a Bruker AV NEO 600 MHz NMR spectrometer (Bruker, Germany). NMR experiments were performed, encompassing one-dimensional (¹H NMR, ¹³C NMR) and two-dimensional (¹H–¹³C HSQC, ¹H–¹H COSY, ¹H–¹³C HMBC) spectroscopic analyses.

2.3.7. Congo red assay

The Congo red assay was employed to probe the potential triple-helix structure of DCP-1-2. The experimental procedure involved incubating the DCP-1-2 solution with Congo red, followed by the sequential addition of NaOH solution (0–0.5 mol/L). The maximum absorption wavelength of each resulting mixture was measured within the 200–800 nm range.

2.3.8. SEM morphological analysis

The morphological characteristics of the polysaccharide powder were examined with a Hitachi S-3400 N scanning electron microscope SEM (Hitachi, Japan). After gold sputtering, samples were imaged at 1000×, 4000×, 10000 × and 40000 × magnification for microstructural analysis.

2.3.9. Content determination

The contents of total sugar, glucuronic acid, and protein in DCP-1-2 were determined using the phenol–sulfuric acid method [22], the sulfuric acid-carbazole method[23], and the Bradford method [24], respectively.

2.3.10. Particle size and Zeta potential determination

The average particle size and Zeta potential of DCP-1-2 in aqueous solution were determined at 25 °C using a Malvern Nano ZS90 laser particle size analyzer (Malvern Analytical, UK).

2.3.11. Thermogravimetric analysis

TGA of DCP-1-2 was performed according to the method described by Guan et al[21]. Specifically, a 10 mg sample was placed in the sample pan of an STA 449 F5 Jupiter thermal analyzer (Netzsch, Germany). The temperature was programmed to increase from 30 to 600 °C at a rate of 10 °C/min. The resulting TGA and DTG curves were obtained to monitor the mass loss during different temperature stages.

2.3.12. Rheological analysis

The rheological analysis of DCP-1-2 was conducted using a TA Instruments Discovery HR-1 rheometer (New Castle, USA) according to the method described by Yang et al[25]. A 10 mg/mL sample solution was prepared for testing. The apparent viscosity was measured across a shear rate range of 1 to 100 s⁻¹. Additionally, the dynamic viscoelastic properties were determined within the linear viscoelastic region by scanning angular frequencies from 1 to 100 rad/s.

2.4. Anti-GLMD activity and mechanism of DCP-1-2 in vitro

2.4.1. Cell culture

HepG2 cells were provided by the Department of Pharmacology, Harbin Medical University. The cells were cultured in high-glucose DMEM supplemented with 10% FBS and 1% penicillin–streptomycin at 37 °C in a humidified atmosphere containing 5% CO₂, and HepG2 cells in a good growth state were used for subsequent experiments.

2.4.2. Preparation of AGEs

AGEs were prepared following the method described by Guo et al [26] . A 0.5 M glucose solution and a 20 g/L BSA solution, both in 0.2 M phosphate buffer (pH 7.4) containing 0.02% sodium azide, were mixed at a 1:2 ratio and incubated at 37 °C for 3 months. The product was then dialyzed, concentrated, and lyophilized to yield AGEs powder.

2.4.3. Cytotoxicity test

HepG2 cells in the logarithmic growth phase were seeded into 96-well plates at a density of 5 × 10³ cells per well and cultured for 24 h. Subsequently, the original medium was discarded, the cells were treated with various concentrations of DCP-1-2 solutions (200 μL per well) for 48 h. Finally, medium containing 5% CCK-8 was added, and after 4 h, the absorbance at 450 nm was measured using a microplate reader. Cell viability was calculated as follows:

$${\text{Cell viability (\%) = [(A}}_{\text{sample}}{\text{ - A}}_{\text{background}}{\text{) / (A}}_{\text{NC}}{\text{ - A}}_{\text{background}}\text{)]×100}$$ (1)

2.4.4. In vitro antioxidant activity evaluation of DCP-1-2

The in vitro antioxidant activities of DCP-1-2 were evaluated by measuring its ability to scavenge ABTS⁺·, ·OH, DPPH·, and O₂^–· [[27], [28]]. Sample solutions of DCP-1-2 were prepared at gradient concentrations, with VC solutions at equivalent concentrations serving as the positive control. After the redox reaction reached equilibrium (incubation at 37 °C for 30–60 min, depending on the radical type), the absorbance at the specified wavelength was measured using a microplate reader to calculate the free radical scavenging rate.

$${\text{Scavenging rate (\%) = [1 - (A}}_{\text{sample}}{\text{ - A}}_{\text{background}}\text{) / (}{\text{A}}_{\text{NC}}{\text{ - A}}_{\text{background}}\text{)}\text{]×100}$$ (2)

2.4.5. Inhibitory effects on α-amylase, α-glucosidase and pancreatic lipase

The inhibitory activities of DCP-1-2 against α-amylase, α-glucosidase, and pancreatic lipase were evaluated as follows [[29], [30]]: Enzyme solutions were mixed with DCP-1-2 sample solutions at gradient concentrations, with acarbose solutions at the same concentration gradients serving as the positive control. After incubation in a 37 °C water bath for 15 min, the corresponding reaction substrates were added. The mixtures were further incubated at 37 °C, and the absorbance was measured at specified wavelengths using a microplate reader. The inhibition rate was calculated by the following formula:

$${\text{Inhibition rate (\%) = [1 - (A}}_{\text{sample}}{\text{ - A}}_{\text{background}}\text{) / (}{\text{A}}_{\text{NC}}{\text{ - A}}_{\text{background}}\text{)}\text{]×100}$$ (3)

2.4.6. Effects of DCP-1-2 on lipid content and oxidative stress in AGEs-Induced HepG2 cells

HepG2 cells were seeded into 24-well plates (500 μL/well) for 24 h, the cells were grouped and treated as follows: NC group, model group, positive Control (Lov group), and DCP-1-2 treatment group, each group was set with three replicate wells, and all treatments were maintained for 48 h under the same culture conditions. After treatment, the intracellular levels of TC, TG, HDL-C, LDL-C SOD and MDA were quantified using commercial kits according to the manufacturer's instructions. The absorbance of each reaction system was measured with a microplate reader at the corresponding characteristic wavelengths. The protein content of each well was determined by the BCA protein assay kit to normalize the lipid index data.

2.4.7. Oil red O staining

Following HepG2 cell seeding, drug treatment, and 48 h of incubation (per Section 2.4.6), Oil Red O staining was performed to visualize intracellular lipid droplets, with detailed experimental procedures described as follows[31]: First, the culture medium in each well was discarded. Then, 4% paraformaldehyde (w/v) was added to each well for fixed at room temperature for 30 min. Next, the Oil Red O working solution was added to cover the cell surface, and the cells were stained in a dark environment at room temperature for 15 min. After staining, the excess dye was discarded, and the cells were differentiated with 60% isopropanol for 30 s to reduce background staining. Finally, a small amount of PBS was retained in the wells to keep the cells hydrated, and imaging of the stained cells was performed using a Nib610-fl inverted fluorescence microscope (Keyence, China). Lipid droplets in the cells were visualized as bright red aggregates, and three random fields of view were captured for each well to ensure the representativeness of the results.

2.4.8. RT-qPCR analysis

Total RNA was extracted from HepG2 cells in the NC group, AGEs model group, and various drug-treated groups using TRIzol (Invitrogen, USA) strictly according to the method described by Zhao et al. [32]. The extracted RNA was then reverse-transcribed into complementary DNA (cDNA) using a commercial kit. RT-qPCR was subsequently performed on a CFX96 Touch system (Bio-Rad, USA). The sequences of the primers used were provided in the supplementary materials.

2.5. Statistical analysis

All experimental data were expressed as mean ± standard deviation (SD). Differences between multiple groups were analyzed by one-way ANOVA, followed by the least significant difference post-hoc test for pairwise comparisons. A p < 0.05 was considered statistically significant.

3. Results and discussion

3.1. Single-factor experiment analysis of DCP-1-2

Fig. 1A presented the schematic diagram (I) and technical flowchart (II) of the extraction and purification process for Dryopteridis crassirhizomatis Rhizoma polysaccharides. Fig. 1B depicted the effects of ultrasonic temperature (I), power (II), and time (III) on the extraction rate. When ultrasonic power and time were held constant, the extraction rate initially increased and then decreased with rising temperature, reaching a maximum of 22.21% at 60 °C. Under constant ultrasonic temperature and time, the extraction rate first increased and then decreased with increasing power, peaking at 24.63% at 200 W. When ultrasonic power and temperature were fixed, the extraction rate increased initially and subsequently decreased with prolonged time, attaining a maximum value of 24.17% at 50 min.

Fig. 1.

Preparation of polysaccharides from Dryopteridis crassirhizomatis Rhizoma. (A) Schematic diagram (I) and flowchart (II) of the extraction and purification process. (B) Results of single-factor experiments on extraction. (C) Response surface plot of the extraction conditions. (D) Contour plot of the extraction conditions.

3.2. Analysis of response surface experiment results

3.2.1. Model fitting and statistical results analysis

The optimization of the ultrasonic extraction process for DCP-1-2 was performed using RSM-BBD. As summarized in Table 2, a total of 17 experiments were designed, including 12 factorial points based on three independent variables and 5 center points. The independent variables investigated were ultrasonic temperature (A), ultrasonic power (B), and ultrasonic time (C). A quadratic polynomial regression analysis was employed to establish the mathematical model between the response variable (Y) and the factors, yielding the following regression equation:

Table 2.

Design matrix of actual versus predicted values for the optimized extraction conditions (Temperature (A), Power (B), and Time (C)) using RSM-BBD.

Run	Factor A: Temperature (℃)	Factor B: Power (W)	Factor C: Time (min)	Response 1 Extraction ratio (%)
1	50	200	60	20.87
2	70	220	50	19.34
3	50	200	40	21.75
4	70	200	60	22.52
5	70	200	40	20.06
6	50	180	50	19.4
7	70	180	50	20.37
8	60	200	50	25.86
9	60	200	50	24.69
10	60	200	50	25.98
11	60	200	50	25.46
12	50	220	50	20.99
13	60	220	40	21.5
14	60	180	40	20.4
15	60	180	60	19.23
16	60	220	60	21.23
17	60	200	50	26.12

Y = 25.62–0.0900A + 0.4575B + 0.0175C − 0.6550AB + 0.8350AC + 0.2250BC − 2.44A² − 3.15B² − 1.88C².

ANOVA was employed to evaluate the significance and adequacy of the established regression model. As shown in Table 3, the regression model was highly significant, with an F-value of 22.77 and a p-value < 0.001. The coefficient of determination R² was 0.9670, indicating that the model could explain 96.70% of the total variation in the extraction rate. The adjusted R² value was 0.9245, further supporting the model's significance. The lack-of-fit test yielded an F-value of 2.00 and a p-value of 0.2555, indicating that it was not significant relative to the pure error and that the model had no obvious systematic deviation. These results collectively demonstrate that the established model is accurate and effective for predicting the response.

Table 3.

Significance test and variance analysis for the regression model of ultrasonic-assisted extraction.

Source	Sum of Squares	Df	Mean Squares	F −value	p-value
Model	97.37	9	10.82	22.77	0.0002	significant
A-Temperature	0.0648	1	0.0648	0.1364	0.7228
B-Power	1.67	1	1.67	3.52	0.1026
C-Time	0.0024	1	0.0024	0.0052	0.9448
AB	1.72	1	1.72	3.61	0.0991
AC	2.79	1	2.79	5.87	0.0459
BC	0.2025	1	0.2025	0.4263	0.5347
A2	25.14	1	25.14	52.92	0.0002
B2	41.87	1	41.87	88.14	＜0.0001
C2	14.86	1	14.86	31.28	0.0008
Residual	3.33	7	0.4751
Lack of Fit	2.00	3	0.6659	2.01	0.2555	not significant
Pure Error	1.33	4	0.3319
Cor Total	100.70	16

Additionally, the table presents the significance (p-values) of individual regression coefficients and their interaction effects. The model indicated that, within the tested variable ranges, the interaction term AC were significant (p < 0.05), the quadratic terms A², B², and C² were all statistically significant (p < 0.001), with B² being highly significant (p < 0.0001). Based on the significance levels of these terms, the influence strength of the factors on the response value followed the order: ultrasonic power (B) > ultrasonic temperature (A) > ultrasonic time (C).

3.2.2. Response surface analysis

To investigate the interactive effects of the factors on the extraction process, 3D response surface plots (Fig. 1C) and 2D contour plots (Fig. 1D) were generated. As illustrated in Fig. 1C, the 3D response surfaces for the pairwise interactions (A × B (I), A × C (II), and B × C (III)) all exhibited pronounced steepness. In the contour plots involving factor A, closely spaced lines with distinct color gradients were observed, forming a relatively elliptical shape. This marked curvature, particularly in relation to ultrasonic power (B), confirmed that it exerted a more significant influence on the response compared to the other variables.

3.2.3. Validation of the ultrasonic extraction prediction model

Based on the response surface analysis conducted with Design-Expert software, the optimal ultrasonic extraction conditions for DCP-1-2 were predicted as follows: ultrasonic temperature 59.72 °C, extraction time 50.03 min, and ultrasonic power 201.52 W, with a corresponding extraction rate of 25.64%. To validate the model's reliability, the predicted conditions were slightly adjusted to practical values of 60 °C, 50 min, and 200 W for operational feasibility. Three parallel validation experiments were conducted under these adjusted conditions, and the average experimental extraction rate of DCP-1-2 was 25.62%, which was in close agreement with the predicted value. The result confirmed that the established response surface prediction model is reliable and stable, and the optimized process parameters are practical for ultrasonic extraction.

3.3. Extraction, separation and purification of DCP

Ultrasonic extraction conditions were first optimized using single-factor experiments combined with the RSM-BBD method. Under these optimized conditions, polysaccharides were extracted from Dryopteridis crassirhizomatis Rhizoma, after removing protein and dialysis, yielding cDCP with a yield of 2.67%. cDCP was subsequently purified using anion- and cation-exchange resin columns, yielding two components: DCP-1 (25.26%) and DCP-2 (4.19%). DCP-1 was further purified by chromatography on a DEAE-52 cellulose column. As shown in Fig. S1, the fraction eluted with 0.1 M NaCl (denoted as DCP-1-2) exhibited the highest yield. Therefore, DCP-1-2 was selected as the target component for subsequent experimental studies.

3.4. Structural and compositional analysis of DCP-1-2

3.4.1. Purity and the molecular weight

HPGPC analysis of DCP-1-2 exhibited a single symmetrical peak with a retention time of 8.714 min (Fig. 2A), indicating it was a homogeneous polysaccharide with high purity. A calibration curve was established using dextran standards, with the regression equation y = − 0.9252x + 10.2656 (R² = 0.9944). Based on this calibration curve, the Mw of DCP-1-2 was calculated to be 159.74 kDa.

Fig. 2.

Structural characterization of DCP-1-2. (A) HPGPC and molecular weight standard curve. (B) UV–Vis spectrum. (C) FT-IR spectrum. (D) SERS spectrum. (E) HPLC chromatograms of monosaccharide standards (I) and DCP-1-2 (II).

3.4.2. UV–Vis spectroscopic analysis of DCP-1-2

The UV–Vis spectrum of DCP-1-2 (Fig. 2B) showed no characteristic absorption at 260 nm or 280 nm, indicating the absence of nucleic acid and protein impurities. These results demonstrated that the Sevag method and the aforementioned purification process were effective in removing protein impurities. Additionally, the high purity of DCP-1-2 confirmed by UV–Vis spectroscopy was consistent with high-performance gel permeation chromatography (HPGPC) results, which were also in line with previous reports[33].

3.4.3. Functional group analysis

FT-IR spectroscopy and SERS spectroscopy were used to analyze the functional groups and structural characteristics of DCP-1-2. As shown in the FT-IR spectrum (Fig. 2C): the absorption peak at 3396 cm⁻1 was assigned to the O-H stretching vibration, the peak at 2925 cm⁻¹ corresponded to C-H stretching vibration, a typical signal of polysaccharides [34], three absorption bands in the range of 1154–1028 cm⁻¹ indicated the presence of a pyranose ring, the peak at 1625 cm⁻¹ was attributed to the bending vibration of bound water, the peak at 832 cm⁻¹ suggested DCP-1-2 contains α-glycosidic linkages [35]. Distinct polysaccharide signals were also observed in the SERS spectrum (Fig. 2D): peak at 868 cm⁻¹ potentially arising from C-C-O-C-O stretching and peak at 1179 cm⁻¹ likely due to C-O-C stretching vibration [36]. The peaks at 1321 cm⁻1, 1455 cm⁻¹ and 1496 cm^-1were possibly related to C-H bending vibration, CH₂ scissoring vibration and C-O-H bending vibration [37].

3.4.4. Monosaccharide composition analysis

HPLC was used to analyze the monosaccharide composition of DCP-1-2, with the chromatogram schematically illustrated in Fig. 2E. By comparing the retention times and peak order of the DCP-1-2 hydrolysate chromatogram (II) with those of the monosaccharide mixed standard chromatogram (I), DCP-1-2 was identified as a neutral polysaccharide. It was composed of four monosaccharides: Man, Glc, Gal, and Ara, with a molar ratio of 1.83:55.55:15.56:27.06. Among these, Glc (55.55%) and Ara (27.06%) were the major monosaccharide components of DCP-1-2, and Ara was likely the primary contributor to the observed bioactivity of DCP-1-2.

3.4.5. Methylation and GC–MS analysis

Methylation analysis was employed to determine the glycosidic linkage patterns in DCP-1-2 by identifying free hydroxyl groups on monosaccharide residues, thereby inferring the specific positions of glycosidic bonds[38]. According to the GC–MS results interpreted with reference to the CCRC spectral database (Fig. 3C, Table 4), at least eight distinct glycosidic linkages were identified in DCP-1-2. These included T-Araf, 1,3-Araf, 1,5-Araf, T-Glcp, 1,3,5-Araf, 1,4-Glcp, 1,6-Glcp, and 1,4,6-Glcp [39], with molar ratios of 11.55:10.10:5.47:5.26:15.28:16.43:22.03:13.84. These findings were consistent with the earlier results from FT-IR spectroscopy and monosaccharide composition analysis, confirming the structural characteristics of DCP-1-2. Furthermore, combining the methylation analysis results with NMR data (detailed in subsequent sections), the specific glycosidic linkages between glycosyl residues in DCP-1-2 were further elucidated.

Fig. 3.

Structural analysis of DCP-1-2 by NMR spectroscopy. (A) 1D NMR spectra: ¹H NMR (Ⅰ) and ¹³C NMR (II). (B) 2D NMR spectra: ¹H–¹³C HSQC (I), ¹H–¹H COSY (II), and ¹H–¹³C HMBC (III). (C) Total ion chromatogram. (D) Proposed structure of DCP-1-2.

Table 4.

Glycosidic linkage analysis of DCP-1-2.

Peak	RT (min)	Sugar derivative	Mass fragments (m/z)	Linkage types	Molar ratio (%)
1	18.148	2,3,5-Me3-Araf	59,71,87,102,118,129,145,161	T-Araf-	11.55
2	20.342	2,5-Me2-Araf	59,71,87,99,113,118,129,160,173,233	1,3-Araf-	10.10
3	21.995	2,3-Me2-Araf	59,71,87,102,118,129,162,173,189	1,5-Araf-	5.47
4	23.673	2,3,4,6-Me4-Glcp	59,71,87,102,118,129,145,162,205	T-Glcp-	5.26
5	24.127	2-Me-Araf	59,74,85,99,118,127,159,173,201,261	1,3,5-Araf-	15.28
6	26.107	2,3,6-Me3-Glcp	59,71,87,99,101,113,118,129,162,233	1,4-Glcp-	16.43
7	27.581	2,3,4-Me3-Glcp	59,71,87,99,101,118,129,162,189	1,6-Glcp-	22.03
8	28.923	2,3-Me2-Glcp	59,74,85,102,118,127,162,187,201,261	1,4,6-Glcp-	13.84

3.4.6. NMR spectroscopic analysis

NMR spectroscopy serves as a powerful tool for elucidating the chemical structure of polysaccharides [40]. The characteristic NMR signals of polysaccharides typically appear in the range of δ ¹H 3.0–5.5  ppm and δ¹³C 60–110 ppm. In the ¹H NMR spectrum of DCP-1-2 (Fig. 3A(I)): signals in the δ¹H 4.9–5.5 ppm region were characteristic of α-anomeric protons, while those in δ¹H 4.3–4.9 ppm correspond to β-anomeric protons, signals in δ¹H 3.0–4.3 ppm were attributed to ring protons [[41], [42]]. Correspondingly, the ¹³C NMR spectrum (Fig. 3A(II)) exhibits anomeric carbon signals in the δ¹³C 98–110 ppm region. Based on reported NMR signal data, and combined with previous monosaccharide composition and methylation analysis results, the anomeric signals of DCP-1-2 were tentatively assigned to α-L-arabinose, α-D-glucose, and β-D-glucose[43].

Further specific assignments were made using ¹H–¹³C HSQC correlation data: the cross-peak at δ¹H/¹³C 4.68/100.72 ppm (Fig. 3B(Ⅰ)) was attributed to H1/C1 of T-β-D-Glcp (residue B), signals at δ¹H/¹³C 4.88/98.68, δ¹H/¹³C 5.31/99.64, δ¹H/¹³C 4.98/99.28, and δ¹H/¹³C 5.07/98.33 ppm were assigned to H1/C1 of T-α-D-Glcp (residue A), 1,4-α-D-Glcp (residue C), 1,6-α-D-Glcp (residue D), and 1,4,6-α-D-Glcp (residue E), respectively, cross-peaks at δ¹H/¹³C 5.27/105.74 and δ¹H/¹³C 5.17/106.10 ppm corresponded to H1/C1 of T-α-L-Araf (residue F) and 1,3-α-L-Araf (residue G), and those at δ¹H/¹³C 5.09/107.41 and δ¹H/¹³C 5.16/109.44 ppm were linked to H1/C1 of 1,5-α-L-Araf (residue H) and 1,3,5-α-L-Araf (residue I) [[44], [45], [46]].

Analysis of the ¹H–¹H COSY spectrum (Fig. 3B(Ⅱ)) showed cross-peaks at δ¹H/¹H 4.68/3.36, 4.88/3.55, 5.31/3.65, 4.98/3.62, 5.07/3.63, 5.27/4.19, 5.17/4.11, 5.09/4.13, and 5.16/4.05 ppm. For residue A, the proton signals at δ¹H 3.55, 3.88, 3.75, 3.72, and 3.81 ppm were assigned to H2, H3, H4, H5, and H6, respectively. A summary of the NMR signal assignments for sugar residues A-I is provided in Table 5 [[20], [47]].

Table 5.

Chemical shifts of main residues of DCP-1-2.

Code	Residues	Chemical Shifts (ppm)
		H1/C1	H2/C2	H3/C3	H4/C4	H5/C5	H6/C6
A	T-α-D-Glcp-(1→	4.88/98.68	3.55/71.70	3.88/76.72	3.75/72.86	3.72/71.18	3.81/60.73
B	T-β-D-Glcp-(1→	4.68/100.72	3.36/71.65	3.78/72.17	3.73/68.26	3.46/73.68	3.81/61.70
C	→4)-α-D-Glcp-(1→	5.31/99.64	3.65/71.72	3.90/73.28	3.55/76.68	3.77/71.55	3.75/60.42
D	→6)-α-D-Glcp-(1→	4.98/99.28	3.62/71.38	3.72/72.67	3.77/70.41	3.97/69.93	3.75/71.12
E	→4,6)-α-D-Glcp-(1→	5.07/98.33	3.63/72.86	3.88/73.30	3.51/78.12	3.82/73.3	3.81/69.96
F	T-α-L-Araf-(1→	5.27/105.74	4.19/80.92	3.97/76.71	4.04/82.15	3.82/61.70
G	→3)-α-L-Araf-(1→	5.17/106.10	4.11/81.58	3.96/82.76	4.05/81.70	3.66/62.15
H	→5)-α-L-Araf-(1→	5.09/107.41	4.13/80.26	3.97/76.31	4.12/82.49	3.80/70.85
I	→3,5)-α-L-Araf-(1→	5.16/109.44	4.05/81.46	3.97/84.17	4.05/81.26	3.82/70.82

Note kindly takecare author citation footnote.

The glycosidic linkage sequence of DCP-1-2 was determined by ¹H–¹³C HMBC spectroscopy (Fig. 3B(III)). The cross-peak between H-1 of residue D and C-4 of residue C (δ 4.98/76.68 ppm) and that between H-1 of residue C and C-5 of residue H (δ 5.31/70.85 ppm) indicated the presence of the repeating unit: [→6)-α-D-Glcp-(1 → 4)-α-D-Galp-(1 → 5)-α-L-Araf-(1 → ]. Additional cross-peaks at δ 4.98/84.17 (D H1/I C3), 5.16/78.12 (I H1/E C4), 4.68/70.82 (B H1/I C5), and 4.88/69.96 ppm (A H1/E C6) suggested other structural fragments: [→6)-α-D-Glcp-(1 → 3,5)-α-L-Araf-(1 → 4,6)-α-D-Glcp-(1 → ], [T-β-D-Glcp-(1 → 5,3)-α-L-Araf-(1 → ], and [T-α-D-Glcp-(1 → 6,4)-α-D-Glcp-(1 → ]. Furthermore, cross-peaks at δ 4.98/82.76 (D H1/G C3), 5.17/84.17 (G H1/I C3), 5.27/76.68 (F H1/C C4), and 5.31/70.82 ppm (C H1/I C5) confirmed segments such as [→6)-α-D-Glcp-(1 → 3)-α-L-Araf-(1 → 3,5)-α-L-Araf-(1 → ] and [T-α-L-Araf-(1 → 4)-α-D-Glcp-(1 → 5,3)-α-L-Araf-(1 → ]. Based on these linkage patterns, three plausible structural units were proposed for DCP-1-2. By integrating molecular weight data, monosaccharide composition, and methylation analysis results, a probable structure was deduced, as illustrated in Fig. 3D.

3.4.7. Congo red assay results

The maximum absorption wavelength (λmax) of the DCP-1-2-Congo red complex did not exhibit a redshift with increasing NaOH concentration (Fig. 4A), indicating the absence of a triple-helix conformation in DCP-1-2.

Fig. 4.

Physicochemical characterization of DCP-1-2. (A) Congo red assay. (B) SEM images of DCP-1-2 at magnifications of 1,000× (I) and 40,000× (II). (C) Standard curves for carbohydrate (I), protein (II), and uronic acid (III) content quantification. (D) Particle size distribution histogram. (E) Zeta potential. (F) TGA and DTG curves. (G) Steady-state flow shear viscosity curve. (H) Dynamic viscoelasticity: storage modulus (G′) and loss modulus (G″).

3.4.8. SEM morphological analysis

SEM micrographs at 1000 × magnification revealed that DCP-1-2 exhibited a partially aggregated, flake-like morphology with smooth surface fractures and irregular features (Fig. 4B(I)). At 40000 × magnification, the surface appeared smooth (Fig. 4B(II)). Based on its chemical composition, zeta potential, and low uronic acid content, DCP-1-2 was classified as a neutral polysaccharide with relatively weak electrostatic interactions and low aggregation tendency, which contributed to its good stability [48].

3.4.9. Content determination

The contents of total sugar, protein, and uronic acid in DCP-1-2 were determined via the phenol–sulfuric acid, Coomassie brilliant blue, and carbazole-sulfuric acid methods to be 76.19 ± 0.91%, ND, and 2.44 ± 2.27%, respectively (Fig. 4C(I-II)).

3.4.10. Particle size and Zeta potential determination

The particle size (Fig. 4D) and Zeta potential (Fig. 4E) of DCP-1-2 were measured using a Malvern particle size analyzer. The results showed that the average particle size of DCP-1-2 was 851 nm and the Zeta potential was −2.98 mV, suggesting that DCP-1-2 was a neutral polysaccharide with large size and a near-neutral charge, which contributes to its stability.

3.4.11. Thermogravimetric analysis

Thermogravimetric analysis was employed to evaluate the thermal stability and decomposition behavior of the polysaccharide DCP-1-2. As shown in Fig. 4F, the initial weight loss occurred around 74 °C, presenting a broad and pronounced peak in the DTG curve, which can be attributed to the loss of adsorbed and bound water. The major decomposition stage took place between 175 °C and 450 °C, where the sample underwent rapid mass loss. A sharp DTG peak appeared at 237 °C, likely corresponding to the cleavage of glycosidic bonds. From 450 °C to 550 °C, the mass loss slowed considerably. The final residue at 550 °C was 42.92%, indicating good thermal stability of DCP-1-2 under the tested conditions[49].

3.4.12. Rheological analysis

As summarized in Fig. 4G, the apparent viscosity of DCP-1-2 decreased rapidly at shear rates below 20 s⁻¹ and stabilized at higher shear rates (≥20 s⁻¹). This shear-thinning behavior was consistent with the characteristics of non-Newtonian pseudoplastic fluids [28]. As shown in Fig. 4H, both the storage modulus (G′) and loss modulus (G″) increased with angular frequency. The crossover point of the polysaccharide's G′ and G″ was observed at a frequency of 7.84 Hz. Before the crossover point, G″ was greater than G′, while beyond this point, G′ exceeded G″. These results confirm that DCP-1-2 exhibits both elastic and viscous characteristics, consistent with viscoelastic fluid behavior[50].

3.5. Anti-GLMD activity and mechanism of DCP-1-2 in vitro

3.5.1. Cytotoxicity test

The effect of DCP-1-2 concentration on HepG2 cell viability is presented in Fig. 5A. Cell viability exhibited a gradual decline with increasing DCP-1-2 concentration. Specifically, at concentrations below 1 mg/mL, the cell viability remained above 90%, with no significant cytotoxicity compared to the NC group. As the concentration of DCP-1-2 increased, the cell survival rate gradually decreased. At concentrations below 1 mg/mL, the survival rate remained above 90%, showing no significant cytotoxicity compared to the NC group. When the concentration reached 2 mg/mL, the cell survival rate was 83.12%, which was considered within a safe range, and a significant difference was observed (p < 0.01). Based on these results, a concentration of 2 mg/mL was selected for subsequent experiments.

Fig. 5.

Antioxidant activity of DCP-1-2 and its effects on key enzymes of glycolipid metabolism. (A) Effect of different concentrations of DCP-1-2 on the viability of HepG2 cells (n = 6, ^** indicated p < 0.01, ^*** indicated p < 0.001 vs. NC group). (B) Antioxidant activity of DCP-1-2, evaluated by scavenging activities against ABTS⁺·(I), ·OH (II), DPPH· (III), and O₂^–· (IV). (C) Hypoglycemic activity of DCP-1-2 in vitro, as measured by the inhibitory activities against α-amylase (I), α-glucosidase (II), and pancreatic lipase (III).

3.5.2. In vitro antioxidant activity of DCP-1-2

The free radical scavenging activities of DCP-1-2 against ABTS⁺·, ·OH, DPPH·, and O₂⁻· were shown in Fig. 5B(I-IV). DCP-1-2 exhibited concentration-dependent scavenging effects on all four types of radicals. At a concentration of 2 mg/mL, the scavenging rates were 87.63 ± 0.17% for ABTS radicals, 71.53 ± 2.81% for hydroxyl radicals, 75.50 ± 7.28% for DPPH radicals, and 75.86 ± 3.45% for superoxide anion radicals, demonstrating strong antioxidant activity.

3.5.3. Inhibitory effects on α-amylase, α-glucosidase and pancreatic lipase

Natural polysaccharides can regulate blood sugar levels by inhibiting the activities of α-amylase and α-glucosidase, thereby delaying carbohydrate digestion and absorption[51]. As shown in Fig. 5C(I-II), DCP-1-2 inhibited both α-amylase and α-glucosidase in a concentration-dependent manner. At 2 mg/mL, the inhibition rates reached 75.80 ± 0.05% for α-amylase and 73.66 ± 2.78% for α-glucosidase. Pancreatic lipase, a key enzyme responsible for fat absorption, is a potential target for preventing lipid disorders induced by excessive dietary fat intake. DCP-1-2 also exhibited a notable concentration-dependent inhibitory effect on pancreatic lipase (Fig. 5C(III)), with an inhibition rate of 34.91 ± 2.93% at 2 mg/mL.

3.5.4. Effect of DCP-1-2 on lipid accumulation in AGEs-Induced HepG2 cells

AGEs are known to impair protein function, promote lipid peroxidation, and contribute to insulin resistance. In this study, HepG2 cells were treated with AGEs to establish a cellular model of GLMD, enabling the integrated analysis of glucose and lipid metabolism within the same cell system (Fig. 6A). As illustrated in Fig. 6B, DCP-1-2 reduced lipid accumulation in AGEs-induced HepG2 cells. Compared with the NC group, the AGEs model group exhibited significantly elevated levels of TC (2.66 ± 0.33 mmol/gprot) (I), TG (2.12 ± 0.32 mmol/gprot) (II), and LDL-C (54.19 ± 3.74 mmol/gprot) (IV), along with a marked decrease in HDL-C (0.47 ± 0.10 mmol/gprot) (III), confirming the successful establishment of the high-fat cell model. Treatment with DCP-1-2 effectively attenuated these alterations, significantly suppressing the increases in TC (1.18 ± 0.16 mmol/gprot), TG (1.49 ± 0.11 mmol/gprot), and LDL-C (38.46 ± 1.61 mmol/gprot), while increasing HDL-C levels (0.70 ± 0.06 mmol/gprot) to an extent comparable to the positive control Lov. These results demonstrate that DCP-1-2 possesses notable anti-GLMD activity and can effectively ameliorate lipid metabolic abnormalities. As shown in Fig. 6C, HepG2 cells were treated with AGEs and exhibited significantly decreased SOD (I) activity and increased MDA (II) content compared with NC group, indicating enhanced oxidative stress and impaired antioxidant capacity. Treatment with DCP-1-2 significantly reversed these changes, increasing SOD activity (from 40.27 ± 3.85 to 56.03 ± 2.64 U/mgprot) and reducing MDA levels (from 6.01 ± 0.58 to 5.15 ± 0.05 U/mgprot).

Fig. 6.

In vitro anti-GLMD activity of DCP-1-2. (A) Schematic diagram of lipid-lowering in vitro. (B) Effects of DCP-1-2 on TC (I), TG (II), HDL-C (III), and LDL-C (IV) levels in AGEs-induced HepG2 cells with hyperlipidemia. (C) Effects of DCP-1-2 on SOD (I) and MDA (II) levels in AGEs-induced HepG2 cells. (D) Partial enlarged view of Oil Red O staining. (E) Representative images of Oil Red O staining in HepG2 cells (NC group, AGEs-induced model group, Lov-positive control group, and DCP-1-2 treatment group). (F) Quantitative analysis of intracellular lipid droplet accumulation rate corresponding to (E). (Data are expressed as mean ± SD, n = 3, significant differences are indicated as ^### indicated p < 0.001, ^## indicated p < 0.01, vs. NC group, ^*** indicated p < 0.001, ^** indicated p < 0.01, * indicated p < 0.05, vs. AGEs model group.).

3.5.5. Oil Red O staining

Oil Red O staining is a well-recognized method for visualizing intracellular lipid accumulation in HepG2 cells, and further enables quantitative evaluation of lipid dysregulation in these cells. (Fig. 6D). As shown in Fig. 6E, compared with the NC group, cells in the AGEs model group exhibited abundant bright red lipid droplets. In contrast, treatment with DCP-1-2 markedly suppressed intracellular lipid deposition, showing an effect comparable to that of the positive control drug Lov. A similar trend was observed in the quantitative analysis of intracellular lipid accumulation rate (Fig. 6F). These morphological findings were consistent with the biochemical measurements of TC, TG, HDL-C, and LDL-C levels in the AGEs-induced HepG2 cell model.

3.5.6. RT-qPCR analysis

The effect of the polysaccharide DCP-1-2 on the expression of key genes in the insulin signaling pathway was evaluated in an AGEs-induced HepG2 cell model (Fig. 7A). As revealed by the RT-qPCR analysis in Fig. 7B, compared with the AGEs-induced group, treatment with DCP-1-2 significantly up-regulated the mRNA expression of Irs1 and Pik3r1 by approximately 51.52% and 38.57%. In contrast, the mRNA level of Akt1 was not significantly altered following DCP-1-2 treatment (p > 0.05), indicating that DCP-1-2 does not affect Akt1 transcription. These results suggested that the ameliorative effect of DCP-1-2 on glucose metabolism in the AGEs-induced HepG2 model may involve regulation upstream of Akt1, potentially through the Irs1/Pik3r1 pathway, rather than through transcriptional modulation of Akt1 itself [[52], [53]].

Fig. 7.

RT-qPCR analysis of DCP-1-2. (A) Schematic diagram of the RT-qPCR experiment. (B-D) Relative mRNA expression levels of genes in AGEs-induced HepG2 cells treated with DCP-1-2: (B) Irs1, Pik3r1, and Akt1; (C) Fas, Ppara, and Srebf1; (D) Rela and JUN. (E) Heatmap of the expression levels for all genes analyzed. (n = 3, Significant differences are indicated as ^### indicated p < 0.001, ^## indicated p < 0.01, ns indicated p ＞ 0.05, vs. NC group, ^*** indicated p < 0.001, ^** indicated p < 0.01, * indicated p < 0.05, ns indicated p ＞ 0.05, vs. AGEs model group.).

The hypolipidemic effect of DCP-1-2 is mediated through the regulation of key lipid-metabolic genes. As shown in Fig. 7C, Downregulation of Srebf1, a master transcription factor for lipid synthesis, suppresses the transcriptional network driving fatty acid and cholesterol production. Reduced mRNA levels of Fas, the rate-limiting enzyme in de novo fatty acid synthesis, lead to decreased intracellular triglyceride accumulation. Furthermore, the inhibition of Ppara, which under pathological conditions can exacerbate lipid deposition, contributes to the restoration of metabolic balance[9]. This concerted downregulation, together with the observed reduction in lipid droplets and triglyceride content, indicated that DCP-1-2 ameliorates AGEs-induced hepatic lipid accumulation by targeting the Srebf1/Fas/Ppara signaling pathway[54].

In this study, Rela and JUN were investigated as core components of the NF-κB and AP-1 transcription factors (Fig. 7D), respectively, which act as key upstream regulators in the AGEs-induced dysregulation of glucose and lipid metabolism in HepG2 cells. Their elevated expression not only indicates a state of cellular inflammation and stress but also directly promotes insulin resistance and lipid accumulation by coordinately modulating inflammatory cytokine production, insulin signaling pathways, and lipid metabolic gene networks[[55], [56]]. Treatment with DCP-1-2 significantly downregulated the mRNA expression of both Rela and JUN. Compared with the model group, the DCP-1-2 treated group showed markedly reduced mRNA levels of Rela and JUN. Fig. 7E showed the heatmap of the aforementioned genes, illustrating the expression levels of the same gene across different groups. The color gradient intuitively reflects the expression intensity, a result that further validates the aforementioned gene expression patterns. These findings reveal that DCP-1-2 has the potential to ameliorate GLMD at the transcriptional level in the AGEs-induced high-fat HepG2 cell model.

4. Conclusion

This study optimized the ultrasonic-assisted extraction of DCP using a combination of single-factor, orthogonal and response surface methodology experiments. Under the optimized conditions, a homogeneous neutral polysaccharide, DCP-1-2, was obtained through purification with ion-exchange and DEAE-52 cellulose chromatography. It exhibited an average molecular weight of 159.74 kDa and was composed primarily of glucose, galactose, and arabinose, with trace amounts of mannose. Structural characterization by SERS spectroscopy, SEM, methylation analysis, and 1D/2D NMR confirmed an arabino-glucan backbone. Biological evaluation demonstrated that DCP-1-2 possesses significant in vitro antioxidant activity and ameliorates GLMD. Based on RT-qPCR analysis, the anti-GLMD mechanism involves upregulation of Irs1 and Pik3r1 expression and downregulation of Srebf1, Fas, Ppara, Rela, and JUN mRNA levels. These molecular changes explained the mechanism of GLMD and the alleviation of cellular glucose metabolism/lipid metabolism imbalance, as well as the mitigation of oxidative stress by DCP-1-2. Collectively, these findings not only lay a solid th5eoretical foundation for the exploitation of Dryopteridis crassirhizomatis Rhizoma as a functional resource, but also offer novel insights into the molecular pathogenesis and potential therapeutic strategies of GLMD.

CRediT authorship contribution statement

Peng Guo: Writing – review & editing, Methodology, Conceptualization. Jiayue Liang: Writing – original draft, Investigation, Formal analysis. Yiyang He: Software, Formal analysis. Yaowen Xing: Validation. Xubin Quan: Data curation. Yuyang Fang: Investigation. Wenqi Zhao: Investigation. Jiahui Liang: Writing – original draft, Investigation, Formal analysis. Pengyu Zhou: Data curation. Xinyi Wang: Software. Yanli Wu: Resources, Methodology, Conceptualization.

Funding

This study was supported by the National Natural Science Foundation of China (81403072), Natural Science Foundation of Heilongjiang Province (LH2020H022), the Postdoctoral Research Fund of Heilongjiang Province (LBH-Q19150), Traditional Chinese Medicine Research Project of Heilongjiang Province (ZHY2024-196), the Harbin Medical University School of Pharmacy Excellent Young Talents Fund (2019-YQ-11) and Student’s Innovation Training Program of Harbin Medical University (202510226100).

Declaration of competing interest

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

Appendix A. Supplementary data

The following are the Supplementary data to this article: