Apple pomaces derived from mono-varietal Asturian ciders production are potential source of pectins with appealing functional properties

Carbohydrate Polymers 264 (2021) 117980 Available online 24 March 2021 0144-8617/© 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Apple pomaces derived from mono-varietal Asturian ciders production are potential source of pectins with appealing functional properties In´ es Calvete-Torre a , b , Nerea Mu˜ noz-Almagro c , M. Teresa Pacheco c , María Jos´ e Ant´ on d , Enrique Dapena d , Lorena Ruiz a , b , Abelardo Margolles a , b , Mar Villamiel c , *, F. Javier Moreno c a Group of Functionality and Ecology of Beneficial Microorganisms (MicroHealth), Dairy Research Institute of Asturias (IPLA-CSIC), Paseo Río Linares s/n, 3300, Villaviciosa, Asturias, Spain b Health Research Institute of Asturias (ISPA), Avenida Hospital Universitario s/n, 33011, Oviedo, Asturias, Spain c Group of Chemistry and Functionality of Carbohydrates and Derivatives, Institute of Food Science Research, CIAL (CSIC-UAM), Nicol´ as Cabrera, 9, Campus de Cantoblanco, Universidad Aut´ onoma de Madrid, 28049, Madrid, Spain d The Regional Agrifood Research and Development Service (SERIDA), Carretera AS-267 PK 19, 33300, Villaviciosa, Asturias, Spain ARTICLE INFO Keywords: Apple pomace Pectin Cider processing Dietary fiber Ultrasound assisted-extraction Phenolic content Carbohydrates ABSTRACT Comprehensive chemical characterization of nine mono-varietal apple pomaces obtained from the production of ciders with PDO is described. They were rich in essential minerals, fibers (35–52.9 %), and polyphenols. High levels in GalA (11.8–21.6 %), revealed the suitability of these apple pomaces as efficient sources of pectins. Extracted pectins showed high variability in monomer composition, with degrees of methylesterification, strongly associated with pectins functional properties, ranging from 58 to 88 %. For a subset of apple pomace varieties, pectin extraction was accomplished by conventional acid heat treatment or ultrasound. Despite ultrasound-assisted extraction did not improve pectin yield, it minimized levels of “non-pectin” components as revealed by the low content of Glc/Man, leading to the obtainment of high-purity pectin. Our work highlights the key role played by the selection of the apple variety to streamline the potential food applications (gelling/ thickening agents or prebiotics) of the extracted pectins that largely depend on their structural features. 1. Introduction Apple pomace is the most abundant by-product of the ciderprocessing industry. Asturias (northern Spain) is one of the largest producers of this alcoholic fermented beverage worldwide, with over 45,000 tons of cider apples processed in odd-numbered years with high production, due to the biannual bearing, which, in turn, generate, approximately, 12,000 tons of apple pomace. Different varieties of apple used for making cider are Protected under some Designation of Origin (PDO), and thereby the products derived from them are marked with a specific logo. The labelling PDO pretends to differentiate its products through marketing strategies, which aims at increasing the perceived quality of products by consumers (Apaolaza, Hartmann, Echebarria, & Barrutia, 2017). Consequently, consumers’ perception of two products with similar flavor and color could be influenced by brand name, driving the choice and loyalty to one of these products marked with PDO (Mora, Elzo-Aizarna, Rozas-Fuertes, Velilla-Echeita, & V´ azquez-Araújo, 2020). In 2002, the PDO “Cider from Asturias” quality mark was granted to cider made in this region with specific local apple varieties grown in the region. In 2019, 7,500 tons of cider apples were used for the production of PDO Cider of Asturias, which generated, approximately, 2,000 tons of apple pomace. Initially, 22 local varieties were recognized under the PDO “Cider from Asturias” (Dapena & Bl´ azquez, 2009) and, recently, 54 other varieties were included. Some studies have been carried out on the biochemical composition, the volatile compounds, and physical-chemical properties, Abbreviations: Ara, arabinose; Aw, Water activity; DM, degree of methyl esterification; DNS, 3,5-dinitrosalicylic acid; DW, dry weight; DPPH, 2,2-diphenyl-1picrylhydrazyl; ELSD, Evaporative Light Scattering Detector; FT-IR, Fourier-Transform Infrared Spectroscopy; Fuc, fucose; Gal, galactose; GalA, galacturonic acid; GC-FID, Gas Chromatography with Flame Ionization Detector; Glc, glucose; HG, homogalacturonan; HPSEC, High Performance Size Exclusion Chromatography; ICPMS, Inductively Coupled Plasma Mass Spectrometry; Man, mannose; Mw, molecular weight; NMR, Nuclear Magnetic Resonance; PDO, Protected Designation of Origin; Rha, rhamnose; RG-I, rhamnogalacturonan I; TFA, trifluoroacetic acid; TPC, total phenolic content; US, ultrasound; Xyl, xylose. * Corresponding author. E-mail address: [email protected] (M. Villamiel). Contents lists available at ScienceDirect Carbohydrate Polymers journal homepage: www.elsevier.com/locate/carbpol https://doi.org/10.1016/j.carbpol.2021.117980 Received 21 December 2020; Received in revised form 17 March 2021; Accepted 19 March 2021 Carbohydrate Polymers 264 (2021) 117980 2 which have allowed the characterization of these varieties of cider apple trees (Arias et al., 2010; Blanco et al., 1992; Díaz et al., 2011; Mangas et al., 1992; Mangas, Rodríguez, Su´ arez, Picinelli, & Dapena, 1999; Pello-Palma et al., 2016). In this context, Di˜ neiro-García, Su´ arez-Valles, and Picinelli-Lobo (2009) studied the antioxidant properties associated to specific polyphenols of apple pomace from the Asturian cider industry and ´ Alvarez et al. (2012) studied the in vitro effect of apple pomace on the control of herpes. However, the potential use of these apple varieties, and the by-product generated from cider-making, as a source of pectins with technological and functional properties of interest in the food and pharma industry have not been studied until now. Apple pomace is normally used as a livestock feed ingredient but is considered an underutilized waste product of the apple industry. According to its compositional analysis, it holds potential of being processed for its technological applications (e.g., as gelling or thickener additive, or as raw material for the production of high value products such as biofuels, antioxidants, pectins or enzymes), and nutritional and pharmaceutical values (Skinner, Gigliotti, Ku, & Tou, 2018; Waldbauer, McKinnon, & Kopp, 2017). Thus, apple pomace has been shown as an appealing alternative for its incorporation into food products due to its viscosity and water holding capacities (Wang, Kristo, & LaPointe, 2020; Yates, Gomez, Martin-Luengo, Iba˜ nez, & Martinez Serrano, 2017). In addition, Yates et al. (2017) also pointed out the multi-valorization potential of apple pomace based on the fact that materials remaining after antioxidant and pectin removal from apple pomace could still be used to design biocompatible scaffolds for osteoblasts and chondrocytes for osseous and cartilage tissue replacement therapies. Specifically, apple pomace has been found to be a good source of pectins, whose yield and properties are largely dependent on the source and the nature of the extraction process used (Kumar & Chauhan, 2010). Thus, yields of pectin extracted from apple pomace have been reported to range from 10 % to 20 % (Dranca, Vargas, & Oroian, 2020; Kumar & Chauhan, 2010; Luo, Xu, & Fan, 2019; Morales-Contreras et al., 2020). Some industries have added pectin extracted from apple pomace collected from the juice processing industry to various food formulations not only for its thickening and texturizing properties, but also for being considered a good active additive due to its biological properties, antioxidant and lipase inhibitor activities (Kumar & Chauhan, 2010; Usman et al., 2020; Wang, Chen, & Lü, 2014). Besides, apple pectin has demonstrated activity to interact both with the intestinal microbiota and the intestinal immune cells, thus acting as a prebiotic substrate capable to promote the intestinal immune barrier (Beukema, Faas, & de Vos, 2020; Licht et al., 2010). Pectin is a natural macromolecular polysaccharide that forms part of the middle lamella and primary cell walls of all higher plants. Structurally, pectins are complex polysaccharides consisting mainly of an α -1,4-linked-D-galacturonic acid (homogalacturonan, HG) backbone, which can alternate with α -1,2-linked-rhamnose units (rhamnogalacturonan I, RG-I). The HG backbone is usually partly methyl-esterified at C-6 position and/or acetyl-esterified at O-2 and/or O-3 positions. Besides, RG-I bears different side chains composed by various neutral sugars, arabinose, galactose and xylose being the most common, although glucose, mannose, fucose and glucuronic acid can also be found (Dranca et al., 2020; Morales-Contreras et al., 2020; Yapo & Koffi, 2014). Rhamnogalacturonan II (RG-II), on the other hand, has an α -1,4-linked-D-galacturonic acid backbone and includes 2-keto-3-deoxy-D-manno-octulosonic acid, 3-deoxy-D-lyxo-2heptulosaric acid, apiose, and aceric acid as side chains. In addition to these main pectin classes, HG can be more or less heavily substituted at O-2 and/or O-3 by monomers or dimers of apiose or xylose leading to apiogalacturonan (AGA) or xylogalacturonan (XGA), respectively (Ropartz & Ralet, 2020). Concretely, XGA is commonly encountered in apple (Schols, Bakx, Schipper, & Voragen, 1995). Industrially, pectin is extracted from apple pomace under acidic conditions using nitric, hydrochloric or sulphuric acids. Although the use of strong acids provides high extraction yield and time-saving advantages, it involves serious environmental problems and also negatively influences the consumer preference. Besides, strong acids can significantly affect the native pectin structure decreasing ester content and disrupting side chains, which can affect their physico-chemical and related biological properties (Kumar & Chauhan, 2010). In this line, citric acid provides several advantages including a lower impact of the pectin extraction process from plant food wastes and by-products on the environment, while better preserving the structural and physical-chemical and biological properties of the apple pomace pectin (Mari´ c et al., 2018; Minjares-Fuentes et al., 2014). In recent years, ultrasound assisted-extraction (UAE) has received considerable attention because it is a green (low energy consumption, non-thermal technology) and appealing technology alternative to conventional chemical extraction processes. Ultrasonic waves generate a cavitation effect in the solvent resulting in faster movement of molecules, and a higher penetration of the solvent into the target material, giving rise to a different type of network that may be accompanied by an enhancement of rheological properties (Anese, Mirolo, Beraldo, & Lippe, 2013; Huang et al., 2018; Moorthy et al., 2017). Only a few recent papers have dealt with the use of UAE for the extraction of pectin from Malus domestica ‘Falticeni’ apple pomace derived from fruit processing (Dranca & Oroian, 2018; Dranca et al., 2020). The use of UAE on this particular apple variety allowed the production of pectin samples with a high GalA content, Mw and high apparent viscosity, although the yield was lower than that obtained from conventional citric acid extraction (Dranca et al., 2020). In this study, we hypothesized that mono-varietal industrial apple pomaces can be valorized through the extraction of compounds with high-added value due to their potential technological and biological properties. Besides, we assumed that different mono-varietal industrial apple pomaces have significant differences in their content of valuable ingredients and, thus, a preliminary screening can help identify the best variety by-product for a specific revalorization approach. For this purpose, we conducted a comprehensive chemical characterization of nine mono-varietal industrial apple pomaces, resulting from the production of ciders with PDO. The selection of the apple varieties to be analysed was based on their high interest, due to the good agronomic behavior, resistance to diseases and quality for the production of cider, being currently some of the most cultivated varieties in Asturias. Likewise, the successful extraction of pectins from all studied apple pomace byproducts and their structural characterization (in terms of monomer composition, molecular weight and degree of methyl esterification) was accomplished, and their potential as functional food ingredients was accordingly discussed. 2. Materials and methods 2.1. Samples and chemicals Nine mono-varietal apple pomaces resulting from the extraction of juice for cider production, by means of a hydraulic press for small batch (Hafico Press, Germany), were supplied by The Regional Agrifood Research and Development Service (SERIDA) (Asturias, Spain). The nine apple varieties, protected by PDO, used were: Solarina (hereafter referred to as “Sol”), Regona (“Reg”), Raxao (“Rax”), Verdialona (“Ver”), De la Riega (“Dlr”), Xuanina (“Xua”), Collaos (“Col”), Perico (“Per”) and Durona de Tresali (“Ddt”). Apple pomace samples were lyophilized, ground with a mill (Letslab Delivering Solutions, S.L.U., Barcelona, Spain) and stored at 4 ◦C until their analysis. Analytical standards as sucrose, rhamnose, galactose, mannose, glucose, fructose, galacturonic acid, pullulan standard set (805−0.34 kDa), hexamethyldisilazane, trifluoroacetic acid, calcium chloride, sodium citrate tribasic dihydrate, 3,5-dinitosalicylic acid, phenol, sulfuric acid, citric acid monohydrate and malic acid were purchased from Sigma (St Louis, MO, USA). Ethanol (99.5 %) was provided by VWR Chemicals (Barcelona, Spain). I. Calvete-Torre et al. Carbohydrate Polymers 264 (2021) 117980 3 2.2. Chemical characterization of apple pomace samples 2.2.1. Total soluble solids, pH, water activity and dry weight determinations Content of total soluble solids of the samples was determined in triplicate using a digital refractometer 30PX/30 GS (Mettler Toledo GmBH, Schwerzenbach, Switzerland). The pH was measured using a pH meter (Mettler Toledo GmBH, Schwerzenbach, Switzerland). Water activity (Aw) was measured with AW Sprint TH-500 instrument (Novasina, Pfa¨ffikon, Switzerland). The dry weight (DW) was gravimetrically determined in an oven at 102 ◦C until constant weight according to the AOAC method 920.151 (Association of Official Analytical Chemists (AOAC), 1990a). 2.2.2. Total fat and protein content determinations Total fat was measured using a Gerhardt soxtherm extractor (Germany) following AOAC method 920.39 (Association of Official Analytical Chemists (AOAC), 2000). Protein content was determined following the Kjeldhal method (Association of Official Analytical Chemists (AOAC), 1990b) for total nitrogen using a conversion factor (total nitrogen ×6.25) to express results as protein. 2.2.3. Total carbohydrates determination Total carbohydrates were determined by the phenol-sulfuric acid method (Dubois, Gilles, Hamilton, Rebers, & Smith, 1956; Masuko et al., 2005). Thus, 167 μ L of phenol-sulfuric solution (5 % w/v) were mixed with the sample solutions. Then, 1 mL of sulfuric acid was added, and samples were mixed using a vortex. The absorbance was measured at 480 nm in a Synergy HT Multi-Mode Microplate reader (BioTek® Instruments, Inc., Winooski, USA) after 30 min of incubation at RT without agitation. Total carbohydrates were determined by comparison of absorbance values to a galacturonic acid (GalA) (G2125 ≥98 % Sigma-Aldrich) standard calibration curve performed in parallel. 2.2.4. Fiber determination Total and insoluble fiber content was measured by the modified enzymatic-gravimetric method (AOAC 985.29 and 991.42) described by McCleary et al. (2010). This method included treatment with heat-stable α -amylase, amyloglucosidase and protease. After the enzymatic treatment, the dry residue is insoluble dietary fiber (IDF). For the total dietary fiber (TDF) estimation, after the enzymatic treatment ethyl alcohol was added and then the dry residue was weighed. Soluble dietary fiber (SDF) was obtained by the difference between the total and insoluble dietary fiber. 2.2.5. Reducing carbohydrates determination Reducing carbohydrates were measured using the method described by Sumner and Graham (1921), by adding 100 μ L of 3,5-dinitrosalicylic acid (DNS) reactive to 100 μ L of the diluted sample previously located in Eppendorf tubes of 1.5 mL. The mixture was stirred and boiled for 5 min. It was then cooled in an ice bath and 750 μ L of milli-Q water was added. After shaking again, 280 μ L of the mixture is transferred to a multi-well plate and the absorbance at 540 nm was measured. The calibration curve was prepared with GalA (up to 5 mg/mL) and the data were expressed as reducing carbohydrates (g/100 g DW). 2.2.6. Mineral content determination The mineral composition of the apple pomace samples was determined in an ICP-MS Elan 6000 Perkin-Elmer Sciex instrument from the Service Interdepartmental Research (SIdI-UAM) in Madrid. Semiquantitative and quantitative analyses of the elements of interest using the external calibration method and internal standards to correct instrumental drift were carried out. 2.2.7. Phenolic content determination Total phenolic content was determined by the Folin-Ciocalteu method (Folin & Ciocalteu, 1927) following the modifications described by Soria et al. (2010). 200 mg of powder sample was added in 5 mL of methanol and the mixture was homogenized with Ultra Turrax (IKA Labortechnik, Janke and Kunkel, Staufen, Germany) at 24,000 rpm for 1 min, heated with continuous stirring (50 ◦C for 20 min), centrifugated (2000 ×g for 15 min) and the supernatant was finally filtrated (1 mL using Sep-Pak Cartridges of 0.45 μ L). Once the methanolic extracts were obtained, 100 μ L of methanol and 100 μ L of Folin-Ciocalteu 2 N were added to 100 μ L of the filtered extract and kept for 5 min at room temperature. Afterwards, 700 μ L of Na 2 CO 3 (75 g/L) were added, and left it in the dark during 20 min. Mixtures were centrifuged and the absorbance was measured in the supernatant at 735 nm using a Synergy HT Multi-Mode Microplate reader (BioTek® Instruments, Inc., Winooski, USA). Total polyphenols were determined by comparison of absorbance values to a standard curve performed with gallic acid solutions. 2.2.8. Soluble carbohydrates and malic acid determination Soluble carbohydrates and malic acid of apple pomace samples were analyzed by gas chromatography (GC) following the method described by Pacheco, Hern´ andez-Hern´ andez, Moreno, and Villamiel (2020). Ethanolic extracts were evaporated under vacuum at 43 ◦C. The dried mixtures were treated with 250 μ L hydroxylamine chloride in pyridine (2.5 % w/v) and incubated at 70 ◦C for 30 min. Subsequently, the samples were persilylated with the addition of 250 μ L of hexamethyldisilazane (HMDS) (Sigma-Aldrich, Madrid, Spain) and 25 μ L of trifluoroacetic acid (TFA), followed by incubation at 45 ◦C for 30 min. Then, the reaction mixtures were centrifuged, and supernatants were analyzed in an Agilent Technologies 7890A gas chromatograph coupled to a flame-ionization detector (Agilent Technologies, Santa Clara, California, USA), equipped with an DB-5HT column (15 m ×0.32 mm × 0.10 μ m) (J & W Scientific, Folsom, California, USA). The oven temperature rose from 150 ◦C to 165 ◦C at a heating rate of 1 ◦C/min and rose again to 380 ◦C at 10 ◦C/min, with a maintenance time of 2 min at this temperature. Data acquisition and integration were performed using Agilent ChemStation Rev. B.03.01 software (Wilmington, DE, USA). 2.2.9. Antioxidant activity (DPPH) Antioxidant activity was measured by 2,2-diphenyl-1-picrylhydrazyl (DPPH) following (Thaipong, Boonprakob, Crosby, Cisneros-Zevallos, & Hawkins Byrne, 2006; Wang et al., 2016) methods with some modifications. Briefly, 300 μ L of apple pomace sample solution (1, 1.5, 2 mg/mL) were added to the same volume of DPPH (2 nM; methanol). After vigorously vortexing, the mixture was incubated in the dark at 4 ◦C with shaking at 600 rpm for 30 min. Then, the absorbance was read (517 nm; using KcJunior Biotek) and DPPH scavenging activity was calculated according to the following equation: Scavenging activity (%) = (C−CB) − (S−SB) C−CB ×100 where, C is the control of the assay; CB, control blank; S, sample; and SB sample blank. 2.3. Pectin extraction from apple pomace Pectins were extracted from apple pomace samples conventionally and by ultrasound (bath and probe). 2.3.1. Conventional pectin extraction Conventional pectin extraction was carried out following the method by (Mierczy´ nska, Cybulska, & Zdunek, 2017) for all apple pomace varieties. A suspension of 20.4 g/L of each variety of apple pomace previously lyophilized and ground (Section 2.1) was prepared with MilliQ water, adjusting the pH to 2.5 with citric acid and keeping at 90 ◦C for 30 min. I. Calvete-Torre et al. Carbohydrate Polymers 264 (2021) 117980 4 2.3.2. Ultrasound-assisted extraction (UAE) Three varieties (Sol, Per and Reg) were selected to be extracted also by UAE at a concentration of 20.4 g/L as described in the previous subsection. The pH of mixtures was adjusted to 2.5 with citric acid and, then, placed in an ultrasonic bath with internal dimensions of 24.0 cm ×14.0 cm ×10.0 cm, a capacity of 3 L and a frequency of 45 kHz (Sonica Sweep System EP 2200, Soltec, Milan, Italy) during 15 and 30 min at 60 ◦C. This temperature was selected as it has been reported to be the maximum temperature for cavitation bubbles to coalesce with vapour bubbles, suggesting that values above this temperature might interfere with the ultrasounds (US) effect on pectin extraction (Hernoux-Villi` ere, Lassi, & L´ evˆ eque, 2013). At the same temperature (i.e., 60 ◦C), pectins from these apple pomace varieties were also obtained using an ultrasonic processor operating at a frequency of 20 kHz with maxima power values of 400 W (“probe mode”) (Digital Sonifier, Branson Ultrasonics Corporation, Danbury, CT, USA). A microtip horn of 12.6 mm diameter was immersed 2 cm in depth with respect to the liquid surface into a 50 mL beaker. Samples were treated at two amplitudes, 30 and 50 % during 30 min and the operating mode was pulsed US (2 s on/2 s off). As a control heating for both UAE protocols, extraction of the same varieties was carried out without ultrasound at 60 ◦C under the same pH conditions. 2.4. Isolation of apple pomace pectins After all extractions, apple pectins were precipitated with two volumes of 96 % acidified ethanol (0.2 % HCl, v/v). The solution was kept at 4 ◦C overnight. After that, the samples were centrifuged at 2,400×g for 5 min and the precipitates were washed with 96 % acidified ethanol (0.04 % HCl v/v) and later with ethanol (96 %). Finally, the recovered pectins were lyophilized and stored at −20 ◦C. Yields were calculated according to the formula: Yield (%) = weight of dried recovered pectin (g) weight of initial powder (g)×100 2.5. Structural characterization of pectins 2.5.1. Neutral sugars and galacturonic acid analysis by Gas Chromatography with Flame Ionization Detector (GC-FID) Monosaccharide composition was analyzed following a method previously described (Mu˜ noz-Almagro, Prodanov, Wilde, Villamiel, & Montilla, 2020). Pectins were hydrolyzed with 2 M trifluoroacetic acid (TFA) at 110 ◦C during 180 min and then were treated with hydroxylamine chloride in pyridine (2.5 %, w/v), hexamethyldisilazane and trifluoroacetic as described in Section 2.2. They were analyzed by GC-FID (7890A gas chromatograph; Agilent Technologies, Wilmington, DE, USA). Injector temperature as 280 ◦C and detector temperature, 350 ◦C; oven temperature was increasing from 150 ◦C to 165 ◦C at 1 ◦C/min and up to 300 ◦C at a heating rate of 10 ◦C/min. For quantitation an internal standard method was carried out using β-phenyl-- glucoside. All analyses were performed in quadruplicate, obtaining relative standard deviation (RSD) values below 10 % in all cases. 2.5.2. Estimation and distribution of relative molecular weight (Mw) by High Performance Size Exclusion Chromatography with Evaporative Light Scattering Detector (HPSEC-ELSD) Molecular weight (Mw) distribution was estimated following the method described by Mu˜ noz-Almagro, Rico-Rodriguez, Villamiel, and Montilla (2018). Samples (1 mg/mL) were filtered and, then, separated by HPSEC-ELSD (Agilent Technologies, Boeblingen, Germany) using TSK-Gel guard column (6.0 mm ×400 mm) and two TSK-Gel columns connected in series, G5000 PWXL (7.8 mm ×300 nm, 10 um), and G2500 PWXL (7.8 mm ×300 nm, 6 um) (Trosoh Bioscience, Stuttgart, Germany). The elution of the samples was carried out using 0.01 M NH 4 Ac as mobile phase at 0.5 mL/min and 30 ◦C during 50 min. 2.5.3. Estimation of the degree of esterification by Fourier transform infrared spectroscopy (FTIR) The degree of methyl esterification (DM) was determined by FTIR analysis. According to the method described by (Mu˜ noz-Almagro et al., 2020), the DM was expressed as the ratio between the peak areas of methylesterified carboxyl groups: COOCH 3 , measured at 1734 cm −1 , and the peak areas of (unmethyl)-esterified carboxyl groups: COO − measured at 1612 cm −1 . Equation: DM =Methyl esterified carboxyl groups Total carboxyl groups ×100 2.6. Statistical analysis Triplicates were performed for each analysis and data were expressed as the mean ±standard deviation. Statistical analysis of data was carried out using the Tukey’s tests (p <0.05), using the program SPSS Statistics 22.0 (IBM Corp., Armonk, NY, USA). 3. Results and discussion 3.1. Overall characterization of apple pomace by-products All samples presented similar ºBrix values, except Xua and Col varieties which exhibited the highest values (≥13◦Brix) (Table 1), probably as a result of a high extractability of soluble sugars during the pressing of apple to obtain ciders. The most predominant mineral in all samples was potassium followed by calcium, magnesium, phosphorus, sodium and iron, with Reg the apple pomace variety with the highest content in these minerals (Table 1). Skinner et al. (2018) found similar values of all elements except in the case of phosphorus and sodium, whose levels were higher than those described in our work. In general, mineral composition might indicate the potential use of apple pomace from these varieties classified as PDO Cider from Asturias as a source of health supplements, since calcium, magnesium and phosphorous are used not only for reducing the risk of osteoporosis but also for neutralizing the acidic effects of low pH foods derived from the diet. Additionally, potassium and iron are also important minerals for lowering blood pressure and improving anemia, respectively (Skinner et al., 2018; Yates et al., 2017). Remarkably, these samples were mostly composed by carbohydrates (64.9–90.6 g/100 g DW) and, to a much lesser extent, by protein and fat in line with previous studies on apple pomace composition from four different apple varieties (i.e., Jonagold, Gala, Idared and Golden Delicious) and derived from cider processing (Miceli-Garcia, 2014). The high variability observed in the levels of these compounds among samples could be attributed to the different varieties of apple as well as to the number of seeds present in each cider apple variety (Sato et al., 2010; Skinner et al., 2018). Our results indicate that apple pomace derived from apple cider varieties used in PDO Cider from Asturias was mainly composed of dietary fiber, with total dietary fiber (TDF) values ranging from 35 to 53 %. Specifically, the soluble dietary fiber (SDF), that normally includes pectin, ranged from 7 % to 23 %, whereas the insoluble dietary (IDF) was between 22 % and 40 % (Table 1). These values suggest that apple pomaces from cider processing could be appealing sources for the elaboration of some food formulations addressing gastrointestinal health problems. Sudha, Baskaran, and Leelavathi (2007) reported similar values in apple pomace derived from juice processing (i.e., 36.5 % IDF and 14.6 % SDF). Col and Dlr were the varieties that exhibited the highest content of malic acid (≥3 mg/g DW), which is beneficial to peristaltic movements in the human intestine as well as to liver and brain functions (Bhushan, Kalia, Sharma, Singh, & Ahuja, 2008; Sato et al., 2010). Moreover, fructose is, among soluble sugars, the one with the highest level in apple pomace samples according to Sato et al. (2010) (glucose:fructose: I. Calvete-Torre et al. Carbohydrate Polymers 264 (2021) 117980 5 sucrose 1.00:1.43:0.56). In case of these PDO apple pomaces, fructose was, indeed, the most predominant soluble sugar (108−142 mg/g DW) and apple pomaces exhibited a ratio of glucose:fructose:sucrose around 1.00:2.48:0.33 mg/g (Table 1). A moderate consumption of fructose has been associated to several beneficial effects, such as the enhancement of mineral balance, among others (Holbrook, Smith, & Reiser, 1989; Rizkalla, 2010). The parameter that probably revealed the greatest variation among samples was total phenolic content (TPC). Waldbauer et al. (2017) found that levels of these bioactive compounds located on the peel could vary up to 30 % in the same cultivar from one year to another one. Sol, Xua, Reg, Ver and Per showed a significantly higher content of TPC with respect to the other varieties, which is in agreement with the results obtained when measuring the inhibitory capacity of the DPPH (Table 1). Acid hydrolysates of all tested apple pomaces contained a variety of neutral sugars that are typically found in the composition of pectins, such as xylose, arabinose, rhamnose, galactose (Kravtchenko, Voragen, & Pilnik, 1992) (Table S1), as well as GalA (Table 1) as pectic monosaccharides. In addition, mannose and glucose levels were also determined, and are indicative of the presence of other polysaccharides as mannans, cellulose and hemicellulose in by-products of apple cider. Since GalA is the main constituent of pectin, the high levels observed ranging from 11.8 to 21.6 %, may indicate that all studied varieties of apple pomace are a good source to obtain pectin. In consequence, dedicated extraction of pectin from the studied apple pomace by-products are addressed in following sections. 3.2. Conventional acid extraction of pectin from nine mono-varietal PDO industrial apple pomace by-products In a first approach, pectin was extracted from the nine varieties of PDO apple pomaces using water acidified with citric acid at 90 ◦C and Table 1 Overall characterization of apple pomace by-products from selected apple varieties with PDO “Cider of Asturias”. Parameter Sol Dlr Rax Xua Reg Ver Col Per Ddt ◦Brix 11.9 ±0.2 abc 11.3 ±0.9 ab 12.1 ±0.06 abc 13.1 ±0.1 bc 11.8 ±0.5 abc 11.1 ±0.4 a 13.5 ±0.07 c 11.9 ±1.1 abc 11.3 ±0.9 ab pH 3.6 ±0.02 c 3.5 ±0.01 bc 3.4 ±0.01 a 3.4 ±0.03 a 3.4 ±0.03 a 3.9 ±0.01 e 3.5 ±0.01 ab 3.6 ±0.06 c 3.7 ±0.04 e DW (%) 18.1 ±0.3 a 17.9 ±0.9 a 20.4 ±0.7 ab 19.9 ±0.8 ab 21.4 ±1.5 ab 16.8 ±0.8 a 18.5 ±0.9 ab 19.1 ±0.7 ab 20.4 ±0.8 b Aw 0.97 ±0.0 a 0.97 ±0.0 a 0.97 ±0.0 a 0.97 ±0.0 a 0.97 ±0.0 a 0.97 ±0.0 a 0.97 ±0.0 a 0.97 ±0.0 a 0.99 ±0.0 a K (mg/100 g) 530.99 554.14 520.81 443.69 595.27 509.64 515.76 434.68 463.11 Ca (mg/100 g) 57.88 51.10 52.12 74.22 86.18 92.13 48.88 73.84 87.59 Mg (mg/100 g) 41.93 38.44 45.05 43.82 52.68 47.16 33.70 38.85 54.51 P (mg/100 g) 21.56 24.28 27.19 23.33 28.44 23.67 19.14 22.74 22.30 Na (mg/100 g) 6.27 5.62 7.17 8.46 7.67 4.68 5.01 5.47 5.86 Fe (mg/100 g) 6.47 4.53 4.47 2.48 3.09 2.95 2.94 3.76 3.56 Total Carb. (%) 79.0 ±5.1 a 72.5 ±12.1 a 64.9 ±9.9 a 90.6 ±10.5 a 88.0 ±12 a 72.5 ±6.8 a 66.7 ±14.3 a 82.8 ±24.2 a 89 ±9 a TDF (%) 52.9 ±4.2 c 42.3 ±3.1 ab 45.1 ±4.1 bc 50.4 ±3.2 bc 47.1 ±0.1 bc 44.9 ±1.5 bc 35 ±1.8 a 48.9 ±4.3 bc 52 ±1.4 bc IDF (%) 31.3 ±0.2 bc 30.5 ±4.1 bc 22.1 ±4.3 a 32.3 ±5.6 cd 40.2 ±1.2 d 35.7 ±4.4 bcd 25.9 ±0.3 ab 39.5 ±0.9 d 38.1 ±2.4 cd SDF (%) 21.6 ±0.2 cd 11.8 ±0.1 abc 23 ±4.1 d 18.1 ±3.1 bcd 6.9 ±1.3 a 9.2 ±2.9 ab 9.1 ±1.5 ab 9.4 ±1.2 ab 13.8 ±1.4 abcd Reducing Carb. (%) 44.3 ±6.9 a 54 ±1.7 a 44.2 ±3.1 a 41 ±2.9 a 52.3 ±2.6 a 44.6 ±1.2 a 39.5 ±2.4 a 41.9 ±1.1 a 32.6 ±2.9 a Protein (%) 3.9 ±0.09 bc 3.7 ±0.07 ab 4.2 ±0.7 bc 3.2 ±0.1 ab 5.1 ±0.7 c 3.8 ±0.03 bc 2.2 ±0.1 a 3.1 ±0.51 ab 4.3 ±0.4 bc Fat (%) 1.4 ±0.03 a 1.2 ±0.18 a 1.8 ±0.2 ab 1.4 ±0.1 a 2.1 ±0.4 ab 1.8 ±0.01 ab 1.3 ±0.3 a 2.4 ±0.4 b 2.2 ±0.3 ab Malic acid (mg/g) 1.9 ±0.05 c 3.2 ±0.03 e 2.3 ±0.2 cd 2.2 ±0.2 cd 2.1 ±0.2 cd 0.6 ±0.01 a 3.0 ±0.05 d 1.1 ±0.1 bc 1.0 ±0.0 b Fructose (mg/g) 138.0 ±0.4 g 128.1 ±0.2 e 125.5 ±1.6 d 142.2 ±1.4 h 117.2 ±1.2 b 135.9 ±0.3 f 122.3 ±0.5 c 126.1 ±1.3 d 108.7 ±0.9 a Glucose (mg/g) 63.1 ±0.9 f 50.5 ±1.0 d 56.4 ±1.2 e 46.2 ±1.5 c 60.6 ±2.1 f 65.3 ±0.0 g 34.4 ±0.2 a 44.3 ±1.4 c 39.6 ±1.2 b Sucrose (mg/g) 15.7 ±0.4 d 10.8 ±0.3 b 13.5 ±0.2 c 34.1 ±0.3 g 18.6 ±0.4 e 17.2 ±0.5 e 4.1 ±0.0 a 27.2 ±0.3 f 12.2 ±0.7 c TPC (g/kg) 5.0 ±0.3 bc 3.5 ±0.2 a 3.6 ±0.3 ab 5.9 ±0.5 cd 6.3 ±0.4 d 6.2 ±0.4 d 4.1 ±0.3 ab 6.5 ±0.6 d 3.7 ±0.2 a % Inhibitory Activity of DPPH 41.8 ±0.9 28.6 ±1.2 31.9 ±0.7 40.4 ±1.1 40 ±0.6 40.9 ±0.7 21.9 ±0.9 40.9 ±0.5 26.5 ±1.4 Galacturonic acid (%) 14 ±2.3 ab 13.9 ±2.7 ab 11.8 ±0.3 a 13.1 ±0.01 a 18.1 ±0.3 bc 19.1 ±1.7 c 12.9 ±0.5 a 20.9 ±0.9 c 21.6 ±0.5 c Neutral sugars (%) 28.3 ±1.8 f 12.5 ±0.3 a 17.3 ±1.1 c 14.7 ±0.5 b 21.9 ±0.1 d 16.6 ±1.5 bc 13.7 ±1.8 b 29.1 ±2.2 f 23.9 ±0.6 e General chemical characterization of nine varieties of apple pomace (data expressed over DW): Sol, Solarina; Dlr. De la Riega; Rax, Raxao; Xua, Xuanina; Reg, Regona; Ver, Verdialona; Col, Collaos; Per, Perico and Ddt, Durona de Tresalli. Abbreviations used: DW (Dry Weight), Aw (Water activity), TDF (Total Dietary Fibre), IDF (Insoluble Dietary Fibre), SDF (Soluble Dietary Fibre) and TPC (Total Phenolic Content). Average of three replicates ±SD. Different letter in the same row indicates significant differences (Tukey p <0.05). Table 2 Yield, sugar composition and molecular parameters of apple pomace pectin samples obtained by conventional extraction (pH 2.5 and 90 ◦C). Sol Dlr Rax Xua Reg Ver Col Per Ddt Yield (%) 10.5 ±1.2 a 10.5 ±0.7 a 10.5 ±0.5 a 10.8 ±0.6 a 11.2 ±0.3 a 9.0 ±0.9 a 9.6 ±0.8 a 8.2 ±0.9 a 11.6 ±1.1 a GalA (%)* 67.9 ±7.3 e 40.3 ±0.4 c 30.6 ±3.5 b 18.3 ±2 a 40.1 ±0.08 c 54.7 ±3.8 d 18.0 ±0.4 a 46.9 ±4.4 c 42.7 ±1.6 c Rha (%) 5.0 ±0.07 a 5.2 ±0.2 a 6.5 ±0.9 a 9.5 ±1 b 24.7 ±2.9 e 10.5 ±0.08 b 9.5 ±0.4 b 18.6 ±0.3 d 15 ±0.2 c Ara (%) 8.5 ±0.8 bc 5.0 ±0.2 a 8.2 ±1.1 bc 7.3 ±0.2 ab 8.5 ±0.8 bc 10.3 ±1.3 cd 8.0 ±0.7 bc 14.5 ±1.7 e 12.3 ±2 de Gal (%) 3.6 ±0.1 b 2.2 ±0.1 a 6.5 ±0.4 e 3.4 ±0.1 b 5.3 ±0.6 d 4.5 ±0.2 b 5.6 ±0.3 d 5.2 ±0.09 d 4 ±0.4 bc Xyl (%) 1.8 ±0.07 c 1.3 ±0.1 b 1.3 ±0.1 b 1.2 ±0.03 b 1.7 ±0.1 c 0.0 ±0.0 a 1.3 ±0.3 b 0 ±0 a 1.7 ±0.08 bc Fuc (%) 2.8 ±0.08 b 2.2 ±0.1 a 2.0 ±0.1 a 4.1 ±0.2 e 2.9 ±0.2 bc 3.4 ±0.4 d 1.9 ±0.1 a 3.2 ±0.04 cd 3.3 ±0.2 d Man (%) 0.2 ±0.02 ab 0.0 ±0.0 a 0.0 ±0.0 a 3.6 ±0.4 c 0.2 ±0.03 ab 0.4 ±0.05 b 0.0 ±0.0 a 0.4 ±0.04 b 0 ±0 a Glc (%) 10.3 ±0.7 a 43.9 ±0.4 c 45.0 ±0.8 d 52.5 ±0.4 e 16.6 ±2.1 b 16.3 ±1.2 b 55.6 ±4.7 e 11.3 ±1.7 a 21.2 ±0.6 c GalA/Rha 13.7 ±1.7 d 7.8 ±0.4 c 4.8 ±1.2 b 2.0 ±0.4 a 1.7 ±0.2 a 5.2 ±0.3 b 1.9 ±0.05 a 2.5 ±0.2 a 2.9 ±0.1 a (Ara þGal)/Rha 2.5 ±0.2 d 1.4 ±0.05 c 2.3 ±0.09 d 1.1 ±0.1 b 0.6 ±0.08 a 1.4 ±0.09 c 1.4 ±0.02 c 1.1 ±0.07 b 1.1 ±0.1 b Relative Mw (kDa) 790 654 654 355 68 102 516 607 514 DM (%) 77 68 74 60 72 88 58 81 56 Nine varieties of apple pomace: Sol, Solarina; Dlr. De la Riega; Rax, Raxao; Xua, Xuanina; Reg, Regona; Ver, Verdialona; Col, Collaos; Per, Perico and Ddt, Durona de Tresalli. Abbreviations used: GalA, galacturonic acid; Rha, rhamnose; Ara, arabinose; Gal, galactose; Xyl, xylose; Fuc, fucose; Man, mannose; Glc, glucose; Mw, molecular weight; DM, Degree of methyl esterification. * Relative carbohydrate monomer content is expressed as percentage of the total carbohydrate content in the apple pomace pectin samples. Average of three replicates ±SD. Different letter in the same row indicates significant differences (Tukey p <0.05). I. Calvete-Torre et al. Carbohydrate Polymers 264 (2021) 117980 6 for 30 min. As it can be observed in Table 2, a yield of about 10 % was achieved under these conditions in all varieties, suggesting their potential use as renewable pectin sources. These results are in line with those reported for apple pectin acidic extraction with hydrochloric, citric and oxalic acids (Morales-Contreras et al., 2020). However, some authors (Dranca et al., 2020; Luo et al., 2019) reported yields 2-fold higher (≥19 %) when pectin was extracted from apple pomace with citric and acetic acids, but it is important to note that the duration of those assays were almost 5 times longer in comparison with those obtained in the present work (30 vs 148 min). Long extraction times combined with high temperature could accelerate molecular motion and facilitate the dissolution of pectin in aqueous solution (Kang, Hua, Yang, Chen, & Yang, 2015). The compositional and structural characteristics of the nine pectins obtained from the corresponding apple pomaces are included in Table 2. The GalA/Rha ratio displayed in the table represents the number of GalA residues per Rha residues, giving an indication of the RG-I backbone with respect to HG content. Thus, a lower value shows a compound richer in RG-I chains. The (Ara +Gal)/Rha ratio indicates the amount of neutral sugar residues attached to the RG-I backbone. With respect to GalA, Sol was the variety that exhibited the highest content of this monosaccharide (67.9 %), this value being within the limit established by the FAO to consider pectin as a food ingredient (E440). Taking into account the content of GalA and neutral sugars of Sol, it is evident that the structure of this pectin was composed mainly of homogalacturonan (HG) as well as only a small part of arabinogalactan, galactan and/or arabinan branches were present as part of the rhamnogalacturonan-I (RG-I) chains (GalA/Rha, ~13.7 %). On the contrary, Reg and Per were highlighted because of the high Rha content (≥18.6 %), suggesting that both pectins were the most enriched in RG-I as compared to the rest of the apple varieties. Per also exhibited the greatest content of arabinose (14.5 %) which could indicate the substitution of the RG-I branching along the HG with numerous arabinans, arabinogalactans and limited galactans side chains (Amorim, Vriesmann, Petkowicz, Martinez, & Noleto, 2016; Yuliarti, Goh, Matia-Merino, Mawson, & Brennan, 2015). In this sense, Schols, Posthumus, and Voragen (1990) indicated that arabinose was the most abundant sugar existing in the hairy region of pectins isolated from apple juice produced by the liquefaction process. Previous studies have demonstrated a correlation between the high content of arabinose and galactose with bifidogenic properties (Di et al., 2017; Ferreira-Lazarte, Kachrimanidou, Villamiel, Rastall, & Moreno, 2018; Manderson et al., 2005). Regarding other monosaccharides (glucose and mannose) likely derived from polysaccharides such as cellulose and hemicellulose, Sol, Reg, Ver and Per revealed the lowest levels of these monosaccharides (10.5–16.8 %). This fact could indicate that the pectin extracted from these varieties could have a greater purity as compared to those extracted from the others. This is a relevant parameter during extraction and purification of pectins because minimizing the levels of “nonpectin” components is important in obtaining good quality pectin (Miceli-Garcia, 2014). In contrast, varieties such as Dlr, Rax, Xua and Col showed remarkable levels of glucose (43.9–55.6 %) indicating the substantial presence of glucose-based polysaccharides, such as cellulose and/or hemicellulose, in addition to pectin. As illustrated in Table 2, in general terms, most of the apple pectins showed high values of estimated Mw (~600 kDa) which suggest that the junction zones are formed by long segments, resulting in an enhancement of rheological properties and, therefore, in a strong potential of these pectins as thickener or gelling agents in the food industry (Morales-Contreras et al., 2020). Unlike this behaviour, Reg exhibited the lowest value of estimated Mw (~68 kDa) which is in line with other commercial modified pectins that have shown to be effective supplements in the treatments of cancer (Morris, Belshaw, Waldron, & Maxwell, 2013). It is well known that molecular weights of pectins largely vary with the starting material and the extraction conditions (Miceli-Garcia, 2014); therefore, a large variability has been reported for apple pectin (10 4 –10 6 Da) (Constenla, Ponce, & Lozano, 2002; Miceli-Garcia, 2014; Rasc´ on-Chu et al., 2009; Rolin, 2002). Likewise, the reduction of Mw in pectin could involve a positive effect on its ability to promote bifidobacteria growth in comparison to unmodified pectin (Ferreira-Lazarte et al., 2018). However, it should be also indicated that not only the Mw but also the monosaccharide composition, RG-I/HG ratio, type of glycosidic linkage between monosaccharides, spatial conformation, degree of esterification and potential presence of non-carbohydrate components can greatly influence the biological activity of pectins (Cui et al., 2021; Larsen et al., 2019). From the analysis of the DM (Table 2), it is noted that, although all samples were high methoxylated pectin, DM varied from 56 to 88 %, which is in agreement with other studies that obtained pectins with a DM ranging from 61 to 82 % in apple pomace resulting from juice extraction (Luo et al., 2019; Morales-Contreras et al., 2020). 3.3. Ultrasound extraction of pectin from three selected mono-varietal PDO industrial apple pomace by-products (Sol, Reg and Per) Taking into account the previous characterization of conventionally extracted pectin, three out of the nine apple pomace varieties were selected for ultrasound-assisted extraction (UAE). Thus, Sol was selected because it exhibited the highest amount of GalA which is associated, in addition to a low acetyl content, to excellent gelling properties in commercial pectins, whilst Reg and Per were chosen because of their high levels in Rha and Gal. In addition, Per was also selected based on its high content in Ara, which could provide a potential role of this pectin as a prebiotic substrate. Moreover, the three selected varieties showed the lowest presence of non-pectic monosaccharides. 3.3.1. Pectin extraction yield Regarding the effect of temperature in conventional acid treatments, an increase in the yields was detected with the increase of temperature (60 ◦C, ~8 %; 90 ◦C,~10 %). The low temperature value set at 60 ◦C was necessary in the assays with ultrasound (US) since according to Hernoux-Villi` ere et al. (2013), the mechanical effect of US is obtained at the lowest temperature possible. No significant differences were found in pectin yields extracted from any of the studied varieties (~ 7 %–8 %) after 30 min in the control heating (conventional acid extraction), US bath and probe at the lowest amplitude (Table 3). Contrary to the results of previously published studies using different food matrices (Hosseini, Khodaiyan, Kazemi, & Najari, 2019; Maran et al., 2017; Ponmurugan et al., 2017), UAE did not increase the yield of pectin as compared to that reached by conventional acid extraction (control heating) under the same temperature. This outcome might be attributed to the low working intensity of the ultrasonic device used in the present study. In this sense, the large amount of bubbles formed during cavitation collapse in an effective way, generating the needed energy to breakdown the biomolecules (Sun & Tomkinson, 2002). Concerning the impact of time in the pectin extracted with an US bath, when data of 15 and 30 min were compared, higher yields were found in the case of the longest time (~4 vs ~7 %). The US waves generate a mechanical effect, when cavitation bubbles collapse on the apple pomace surface, leading to greater penetration of the solvent into the solid matrix. Therefore, affecting the separation between the solid and liquid phases, a second ultrasound treatment may be efficient in dissolving the pectin previously absorbed in the residue (Dranca & Oroian, 2018). These factors, alongside the absence of periodical agitation meant to keep the mixture evenly distributed (Xu et al., 2014) may also explain why the combination of heating and ultrasound treatment did not lead to a higher pectin yield. Finally, yields were considerably lower when pectin was extracted with probe, regardless of the amplitude used. I. Calvete-Torre et al. Carbohydrate Polymers 264 (2021) 117980 7 3.3.2. Monomer composition and molecular weight of extracted pectins The method of extraction can have influence on the structure of pectin (Dranca et al., 2020). As illustrated in Table 3, it is noteworthy that although the sonication did not improve the yield, an increase in the content of GalA was found (bath mode) due to lower content of non-pectic polysaccharides (glucose and mannose), with the exception of the mannose content in Reg. These results indicated that as consequence of the sonication (bath mode), pectins of Sol and Per varieties were enriched in HG structure as compared to the respective pectins extracted without assistance of US. This finding was also supported by the notable increase in the GalA/Rha ratio found in these two varieties following sonication under bath mode conditions. Reg was still the variety with the highest amount of Rha (12.8 % under bath mode) suggesting that pectin extracted from this variety was the most enriched in RG-I as compared to the rest of the apple varieties. Likewise, the observed decrease of the GalA/Rha ratio in Reg variety following sonication (under bath and probe modes) strengthened its enrichment in RG-I. On the other hand, depending on the variety and the applied amplitude, the content in galactose, arabinose and fucose increased significantly in some of the pectins extracted with probe US with respect to their corresponding control heating (Table 3). This fact is relevant because these carbohydrates could promote the growth of beneficial bacteria in the gut microbiota (Cui et al., 2019). Regardless of the extraction method used, all HPSEC-ELSD chromatographic profiles of apple pectin extracted from Reg, Per and Sol (Fig. 1) showed the most abundant fraction at a retention time (t r ) of 23 min, which could be identified mainly as pectin. However, according to the mannose and glucose levels reported in Table 3, it cannot be ruled out the co-elution of hemicellulose remnants that could also be extracted from the cell walls. Unlike control heating and bath ultrasound treatments, in general, a delay in the elution of the main fraction in US probe treatments was found as result of the severity of the treatment. Remarkably, this reduction of Mw could suggest that hemicellulose glued to pectin was considerably decreased, corroborating the low contents of mannose and glucose observed in these samples in the monomer composition (Table 3). Furthermore, two chromatographic peaks were also detected around 35 and 39 min, which could be attributed to modified pectins of different Mw and salts from citric acid used during extraction, respectively. The modifications monitored by SEC-ELSD regarding the estimation of Mw of the different fragments and their corresponding relative abundances (%) are shown in Fig. 2. No substantial differences in the average Mw of the pectin fractions were detected among all of them extracted by conventional acid treatment (control heating) and those extracted using bath ultrasound unit (Fig. 2 and Table S2), demonstrating the absence of interference of this assisted-extraction method in the recovery of this polysaccharide. However, the estimated number average molecular weight (Mn) remarkably increased in pectins extracted from Sol and Reg varieties using bath ultrasound. This fact led to lower polydispersity indexes as compared to the conventional extraction method, which is indicative of a narrower distribution of the pectins extracted with the assistance of bath ultrasound (Table S2). On the contrary, all pectins extracted by probe exhibited an important reduction in the average Mw (Fig. 2 and Table S2). Likewise, in the case of the pectin extracted from Sol variety, this reduction was more notable as intensity increased. Finally, the reduction of the average Mw in pectins from Sol and Per varieties extracted by US probe was sharper than the corresponding Mn values, giving rise to an approximately 2fold decrease in the polydispersity indexes as compared to conventional extraction (Table S2). The pectin extracted by US probe for Reg variety underwent an inverse behaviour (that is, a sharper decrease in Mn than in Mw) showing, thus, a wider range of molecular mass than that observed following a conventional extraction (Table S2). On the other hand, as it was expected, the Mw of all pectins extracted under acidic conditions (control heating) was higher at 60 ◦C in Table 3 Yield, sugar composition and molecular parameters of apple pomace pectin samples of Sol, Per and Reg varieties extracted by US (bath and probe modes). Extractions Time (min) Yield (%) GalA (%)* Rha (%) Gal (%) Xyl (%) Ara (%) Fuc (%) Man (%) Glc (%) GalA/Rha Ara +Gal/Rha DM (%) Sol Control heating 30 7.8 ±1.1 e 60 ±2.8 a 4.2 ±0.3 abcd 3.1 ±0.2 a 2.1 ±0.2 cd 9.6 ±1 abc 2.3 ±0.3 ab 3.1 ±0 d 15.6 ±1.4 c 7.7 3.02 50 Bath US 15 3.9 ±0.1 b 70.1 ±1.2 bc 4.1 ±0.4 abcd 2.9 ±0.04 a 1 ±0.3 ab 6.8 ±1 a 2.3 ±0.2 ab 0.2 ±0 a 5.1 ±1 ab 16.8 2.37 52 30 7.7 ±0.3 e 71.0 ±2.8 bc 2.8 ±0.3 a 3.3 ±0.3 ab 1.7 ±0.5 bcd 9.8 ±1.5 abc 2.7 ±0.3 abc 3.3 ±0.3 de 5.4 ±0.6 ab 25.2 4.68 55 Probe US 30 % amplitude 30 7.9 ±0.7 e 66.7 ±1.9 abc 7.9 ±0.9 cd 6.4 ±0.2 d 2.7 ±0.2 de 9.6 ±1.4 abc 3.1 ±0.2 bcd 0.9 ±0.1 abc 2.7 ±0.2 a 5.55 2.20 45 50 % amplitude 30 4.7 ±0.3b c 68.9 ±2.3 abc 6.9 ±0.5 abcd 4.3 ±0.3 ab 0 ±0 a 11.5 ±0.05 bc 3.9 ±0.9 d 0.3 ±0.01 ab 4.2 ±0.4 a 8.9 2.3 50 Per Control heating 30 7.6 ±0.7 e 59 ±2.6 abc 3.2 ±1.1 ab 3.2 ±0.6 a 2.7 ±0.05 de 9.7 ±1.1 abc 2.7 ±0.2 abc 4.6 ±0.2 f 14.9 ±1.5 c 9.2 1.8 53 Bath US 15 3.8 ±0.3 ab 75.4 ±0.6 c 3.8 ±0.3 abc 3.1 ±0.2 a 1.7 ±0.8 bcd 8.8 ±0.01 ab 2.6 ±0.03 abc 3.1 ±0.2 d 4.2 ±0.08 a 19.6 3.5 74 30 6.5 ±0.03 de 70.7 ±4.5 bc 5.4 ±3.3 abcd 3.1 ±0.1 a 1.5 ±0.2 bc 10.9 ±0.7 bc 2.6 ±0.2 abc 3.1 ±0.1 d 5.3 ±1.8 ab 13.1 2.9 52 Probe US 30 % amplitude 30 4.0 ±0.2 c 64.3 ±3.8 abc 8.4 ±0.6 de 6.1 ±0 cd 4.3 ±0.2 g 8.8 ±1.5 ab 2.8 ±0.03 abc 1 ±0.1 bc 4.2 ±1.3 a 3.07 1.17 51 50 % amplitude 30 5.0 ±0.5 cd 65.8 ±0.7 abc 8 ±1 cd 6.1 ±0.5 cd 3.2 ±0.3 ef 9.3 ±0.7 abc 3.5 ±0.2 cd 0.9 ±0.07 abc 3.2 ±0.2 a 7.8 2.3 50 Reg Control heating 30 7.0 ±0.5 e 67 ±1.1 abc 7.6 ±2 bcd 3.4 ±0.1 ab 1.3 ±0.1 bc 8.79 ±0.1 ab 2.4 ±0 ab 0.2 ±0 a 9.3 ±1 b 9.2 1.8 88 Bath US 15 4.2 ±0.4 b 69.1 ±4 bc 8.4 ±1.8 de 3.9 ±0.2 ab 1.7 ±0.1 bcd 7.4 ±0.4 a 2.1 ±0.01 a 3.9 ±0.7 ef 7.1 ±0.3 ab 8.3 1.6 83 30 6.8 ±0.5 de 68.4 ±3.2 abc 12.8 ±2 ef 3.2 ±0.3 a 1.2 ±0.4 bc 7.8 ±0.7 a 2 ±0.3 a 3.1 ±0.4 d 4.3 ±5 a 4.9 1.0 81 Probe US 30 % amplitude 30 5.5 ±0.3 cd 48.2 ±3.1 ab 14 ±1.8 f 9.3 ±1.7 e 6.5 ±0.6 h 12 ±2 c 3.9 ±0.3 d 1.4 ±0.2 c 4.7 ±0.5 a 2.66 1.98 43 50 % amplitude 30 2.43 ±0.2 a 53.7 ±1.7 abc 22 ±1.7 g 4.8 ±0.4 bc 4 ±0.3 fg 7.4 ±0.2 a 2.7 ±0.2 abc 0.8 ±0.06 abc 4.8 ±0.2 ab 2.3 0.7 53 Three varieties of apple pomace: Sol, Solarina; Reg, Regona; Per, Perico. Abbreviations used: GalA, galacturonic Acid; Rha, rhamnose; Ara, arabinose; Gal, galactose; Xyl, xylose; Fuc, fucose; Man, mannose; Glc, glucose; DM, Degree of methyl esterification. * Relative carbohydrate monomer content is expressed as percentage of the total carbohydrate content in the apple pomace pectin samples. Average of three replicates ±SD. Different letters in the same column indicates significant differences among extraction methods within the same apple variety (Tukey p <0.05). I. Calvete-Torre et al. Carbohydrate Polymers 264 (2021) 117980 8 comparison with the conventional extraction carried out at 90 ◦C for 30 min. 3.3.3. Degree of methyl esterification of extracted pectins With respect to the changes in the functional groups and the bonding configurations of the apple pectins under the experimental conditions assayed, in general, only slight changes were observed in the spectral profiles of the extracted samples, regardless of the time period and power density applied (Fig. 3). The wide absorption band at 3415 cm −1 corresponded to O – H stretching vibration of hydroxyl groups whereas the band that appears on 2936 cm −1 was assigned to the vibrations of C – H ( – CH, – CH 2 , – CH 3 ) (Oliveira et al., 2016). The bands around 1734 cm −1 and 1612 cm −1 , belonging to the vibration of asymmetric tension of methylesters of the carbonyl groups and symmetrical tension of the carboxylate ions, respectively, changed as a result of the variation of the DM. In general, the reduction of the peaks 1145 and 1051 cm −1 , corresponding to the stretching vibrations C – OH and C – O – C glycosidic bond vibration of galacturonic acid molecules belonging to homogalacturonan region, as well as to the minor presence of glycosidic linkages between sugar units (Zhang et al., 2018). Other bands were maintained regardless of the type of extraction such as 962 cm −1 (C – O bending), 914 cm −1 (rocking mode of – CH 3 ) and the 832 cm −1 related to the – COOH bending out of the plane at the C-6 position (Wang & Lu, 2014). In the analysed samples, these qualitative changes affected the estimated DM values (Table 3), the most important molecular parameter determined by FT-IR. Regardless of the extraction method, no changes were detected in this parameter in Sol and Per, their values being Fig. 1. HPSEC-ELSD profiles of apple pectins extracted from a) Reg b) Per c) Sol using conventional acid treatment (control heating), or ultrasound (bath and probe units) for 15 and 30 min. fx1 I. Calvete-Torre et al. Carbohydrate Polymers 264 (2021) 117980 9 slightly over 50 %. On the contrary, in the case of Reg, the values of DM decreased from 88 to 53 % when ultrasound with probe was applied as consequence probably of depolymerisation that occurs as time and intensity of US increases. However, other possible mechanisms should not be discarded since depolymerisation can be accompanied by other effects resulting in structural changes. 4. Conclusions The comprehensive chemical characterization of nine mono-varietal industrial apple pomaces resulting of cider apples used in the production of ciders with protected designation of origin has reinforced the potential valorization of this underutilized by-product typically used for animal feed. Thus, the high content in essential minerals, including K, Ca, Mg and P, fibers (whose content ranged from 35 % to 52.9 %) and polyphenol described in this work warrants further functional studies dealing with the utility of apple pomace as a source of human dietary supplements and/or food additives. Likewise, the high level of galacturonic acid found in all tested apple pomaces (ranging from 12 to 22 %) revealed their suitability as a source of pectin. Consequently, pectin was extracted from all apple pomaces by conventional acid heat treatment and a comparison was performed by assisting the extraction with Fig. 2. Estimation of average Mw (kDa) of pectins (Peak 1) and other fractions (Peaks 2 and 3) formed (%) after extraction from three different apple pomace varieties (Reg, Per and Sol) using conventional acid treatment (control heating) or ultrasound (bath and probe units) for 15 and 30 min. I. Calvete-Torre et al.