Comparative characterization of pectin fractions from pomegranate peel: Structure–Function relationships

Abstract

This study evaluated structure-function relationships of four pectin derived from pomegranate peel: alcohol-insoluble residue (AIR), water-soluble pectin (WSP), homogalacturonan (HG)-enriched pectin, and rhamnogalacturonan-I (RG-I)-enriched pectin. AIR and WSP exhibited higher molecular weight (M_w) and moderate degree of methylation/acetylation (DM/DA), showing superior water- and oil-holding capacities (13.88–20.90 g/g and 4.14–4.65 g/g respectively), excellent emulsifying activity index and emulsion stability index (EAI: 12–16 m²/g, ESI: 80–91%), and strong antioxidant activity. HG-enriched pectin, rich in galacturonic acid (GalA) (74.24–89.94%) and low DM (<5%), formed strong calcium-induced gels (gel strength up to 190 g) but had limited emulsifying and holding capacities. RG-I-enriched pectin, with abundant neutral sugar branches, demonstrated moderate overall functionality. Structure-function correlation analysis revealed that high M_w, moderate DM/DA, and branching enhanced emulsifying, water-holding, oil-holding, and antioxidant properties, while linear HG structures promoted gel formation. These results support the targeted use of pomegranate peel pectin fractions as functional ingredients in food applications.

Highlights

• •High DM and DA improved emulsification but weaken the gel-forming ability.
• •High DA and branching enhance the antioxidant capacity of pectin significantly.
• •AIR and WSP showed strong emulsification, WHC, OHC and antioxidant properties.
• •HG-enriched pectin exhibited gel strength in calcium-induced systems.
• •Pectin functional depended on Mw, DM, DA, MR3 and GalA content.

1. Introduction

Today, pomegranate (Punica granatum L.) is widely cultivated across numerous countries, including Israel, Afghanistan, China, Japan, India, Pakistan, Egypt, Australia, South Africa, and the United States (Holland et al., 2009). Pomegranate fruits are abundant in anthocyanins, polyphenols, flavonoids, vitamin C, and other natural bioactive compounds (Al-Said et al., 2009). These compounds are renowned for their health-regulating properties and multifunctional medicinal benefits, which include antioxidant, anticancer and anti-inflammatory effects. Consequently, pomegranates have become highly popular among consumers. Over the past decade, there has been a marked increase in the demand for pomegranate-derived products, including juices, wines, jams, and jellies. However, pomegranate processing also generates a substantial number of by-products (Giri et al., 2023). Studies report that processing 1 ton of fresh pomegranate fruit in the juice industry produces approximately 669 kg of by-products, comprising 78% peel and 22% seeds (Qu et al., 2009). Among these by-products, pomegranate peel stands out as one of the most valuable in the food industry. It is particularly rich in pectin (10%–20%), as well as polyphenols, proteins, organic acids, and alkaloids (Rahmani et al., 2017), highlighting its potential as a valuable biological resource.

Pectin is a vital acidic heteropolysaccharide located in plant cell walls (Harris and Smith, 2006). It is composed of approximately 17 different monosaccharides and characterized by a backbone of α-1,4-linked D-galacturonic acid (GalA) residues (Kumar et al., 2023). Structurally, pectin is generally divided into three principal domains: homogalacturonan (HG), rhamnogalacturonan-I (RG-I), and rhamnogalacturonan-II (RG-II). Among these, HG is the most abundant domain, comprising 57% to 70% of total pectin. It consists of a linear chain of linear chain of α-1,4-linked GalA residues, forming the “smooth zone” (Kumar et al., 2023). RG-I, the second most prevalent domain in pectin, comprises 20%–35% of total pectin and features a repeating disaccharide backbone of [→2)-α-L-Rhap-(1 → 4)-α-D-GalpA-(1→] (Niu et al., 2024). RG-II, the most structurally complex pectin domain, is highly conserved across plant species and constitutes about 10% of pectin. Both RG-I and RG-II possess highly branched side chains rich in neutral sugars and are collectively referred to as “hairy region” (Kumar et al., 2021).

Pectin exhibits significant biological activities (e.g., anti-inflammatory, immunomodulatory, and hypoglycemic effects) as well as important functional properties such as emulsification and gelation. These characteristics are intrinsically linked to its botanical origin, extraction methods, and molecular structure, including molecular weight, degree of methylation (DM), degree of acetylation (DA), and D-galacturonic acid (GalA) content (Kumar et al., 2021; Pang et al., 2024; Qiao et al., 2024; Wu et al., 2020). The industrial functionality of pectin is strongly influenced by its structural domains. Homogalacturonan (HG)-rich pectin forms stable gel networks extensively utilized in the food and pharmaceutical industries. For example, hawthorn-extracted HG-LMP produces denser and more stable hydrogels than commercial counterparts (Bu et al., 2022; Willats et al., 2006). In contrast, rhamnogalacturonan-I (RG-I)-enriched pectin exhibits enhanced immunomodulatory and antioxidant activities and performs well in pH-induced, sugar-free gel systems. For instance, apple-derived RG-I pectin shows improved gel strength, storage modulus (G′), and water-holding capacity (Liu et al., 2024; Zhang, He, et al., 2021). In addition, RG-I-enriched pectin shows promising potential in emulsion-based foods, as its abundant neutral sugar side chains promote molecular cross-linking and form thicker oil-water interfacial layer, thereby improving long-term emulsion stability (Niu et al., 2023; Qiu et al., 2024).

Previous studies on pomegranate peel pectin have primarily focused on evaluating individual functional properties using bulk or partially purified extracts. For example, Yang et al. (2018) reported that pomegranate peel pectin can function as an effective emulsifier whose performance dependent on pH and ionic conditions, while Abid et al. (2017) characterized Ca²⁺-mediated gelation behaviors across several cultivars and reported cultivar-dependent differences. However, these studies largely followed a “single extract–single functionality” framework. More recently, a limited number of studies have begun to investigate domain-related fractions (e.g., WSP, HG-, and RG-I-enriched pectin) for specific processing-related questions, such as the effects of high-pressure homogenization on anthocyanin-pectin interactions (Li, Li, Peng, et al., 2024), or ethanol-based fractionation combined with structural characterization (Balli et al., 2022). Nevertheless, an explicit and application-oriented structure-function mapping for pomegranate peel pectin remains limited, lacking systematic comparisons of different hierarchical components (AIR, WSP, HG, and RG-I enriched pectin).

Therefore, this study aims to comprehensively investigate the physicochemical characteristics of pomegranate peel pectin with distinct structural domains and to explore the relationships between molecular structure and functional properties. To this end, four fractions (AIR, WSP, HG-, and RG-I-enriched pectin) were extracted from four pomegranate cultivars and thoroughly analyzed for their structural features and key functional properties (water- and oil-holding capacities, emulsifying and gelation abilities, and antioxidant properties). By establishing a domain-to-function linkage across multiple fractions and cultivars, this study offers a scientific basis for the targeted valorization of pomegranate peel pectin and the selection of specific germplasm resources for application-driven processing. Moreover, the domain-based functional differentiation identified herein may guide future fractionation and structural tailoring strategies to optimize pectin performance for specific industrial applications.

2. Materials and methods

2.1. Materials and chemicals

Pomegranate fruits from four varieties at the ripening stage were collected from Yunnan Province, China: ‘Guang Yan’ (GY), ‘Lv Zi’ (LZ), ‘Suan Tian’ (ST), and ‘Tunisia’ (TS). The pericarp was manually harvested and stored at −80 °C. Alcohol oxidase and monosaccharide standards, comprising arabinose (Ara), xylose (Xyl), fucose (Fuc), galactose (Gal), galacturonic acid (GalA), rhamnose (Rha), glucose (Glc), mannose (Man), and glucuronic acid (GlcA), were obtained from Beijing Solarbio Science & Technology Co., Ltd. (Beijing, China). The methanol standard was obtained from Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Penta-O-acetyl-β-D-glucose and vitamin E standards were obtained from Shanghai Yuanye Biotechnology Co., Ltd. (Shanghai, China). All other chemicals and reagents were purchased from Aladdin Biochemistry and Technology Co., Ltd. (Shanghai, China) and Macklin Biochemical Co., Ltd. (Shanghai, China). All experimental procedures were carried out using distilled water.

2.2. Pectin extraction from pomegranate peels

2.2.1. Alcohol insoluble residue (AIR)

Based on a previously reported method (Abid et al., 2017) with slight modifications, thawed pomegranate peels were first subjected to a 90 °C water bath for 3 min to deactivate pectinase enzymes. Subsequently, the peels were oven-dried at 50 °C until reaching a constant weight, then ground into a powder using a pulverizer (BJ-800 A, Hangzhou Bai Jie Technology Co., Ltd., Hangzhou, China). The resulting powder was dispersed in distilled water at a solid-to-liquid ratio of 1:20 (w/v), and the pH was adjusted to 1.70 using a 1 M HNO₃ solution. Extraction was carried out in a water bath at 86 °C for 80 min. The mixture was subsequently cooled and centrifuged at 8000 ×g for 5 min (Sorvall LYNX 6000, Thermo Fisher Scientific Inc., Waltham, USA), and the supernatant collected. Pectin was precipitated using ethanol (95%, 3:1, v/v), and let stand for 12 h, then centrifuge at 8000 ×g at 4 °C for 15 min to recover the precipitate, followed by freeze-drying at −80 °C for 48 h using an LGJ-25JY freeze dryer (Ba Fang Zhong Da Science and Technology Development Co., Ltd., Beijing, China). The dried AIR was stored in airtight containers under desiccated conditions.

2.2.2. Water-soluble pectin (WSP)

WSP was extracted using a modified method reported by Zhang, Xu, et al. (2022). AIR was dissolved in distilled water at a ratio of 1:200 (w/v), and the solution was centrifuged at 15,000 ×g for 15 min. The resulting supernatant was dialyzed in distilled water by a 3.5 kDa dialysis membrane (Beijing Solarbio Science & Technology Co., Ltd.) for 72 h at 4 °C. The solution was freeze-dried at −80 °C for 72 h and stored in dry containers.

2.2.3. Homogalacturonan (HG)-enriched pectin

HG-enriched pectin extraction was conducted following Santiago et al. (2018) with some modifications. AIR was dissolved in distilled water at a ratio of 1:200 (w/v), the pH was adjusted to 12.00 with pre-cooled 1 M NaOH solution and stirred continuously for 6 h at 4 °C. This alkaline pretreatment was applied to saponify ester bonds, thereby facilitating the isolation of pure polygalacturonic acid-rich HG fractions. Subsequently, the pH was adjusted to 1.00 with 1 M HCl solution incubated in a water bath at 80 °C for 24 h. The acid treatment selectively cleaves portions of the RG-I regions, yielding fractions predominantly composed of linear HG (smooth) domains. After extraction, 95% ethanol was added at a ratio of 4:1 (v/v), followed by thorough mixing and centrifugation at 15,000 ×g for 15 min. The obtained precipitate was dried at 40 °C for 12 h, redissolved in distilled water at a ratio of 1:100 (w/v), stirred overnight, then centrifuged at 15,000 ×g for 15 min. The supernatant was collected, and its pH was adjusted to 6.00 using 1 M LiOH solution. The final solution was dialyzed, freeze-dried, and stored in desiccated containers.

2.2.4. Rhamnogalacturonan-I (RG-I)-enriched pectin

RG-I-enriched pectin was extracted following the method by Santiago et al. (2018) with some modifications. AIR was dispersed in distilled water at a ratio of 1:200 (w/v), then water-bathed at 90 °C for 30 min. The pH of the suspension was then adjusted to 12.00 using preheated 1 M NaOH solution, and then the solution was water-bathed at 90 °C for 2 h under gentle stirring. The solution was cooled, ethanol (4:1, v/v) was added, and the mixture was centrifuged at 15,000 ×g for 15 min to collect the precipitate. The obtained precipitate was dried at 40 °C for 12 h, redissolved in distilled water at a ratio of 1:100 (w/v), and stirred overnight at room temperature. The solution was then centrifuged again at 15,000 ×g for 15 min to collect the supernatant. Then, its pH was adjusted to 6.00 using 1 M HCl solution. The final solution was dialyzed, freeze-dried, and stored in desiccated containers.

2.3. Composition and structural analysis of pectin

2.3.1. Galacturonic acid (GalA) content

The m-hydroxybiphenyl colorimetric method was used to quantify GalA (Blumenkrantz & Asboe-Hansen, 1973). A Pectin (0.1 mg/mL) solution was mixed with 6 mL of H₂SO₄ containing 0.0125 M Na₂B₄O₇, and then cooled in an ice-water bath. The mixture was vortexed thoroughly and heated in an oil bath at 100 °C for 5 min, cooled, and reacted with 40 μL of 0.125 M NaOH solution containing 0.009 M 3-phenylphenol. After 5 min, the absorbance at 520 nm was recorded using a UV–Vis spectrophotometer (T9CS, Beijing Puxi General Instrument Co., Ltd., Beijing, China). A standard curve was generated using GalA standards concentrations ranging from 0 to 0.1 mg/mL.

2.3.2. Monosaccharide composition

Monosaccharide composition was analyzed using high-performance liquid chromatography (HPLC) after derivatization with PMP, following a previous method reported by Qiao et al. (2024). Twenty milligrams of pectin were dissolved in 2 mL of 2 M TFA, sealed under nitrogen, and hydrolyzed at 120 °C for 2.5 h. Then, the solution was evaporated with a nitrogen stream, redissolved, and washed three times with methanol to remove residual acid. The final product was dissolved in 1 mL distilled water, centrifuged at 15,000 ×g for 15 min, and derivatized by sequentially adding 100 μL of 0.6 M NaOH solution and 100 μL of 0.5 M PMP methanol solution at 70 °C for 100 min. The reaction mixture was then neutralized with 100 μL of 0.6 M HCl solution, and excess PMP was removed by three extractions with 1 mL portions of chloroform. The obtained solution was filtered by a 0.45 μm aqueous membrane prior to HPLC analysis. The analysis was conducted on an HPLC system (Vanquish, Thermo Fisher Scientific Inc., Waltham, USA) equipped with an Agilent 5 TC-C18 (2) column (4.6 mm × 250 mm, 5 μm, Agilent). Mobile phase A comprised 15% acetonitrile and 85% potassium dihydrogen phosphate (0.1 M, pH 6.9), while mobile phase B comprised 40% acetonitrile and 60% potassium dihydrogen phosphate (0.1 M, pH 6.9). A linear gradient was applied, with phase A decreasing from 100% to 60% and phase B increasing from 0% to 40% over 50 min. The flow rate was set at 1.0 mL/min, column temperature was set at 25 °C, and detection wavelength was set at 250 nm. Standard monosaccharides, including xylose (Xyl), mannose (Man), arabinose (Ara), glucuronic acid (GlcA), galactose (Gal), glucose (Glc), fucose (Fuc) and rhamnose (Rha) were used for identification and quantification.

2.3.3. Molar mass distribution

Molecular weight was analyzed by high performance size-exclusion chromatography (HPSEC) coupled with multi-angle laser scattering (MALLS) (Dawn Heleos II, Wyatt, USA), a refractive index (RI) detector (OptilabrEX, Wyatt, USA) and ultraviolet detector (L-2400, Hitachi, Japan) (Yang et al., 2018). A 1 mg/mL pectin solution was eluted with 0.1 M NaCl solution at 0.5 mL/min (30 °C). The refractive index increment (d_n/d_c) was determined to be 0.138 mL/g. ASTRA 5.3.4 software was applied to calculate the weight-average molecular weight (M_w), number-average molecular weight (M_n), and polydispersity index (M_w/M_n).

2.3.4. Degree of methylation (DM)

The alcohol oxidase method was used to measure DM as described by Klavons and Bennett (1986). In brief, pectin was hydrolyzed with 2 M NaOH solution. Following hydrolysis, 1 mL of hydrolyzed pectin solution and 1 mL of alcohol oxidase solution (1 U/mL) were added sequentially to a test tube. The mixture was mixed thoroughly and incubated at 25 °C for 15 min in a water bath. After the reaction, 2 mL of acetylacetone solution, prepared from 2 M ammonium acetate, 0.02 M 2,4-acetylacetone, and 0.05 M acetic acid, was added. The solution was vortexed and further incubated at 58 °C for 15 min. Two milliliters of distilled water was added to the solution after cooling to room temperature, and mixed thoroughly. The absorbance was recorded at 412 nm using a UV–Vis spectrophotometer (T9CS, Beijing Puxi General Instrument Co., Ltd., Beijing, China). A standard curve was generated by using methanol standards (0–50 μg/mL). The DM value of pectin was calculated as the molar ratio of methoxy groups to galacturonic acid (GalA).

2.3.5. Degree of acetylation (DA)

DA was analyzed using the hydroxylamine colorimetric method (Hestrin, 1949). Briefly, 1 mL of 3 mg/mL pectin solution, 1 mL of 0.1 M hydroxylamine hydrochloride solution, and 1 mL of 1.5 M NaOH solution were added sequentially to the test tube. The mixture was thoroughly mixed and left to stand for 20 min. Subsequently, 0.7 mL of 2 M HCl solution was added, and the mixture was stirred for another 20 min. Finally, 2 mL of a 0.37 M FeCl₃ solution was added, followed by stirring for 10 min. Absorbance was recorded at 500 nm using a UV–Vis spectrophotometer (T9CS, Beijing Puxi General Instrument Co., Ltd., Beijing, China). A standard curve was prepared using β-D-glucose pentaacetate solution with concentrations ranging from 0 to 300 μg/mL. The DA value of pectin was calculated as the molar ratio of acetyl groups to galacturonic acid (GalA).

2.3.6. Z-average size (Z-avg) and zeta potential (ζ-potential)

A nanoparticle size and potential analyzer (Nano-ZS90, Malvern Instruments, UK) was used to analyze the Z-average size and ζ-potential (Panwar et al., 2023). A 0.1% (w/v) pectin solution was prepared, and its pH was adjusted to 4.00 using either 0.1 M HCl or NaOH solution, as necessary. Then the sample was transferred into a cuvette for measurement, with the refractive index set to 1.33 and the temperature controlled at 25 °C.

2.3.7. Fourier transform infrared spectroscopy (FT-IR)

FT-IR spectra were recorded following the method of Liu et al. (2024) with some modifications. Freeze-dried pectin samples were mixed with dried KBr powder (spectral grade, 1:100, w/w), ground to a fine powder using an onyx mortar and pressed into pellets using a mold. Then, the samples were scanned by a FT-IR spectrometer (BRUKER VERTEX 70, Germany) ranged at 4000–400 cm⁻¹ with a resolution of 4 cm⁻¹and 32 scans per sample.

2.3.8. X-ray diffraction (XRD)

XRD analysis was conducted using a powder X-ray diffractometer (Bruker D8 Advance, Germany), according to the method of Li et al. (2023), with small modifications. Freeze-dried pectin samples were ground into a fine powder using an agate mortar. The powder was then scanned at 2°/min from 10° to 60°. The degree of crystallinity (I%) was calculated using eq. (1):

$$I\left(\%\right)=\left(\frac{A_c}{A_c+A_a}\right)\times 100$$ (1)

Where A_c and A_a are the diffraction peak area of the crystalline region and the diffraction area of the non-crystalline part, respectively.

2.3.9. Scanning electron microscopy (SEM)

Microtopographic images were obtained using SEM (TESCAN MIRA LMS, TESCAN Trading Ltd., Czech Republic) following the method of Li et al. (2025), with slight modifications. The freeze-dried pectin samples were mounted on a copper stub using conductive adhesive for gold spraying, and then placed in the SEM for observation. The images were captured at an accelerating voltage of 2–5 kV.

2.3.10. Differential Scanning Calorimetry (DSC) and Thermogravimetric Analysis (TGA)

The thermal properties of pectin samples were determined using DSC and TGA according to the methods of Panwar et al. (2023), with small modifications. For DSC analysis, 4 mg pectin samples were accurately weighed and sealed in aluminum trays. Then, the samples were transferred to a PerkinElmer DSC 4000 (USA) with an empty aluminum pan as the reference. Thermal analysis was performed across a temperature range of 25 to 400 °C with a nitrogen flow rate of 50 mL/min, with a uniform heating rate of 10 °C per minute. For TGA, pectin samples were analyzed using a TGA 4000 PerkinElmer thermogravimetric analyzer (USA). The temperature range was set from 30 to 600 °C under a nitrogen atmosphere at 50 mL/min with a heating rate of 10 °C/min.

2.3.11. Determination of total phenol content and viscosity

The total phenol content (TPC) of the crude pectin extract was determined using the Folin-Ciocalteu method. Briefly, 1 mL of pectin solution (1 mg/mL) was mixed with 1.5 mL of Folin-Ciocalteu reagent and 1 mL of 75% sodium carbonate solution. The mixture was incubated in the dark at room temperature for 1 h, after which the absorbance was measured at 765 nm. the TPC was quantified using a gallic acid GA standard curve and expressed as GA equivalents (GAE) (He et al., 2025).

Rheological properties were evaluated by measuring the apparent viscosity of a 2% pectin solution over a shear rate of 0.01–1000 s⁻¹ at 25 °C using a rheometer (Anton Paar MCR 102, Austria). The flow behavior was recorded to characterize the rheological properties of the pectin samples (Sun et al., 2025).

2.4. Functional properties analysis of pectin

2.4.1. Water holding capacity (WHC)

The WHC of pectin was determined following the methods of El Fihry et al. (2024) with some modifications. Freeze-dried pectin samples were dispersed in distilled water at a ratio of 1:20 (w/v) and vortexed for 2 min. The mixtures were then centrifuged at 4000 ×g for 30 min, after which the supernatant was carefully decanted. The remaining residue was then weighed, and WHC was calculated as the amount of water retained per gram of pectin (g/g), using the following eq. (2):

$$\mathit{WHC}\left(g/g\right)=\frac{m_2-m_1}{m_0}$$ (2)

where m₀ is the weight of the pectin sample, m₁ is the weight of the sample and the centrifuge tube, and m₂ is the weight of the centrifuge tube and residue after centrifugation.

2.4.2. Oil holding capacity (OHC)

The OHC was analyzed according to method of El Fihry et al. (2024) with some modifications. Freeze-dried pectin samples were dispersed in olive oil at a ratio of 1:10 (w/v) and vortexed for 2 min. The mixture was then centrifuged at 4000 ×g for 30 min, and the supernatant was carefully removed. The remaining residue was weighed, and the OHC was calculated as the amount of oil retained per gram of pectin (g/g) using eq. (3):

$$\mathit{OHC}\left(g/g\right)=\frac{m_4-m_3}{m_0}$$ (3)

where m₀ is the weight of the pectin sample, m₃ is the weight of the sample and the centrifuge tube, and m₄ is the weight of the centrifuge tube and residue after centrifugation.

2.4.3. Emulsifying properties

The emulsification properties were assessed using the methodology reported by Yang et al. (2023), with some modifications. A 1% (w/v) pectin solution was prepared, and its pH was adjusted to 4.00 using 0.1 M NaOH or HCl solution as needed. An emulsion was prepared by combining 15 mL of pectin solution and 5 mL of olive oil, followed by high-speed shear (T25, IKA, Germany) mixing at 15,000 rpm for 3 min. Fifty microliters of the emulsion were then immediately added to 10 mL of 0.1% (w/v) aqueous sodium dodecyl sulfate (SDS) solution, thoroughly mixed, and the absorbance was measured at 500 nm (A₀). After the emulsion was allowed to stand for 24 h, another 50 μL emulsion was added to 10 mL of SDS solution, and the absorbance was recorded again at 500 nm (A₂₄). SDS solution served as the blank control. The Emulsifying Activity Index (EAI) and Emulsifying Stability Index (ESI) were calculated using eqs. (4), (5), respectively.

$$\mathit{EAI}\left(m^2/g\right)=\frac{2\times 2.303\times A_0\times D}{C\times \left(1-\mathrm{\Phi }\right)\times 10000}$$ (4)

$$\mathit{ESI}\%=\frac{A_{24}}{A_0}\times 100$$ (5)

where D is the dilution factor, C is the pectin concentration (g/mL) before emulsion formation, Φ is the oil volume fraction, A₀ is the absorbance at 0 h, and A₂₄ is the absorbance at 24 h.

2.4.4. Gelling properties

The gelling properties of pectin were analyzed based on the method reported by Liu et al. (2024) with some modifications. Pectin (2%, w/v) and sucrose (70%, w/v) were dissolved in distilled water under continuous magnetic stirring at 50 °C. The pH was adjusted to 3.50 with 0.5 M citric acid and sodium citrate solution. The solution was then heated and maintained at 80 °C for 20 min. Subsequently, 1.2 mL of 0.5 M CaCl₂ was added to achieve a Ca²⁺ concentration of 120 mg/g of pectin. After vigorous stirring to ensure homogeneous distribution of calcium ions. The pH of the mixture remained at 3.50, as the small volume of added CaCl₂ did not significantly alter the acidity. The mixture was then stored at 4 °C for 24 h to allow gel formation. Gel strength was determined using a TA. XT Plus texture analyzer (Stable Micro Systems, UK). The gels were compressed using a cylindrical probe (P 0.5, diameter 12.7 mm) at a constant speed of 1 mm/s until 50% deformation of the gel height, and the corresponding force was recorded as the gel strength (Li et al., 2023).

2.4.5. Antioxidant properties

The antioxidant properties of pectin were analyzed following the methods of Qiao et al. (2024) with minor modifications. A 1% (w/v) pectin solution and a 0.2 mM DPPH methanol solution (with an absorbance value of approximately 0.9 at 517 nm) were prepared. Subsequently, 900 μL of the DPPH solution was combined with 100 μL of pectin solution, and the mixture was placed in the dark at 25 °C for 1 h. Following the reaction, absorbance at 517 nm was recorded using a UV–Vis spectrophotometer. A standard curve was constructed using vitamin E solutions with concentrations ranging from 0 to 300 μmol/L.

Additionally, a 7 mM ABTS stock solution was mixed with an equal volume of 2.45 mM potassium persulfate (K₂S₂O₈) solution, followed by dark reaction for 12 to 16 h. Prior to use, the resulting mixture was then diluted 20-fold to achieve an absorbance of approximately 0.9 at 734 nm. Then, 900 μL of the ABTS mixture was combined with 100 μL of pectin solution. The reaction was conducted in the dark at 25 °C for 6 min, and the absorbance at 734 nm was recorded by a UV–Vis spectrophotometer. A standard curve was also prepared using vitamin E solutions with concentrations ranging from 0 to 300 μmol/L.

2.5. Statistical analysis

All experiments were conducted in triplicate, and the results were presented as the mean ± standard deviation (SD). Data analysis was conducted using SPSS software (version 27; SPSS Inc., Chicago, U.S.A.). One-way analysis of variance (ANOVA), followed by Duncan's test, was used to compare the sample groups, and differences with p < 0.05 were considered statistically significant. Graphs were generated using Origin 2021 (Origin Lab Corp., MA, USA).

3. Results and discussion

3.1. Physicochemical properties of pectin

3.1.1. Yield and molecular weight

Four pectin fractions, including AIR, WSP, HG-enriched pectin, RG-I-enriched pectin, were extracted from four pomegranate varieties (GY, LZ, ST, and TS). The extracted fractions were designated as AIR-GY, WSP-GY, HG-GY, and RG-I-GY, with similar nomenclature applied to the other varieties. The yields of each pectin component, presented in Table 1, showed significant differences among the fractions within the same pomegranate variety. Overall, AIR had the highest yield (6.05%–7.82%), followed by WSP (5.09%–6.98%) and RG-I-enriched pectin (2.68%–3.67%), and HG-enriched pectin had the lowest yield (0.43%–0.59%). These findings are consistent with previous studies. For instance, Abid et al. (2016) reported AIR yields of 6.4% to 11% from pomegranate peel using acid hydrolysis. Similarly, Li, Li, Peng, et al. (2024) reported extraction yields of 6.83%, 1.69%, and 2.82% for WSP, HG-enriched pectin, and RG-I-enriched pectin from pomegranate peel, respectively, closely matching our results. Additionally, notable cultivar-dependent variations were observed. For example, LZ yielded more AIR (7.8%) than ST (6.1%), likely reflecting differences in genetic backgrounds and environmental growth conditions (Liu et al., 2022).

Table 1.

Yield, molecular weight parameters and chemical composition of different pectin fractions from pomegranate peels.

	GY				LZ				ST				TS
	AIR	WSP	HG	RG-I	AIR	WSP	HG	RG-I	AIR	WSP	HG	RG-I	AIR	WSP	HG	RG-I
Yield (%)	6.62± 0.20dAB	5.83± 0.15cB	0.43± 0.04aAB	3.19± 0.11bB	7.82± 0.36dB	6.98± 0.22cC	0.59± 0.05aB	2.68± 0.22bA	6.09± 0.37dA	5.09± 0.14cA	0.38± 0.12aA	3.39± 0.05bBC	7.05± 0.11dAB	6.05± 0.22cB	0.55± 0.01aAB	3.67± 0.13bC
Mw (kDa)	920.85 (±1%)	1086.50 (±1%)	12.86 (±7.71%)	110.89 (±0.73%)	1216.00 (±2%)	1231.00 (±1%)	11.57 (±10%)	416.4 (±0.7%)	1267.21 (±1%)	716.90 (±0.8%)	12.28 (±2%)	266.15 (±0.97%)	1347.62 (±2%)	1092.44 (±1%)	11.25 (±2%)	48.86 (±2.36%)
Mn (kDa)	838.21 (±1%)	842.33 (±1%)	8.17 (±7.51%)	41.53 (±0.82%)	773.8 (±1%)	716.7 (±1%)	7.77 (±9%)	93.65 (±0.7%)	1150.13 (±1%)	470.9 (±0.8%)	9.42 (±7%)	180.15 (±0.97%)	1190.46 (±2%)	858.56 (±1%)	6.57 (±2%)	28.52 (±3,18%)
Mw/Mn	1.07 (±2%)	1.22 (±1%)	1.57 (±10%)	2.68 (±1.09%)	1.57 (±2%)	1.72 (±2%)	1.49 (±14%)	4.45 (±1%)	1.08 (±2%)	1.52 (±1%)	1.31 (±10%)	1.47 (±1%)	1.10 (±2%)	1.21 (±1%)	1.71 (±3%)	1.66 (±4.18%)
DM (%)	50.48± 0.45dC	48.38± 0.13cC	4.09± 0.12aB	6.55± 0.12bD	42.19± 0.34dB	43.39± 1.14cA	4.00± 0.10aAB	5.90± 0.07bC	48.85± 0.18dA	40.17± 0.18cA	4.65± 0.27aC	5.88± 0.27bB	48.90± 0.18dC	37.83± 0.12cB	3.88± 0.04aA	5.38± 0.19bA
DA (%)	24.79± 0.74dB	16.92± 0.86cB	1.33± 0.11aB	9.50± 0.57bA	19.68± 0.54dA	11.51± 0.28bA	0.34± 0..14aA	16.76± 0.54cB	27.73± 0.22dC	18.74± 0.61cC	1.03± 0.29aB	9.56± 0.42bA	29.07± 0.31dD	16.53± 0.23cB	0.98± 0.11aB	10.19± 0.51bA
Z-avg	1816.3 ± 59.50dC	1566.0 ± 63.51cC	1000.60± 42.27aC	1232.00 ± 41.68bD	1414.3 ± 107.22dA	996.30 ± 51.54cAB	622.63 ± 104.84aB	829.50 ± 31.16bC	1427.6 ± 97.01cA	1052.6± 41.10bB	557.73± 21.77aA	598.80± 51.11aA	1582.3 ± 96.46dB	955.23± 17.97cA	598.23± 20.32aAB	685.40± 69.30bB
ζ-potential	−28.56± 1.30aA	−32.61± 2.01bB	−33.90± 1.48aB	−28.84± 1.48bB	−30.87± 1.80bB	−31.92± 2.32bAB	−34.00± 1.25cC	−27.15± 0.94aA	−31.98± 1.41bC	−33.36± 1.12cC	−33.72± 1.18cC	−26.59± 0.58aA	−31.89± 0.71bC	−31.16± 0.34bA	−32.16± 0.70bA	−26.94± 0.52aA
TPC (mg GAE/g)	32.50± 0.18dC	19.97± 0.53cB	7.54± 0.23aC	15.67± 0.19bA	24.70± 0.22dA	19.07± 0.22bA	6.81± 0.42aB	19.57± 0.08cD	27.81± 0.30dB	22.48± 0.16cC	5.82± 0.16aA	18.78± 0.06bC	39.43± 0.32dD	26.81± 0.15cD	7.08± 0.23aB	18.38± 0.18bB
Monosaccharides Composition (%)
Man	0.20± 0.00bcC	0.09± 0.00aAB	0.27± 0.06cB	0.15± 0.01abB	0.07± 0.00aA	0.14± 0.03bcB	0.17± 0.01cAB	0.10± 0.03abA	0.20± 0.01abC	0.12± 0.01aAB	0.28± 0.05cB	0.20± 0.00bC	0.14± 0.01bcB	0.08± 0.01aA	0.11± 0.01abA	0.16± 0.01cBC
Rha	2.19± 0.00bB	1.39± 0.01aB	4.32± 0.26cB	4.14± 0.25cD	0.89± 0.07aA	1.71± 0.08bC	2.57± 0.04cA	1.65± 0.02bA	2.00± 0.21bB	1.60± 0.05aC	5.25± 0.14dC	3.08± 0.06cB	1.89± 0.08bB	1.04± 0.05aA	2.29± 0.04cA	3.74± 0.03dC
GlcA	0.60± 0.11abB	0.51± 0.05aB	0.75± 0.02bB	1.13± 0.00cB	0.18± 0.00aA	0.51± 0.02bB	0.32± 0.01aA	0.62± 0.10bA	0.65± 0.16bB	0.35± 0.01aA	0.64± 0.10bB	0.51± 0.04abA	0.47± 0.09aAB	0.35± 0.00aA	0.41± 0.00aA	0.64± 0.08bA
GalA	64.42± 1.62bAB	69.37± 2.27cA	79.02± 0.93dB	45.46± 0.72aA	69.45± 1.15bB	77.74± 1.17cB	89.94± 1.05dC	50.70± 1.90aB	60.26± 1.24bA	70.81± 1.86cA	74.24± 1.19cA	51.69± 1.30aB	69.12± 3.46bB	76.34± 1.60cB	80.83± 1.63cB	60.51± 1.61aC
Glc	7.44± 0.94aB	15.25± 0.21bC	6.13± 0.49aAB	15.95± 0.56bB	23.27± 0.03bC	3.19± 0.09aA	3.49± 0.01aA	34.28± 0.59cD	8.18± 0.15aB	17.72± 0.52bD	6.97± 2.01aBC	24.46± 0.65cC	3.97± 0.49aA	10.01± 0.58cB	9.28± 0.53cC	7.89± 0.09bA
Gal	9.35± 0.37bB	7.88± 0.01aC	7.72± 0.24aB	19.05± 0.09cC	3.50± 0.17bA	7.87± 0.23cC	2.17± 0.05aA	8.30± 0.31cA	11.60± 0.88bcC	6.50± 0.31aB	10.37± 1.88bC	14.11± 0.22cB	8.24± 0.58bB	4.98± 0.20aA	5.71± 0.20aB	14.17± 0.09cB
Xyl	1.72± 0.10cC	0.65± 0.03aB	1.25± 0.11bAB	1.07± 0.06bD	0.63± 0.07bA	0.89± 0.02cC	0.92± 0.02cA	0.41± 0.04aA	1.88± 0.03dC	0.50± 0.06aA	1.57± 0.01cC	0.88± 0.06bC	1.29± 0.06bB	0.47± 0.01aA	1.09± 0.25bA	0.58± 0.01aB
Ara	13.50± 1.21cB	4.54± 0.12bB	0.18± 0.03aB	12.37± 0.65cB	1.84± 0.02bA	7.56± 0.06dD	0.17± 0.00aB	3.60± 0.38cA	14.41± 0.25dB	2.12± 0.23bA	0.20± 0.01aB	4.58± 0.25cA	14.34± 0.84dB	6.46± 0.40bC	0.09± 0.01aA	11.62± 0.26cB
Fuc	0.56± 0.10bA	0.32± 0.01aB	0.38± 0.02aB	0.70± 0.00bC	0.17± 0.02aA	0.40± 0.03cC	0.24± 0.01bA	0.35± 0.04cA	0.82± 0.13aA	0.27± 0.01aA	0.48± 0.08aB	0.48± 0.02aB	0.53± 0.15bA	0.27± 0.01aA	0.19± 0.01aA	0.69± 0.01bC
HG%	62.23± 0.00bB	67.98± 0.01cA	74.70± 0.26dB	41.32± 0.25aA	68.57± 0.07bD	76.03± 0.08cD	87.37± 0.04dD	49.05± 0.02aC	58.26± 0.21bA	69.21± 0.05cB	68.98± 0.14cA	48.60± 0.06aB	67.23± 0.08bC	75.30± 0.05cC	78.55± 0.04dC	56.77± 0.03aD
RG-I%	27.24± 0.86bB	15.20± 0.11aC	16.54± 0.73aB	39.70± 0.24cD	7.11± 0.06aA	18.84± 0.02cD	7.49± 0.03aA	15.19± 0.73bA	30.02± 0.21dC	11.83± 0.45aA	21.07± 2.17bC	24.86± 0.60cB	26.36± 0.10cB	13.52± 0.49bB	10.37± 0.29aA	33.26± 0.21dC
MR1	2.36± 0.09bB	4.69± 0.05cA	5.71± 0.23dB	1.22± 0.02aA	9.89± 0.05cD	4.22± 0.01bA	14.80± 0.01dD	3.55± 0.18aD	1.96± 0.00aA	6.45± 0.32cC	4.18± 0.46bA	2.24± 0.06aC	2.63± 0.04aB	5.78± 0.25bB	8.64± 0.47cC	1.97± 0.01aB
MR2	0.03± 0.00bC	0.02± 0.00aB	0.05± 0.00cB	0.09± 0.01dC	0.01± 0.00aA	0.02± 0.00bC	0.03± 0.00cA	0.03± 0.00dA	0.03± 0.00bC	0.02± 0.00aC	0.07± 0.00dC	0.06± 0.00cB	0.03± 0.00bB	0.01± 0.00aA	0.03± 0.00bA	0.06± 0.00cB
MR3	10.42± 0.37dB	8.93± 0.17cB	1.83± 0.06aB	7.61± 0.64bB	6.05± 0.69bA	9.04± 0.50cB	0.91± 0.03aA	7.21± 0.35bB	13.08± 1.69cB	5.38± 0.49bA	2.01± 0.30aB	6.06± 0.03bA	11.96± 0.64cB	11.04± 1.12cC	2.53± 0.05aC	6.90± 0.00bAB

Mean ± standard deviation (n = 3). Different lowercase letters indicate significant differences between different fractions of pectin from the same variety, and different uppercase letters indicate significant differences between the same fractions of pectin from different varieties (p < 0.05).

M_w is the weight-average of molecular mass; M_n is the number-average of molecular mass. DM is the degree of methylation, DA is the degree of acetylation. Z-avg is the Z-average size. TPC is the total phenolic content.

Man, Rha, GlcA, GalA, Glc, Gal, Xyl, Ara and Fuc are mannose, rhamnose, glucuronic acid, galacturonic acid, glucose, galactose, xylose, arabinose and fucose, respectively.

HG% = GalA - Rha; RG-I% = (GalA - HG%) + Rha + Gal + Ara; MR1 = GalA / (Rha + Ara + Gal + Xyl + Fuc); MR2 = Rha / GalA; MR3 = (Ara + Gal) / Rha.

The molecular weight (M_w) distribution of pectin plays a key role in its bioactivity, emulsification capacity, and rheological behavior (Ke et al., 2020). Table 1 shows that the M_w of four pectin fractions varied significantly. AIR and WSP had the highest M_w (920.85–1347.62 kDa and 716.90–1267.21 kDa, respectively), while HG-enriched pectin (11.25–12.86 kDa) and RG-I-enriched pectin (48.86–416.4 kDa) had significantly lower M_w. The relatively low M_w of HG-enriched and RG-I-enriched pectin might be due to depolymerization during extraction, as high temperature and prolonged extraction times can degrade pectin structure (Gutöhrlein et al., 2020). Additionally, pectin's M_w is primarily influenced by the architecture of its backbone and the length and complexity of its side-chains. Previous research demonstrates that the RG-I region of potato pectin, characterized by extensive neutral sugar side chains, possesses a markedly higher M_w compared to the more linear and less branched HG domain (Yang et al., 2018). This is consistent with our findings, where extracted HG-enriched pectin had a lower M_w than RG-I-enriched pectin.

The polydispersity index (M_w /M_n) represents the molecular weight distribution of polymers, with a value of 1.0 indicating a monodisperse polymer and higher values representing a broader distribution (Liu et al., 2022). In this study, RG-I-enriched pectin exhibited relatively high M_w /M_n values (1.47–4.45), indicating a wide molecular weight distribution and significant heterogeneity among its components. This broad distribution may arise from multiple factors. First, the RG-I domain is inherently heterogeneous, characterized by variable backbone lengths and diverse side-chain branching patterns. Second, partial co-extraction of other cell wall polysaccharides, such as hemicelluloses (e.g., xyloglucans), which are closely associated or cross-linked with RG-I regions, may occur during alkaline extraction (Chandel et al., 2022). Incomplete separation of these components could further increase the polydispersity and contribute to the broadened apparent M_w distribution. In contrast, WSP and HG-enriched pectin showed more moderate M_w /M_n values, ranging from 1.21 to 1.72 and 1.31 to 1.71, respectively, indicating narrower molecular weight distributions and relatively uniform molecular structures. This more homogeneous distribution in WSP may result from its higher degree of polymerization and consistent molecular size. Meanwhile, the linear and less branched structure of HG-enriched pectin likely facilitated more uniform degradation and extraction, contributing to its narrower distribution (Mao et al., 2019). Among all the samples, AIR exhibited the lowest M_w /M_n values (1.07–1.58), approaching that of a monodisperse polymer. This suggests that AIR contained polymer with the most uniform molecular weight, possibly due to a high degree of polymerization, an intact molecular structure, and minimal degradation during extraction.

3.1.2. Monosaccharide composition

The ratio of monosaccharides is a key parameter for understanding the structural characteristics of pectin. As shown in Table 1, nine monosaccharides were detected across all samples, with GalA being the predominant component. GalA content varied significantly among different fractions, with the highest GalA levels in HG-enriched pectin (74.24%–89.94%), followed by WSP (69.37%–77.74%) and AIR (60.26%–69.45%), while RG-I-enriched pectin had the lowest GalA levels (45.46%–60.51%). These findings align with previous reports on pectin from red raspberry, where the GalA content was 88.32% in HG-enriched pectin, 68.19% in WSP, and 34.47% in RG-I-enriched pectin (Zhang, Xu, et al., 2022). In addition to GalA, eight neutral monosaccharides were detected, in descending order of overall abundance: glucose (Glc), galactose (Gal), arabinose (Ara), rhamnose (Rha), xylose (Xyl), glucuronic acid (GlcA), fucose (Fuc), and mannose (Man). Among them, Glc was the dominant neutral sugar in pomegranate peel pectin, possibly derived from residual soluble sugars after extraction or co-extracted non-pectin polysaccharides (Wang et al., 2024). Previous studies have reported that pomegranate peel pectin typically contains 61.15% GalA, 26.81% Glc, 6.57% Ara, and 4.13% of Gal (Zhuang et al., 2019). Notably, varietal differences were evident. For example, the HG-enriched fraction from TS had the highest GalA content (90%), whereas that from ST showed the lowest (74%). Such variability highlights the influence of cultivar-specific genetic and physiological factors on pectin composition.

Structurally, AIR and WSP showed high GalA content (>50%) alongside moderate Ara and Gal levels, indicating the presence of both RG-I and HG domains. HG-enriched pectin exhibited the highest GalA content (>70%) and minimal amounts of other monosaccharides, indicating a typical linear HG structure with few or no side chains. Conversely, the RG-I-enriched pectin showed lower GalA but high Rha content, consistent with its Rha-GalA repeating backbone. The elevated levels of Ara and Gal further support the presence of abundant branched side chains, which are characteristic of RG-I domain (Gao et al., 2013). During extraction, both HG- and RG-I-enriched pectin underwent alkaline hydrolysis, but HG-enriched pectin was also subjected to acid hydrolysis, where Gal and Ara were hydrolyzed, resulting in the lowest Gal and Ara content in HG-enriched pectin.

As presented in Table 1, to further elucidate pectin structure, three molar ratios (MR1, MR2, and MR3) were obtained from the monosaccharide composition. MR1 represents the degree of linearity, MR2 indicates the proportion of RG-I in the pectin structure, and MR3 reflects the degree of branching (Hu et al., 2021). The MR1 of HG-enriched pectin was higher than that of AIR, WSP and RG-I-enriched pectin, with HG-LZ having the highest MR1 (14.80), indicating the greatest linearity. Previous studies suggest that pectin with MR2 values between 0.05 and 1 are primarily composed of the RG-I region (Liu et al., 2022). All RG-I fractions, except RG-I-LZ, had MR2 values exceeding 0.05, confirming the presence of RG-I domains. These findings supported the expected structural characteristics: HG-enriched pectin predominantly exhibited a “linear structure,” while RG-I-enriched pectin featured a “hairy structure.”

3.1.3. Degree of methylation (DM) and degree of acetylation (DA)

DM and DA are critical factors influencing the pectin's functional properties, including emulsification and gelling behavior. High-methoxyl pectin (HMP, DM > 50%) forms gels in acidic conditions (pH < 3.50) with high concentrations of soluble solids, stabilized by hydrogen bonding and hydrophobic interactions between the methoxyl groups (Chan et al., 2017). In contrast, low-methoxyl pectin (LMP, DM < 50%) gels in the presence of calcium or other multivalent cations (e.g., Ca²⁺) over a broad acidic pH range (2.0–6.0), with gelation depending on both pH and soluble solids concentration (Li, Li, Wang, et al., 2024). In this study, the DM of AIR, WSP, HG-enriched pectin, and RG-I-enriched pectin ranged between 42.19%–50.48%, 37.83%–48.38%, 3.88%–4.65%, and 5.38%–6.55%, respectively. These results suggested that all pectin fractions extracted from pomegranate peels, except AIR-GY, fell within the LMP category. The significantly lower DM values observed in HG- and RG-I-enriched pectin compared to AIR and WSP might be due to demethylation under alkaline extraction conditions. Similar trends were reported by Zhang, Xu, et al. (2022), who found higher DM in WSP (38.71%) than in HG (3.57%) and RG-I (7.79%) extracted from red raspberry. In contrast, EI Fihry et al. (2022) reported a higher DM value of 69% for pectin extracted from pomegranate peels under milder conditions (70 °C water extraction for 2 h), likely due to reduced demethylation.

Higher acetylation levels have been shown to enhance pectin's emulsifying ability but reduce its affinity with cations, thereby hindering intermolecular associations and gel formation. Among the fractions, AIR exhibited the highest DA values (19.68%–29.07%), significantly greater than those of WSP (11.51%–18.74%) and RG-I-enriched pectin (9.50%–16.76%). HG-enriched pectin showed the lowest DA values, ranging from 0.98% to 1.33%. It is worth noting that the same pectin from different varieties also shows significant differences: for example, the DA of RG-I-LZ is 16.76%, which is significantly higher than the other three varieties. The DA of HG- and RG-I-enriched pectin is significantly lower than that of AIR and WSP, because HG- and RG-I-enriched pectin underwent deacetylation during the extraction process under alkaline conditions. Additionally, previous studies have found that the acetylation of pectin is mainly related to the presence of acetylated GalA in the RG-I region (Broxterman et al., 2017), which further explains why RG-I-enriched pectin has a much higher DA than HG-enriched pectin. Previous studies have reported DA values of 12% to 15% for pomegranate peel pectin (Zhuang et al., 2019). Kotani et al. (2024) reported DA values of 15%, 5%, and 44% for WSP, HG-, and RG-I-enriched pectin from cassava, respectively, highlighting the high acetyl group content in RG-I-enriched pectin.

3.1.4. Z-average size and ζ-potential

As shown in Table 1, significant variations were observed among the different fractions, with AIR exhibiting the largest Z-average size (1414.33 nm–1816.33 nm), followed by WSP (955.23 nm–1516 nm) and RG-I-enriched pectin (598.80 nm–1232 nm), while HG-enriched pectin had the smallest Z-average size (557.73 nm–1000.60 nm). Previous studies have established a positive correlation between pectin Z-average size, molecular weight, and intrinsic viscosity, indicating that pectin with higher molecular weights tend to exhibit larger Z-average size (Chen et al., 2019). Our study supported this relationship, as the Z-average size trends aligned with the molecular weight results discussed in Section 3.1.

ζ-potential, an important indicator of colloidal system stability, was also evaluated. As shown in Table 1, all pectin fractions exhibited negative surface charges, with ζ-potential values ranging from −28.86 mV to −31.98 mV for AIR, −31.16 mV to −33.36 mV for WSP, −32.16 mV to −34.00 mV for HG-enriched pectin, and − 26.59 mV to −28.84 mV for RG-I-enriched pectin. Previous studies have shown that the magnitude of negative ζ-potential was positively correlated with GalA content and decreases with lower DM values (Bachari et al., 2024). The lower ζ-potential observed in HG-enriched pectin might be explained by its high GalA content and low DM. Notably, RG-I-enriched pectin had the lowest absolute value of ζ-potential among all fractions. This could be due to its high content of neutral sugar side chains (such as Ara and Gal), which contribute little to net surface charge (Humerez-Flores et al., 2021). Additionally, these flexible side chains might form a “branched” hydration layer that increased the hydrated radius of the particle surface. As a result, the shear plane of the electrical double layer moved further away from the actual charged groups, reducing the measured absolute value of the ζ-potential (Xu et al., 2025). A similar result was found by Neckebroeck et al. (2020), who reported that HG-enriched pectin extracted from carrots exhibited a lower ζ-potential than WSP and RG-I-enriched pectin. Moreover, polysaccharides with smaller particle sizes generally exhibit more negative zeta potential values (Panwar et al., 2023). A lower ζ-potential can enhance the sensitivity of pectin to Ca²⁺ ions, which is particularly important for the formation and stability of pectin-based hydrogels (Bu et al., 2022).

3.1.5. Thermal properties

Thermal analysis is crucial for evaluating the stability of biopolymers in food processing, including baking, extrusion and drying (Li, Li, Wang, et al., 2024). DSC results (Fig. 1A–D) showed that all pectin fractions followed a similar thermal behavior pattern, displaying an endothermic and an exothermic peak. The first endothermic peak appeared between 82.47 °C and 105.51 °C, corresponding to the removal of bound water from pectin (Liu et al., 2022). The second exothermic peak appeared between 252.05 °C and 263.03 °C, marking the onset of pectin depolymerization. This process may be affected by inter-unit hydrogen bonding between GalA units and GalA conformational transition (Zhang, Fan, et al., 2020). Compared to commercial citrus pectin (T_p = 243 °C) (Liu et al., 2022), the samples showed higher thermal stability with higher T_p. Among the fractions, HG-GY pectin displayed the highest T_p value of 265.03 °C. Moreover, narrower and sharper exothermic peaks indicated a narrower and shorter melting range of pectin, a more concentrated and uniform molecular weight distribution, and a more ordered molecular arrangement (Dranca & Oroian, 2019). According to our study, the exothermic peaks of AIR, WSP, and HG-enriched pectin were narrower and sharper than those of RG-I-enriched pectin. This indicates that the molecular weight distribution of AIR, WSP, and HG-enriched pectin was concentrated, which was consistent with the molecular weight distribution measured above.

Fig. 1.

Differential Scanning Calorimetry (DSC) curves (A–D), Thermogravimetric Analysis (TG) curves (E–H), and Differential Thermogravimetric (DTG) curves (I–L) of different pectin fractions from pomegranate peels.

The TG and DTG curves of pectin (Fig. 1E–L) revealed similar thermal behavior across all pectin samples within the temperature range of 30 °C–600 °C. The thermal degradation of pectin can be roughly divided into three phases. In the first phase, occurring between 30 °C and 180 °C, a slight weight loss was observed in the TG curves, probably due to evaporation and desorption of water from the pectin (Zhu et al., 2024). Correspondingly, the same heat absorption reaction was observed on the DSC curve. The second phase involved a more significant weight loss, occurring between 180 °C and 278 °C for WSP and between 180 °C and 330 °C for AIR, HG-, and RG-I enriched pectin. During this stage, the degradation of galacturonic acid chains led to the release of carbon dioxide, carbon monoxide, water, and aliphatic compounds (Qin et al., 2022). In the third phase, which occurred between 330 °C and 600 °C, the weight of pectin decreased slightly due to the thermal decomposition of carbon (Ma et al., 2022). The temperature corresponding to the maximum thermal decomposition rate (T_max) serves as a key indicator of thermal stability, with higher T_max reflecting greater thermal stability (Zhang, Wang, et al., 2020). The results showed that HG- and RG-I enriched pectin exhibited higher T_max values compared to AIR and WSP pectin, indicating these fractions possessed superior thermal stability. Previous studies indicate that pectin thermal degradation is closely associated with β-elimination, which is promoted by methyl ester groups at the C-6 position of galacturonic acid residues (Krall & McFeeters, 1998). In this study, HG-enriched pectin obtained though alkaline treatment exhibited an extremely low degree of methylation (DM < 5%), which likely suppressed β-elimination and shifted degradation toward glycosidic bond cleavage, a pathway requiring higher activation energy. In addition, the linear and unbranched homogalacturonan backbone favored a more ordered molecular arrangement. Consistently, X-ray diffraction (XRD) revealed relatively higher crystallinity in the HG-enriched fraction, suggesting tighter aggregates stabilized by extensive intra- and intermolecular hydrogen bonding. These structural features enhanced molecular cohesion and thermal resistance, resulting in a higher maximum decomposition temperature (T_max). In contrast, highly branched or highly methylated pectins generally exhibit more amorphous and loosely packed structures and therefore lower thermal disruption. This interpretation is consistent with Liu et al. (2022), who reported a positive association between higher GalA content and improved thermal stability.

3.2. Structural characterization of pectin

3.2.1. FT-IR analysis

FTIR spectra of pectin samples were recorded within the wavenumber range of 400–4000 cm⁻¹ (Fig. 2A–D), revealing similar functional groups and structural features across all fractions. A broad absorption band between 3100 and 3600 cm⁻¹, centered at 3400 cm⁻¹, corresponded to O-H stretching vibrations resulting from intra- and intermolecular hydrogen bonding (Bu et al., 2022). The peak near 2930 cm⁻¹ was due to C-H stretching, involving CH, CH₂ and CH₃ bending vibrations (Zhang, He, et al., 2021). The peak at 1745 cm⁻¹ corresponded to the C=O stretching vibration of esterified carboxyl groups (COO-R), while strong and slightly weaker peaks at 1620 cm⁻¹ and 1425 cm⁻¹, respectively, represented the asymmetric and symmetric stretching vibrations of (COO⁻) (Qin et al., 2022). Notably, in RG-I-enriched pectin, the absence of absorption at 1745 cm⁻¹, coupled with prominent peaks at 1620 cm⁻¹ and 1425 cm⁻¹, suggested that β-elimination reaction and chemical de-esterification occurred during extraction (Zhang, Xu, et al., 2022). All pectin samples showed three absorption peaks in the 1010–1150 cm⁻¹ region, confirming the presence of a pyranose ring structure. The region between 1300 cm⁻¹ and 800 cm⁻¹ for all pectin fractions, known as the carbohydrate fingerprint region, is typically complex and challenging to interpret in detail (Hu et al., 2021).

Fig. 2.

Fourier Transform Infrared (FT-IR) spectra (GY (A), LZ (B), ST (C), and TS (D)) and X-ray diffraction (XRD) spectra (GY (E), LZ (F), ST (G), and TS (H)) of different pectin fractions from pomegranate peels.

3.2.2. Crystal structure characterization

The X-ray diffraction patterns of pectin provide insights into its crystalline or amorphous characteristics, which directly affect its solubility and functional behavior in food systems. A higher degree of amorphousness facilitated the entrapment of solutes and the formation of polymeric networks within the pectin matrix, thereby enhancing viscoelasticity and contributing to the stabilization of food texture (Sucheta and N. N., & Yadav, S. K., 2020). In general, sharp diffraction peaks are indicative of crystallinity regions, while their absence reflects an amorphous structure. As illustrated in Fig. 2E–H, all pectin samples exhibited broad characteristic peaks near 13.1° and 21.5°. The broad peak at approximately 13.1° was mainly attributed to scattering from the amorphous matrix of pectin molecules, whereas the peak near 21.5° suggested the presence of crystalline domain with some degree of regular stacking or periodic arrangement of the polymer chains. The presence of both features indicated that the pectin sample possessed a semi-crystalline, semi-amorphous nature, which is a structural characteristic commonly observed in pectin (Jiang et al., 2018). Similar findings were reported by Niu et al. (2023), who observed comparable semi-crystalline and semi-amorphous structures in HG- and RG-I-enriched pectin extracted from sugar beet.

Quantitative analysis further revealed that HG-enriched pectin exhibited the highest crystallinity, ranging from 11.58% to 16.73%, which was significantly greater than that of AIR (7.66%–9.98%), WSP (7.44%–13.72%), and RG-I-enriched pectin (7.2%–10.4%). This higher crystallinity in HG-enriched pectin might be attributed to the structural features of the HG domain, which primarily consists of linear chains of GalA. These linear chains exhibited a more uniform structure and facilitated regular chain packing, contributing to higher crystallinity. In contrast, the RG-I domain consists of an alternating Rha-GalA backbone with a large number of randomly branched neutral sugar side chains. This “hairy” and irregular architecture disrupted the orderly alignment of polymer chains, thereby lowering crystallinity. Additionally, previous studies have shown that pectin with a low DA retained freer hydroxyl and carboxyl groups, which enabled the formation of stable intermolecular hydrogen bonds and promoted the development of crystallinity or partially ordered regions (Shahin et al., 2023). In contrast, high acetylation can hinder hydrogen bond formation due to steric hindrance and a lack of available bonding sites, leading to less orderly chain packing and increased amorphousness (Nasseri et al., 2021). This also provided a structural explanation for the relatively higher crystallinity observed in HG-enriched pectin, which had the lowest DA among the fractions analyzed.

3.2.3. Microstructure characterization

The surface morphology of the different pectin fractions was observed using a scanning electron microscope (Fig. 3), revealing distinct structural variations among AIR, WSP, HG-, and RG-I-enriched pectin. AIR (Fig. 3A, E, M and I) exhibited a rough, dense surface with a prominent fiber network and compact structure. Distinct folds and irregular fiber intertwining were observed, likely due to the high cellulose and hemicellulose contents in the AIR. In contrast, WSP (Fig. 3B, F, J and N) showed a smooth, continuous sheet-like morphology with noticeable folding or curling and a small number of filamentous structures, indicating good water solubility and relatively homogeneous molecular characteristic.

Fig. 3.

Representative SEM images of different pectin fractions from pomegranate peels: GY (A–D), LZ (E–H), ST (I–L), and TS (M–P).

HG-enriched pectin (Fig. 3C, G, K, and O) exhibited long-chain filamentous structures, forming a loose and disorganized mesh, accompanied by a few irregular granules. This morphology may be associated with its highly linear backbone and low degree of branching. In comparison, RG-I-enriched pectin (Fig. 3D, H, L, and P) exhibited a compact, densely branched surface with some fragmented, flake-like structures, consistent with its high content of neutral sugar side chains and branched, heterogeneous architecture. These observations align with findings by Santiago et al. (2018), who reported that AIR from carrot puree exhibited branched structures, with RG-I-enriched pectin showing dense short branches, and HG-enriched pectin displaying extended, chain-like features with minimal branching. Similarly, Zhang, Xu, et al. (2022) found that WSP from clear red raspberry juice presented a highly branched microstructure. Their study described RG-I-enriched pectin as having abundant dendritic features resembling “hairstyles,” while HG-enriched pectin exhibited numerous elongated strands characteristic of the so-called “smooth region.”

3.3. Characterization of the functional properties of pectin

3.3.1. Antioxidant properties

Antioxidants play a critical role in maintaining food quality, prolonging shelf life, and offering potential health benefits. Fig. 4A shows that the DPPH radical-scavenging activities of AIR, WSP, and RG-I-enriched pectin ranged from 0.20 to 0.30 μmol TE/mg, 0.13 to 0.23 μmol TE/mg, and 0.10 to 0.22 μmol TE/mg, respectively, which were significantly higher than that of HG-enriched pectin, ranged only from 0.02 to 0.03 μmol TE/mg. Additionally, except for AIR-ST, AIR samples showed remarkably higher DPPH radical scavenging activities, than both WSP and RG-I-enriched pectin. For comparison, Esther del Amo-Mateos et al. (2024) reported DPPH radical scavenging activities of 0.056 μmol TE/mg and 0.17 μmol TE/mg for pectin extracted from beet pulp and beet, respectively. These results suggested that pomegranate peels pectin exhibited superior antioxidant activities. As illustrated in Fig. 4B, the ABTS radical scavenging activities followed a trend similar to that of the DPPH assay. Although the two assays showed a broadly consistent ranking, it should be emphasized that both DPPH and ABTS primarily evaluated the radical scavenging ability of in vitro electron transfer under specific conditions, which may not fully reflect the efficacy of antioxidants in complex biological lipid/water interfaces.

Fig. 4.

DPPH scavenging activity (A), ABTS scavenging activity (B), Water holding capacity (WHC) (C), and Oil holding capacity (OHC) (D) of different pectin fractions from pomegranate peels. Values are the mean of three replicates with error bars representing standard deviations. Different lowercase letters indicate significant differences between different fractions of pectin from the same variety, and different uppercase letters indicate significant differences between the same fractions of pectin from different varieties (p < 0.05).

To further explore the structural basis for antioxidant capacity, Pearson correlation analysis was conducted (Fig. 7A). The analysis revealed significant correlations between antioxidant performance of pectin and several structural parameters. Specifically, M_w, DM, DA, MR3, and Ara were significantly positively correlated with antioxidant capacity, while GalA and Rha were significantly negatively correlated. These findings suggested that pectin fractions with higher MR3, DM, and DA tend to exhibit stronger free radical scavenging abilities. Our study found that, AIR and WSP fractions exhibited stronger antioxidant activity compared to HG-enriched pectin, likely due to the retention of phenolic substances or covalently bound groups during extraction. Xiang et al. (2024) reported that beet pectin contained esterified aromatic acids (such as ferulic acid) and protein residues, which could substantially enhance its antioxidant potential. Notably, phenolic acids bound to side chains of beet pectin also acted as effective antioxidants. In contrast, the extraction of RG-I-enriched pectin involved alkaline hydrolysis, while HG-enriched pectin underwent both alkaline and acid hydrolysis, which led to the removal of phenolic compounds and consequently reduced antioxidant capacity. This interpretation is supported by our total phenol content (TPC)result. AIR exhibited the highest TPC (24.70–39.43 mg GAE/g), significantly higher than WSP (19.07–26.81 mg GAE/g) and RG-I-enriched pectin (16.67–19.57 mg GAE/g), while HG-enriched pectin has the lowest TPC (5.82–7.54 mg GAE/g).

Fig. 7.

Relationship between physicochemical properties, structural characteristics and functional properties of pectin samples: Pearson correlation coefficient analysis (A) and Principal component analysis (p < 0.05) (B). And schematic structures of pectin and schematic diagrams of the antioxidant, emulsification and gelation mechanisms of pectin (C). In the figure, the color shades of the circles corresponding to different pectin represent the strength of the functional properties of the pectin; the darker the color, the stronger the corresponding functional properties.

Moreover, the abundant neutral sugar side chains (high MR3) present in RG-I-enriched pectin may contribute to its enhanced antioxidant capacity. This branched structure not only provided more hydroxyl sites susceptible to oxidation but also facilitated the physical entrapment of free radicals (Niu et al., 2024). He et al. (2025) similarly reported that the RG-I branching structure in kiwifruit pectin enhanced antioxidant activity by increasing molecular surface area and interaction sites. In contrast, HG-enriched pectin lacked both extensive branching and phenolic-bond structures, and its linear backbone was devoid of effective antioxidant functional groups, resulting in the weakest antioxidant performance.

Acetylation may also contribute indirectly to antioxidant behavior. It has been reported that acetyl groups disrupt the intermolecular hydrogen bond network within pectin, thereby increasing its molecular flexibility and solubility. This more “open” conformation may enhance the exposure of hydroxyl groups, thereby facilitating radical-scavenging interactions (Zhang et al., 2014). In our study, the DA value of AIR was higher than that of WSP, RG-I-, and HG-enriched pectin, while the DA value of HG-enriched pectin was the lowest, which further supported its limited antioxidant capacity. However, the literature contains contradictory findings. For example, Chen et al. (2024) observed a negative correlation between M_w and DM with antioxidant capacity, and a positive correlation with GalA. Furthermore, other researchers have reported no significant relationship between pectin's antioxidant capacity and M_w (Niu et al., 2024; Qiao et al., 2024). These inconsistencies highlight that pectin's antioxidant capacity of pectin is multifactorial and does not show a simple linear correlation with structural parameters such as Mw and DM. Instead, this apparent antioxidant activity is often influenced by co-extracted components, such as bound phenolics and proteins (Yan et al., 2023). Meanwhile, M_w and DM exert indirect effects by influencing polymer solubility and molecular conformation. In addition, antioxidant evaluations is assay-dependent, reflecting different mechanisms such as radical scavenging or metal chelation. Therefore, the antioxidant mechanisms of pectin require further investigation.

3.3.2. Water holding capacity (WHC) and Oil holding capacity (OHC)

WHC and OHC are essential functional properties of polysaccharides. WHC reflects the capacity of a sample to absorb and retain water, which plays an important role in determining the texture, sensory attributes, and stability of food products (Kumar et al., 2021). As depicted in Fig. 4C, the WHC of AIR and WSP was the highest, ranging from 13.88 to 19.01 g/g and 12.04 to 20.90 g/g, respectively. These values were higher than those of RG-I-enriched pectin (4.16–8.44 g/g) and HG-enriched pectin (4.91–8.44 g/g). The WHC values observed in this study surpassed those previously reported for pomegranate pectin (2.87 g/g) (El Fihry et al., 2024), but remained lower than that of potato pectin (37.84 g/g) (Yang et al., 2018).

Correlation analysis (Fig. 7A) revealed that WHC was significantly positively associated with several structural features, including M_w, DM, Ara, and MR3. High M_w pectin molecules can form extensive polymer networks or colloidal systems that trap more water molecules, thereby enhancing WHC (Qin et al., 2019). The relatively low WHC of HG- and RG-I-enriched pectin was likely due to depolymerization during extraction, which decreased M_w and reduced network-forming capacity (Xu et al., 2020). Moreover, a higher MR3 contributed to the formation of a complex, three-dimensional “sugar block” network that effectively retained water (Méndez-Albiñana et al., 2025). The role of DM in WHC is complex. While low DM increases the availability of free carboxyl groups and enhances hydrophilicity, excessive charge density can lead to strong electrostatic repulsion between molecules, hindering network formation (Nguyen et al., 2019). In contrast, moderate methylation could reduce repulsion, promote closer molecular packing, and facilitate hydrogen bonding and hydrophobic interactions, thereby enhancing water retention (Singthong et al., 2004). This mechanism might explain the higher WHC observed in high DM pectin sample in this study. It was worth noting, however, that even the highest DM value observed in WSP was only around 50%, indicating that this level of methylation may be optimal for achieving a balance between hydrophilicity and network-forming ability.

Notably, the thermal property parameters T_pDSC and T_pDTG were significantly negatively correlated with WHC, indicating that thermally stable pectin tend to have lower WHC. In addition, crystallinity was found to be negatively correlated with WHC. Higher crystallinity implied a more rigid and ordered molecular structure, which limited swelling and reduced the ability to bind water (Li et al., 2019). Conversely, a more amorphous and flexible pectin conformation allowed water to penetrate and be retained within molecular voids. Additionally, the elevated WHC observed in AIR and WSP may be attributed to residual cellulose and hemicellulose in these samples, both of which are known for their strong water-binding capacities (He et al., 2023).

OHC reflects the ability of polysaccharides to absorb and retain oil, making pectin with high OHC values as an emulsifier or stabilizer in high-fat food formulations (Spinei & Oroian, 2023). Fig. 4D showed that the OHC of AIR and WSP ranged from 4.14 to 4.41 g/g and 4.82–4.65 g/g, respectively, significantly exceeding that of RG-I-enriched pectin (2.64–3.12 g/g) and HG-enriched pectin (1.84–2.45 g/g). These results were comparable to previously reported OHC values for pomegranate pectin (2.42 g/g), and potato pectin (4.40 g/g) (El Fihry et al., 2024; Jeddou et al., 2016).

According to the correlation analysis in Fig. 7A, OHC was strongly associated with hydrophobic/hydrophilic structural characteristics. M_w, DM, DA, and MR3 all exhibited significant positive correlations with OHC. High M_w pectin molecules formed thicker surface films on oil droplets, enhancing coating ability and adhesion, which improved OHC (Pourramezan et al., 2022). Furthermore, these large molecules could form gel-like networks that physically trapped oil (Qin et al., 2019), which partly explained the higher OHC of AIR and WSP. Both DM and DA are critical in enhancing oil retention because methyl and acetyl groups increased the hydrophobicity and amphiphilicity of pectin molecules (Chen et al., 2021). Pectin enriched with those groups could more readily bind to oil droplets or retain oil within hydrophobic domains of the polysaccharide matrix. Additionally, crystallinity and thermal stability (as measured by DSC and TG) were negatively correlated with OHC. This might be because high structural rigidity limited the exposure of hydrophobic sites, thereby reducing oil-binding potential (He et al., 2022).

3.3.3. Emulsifying properties

The emulsifying property of pectin is a critical factor in its application as an emulsifier and stabilizer in food systems. As shown in Fig. 5A, visual observations of freshly prepared emulsions and emulsions stored under standard conditions for 24 h revealed clear differences among the pectin fractions. Emulsions prepared with AIR and WSP exhibited a uniform milky white appearance immediately after preparation. In contrast, emulsions with RG-I-enriched pectin appeared uniformly light yellow, while those containing HG-enriched pectin were slightly cloudy with a yellow hue and visible oil separation on the surface. After 24 h of storage, AIR and WSP emulsions largely maintained their stability, with no visible phase separation–except for AIR-LZ, which exhibited slight oil floating. In contrast, all emulsions containing RG-I-enriched pectin showed clear phase separation, and those with HG-enriched pectin underwent severe phase separation and emulsion breakdown. These observations suggested that AIR and WSP fractions possessed superior emulsifying abilities compared to RG-I- and HG-enriched pectin, with the latter showing the weakest emulsification performance.

Fig. 5.

Digital photographs of fresh oil-in-water emulsions of different pectin fractions from pomegranate peels and after 24 h of storage samples and storage for 24 h (A), emulsification activity (B), and emulsion stability (C). Values are the mean of three replicates with error bars representing standard deviations. Different lowercase letters indicate significant differences between different fractions of pectin from the same variety, and different uppercase letters indicate significant differences between the same fractions of pectin from different varieties (p < 0.05).

To quantitatively assess emulsification properties, the EAI and ESI were measured. As shown in Fig. 5B, AIR exhibited EAI values ranging from 11.54 to 15.97 m²/g, while WSP ranged from 12.40 to 14.13 m²/g. These values were significantly higher than those of RG-I-enriched pectin (11.47–12.19 m²/g) and HG-enriched pectin (3.34–5.01 m²/g). Emulsification stability (Fig. 5C) followed a similar trend: AIR showed the highest ESI (78.10%–91.53%), followed by WSP (77.69%–85.53%), RG-I-enriched pectin (56.53%–69.52%), and HG-enriched pectin (40.00%–60.59%).

Correlation analysis (Fig. 7A) further revealed that pectin's emulsifying properties were closely related to structural parameters, including DM, DA, M_w, MR3, GalA, and Rha. Among these, M_w, DM, DA, and MR3 were significantly positively correlated with EAI and ESI, while Rha and GalA showed significant negative correlations. These results indicate that pectin with higher degrees of hydrophobic substitution, abundant side chains, and large molecular sizes are more capable of forming stable emulsifying films at the oil-water interface. High DM and DA levels enhance emulsification by introducing hydrophobic methoxy and acetyl groups into the pectin molecules, conferring amphiphilic properties. These hydrophobic moieties acted as anchoring points at the oil-water interface, reducing interfacial tension and enhancing droplet dispersion (Li et al., 2024). In this study, the DM and DA values of the pectin followed the order AIR > WSP > RG-I > HG, consistent with observed EAI values.

Studies have demonstrated that viscosity plays a vital role in emulsion stability; pectin with higher viscosity can effectively restrict the mobility and aggregation of oil droplets, thus improving emulsification properties (Ren et al., 2020). In contrast, low M_w pectin tend to exhibit reduced viscosity, which increased droplet mobility and the likelihood of flocculation and coalescence, ultimately compromising emulsification efficiency (Humerez-Flores et al., 2022). Our viscosity measurements support this trend, with the apparent viscosity decreasing in the order: AIR > WSP > RG-I-enriched pectin > HG-enriched pectin. Correspondingly, AIR and WSP fractions exhibited higher viscosity and superior emulsifying stability. In contrast, the HG- and RG-I-enriched pectin underwent partial depolymerization during extraction, resulting in lower molecular weight and viscosity. This decrease in viscosity weakened the steric hindrance at the water-oil interface, making oil droplets more susceptible to aggregation and explaining the lower ESI observed in these pectin fractions.

Additionally, pectin with highly branched structures generally showed superior emulsifying activity. RG-I-enriched pectin often contained covalently or non-covalently bound protein fragments, which were hydrophobic and acted as interfacial surfactants when absorbed onto oil droplet surfaces (Chen, Fu and Luo, 2016). Although the covalently bound protein content in pomegranate peel pectin was relatively low, the longer and branched neutral sugar side chains could still stabilize oil droplets through physical entanglement and steric hindrance, thereby enhancing emulsifying activity of RG-I-enriched pectin compared to HG-enriched pectin. Moreover, the RG-I side chains of pectin were highly flexible and hydrophilic. Upon adsorption onto the oil droplets surface, these side chains extend into the aqueous phase like “bristles,” forming a thick hydrated layer. This hydrated polysaccharide barrier effectively prevents the oil droplets coalescence through spatial site resistance, thus greatly improving the ESI of the RG-I-enriched emulsion (Niu et al., 2024). This mechanism further explained the higher ESI of RG-I-enriched pectin compared to HG-enriched pectin. Consistently, Ma et al. (2024) reported that highly branched watermelon rind pectin exhibited significantly better emulsifying stability than linear citrus pectin. However, the ESI of RG-I-enriched pectin was still lower than that of AIR and WSP. This may be related to its relatively less negative ζ-potential. As discussed in Section 3.1.4, the abundance of neutral sugar side chains in RG-I-enriched pectin contributes limited charge, resulting in a less negative ζ-potential compared to other fractions. Weaker electrostatic repulsion between droplets increases their susceptibility to aggregation or coalescence over time, potentially leading to phase separation (McClements & Gumus, 2016).

Furthermore, previous studies have reported that a higher proportion of Ara side chains significantly improved the emulsifying stability of pectin (Bindereif et al., 2021). The Ara molar ratio of AIR-LZ was only 1.84%, which was considerably lower than that of other AIR fractions, explaining its relatively poor ESI performance. Based on the results of this study, the AIR and WSP fractions extracted from pomegranate peel demonstrated excellent emulsifying properties and showed promise as potential natural emulsifiers for use in the food industries.

3.3.4. Gel properties

The gelation properties of pectin are critical to its application in the food industry. As shown in Fig. 6A, notable differences were observed in the macroscopic gelation behaviors among the different pectin fractions. Gels formed by AIR and WSP exhibited visible mobility, indicating that the intermolecular interactions were insufficient to establish a robust three-dimensional network. In contrast, all HG- and RG-I-enriched pectin fractions, except RG-I-LZ, formed stable gels that maintained their shape when inverted, suggesting the formation of a denser and more continuous gel matrix. Gel strength, a key indicator of hydrogel integrity, was measured for each pectin sample (Fig. 6B). Among all fractions, HG-enriched pectin exhibited the highest gel strength, ranging from 41.21 g to 190.15 g, significantly surpassing that of other fractions. Notably, under the same Ca²⁺-induced gelation conditions, gel performance varied markedly among cultivars, with the following order observed: HG-LZ > HG-TS > HG-GY > HG-ST, while RG-I-LZ failed to form a stable gel. These results indicate that varietal differences and growth conditions influence the compositional and structural characteristics of pectin, thereby modulating its gelation behavior and functional performance.

Fig. 6.

Digital photographs of gels of different pectin fractions from pomegranate peels (A), gel strength (B) and shear rate-viscosity curve (GY (C-1), LZ (C-2), ST (C-3), and TS (C-4)) of different pectin fractions from pomegranate peels. Values are the mean of three replicates with error bars representing standard deviations. Different lowercase letters indicate significant differences between different fractions of pectin from the same variety, and different uppercase letters indicate significant differences between the same fractions of pectin from different varieties (p < 0.05).

The correlation analysis (Fig. 7A) revealed that GalA content was significantly and positively correlated with gel strength, whereas DM, DA, and M_w were negatively correlated. These findings aligned with the well-established understanding of LMP gelation: LMP forms strong calcium-mediated “egg-box” structures in the presence of Ca²⁺, while high-DM pectin typically requires acidic conditions for gelation, often resulting in weaker gels (Chan et al., 2017). Additionally, the molar ratio MR1 was strongly positively correlated with gel strength, while MR3 was significantly negatively correlated. This suggested that pectin with higher HG content and fewer RG-I side chains possessed superior gel-forming capacity. Higher RG-I content, along with longer and more abundant side chains, weakened gel strength and, in extreme cases, prevented gelation altogether. Structurally, the RG-I domain acted as a flexible “hinge” or discontinuity within the HG backbone. The resulting bending and steric hindrance reduced the alignment of HG segments, disrupting the formation of ordered inter-chain associations necessary for gelation (Zhang, Zheng, et al., 2022). Particularly, a high MR3 introduced significant steric effects that interfering with crosslinking between HG regions.

Although this study observed a negative correlation between M_w and gel strength, this relationship is more likely associated with the concurrent increase in esterification levels at higher M_w, rather than reflecting an inherently inhibitory effect of molecular size on gelation. Specifically, high-molecular-weight AIR and WSP fractions formed weaker gels, which may be attributed to several factors. First, their higher degrees of methylation (DM) and acetylation (DA) relative to HG-enriched pectin result in fewer free GalA carboxyl groups available for Ca²⁺-mediated cross-linking, thereby limiting the formation of “egg-box” junction zones (Li, Fan, et al., 2024). Second, as less refined fractions, AIR and WSP may retain residual cellulose or other structural polysaccharides. These co-extracted components can dilute the effective pectin concentration and disrupt gel continuity, ultimately weakening the gel network (Gawkowska et al., 2018). In contrast, the HG-enriched pectin exhibited the highest GalA content and the lowest DM and DA values, resulting in more dissociated carboxyl groups and minimal steric hindrance. Together, these factors facilitated the formation of strong calcium cross-links, contributing to a superior gelation capacity (Zhang, Wang, et al., 2021). Moreover, a higher DA value is recognized as a major factor impeding pectin gelation. Notably, the RG-I-LZ sample failed to form a gel, likely due to its relatively high DA value of 16.76%, markedly higher than other RG-I-enriched pectin fractions.

3.4. Principal component analysis (PCA)

The PCA results are shown in Fig. 7B. The pectin samples exhibited distinct clustering on the two-dimensional score plot: AIR was primarily distributed positively along principal component 1 (PC1); WSP tended to cluster toward the positive end of PC1 and the negative end of principal component 2 (PC2); HG-enriched pectin was mainly located negatively along PC1; and RG-I-enriched pectin clustered positively along PC2. Pectin from the four structural domains showed distinct clustering trends in the score plot, reflecting significant structural differences among the pectin types. The cumulative variance explained by PC1 and PC2 was 70%, with PC1 accounting for approximately 45.4% of the total variance, primarily associated with functional properties. Variables such as gel strength and thermal stability showed strong negative contributions to PC1, whereas antioxidant activity (DPPH, ABTS), emulsifying properties (EAI, ESI), OHC, and WHC contributed positively. PC2 explained approximately 25.6% of the variance and was mainly associated with structural composition. GalA content and MR1 contributed strongly negatively to PC2, while neutral sugars such as Gal, GlcA, and Fuc contributed positively. In the PCA loading plot, variables positioned closer together indicated stronger correlation. ABTS, DPPH, WHC, OHC, and EAI were closely associated with DA, DM, M_w, whereas gel strength clustered near crystallinity, GalA, and MR1, indicating strong correlation coefficients with the correlation analysis results.

3.5. Antioxidant, emulsification and gelation mechanisms of pectin

The structural diagrams of AIR, WSP, HG-, and RG-I-enriched pectin are presented in Fig. 7C. AIR was mainly composed of pectin chains, with minor residues of cellulose and hemicellulose. WSP comprised three domains: HG, RG-I, and RG-II. HG-enriched pectin consisted mainly of long HG chains composed of GalA, with limited side chains of Rha and Gal. In contrast, RG-I-enriched pectin featured a backbone of alternating GalA-Rha residues, with abundant side chains primarily composed of Gal and Ara. Its antioxidant activity was closely related to the presence of neutral sugar side chains and acetylation. Specifically, the complex branched side chains enhance the physical ability of pectin to trap free radicals, while phenolic acids covalently attached to these side chains can directly scavenge free radicals. The emulsification mechanism was primarily influenced by molecular weight, neutral sugar side chains, methylation, and acetylation. Methoxy and acetyl groups act as hydrophobic anchors on the surface of oil droplets, facilitating interfacial adsorption. Furthermore, neutral sugar side chains may carry covalently or non-covalently bound protein fragments, which contribute to emulsification. These extended side chains also promote interfacial stabilization through physical entanglement and steric hindrance, thereby enhancing emulsifying properties. Regarding gelation mechanism, in this study, the extracted pectin was classified as LMP, and its gelation was primarily calcium-induced. In pectin with high GalA content and low DM and DA values, Ca²⁺ ions bridge de-esterified GalA residues on parallel HG chains, forming a stable three-dimensional network known as the “egg-box” structure, resulting in strong gel formation.

4. Conclusion

This study systematically compared the physicochemical properties, structural characteristics, and functional properties of different pectin fractions extracted from pomegranate peel—namely AIR, WSP, HG- and RG-I-enriched pectin. Each fraction exhibited distinct structural characteristics that were closely associated with its functional performance. HG-enriched pectin rich in linear homogalacturonan and low in methylation and acetylation, showed strong calcium-induced gelation. RG-I-enriched pectin, with abundant branched neutral sugar side chains, exhibited moderate emulsifying and antioxidant properties. AIR and WSP, with mixed HG and RG-I domains and higher molecular weight, exhibited excellent water/oil retention and antioxidant activity. These findings underscored the critical role of structural attributes such as degree of methylation, molecular weight, and the extent of branching in determining the functional performance of pectin from pomegranate peel. Tailoring these structural features through selective extraction and fractionation provides a promising strategy for designing pectin-based ingredients with targeted functionalities. Nevertheless, several limitations remain. The established structure-function relationships, currently relying on chemical assays, necessitate further validation through advanced techniques such as NMR spectroscopy, rheology, and molecular dynamics simulations. Furthermore, exploring the synergistic mechanisms between pectin and co-extracted phenolics, alongside employing targeted structural modifications (e.g., controlled de-esterification), will be essential to fully harness their functional potential.

CRediT authorship contribution statement

Rongping Li: Writing – review & editing, Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Wenhui Zou: Methodology, Data curation. Yunxia Yang: Methodology, Data curation. Xueyan Zhang: Methodology, Data curation. Linyan Zhou: Writing – review & editing, Supervision, Resources, Funding acquisition, Conceptualization.

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.

Data availability

Data will be made available on request.