2.1 Materials, pectin extraction and ethanol gradient precipitation

Dried hawthorn was obtained from Laiwu Wanbang Food Co., Ltd. All reagents used in the experiments were analytical reagents unless otherwise noted.

Fifty gram dried hawthorn was beaten with liquid at 1:30, pH was adjusted to 2.5 with the solution of 2 mol/L citric acid, and then the mixture was kept in 85°C water bath for 120 min. The extract solution was obtained by centrifuging the mixture for 15 min. The extract solution was concentrated by a rotary evaporator and then four times the volume of 95% edible alcohol was added to the concentrated solution after it was cooled to room temperature. After stirring for 10 min and depositing at 4°C for 24 h, the crude pectin was obtained by centrifugation at 5000 × g for 15 min.

The crude pectin was washed two to three times with 95% edible alcohol and dissolved in deionized water after removing excess alcohol. The solution was dialyzed for about 48 h, and the water was changed every 4 h. The purified pectin was finally obtained by rotary evaporation and freeze drying after the dialysis.

Five gram freeze-dried samples were dissolved into deionized water to a concentration of 1%. Then, 95% edible alcohol was added to the sample solution to make the ethanol content in the system 30%. Thereafter, the mixture was stirred and refrigerated (about 4°C) for 24 h, then centrifuged for 15 min at 5000 × g, and passed through a 100 mesh sieve to obtain 30% ethanol-precipitated pectin. The supernatants containing 40%, 50%, and 60% ethanol were prepared by adding edible alcohol to obtain ethanol-precipitated pectin.

2.2 Physicochemical analysis

2.3 Molecular weight determination

The molecular weight and distribution of pectin samples were analyzed using high-performance liquid chromatography equipped with a guard column of TSK (35 mm × 46 mm), columns of TSK gel G3000PWxl (300 mm × 4.6 mm), and TSK gel G4000PWxl (300 mm × 4.6 mm). The mobile phase was 0.1 mol/L NaNO₃ solution containing 0.05% NaN₃. The 3 mg/mL pectin solution was prepared using 0.05% NaN₃ solution and was filtered by 0.45 μm water filter. The column temperature, the flow rate, the injection volume, and the elution time were 40°C, 0.4 mL/min, 20 μL, and 100 min, respectively. P-82 pullulans standard (2 mg/mL) served as markers. The standard curve equation of P-82 pullulan is $\mathrm{f}\left(\mathrm{x}\right)=-0.00036{\mathrm{x}}^2+0.0413{\mathrm{x}}^2-1.703\mathrm{x}+29.37$, and R² = 0.9997 (Mw 1080, 6100, 9600, 21,100, 47,100, 107,000, 194,000, 337,000 Da).

2.4 Monosaccharide composition determination

The gas chromatographic method was used to determine the monosaccharide composition of pectin fractions (Qin et al., 2022).

2.5 FTIR analysis

The differences in functional groups of pectin samples were determined using the Nicolet iS10 FTIR spectrometer (Thermo Fisher Scientific; Li et al., 2018). The pectin fractions were mixed and pressed into a thin pallet with KBr, then the infrared spectrograms of samples were gained in the wavelength range 4000–400 cm⁻¹.

2.6 Congo red

Two microliter of 2.5 mg/mL pectin sample solution was mixed evenly with 2 mL Congo red solution (80 μmol/mL), and then different volumes of NaOH solution (1 mol/mL) were slowly added into the mixture so that the concentration of NaOH in the mixture was 0.1, 0.2, 0.3, 0.4, and 0.5 mol/mL. After 10 min at 25°C, the absorption spectra in the wavelength range 190–600 nm were scanned by UV spectrophotometer and the maximum absorption wavelength was recorded.

2.7 Thermal analysis

Differential scanning calorimetry (DSC) was applied to analyze the thermal behavior of pectin samples (200F3, Netzsch, Germany) analysis (Sharma et al., 2015). Add 20 μL deionized water to the 5 mg pectin sample into a crucible and press the crucible lid with a crucible press, and then let it stand for about 12 h to fully wet the sample. The sample was placed in the DSC test chamber with an empty crucible as reference. Nitrogen was used to purge the crucible at a speed of 20 mL/min. The test temperature and the heating rate were 30–300°C and 10°C/min, respectively.

Thermogravimetric (TG) of pectin fractions was determined using Thermogravimetric Analyzer (DTG60A, Shimadzu, Japan). Pectin samples (5–10 mg) were placed in the sample dish and heated from 30°C to 600°C at 10°C/min. Derivative TG (DTG) curves were obtained by differentiating TG values.

2.8 In vitro antioxidant activity assays of HP

2.9 Preparation of emulsion

Five microliter of 2% pectin solution was mixed with 5 mL corn germ oil, and the emulsion was gained by homogenization at 12,000 rpm for 2 min.

2.10 Statistical analysis

The data gained in triplicate was marked as means ± standard deviation, and processed by SPSS 26.0. Values of p < .05 were identified as statistically significant.

3.1 Physicochemical property and surface morphology

The DE value, protein content, and GalA content of four pectin fractions (CAHP30, CAHP40, CAHP50, and CAHP60) are shown in Table 1. The pectin fractions precipitated by ethanol classification were all low methoxy pectin. With the progress of ethanol classification, the DE values only changed in a small range and did not change regularly. The content of GalA in pectin samples showed a decreasing trend with the progress of ethanol gradient precipitation. After ethanol fractionation, the protein content of pectin samples was significantly decreased with the increase in ethanol concentration. The results of UV spectrum (Figure 1a) showed that pectin fractions had no absorption peak at 260 nm, which indicates that the four pectin samples did not contain nucleic acid; the absorption peak appeared at 280 nm indicating that the sample contained a small amount of protein. The above results showed that ethanol fractionation has a certain effect on the purification of pectin samples, but the protein in pectin samples cannot be completely removed (Hui & Gao, 2022).

From Figure 2, CAHP30 had a relatively intact large-scale structure with serrated cracks and holes on surface. CAHP40, CAHP50, and CAHP60 were incomplete pieces, and the forms have become fragmented compared to CAHP30. Among them, CAHP40 had large holes on the surface but with a relatively flat sheet structure, but all of them had been broken into different sizes. The CAHP50 and CAHP60 were no longer visible in their full mass form, becoming crumpled and curly. Therefore, we inferred that the surface morphology of pectin could be changed by ethanol fractional precipitation, and the degree of pectin surface integrity was negative with precipitation concentration (Sun et al., 2020).

3.2 Molecular weight and distribution of HP

Molecular weight of pectin (Table 1) decreased with the ethanol grading, indicating that HP with lower Mw can be obtained at high ethanol concentration. The results were similar to the results gained by Hui and Gao (2022). The polydispersity coefficient can be used to evaluate the molecular weight distribution of pectin samples. The larger the polydispersity coefficient, the wider the molecular weight distribution of the samples; otherwise, the narrower (Jiang et al., 2018). As shown in Table 1, the Mw/Mn of CAHP30, CAHP40, CAHP50, and CAHP60 were 3.29, 2.57, 2.66, and 2.77, respectively, indicating that CAHP30 had the widest molecular weight. The Mw/Mn of CAHP40, CAHP50, and CAHP60 ranged from 2.57 to 2.77, which proved the homogeneity of these three fractions (Xu et al., 2014).

3.3 Monosaccharide composition of pectin fractions

From Table 1, the content of GalA in the four fractions was above 80%, CAHP30 and CAHP50 had the same monosaccharide composition, containing rhamnose (Rha), arabinose (Ara), mannose (Man), galactose (Gal), and glucose (Glc), but the content of each monosaccharide was different. Compared with CAHP30 and CAHP50, CAHP40 contained xylose (Xyl). In addition, the branching degree and the length of pectin can be expressed by (Ara + Gal)/Rha (Qin et al., 2022). The ratio of (Ara + Gal)/Rha of CAHP30 (1.73), CAHP40 (1.82), CAHP50 (1.57), and CAHP60 (1.48) showed that the four ethanol-precipitated pectin samples had short side chains and low degree of branching. The HG content of the four pectin samples was above 99%, indicating that all the pectin obtained by ethanol precipitation was HG type. The results were different from the HP gained previously the proportion and composition (Bu et al., 2022; Chen et al., 2019; Li, Zhang et al., 2022; Qin et al., 2022; Roman et al., 2021; Sun et al., 2020; Zhang et al., 2022), which indicated that ethanol precipitation has a certain effect on the monosaccharide conformation of pectin and CAHP30, CAHP40, CAHP50, and CAHP60 were new pectin polysaccharides.

3.4 Thermal analysis

The exothermic peaks of the four pectin samples were different (Figure 3a), which reflected the unique degradation characteristics of the samples. DSC results showed that the degradation temperature of different samples (CAHP30: 119.6°C, CAHP40: 123°C, CAHP50: 122.9°C, and CAHP60: 128.3°C) and their enthalpy change (CAHP30: 3450 J/g, CAHP40: 2916 J/g, CAHP 50: 4120 J/g, and CAHP60: 4234 J/g) were different. It showed that the degradation temperature of pectin fractions increased with the increase in ethanol precipitation concentration, indicating that ethanol fractionation can improve the thermal stability of pectin, and the thermal stability was positively correlated with ethanol concentration. The reason for the change in degradation temperatures may be that the chemical composition of pectin samples was changed by ethanol fractionation, which was related to the DE value, Mw, and GalA content of the pectin samples (Sun et al., 2020). It was worth mentioning that CAHP40 had the sharpest exothermic peak, which indicated that it had the shortest melting range and concentrated molecular weight distribution. Previous data on molecular weight (Table 1) corresponded to this result. To sum up, CAHP60 was more resistant to high temperatures, which made it a potential additive for heating food.

The TG curves of the pectin samples are shown in Figure 3b. Three degradation stages were shown in the range 30°C–600°C. The first stage occurred between 40°C and 180°C, and the mass loss of pectin samples was around 18%–22%, which may be due to the evaporation of free and bound water in the sample (Qin et al., 2022). The mass loss of pectin was the largest in the second stage, which occurred between 200°C and 330°C. The depolymerization of pectin led to the degradation of the pectin galacturonic acid chain and side chain (Jiang, Qi et al., 2020; Xu et al., 2022). The third stage was between 330°C and 600°C, and the quality reduction related to the burning of aromatic carbon (Norcino et al., 2020). The mass residue at 600°C of CAHP60 was about 21.72%, the residual amount of CAHP40 and CAHP50 was about 18.58% and 17.95%, and the residual amount of CAHP30 was 2.54%, which was significantly lower than other samples.

According to the DTG curve (Figure 3c), in the first degradation stage, the peak of CAHP40 was at 65°C, which was significantly lower than that of other samples, indicating that more water was combined in CAHP40 and more mass was lost by water evaporation. The peak temperature of CAHP30 was 80°C, indicating that it had less water content. The decrease in peak temperature in the second stage was generally caused by the decrease in molecular weight due to the destruction of pectin molecules, and lower molecular weight corresponds to lower peak temperature (Jiang, Qi et al., 2020). The molecular weights of the four pectin fractions were different, but their peak temperatures varied in a very small range. We speculated that the ethanol fractional precipitation did not destroy the pectin molecules and could not cause a significant change in the peak temperature.

3.5 FTIR spectroscopy

Infrared spectroscopy is one of the important techniques for analyzing the primary structure of pectin. According to Figure 1b, four fractions of pectin obtained by ethanol fractional precipitation showed similar characteristic absorption peaks of polysaccharides, and the peak intensity showed a gradually increasing trend. The absorption peaks of the four pectin at 3354 cm⁻¹ were caused by the tensile vibration of O–H (Li, Zhang et al., 2022). The absorption peak at 2944 cm⁻¹ was the C–H stretching vibration (CH, CH₂, and CH₃ groups; Jiang, Xu et al., 2020). Both of these were typical of polysaccharide absorption. The absorption peaks near 1739 and 1612 cm⁻¹ were the tensile vibration of esterified carbonyl C=O and the free carboxyl COO⁻. The DE values of the previous four pectin grades were calculated from these two peak areas (Muñoz-Almagro et al., 2017). The peaks at 1440 cm⁻¹ are due to the bending vibration of C–H. The smaller peaks between 900 and 1200 cm⁻¹ represented the “fingerprint” region of carbohydrates (Chen et al., 2021). Weak absorption peaks at 881 cm⁻¹ indicated α-glycosidic and β-glycosidic bonds. The results showed that ethanol grading did not change the group type of pectin (Yang et al., 2015).

3.6 Congo red

Pectin cannot form secondary structures like α-helix and β-pleated sheet, which is different from proteins. But some pectin molecules are able to form triple helix structure, which affects the biological activity of pectin. Therefore, it is necessary to figure out whether HP has a helix structure and the effect of ethanol fractionation on its triple helix structure. Congo red can be used for judging the triple helix structure of HP. Congo red can form a complex with pectin with a triple helix. In low-concentration alkaline solution, the maximum absorption wavelength (λ_max) of the composite shifts to a longer wavelength in the visible light range, it is called a red shift phenomenon occurs (Giese et al., 2008).

As shown in Figure 1c, compared with the Congo red solution, the pectin samples obtained by ethanol fractionation precipitation formed complexes with Congo red, and their λ_max increased. The λ_max of CAHP30 reached the maximum at the concentration of 0.1 mol/L of NaOH, and that of CAHP40, CAHP50, and CAHP60 reached the maximum at the concentration of 0.2 mol/L of NaOH. Then, with the increase in NaOH concentration, the λ_max of the four fractions decreased, but it was always higher than the blank group. The results showed that the four different fractions had triple helix structure. Further analysis of Figure 1c showed that among the four samples, the wavelength of the red shift of CAHP60 was 7.5 nm, which was greater than CAHP30 (4.5 nm), CAHP40 (6 nm), and CAHP50 (6.5 nm). Results showed that ethanol fractionation had an effect on the molecular conformation of pectin samples, mainly because the conformation of the pectin molecule was related to its molecular weight distribution (Wang & Li, 2019), and ethanol fractionation affected the molecular weight distribution of HP.

3.7 Emulsifying properties of pectin samples

3.7.1 Appearance, microscopy, and particle size analysis of emulsion

As shown in Figure 4a, no demulsification occurred during the preparation of all emulsions. E-CAHP40 and E-CAHP50 showed a tendency of phase separation after 1 h of preparation. After the storage of 12 h, the phase separation of E-CAHP40 and E-CAHP50 was slight. The degree of phase separation of E-CAHP40 and E-CAHP50 showed an increasing trend after 1 day of storage, while E-CAHP30 and E-CAHP60 had no significant change. After storage for 30 days, E-CAHP40 and E-CAHP50 had obvious continuous-phase precipitation, and E-CAHP30 and E-CAHP60 remained stable throughout the observation period. The results showed that CAHP30 and CAHP60 had excellent emulsifying ability, while CAHP40 and CAHP50 were not suitable for use as emulsifiers.

Microscopic images of the emulsions after storage for 30 days are shown in Figure 4b. We found that E-CAHP40 and E-CAHP50 had a demulsification phenomenon, which indicated that CAHP40 and CAHP50 had poor emulsifying properties. E-CAHP30 had smaller droplet size than E-CAHP60, and the study demonstrated that the smaller droplet size is beneficial to the stability of the emulsion (Jia et al., 2015). The particle size of all sample emulsions after storage for 30 days was less than 50 μm, and the droplet size distribution was generally consistent with the stability and microstructure of the emulsion (Figure 5a).

3.8 Rheological properties of emulsion

As shown in Figure 6a, the emulsions prepared from different pectin samples have different viscosities, and the viscosity curves of the four sample emulsions all show shear-thinning characteristics. But for E-CAHP40, with increasing shear rate, we observed three stages, the shear thinning-shear thickening-shear thinning curve (Jia et al., 2015). Ren et al. (2020) demonstrated that high-viscosity continuous phase can inhibit droplet motion and phase separation, resulting in better stability of the resulting emulsion. Increasing the concentration of protein in the aqueous phase or the rearrangement of droplets caused by excessive shear forces can enhance the intermolecular interactions, resulting in a corresponding increase in the apparent viscosity of the emulsion (Jiang, Xu et al., 2020; Liu et al., 2019). Pectin with high molecular weight had a longer molecular chain than low-molecular-weight pectin, it had more active binding sites, and it can form an osmotic network structure with high viscosity and elastic modulus faster (Cao et al., 2020).

According to Figure 6b, it can be seen that with the change in ethanol concentration, G′ and G′′ of E-CAHP30, E-CAHP50, and E-CAHP60 all intersected at lower frequency. Below this frequency, G′ was higher than G′′, and above this frequency, G′′ was greater than G′, indicating that the emulsion was transformed from elastomer to solution. The G′ of E-CAHP40 was higher than that of G′′, but there were intersection points between G′ and G′′ at high frequency, which indicated that E-CAHP40 exhibited a property between the entangled network structure and the weak gel structure, and the weak gel structure was easily destroyed into the entangled network structure at high frequency. The loss tangent angle (tan δ) was the ratio of G′ to G′′. When tan δ < 1, the system mainly exhibited elastic characteristics, and when tan δ > 1, the system mainly exhibited viscous characteristics. Figure 6c showed that E-CAHP50 almost presented solution characteristics in the process of increasing frequency, E-CAHP40 showed that weak gel structure was destroyed, and E-CAHP30 and E-CAHP60 showed a relatively stable trend of transition from elastomer to solution.

3.9 In vitro antioxidant activity assays of HP

As shown in Figure 7a, the hydroxyl radical scavenging rate of all pectin fractions showed an escalating trend with the rise of concentration of pectin samples from 0.2 to 2 mg/mL. The IC₅₀ values of pectin samples were 0.041 mg/mL (VC), 0.831 mg/mL (CAHP30), 0.697 mg/mL (CAHP40), 0.623 mg/mL (CAHP50), and 0.551 mg/mL (CAHP60), indicating that the hydroxyl radical scavenging rate was improved with the ethanol grading. Research by Li et al. (2013) and Wang et al. (2022), respectively, demonstrated that polysaccharides with low molecular and high GalA content had better hydroxyl radical scavenging ability. Polysaccharides with low molecular weight cause more hydroxyl groups to be exposed in pectin, and the higher content of GalA made the electrophilic group of pectin more abundant, which was beneficial to the release of hydrogen in O–H bonds (Wang et al., 2022; Yan et al., 2016).

The ABTS radical clearance rate of pectin presented concentration dependence (Figure 7b). The IC₅₀ values of the samples were 0.016 mg/mL (VC), 2.547 mg/mL (CAHP30), 1.944 mg/mL (CAHP40), 1.618 mg/mL (CAHP50), and 1.188 mg/mL (CAHP60), which indicated that ethanol fractionation could significantly promote the ABTS clearance rate, and the ABTS scavenging rate was positively correlated with ethanol concentration. The ABTS scavenging rate of pectin was correlated with the molecular weight, mainly because the lower molecular weight exposes some of the active groups of pectin, producing more hydrogen that helps to scour free radicals (Wu et al., 2018).

The absorbance of pectin was positively correlated with its reducing power, and higher absorbance indicated strong reducing power and antioxidant power (Chen & Huang, 2019). As shown in Figure 7c, the absorbance value of samples reached the maximum at 10 mg/mL and absorbance value of CAHP60 was 0.299, the results showed that CAHP60 had better antioxidant capacity, indicating that the reduction of molecular weight of pectin promoted the improvement of its reducing power (Chen, Shi et al., 2016), the reason is that the reduction in molecular weight of pectin causes structural changes that expose more reducing ends and produce more reductones, which exhibit antioxidant activity by providing hydrogen atoms (Chen, Shi et al., 2016; Li et al., 2013; Sun et al., 2020).