The protocol of hot water extraction of polysaccharides was according to the previous one with some modifications (32). The extract was refluxed twice with hot water at a material-to-liquid ratio of 1:20 (w/v) for 45 min each time, and the clarified solution was combined and extracted by filtration with a cloth funnel with diatomaceous earth while hot, concentrated at 65°C, alcoholic precipitation in 95% ethanol solution, followed by centrifugation at 10,000 × g for 10 min and freeze-dried to collect the precipitate. Then, the obtained polysaccharides were defatted and deproteinated by petroleum ether and Sevage reagent (chloroform: n-butanol volume ratio = 4:1). After that, the polysaccharide was alcoholically precipitated in 95% ethanol, centrifuged after 12 h in the refrigerator at 4°C, and the precipitate was used for freeze-drying to obtain the defatted and deproteinized crude polysaccharide of elderberry, which was EFP.

The polysaccharide separation and purification process were as shown in Figure 1. About 5.00 g of EFP was dissolved thoroughly in 100 mL of deionized water, followed by centrifugation at 10,000 × g for 10 min, and filtered through a 0.45-μm microporous filter membrane. The filtered crude polysaccharide solution was slowly added to a pre-equilibrated DEAE-cellulose 52 anion-exchange column (50 mm × 1,000 mm) and eluted with deionized water and 0.1, 0.2, 0.3, 0.4, and 0.5 mol/L NaCl solution, respectively, (33). Each tube was 50 mL with a flow rate of 5 mL/min, and 50 tubes of each were collected. The polysaccharide content of each tube was detected at 490 nm using the phenol-sulfuric acid method according to Ji et al. (33), and the elution curve was plotted with the number of tubes as the horizontal coordinate and the absorbance value as the vertical coordinate, and the polysaccharide eluates corresponding to the peaks (EFP-0, EFP-1, EFP-2, EFP-3, and EFP-4) were combined according to the peaks of the elution curve and spun off for concentration. 100 mL EFP-0 solution (0.5 mg/mL) was slowly added to Sephadex G-100 Sephadex column (16 mm × 2000 mm), eluted with deionized water, and 30 tubes of polysaccharide solution were collected at 10 mL per tube. The collected polysaccharides were lyophilized. EFP-1, EFP-2, EFP-3, and EFP-4 were purified by the above method.

As previously described and improved upon Olawuyi et al. (34), the heavy mean molecular weight (Mw) distribution and polydispersity (Mw/Mn) of the five polysaccharide fractions were determined by high-performance gel permeation chromatography (HPGPC; 35). The samples and standards were weighed precisely, and the samples were prepared into a 5-mg/ml solution, centrifuged at 12,000 rpm for 10 min, then the supernatant was filtered through a 0.22-μm microporous membrane and the samples were transferred into a 1.8-mL injection vial. After that, 20 μL of sample solution was injected into a BRT105-104-102 series gel column (8 × 300 mm) with 0.05 mol/L NaCl solution as the mobile phase and the column temperature was 40°C at a flow rate of 0.6 mL/min. Using the detector RI-10A, the molecular weight size of each sample was calculated based on the standard curve.

The monosaccharide composition of purified polysaccharides was determined by ion chromatography (Thermo Fisher ICS5000, United States; 36). About 10 mg of lyophilized polysaccharide powder was placed in an ampoule, 10 mL of 3 mol/L TFA was added and hydrolyzed at 120°C for 3 h, and then the residual TFA was removed by rotary evaporator. A 10-mL acid hydrolyzed solution was accurately aspirated and blown dry in a tube, 10 mL of deionized water was added with vortex mixing, 100 μL of deionized water was aspirated, and 900 μL of deionized water was added and then centrifuged at 12,000 rpm for 5 min. The analysis was performed on a Dionex CarbopacTMPA20 (3 × 150) column with H₂O and 15 mol/L NaOH and 100 mol/L NaOAC as mobile phases at a flow rate of 0.3 mL/min, column temperature of 30°C, and injection volume of 5 μL. The retention times and peak areas of the 16 monosaccharide standards were compared, and the different sugars were identified and quantified separately.

The infrared spectral analysis was conducted following the methods described and improved by previous authors (20). The extracted and isolated purified elderberry polysaccharide was pressed into thin slices, and the slices were subjected to Thermo Nicolet IS10 Fourier transform infrared spectroscopy (FT-IR) with the setting of mid-infrared mode scan range of 4,000 to 400 cm^–1, and the analysis was performed using OMSNIC 8.2 (Thermo Scientific™, United States) software.

The maximum purity of polysaccharides in EFP-2 was obtained by isolation and purification, so the EFP-2 fraction was selected to investigate the structure and antioxidant activity of polysaccharides. As previously stated and modified (37), an X’Pert Pro X-ray diffractometer (PANalytical, Netherlands) was used to measure the crystallization properties of EFP-2 under 40 kV and 40 mA of radiation at Cu Kα (λ = 0.154 06 nm) on a copper target. The X-ray intensity was measured using a NaI crystal scintillation counter (scintillation counter) with a scanning range of 5° to 60°, a step size of 0.02°, and a scanning speed of 4°/min.

The study of the microstructure of elderberry polysaccharides was conducted by scanning electron microscopy (38). About 5 mg of the dried sample was taken, adhered to a conductive carbon film containing a double-sided adhesive, placed in the sample chamber of the ion sputterer, and sprayed with gold for about 40 s. After the sample was removed, it was placed in the scanning electron microscope observation chamber with an acceleration voltage of 2 kV for observation.

The methylation steps were mainly referred to Yang et al. (39) with some modifications. About 2 to 3 mg of EFP-2 sample was added to 1 mL of anhydrous DMSO, then the anhydrous base solution and iodomethane solution are quickly added and reacted for 60 min at 30°C in a magnetic stirring water bath, and finally 2 mL of ultrapure water was added to terminate the methylation reaction. The methylated polysaccharide was taken and hydrolyzed by adding 1 mL of 2 M TFA for 90 min and evaporated dry by a rotary evaporator. About 2 mL of double-distilled water and 60 mg of sodium borohydride have been added to the residue to reduce for 8 h, then, glacial acetic acid was added to neutralize the residue, 1 mL acetic anhydride was added to acetylate the reaction at 100°C for 1 h, and cooled. After that, 3 mL of toluene was added to remove the excess acetic anhydride. The acetylated product was dissolved with 3 mL of CH₂Cl₂ and the upper aqueous layer was removed. After, the CH₂Cl₂ layer was dried with an appropriate amount of anhydrous sodium sulfate, and the volume was fixed at 10 mL and put into a liquid phase vial. Finally, an RXI-5 SIL MS column of 30 m × 0.25 mm × 0.25 μm was used, the starting temperature was 120°C, ramped up to 250°C/min at 3°C/min, and held for 5 min. The injection port temperature was 250°C, the detector temperature was 250°C/min, and the carrier gas was helium at a flow rate of 1°mL/min.

About 40 mg of EFP-2 was dissolved in D₂O and then loaded in an NMR tube. The NMR spectra were scanned on a high-resolution 700 MHz Bruker AVANCE III NMR spectrometer (Bruker, Germany). One-dimensional NMR spectra (¹H and ¹³C) and two-dimensional NMR spectra [correlation spectroscopy (COSY), heteronuclear single quantum coherence (HSQC), and heteronuclear multiple bond coherence (HMBC)] were adopted to analyze the structural features of polysaccharides (40).

RAW 264.7 cell line was obtained from the Institute of Cell Biology, Chinese Academy of Sciences (Shanghai, China) and cultured in DMEM medium (with 10% fetal bovine serum and 100 U/ml each of penicillin and streptomycin) at 37°C, 5% CO₂, and saturated humidity in an incubator (41).

The viability of RAW 264.7 cells was determined by a standard assay using the MTT method (42). Cells of logarithmic growth phase were collected, washed with PBS buffer, trypsin digested, and centrifuged after the termination of digestion, and the cells were made into cell suspension with culture medium, inoculated into 96-well plates at a density of 5 × 10⁴ cells/well, and cell dilution was added at 100 μL/well and incubated for 24 h at saturated humidity, 37°C, and 5% CO₂. The samples were diluted with culture medium separately, and 100 μL was added to each well; the solvent model group and control group (EFP-2 solution with different solubility) were also set up, and the final concentration of 1.2 mM H₂O₂ was added, respectively. After 4 h incubation, MTT solution was taken and added to the cells at 10 μL/well, and incubation was continued in the incubator for 4 h. After 4 h, the liquid in the wells was discarded, and 150 μL DMSO was added to each well to dissolve it. The absorbance at 490 nm was measured with an enzyme marker, and the survival rate was calculated according to the following Eq. (1).

The differences in cell morphology and number between the different treatment groups were compared by fluorescence microscopy. RAW264.7 cells were inoculated in 6-well plates (5 × 10⁵ cells/mL) at 1 mL per well, and blank, VC (100 μg/mL), and control (12.5–200 μg/mL) were set up in the same way as the MTT assay, and after 24 h of incubation, respectively. Cells were observed and photographed under a 200 × microscope using a fluorescent microscope to compare differences in cell morphology and number between different EFP-2 concentrations.

The 12-well plates were seeded with good cells plastered overnight (5 × 10⁵ cells/well), and VC, EFP-2 in concentration gradient, and 1.2 Mm H₂O₂ were added. Another sample was set as blank control and 1.2 mM H₂O₂ was used as the positive control, and the cells were incubated in the incubator for 4 h.

We transferred the collected cells to a 1.5 mL EP tube with PBS and divided it into two parts, one for the blank control group and the other part for the same volume as the sample tube, as the No stain group. Dilute 10 × Binding Buffer with PBS, dilute BV421 Annexin V with Binding Buffe: BV421 Annexin V = 50:1 volume ratio, add 70 μL per tube to the sample tube, and place on ice for 30 min away from light. The nucleic acid dye 7-AAD was added to 1 μL per tube and left for 10 min at room temperature and protected from light (BV421 Annexin V and 7-AAD can also be stained at the same time, considering that BV421 Annexin V’s staining ability is lower than that of 7-AAD, BV421 Annexin V is stained first). Add PBS 800 μL/tube, centrifuge, discard the supernatant, and add 200 μL PBS to make cell suspension. Filter on the machine and flow cytometry for testing.

Zebrafish parents are kept in the zebrafish breeding system, quality control reference “B/T 39649-2020 experimental animals Experimental fish quality control.” Ten pairs of parental fish, male and female, were temporarily reared in 3 L paired boxes, and fertilized eggs with basically synchronized development were obtained using the photoinduction method; 3 dpf embryos were selected as experimental organisms. The experiment was set up the control group (blank control group), the GSH group (positive drug control group, 100 μM) and the subject group (200 μg/L). The incubation was continued for 24 h after administration. The rest of the operation was performed according to the standard “Rapid assay for zebrafish model of antioxidant function of T/ZHCA health food (ROS method).” After the experiment, the live imaging of each group of pups was performed using a Nikon fluorescence microscope (Ci-S), and no less than 10 fish were imaged in each group; the image grayscale values were extracted using ImageJ software, and the ROS clearance rate was calculated.

Fourier transform infrared spectroscopy spectra revealed the main functional groups of polysaccharides. The FT-IR spectra as shown in Figure 3 indicated that the IR spectra of all five components contained similar absorption peaks in the range of 4,000 to 500 cm^–1, only the absorption intensities differ. EFP-0, EFP-1, EFP-2, EFP-3, and EFP-4 can observe a broad and intense stretching peak in all FTIR spectrum around 3,200 to 3,400 cm^–1, which was mainly caused by polysaccharide molecules or intermolecular O-H stretching vibrations, indicating the presence of intermolecular hydrogen bonds (49, 50). The weak absorption bands around 2,900 to 2,800 cm^–1 (Figures 3A–E) were attributed to C-H, -CH₂-, and -CH₃- asymmetric stretching vibrations (51), and the presence of these two absorption peaks suggests that all five components purified from elderberry polysaccharides are carbohydrates, but EFP-0 is relatively strong (43). The absorption peak near 1,400 to 1,700 cm^–1 was due to the -COOH bending vibration or symmetric stretching vibration of C-O, indicating the presence of glucuronic acid in EFP and the presence of -COOH groups; in particular, EFP-2 has the largest carboxyl peak (52–54). EFP-3 has an absorption peak at 1,597.92 cm^–1, indicating the presence of N-H variable angle vibration in EFP-3 (49). The absorption peaks at 1,000 to 1,200 cm^–1 were related to the C-O-C stretch of the glycoside bonds, which have different spectral shapes for polysaccharides composed of different monosaccharides, where multiple absorption bands appeared in EFP-2, EFP-3, and EFP-4, but only 1–2 absorption bands in EFP-0 and EFP-1 (Figures 3A,B); and it is noteworthy that EFP-0, EFP-2, EFP-3, and EFP-4 (Figures 3A,C–E) have absorption peaks at 1,000 to 1,050 cm^–1, indicating the presence of the pyranose ring in these four polysaccharide fractions (55), and 1,024 cm^–1 is the peak of C-O-C stretching vibration (56). A peak around 900 cm^–1 indicates the presence of β-glycoside bonds in the polysaccharide chain, but the peaks of EFP-0 and EFP-1 are weaker, and there is an absorption peak around 850 cm^–1 indicating the presence of α-glycoside bonds in polysaccharides, but only EFP-0 has no absorption peak here (45, 57, 58) and (59). Thus, the FTIR spectra of these purified polysaccharide components all have typical absorption peaks of heteropolysaccharides.

EFP-2 was determined by X-ray diffraction to be an amorphous or crystalline structure. Generally, polysaccharides with sharp narrow diffraction peaks are crystalline structures, while polysaccharides with broad diffraction peaks are amorphous structures (60). As shown in Figure 4, the XRD profile of EFP-2 has a strong broad peak at 2θ = 20°, and a similar peak pattern was observed in the 2θ region by Zhao et al. (61) and Ji et al. (50). A new sharp crystal peak at approximately 2θ = 32, 46, and 56° each indicates that EFP-2 had an ordered crystal conception, and similar results were observed by Yuan et al. (62) and Bai et al. (63), which may be related to the content of glyoxylate, an acidic polysaccharide containing three peaks with sharp and narrow diffraction (64). This inference was consistent with the results for monosaccharide composition.

Scanning electron microscopy can be used as a qualitative tool to analyze the surface morphology of polysaccharides (64). As can be seen from Figure 5, the polysaccharide structure appears fragmented at 500 × magnification and had a rough surface, which may be caused by various hydroxyl and carboxyl groups (64). At 1,000 × magnification, the EFP-2 polysaccharide surface showed agglomerates or lamellar aggregates, but no regularity overall, which may be due to the magnetic field enhancement leading to different degrees of cellular tissue damage, resulting in wrinkling, cracking, and rupture (21). At 3,000 × magnification, the surface of EFP-2 polysaccharide was a smooth sheet-like structure with tighter binding and surface distribution, possibly due to stronger interactions between polysaccharide molecules (65, 66).

Methylation analysis is the most widespread and effective method for determining the type and linkage of polysaccharide glycosidic bonds (44, 67). To analyze the linkage mode of monosaccharides in EFP-2, the polysaccharide structure was analyzed by methylation, the reaction products were hydrolyzed and acetylated sequentially, and the resulting final products were analyzed by GC-MS to obtain information on the linkage mode of monosaccharides.

Figure 6 showed that the total ion flow diagram of the methylation products and the analysis of the polysaccharide methylated glycol acetates results were listed in Table 3. It can be seen that EFP-2 included 12 glycosidic bond types, and 1,4-Glcp (34.90%), 1,4-Manp (32.70%), and 1,4,6 -Glcp (10.80%) were the three major glycosidic bond linkage mode, with small amounts of t-Galp (5.10%), t-Manp (4.40%), t-Araf (2.50%), t-Glcp (2.40%), 1,4,6-Galp (1.90%), 1,5-Araf (1.90%), 1,3-Galp (1.30%), 1,6-Galp (1.00%), 1,3,6-Galp (1.00%), and 1,3,6-Galp (0.50%) residues. It can be seen that the EFP-2 backbone may be joined by 1,4-α-Glcp and 1,4-α-Manp, and that 1,4-α-Glcp produced a branched chain at the C-6 position.

As can be seen in the methylation results, the number of glycosidic bonds differs from the composition of the monosaccharides measured, mainly due to the fact that methylation analysis is more qualitative than quantitative, and the percentage content of GC-MS is different due to the lower sensitivity of ion chromatography. In addition, EFP-2 has good solubility, and the use of DMSO reagents can avoid the problem of insufficient methylation of polysaccharides, so the glycosidc bond type determined by the methylation experiment is reliable (68).

According to the ¹H NMR spectrum (Figure 7A), EFP-2 had five main anomeric proton signals at δ 5.22, 5.10, 5.09, 4.70, and 4.45, which were labeled A, B, C, D, and E, respectively. The chemical shifts from 3.21 to 4.44 ppm in the ¹H NMR spectrum were assigned to protons from C-2 to C-6 in the residues. The corresponding anomeric carbon signals in the ¹³C NMR spectrum were labeled with reference to the HSQC data. All the ¹H and ¹³C NMR (Figure 7B) of labeled residues were assigned with data from the ¹H-¹HCOSY (Figure 7C) and HSQC spectra (Figure 7D). In ¹³C NMR spectrum(Figure 7B), four isomers with heteropolycarbon resonance signals were at δ_C 90 to 110 ppm (107.30, 101.59, 107.48, 104.03, and 102.91 ppm). This indicated that EFP-2 contained both α and β configurations in its structure (69), which was consistent with the results of the FTIR spectrum. All the ¹H and ¹³C signals were assigned as completely as possible according to 2D NMR analysis and literature values. The chemical shift in the anomeric proton of residue A was δ 5.22. The corresponding chemical shift in the anomeric carbon was δ 107.30. The other protons of residue A were assigned from the COSY spectrum. The other corresponding carbon and hydrogen signals were δ 79.65 (4.30), 76.38 (4.11), 84.03 (4.10), and 62.54 (3.82, 3.72), determined with HSQC (Figure 7D). Based on these NMR data, we inferred that T-α-L-Araf (1→ was a constituent unit of EFP-2 Combined with the analysis, we inferred that the →4) –Manp (1→ was assigned as residue B, →5) -α-L-Araf (1→ as residue C, T-β-D-Galp (1→ as residue D, and →4) -β-D- Glcp (1→ as residue E (71–73). Taking the results of methylation analysis into consideration, it was supposed that the presence of two different terminal signals was suggested as T-α-L-Araf (1→ and T-β-D-Galp (1→ in EFP-2, as well as comparing their NMR data with those in references (Table 4; 70). In the HMBC spectrum, some inter-residual cross-peaks were observed: B-H-1 to E-C-4 (Figure 7E). In the NOESY spectrum, some inter-residual cross-peaks were observed: B-1 to E-4/D-3 and C-5; E-1 to C-5/D-3 and E 4; D-1 to C-5/D-3 and E 4 (Figure 7F). The determination of the monosaccharide composition analysis confirmed that EFP-2 was mainly composed of galactose, arabinose, glucose, and mannose. Based on its monosaccharide composition and 1D and 2D NMR spectroscopy results, EFP-2 was proposed to comprise two units of →4) –Manp (1→4) -β-D-Glcp (1→ and →4) -β-D-Glcp (1→ 5) -α-L-Araf (1→.

Compared with the model group (Figure 8A), the survival rate of cells under 1.2 mM H₂O₂ conditions both increased with the increase of polysaccharide concentration, which acted as a protective anti-free radicals in vitro, indicating that EFP-2 had a strong protective effect on RAW264.7 cells with H₂O₂-induced damage. It was shown that cell viability was enhanced with increasing concentrations, suggesting a dose-dependent relationship between polysaccharide concentration and cellular activity (12.5–200 μg/mL), which was consistent with the studies reported by Wang et al. (74) and Xie et al. (75). In Figure 8B, the cells were mostly round, with smooth surfaces and uniform size in clusters (76). Cell density and cell viability increased significantly after 24 h treatment with EFP-2 (12.5–200 μg/mL), which became more pronounced with increasing EFP-2 concentration, suggesting that FFP-2 could inhibit morphological changes in RAW264.7 cells and protect them from H₂O₂-induced oxidative stress. This may be because the inhibitory effect of EFP-2 on cellular death induced by H₂O₂ oxidative stress is determined by monosaccharides, which can provide the ability of cells or control oxidative stress (77). Studies have shown that many plant polysaccharides can scavenge excess free radicals and play an antioxidant role, and Li et al. (78) have concluded that red mushroom polysaccharides also have a significant protective effect on the activity of H₂O₂-induced RAW264.7 cells.

Studies have shown that H₂O₂-induced oxidative stress can induce DNA damage, inflammation, tissue damage, and apoptosis, which is a key target of most chemotherapy drugs because it is the primary mechanism for clearing damaged cells (75). As shown in Figure 9, the percentage of apoptosis in the H₂O₂ group was 0.56%, the percentage of apoptosis in RAW264.7 cells treated with EFP-2 was significantly reduced to less than 0.50%, and the percentage of apoptosis in RAW264.7 cells gradually decreased to 0.48, 0.46, and 0.37% with the increase of EFP-2 concentration (12.5–50 μg/mL). Compared with VC, RAW264.7 cells treated with EFP-2 (<25 μg/mL) had a weaker apoptotic activity, which was enhanced when the concentration of EFP-2 (50 μg/mL) was increased to a certain concentration. These results suggested that EFP-2 can participate in the process of programmed suicide to remove non-functional, unwanted, abnormal, and harmful cells, reducing the death of RAW264.7 cells. Thus, EFP-2 was shown to protect cells from H₂O₂-induced oxidative stress.

Zebrafish are often used as biological models for in vivo studies of small molecules and drugs because of their fast growth rate, small size, transparency, and ease of handling (77, 78). Glutathione, as a key antioxidant element, with the formation of its disulfide dimer, can respond to chemical stress and clear ROS, thereby balancing the intracellular redox homeostasis, helping to prevent oxidative stress in cells, and maintain a normal immune system (74, 79). The intensity of green fluorescence is positively correlated with the level of ROS, and the higher the brightness, the higher the level of ROS. When the level of ROS increases in the body, it will consume the antioxidant substances in the body and disrupt the homeostasis of the antioxidant defense system in the body. If the synthesis rate of antioxidant substances in the body is slow or the clearance rate of ROS is not fast, it will cause the peroxidation level of lipids and proteins in the body to increase, leading to the occurrence of chronic diseases and causing aging and other physiological phenomena related to peroxidation in the body (22, 80). Compared with the blank control group (Control group), both the positive drug control group (GSH group) and the subject group (EFP-2 group) could significantly reduce the overall ROS level of zebrafish litter, and the ROS clearance rate was 50.51 and 44.62% (Table 5), respectively. After the addition of EFP-2, the fluorescence intensity of zebrafish was significantly reduced (Figure 10), and the ROS level in zebrafish was decreased, which improved its antioxidant level, directly reflecting that EFP-2 has an antioxidant ability. In addition to the direct antioxidant response, polysaccharides may also activate some signaling pathways that protect zebrafish from oxidative stress damage. For example, Yang et al. (40) added HJP-1a at doses of 5, 25, and 50 μg/mL, which can increase the fluorescent spots of zebrafish embryos by 60.8 to 83.3%, and then achieve the best effect on the protective effect of oxidative damage cells.