- Email: [email protected]
Polysaccharide gels as oil bulking agents: Technological and structural properties
Food Hydrocolloids 36 (2014) 374e381
Contents lists available at ScienceDirect
Food Hydrocolloids journal homepage: www.elsevier.com/locate/foodhyd
Polysaccharide gels as oil bulking agents: Technological and structural properties A.M. Herrero a, *, P. Carmona b, F. Jiménez-Colmenero a, C. Ruiz-Capillas a a b
Instituto de Ciencia y Tecnología de Alimentos y Nutrición (ICTAN-CSIC), (Formerly, Instituto del Frío), Ciudad Universitaria, 28040 Madrid, Spain Instituto de Estructura de la Materia (CSIC), Serrano 121, 28006 Madrid, Spain
a r t i c l e i n f o
a b s t r a c t
Article history: Received 17 January 2013 Accepted 8 August 2013
Lipid and polysaccharide interactions in various polysaccharide gels prepared for use as oil bulking agents were investigated using Raman spectroscopy. Two different polysaccharide gel matrices containing olive oil were prepared using mixed biopolymer systems of alginate with inulin (OM-A/I) or dextrin (OM-A/D). Stability and textural properties were also evaluated in these matrices. Thermal stability of the different matrices was optimal. The textural behaviour of polysaccharide matrices with olive oil incorporated differed depending on the polysaccharide gels used in their formulation. OM-A/I presented the highest (P < 0.05) hardness, adhesiveness, and chewiness. These matrices were stabilized by hydrogen bonding between oil carbonyl groups and water and/or carbohydrate molecules. Lipid acyl chains are relatively fixed through hydrogen bonding and intermolecular order upon micelle formation. Raman spectroscopic results also showed carbohydrate-water hydrogen bonding, in which inulin seemed to be more strongly bonded to water than dextrin. This difference in the structural behaviour of inulin and dextrin in terms of hydrogen bonding to water may explain the differences in textural properties. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Polysaccharide Olive oil Texture Raman spectroscopy Structural characteristics
1. Introduction Improving the fat content of foods based on processing strategies (reformulation) is one of the most important current approaches to the development of potential functional foods, especially in the case of meat-based functional foods. Healthier lipid reformulation processes are generally based on replacement of the animal fat normally present in the product with another fat whose characteristics are more in line with health recommendations (Jiménez-Colmenero, 2007). A variety of non-meat fats of plant and marine origin have been used in product formulation (as non-meat ingredients) to partially replaces meat fats (mainly pork or beef). These oils have been incorporated in meat products in liquid and solid forms, encapsulated, or as oil-in-water emulsions (Jiménez-Colmenero, 2007). Strategies for incorporation of healthy oils in a gel-like matrix to form oil’s bulking agent (in which this new ingredient acts as an animal fat replacer) could offer new possibilities for improving the fat content of meat products. Polysaccharides, used either individually or in combination, can be used to create a variety of gel structures which may be suitable for immobilizing oil droplets and thus to acting as oil’s bulking agents.
* Corresponding author. Tel.: þ34 915616800; fax: þ34 91 5645557. E-mail address: [email protected] (A.M. Herrero). 0268-005X/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.foodhyd.2013.08.008
In this regard, healthy oils in a konjac matrix have been used to improve fat content in dry fermented sausage (Ruiz-Capillas, Triki, Herrero, Rodríguez-Salas, & Jiménez-Colmenero, 2012), fresh merguez sausages (Triki, Herrero, Jiménez-Colmenero, & Ruiz-Capillas, 2013) and frankfurters (Salcedo-Sandoval, Cofrades, Ruiz-Capillas, Solas, & Jiménez-Colmenero, 2013). Alginate gels may offer interesting possibilities as oil bulking agents. Alginates are of particular interest for their ability to form gels in the presence of calcium salt (Roopa & Bhattacharya, 2010; Zhang, Daubert, & Foegeding, 2005). These gels consist of polymeric molecules cross-linked to form a three-dimensional macromolecular network containing a large fraction of water within their structure which displaying mechanical rigidity. The properties of the gel are the net result of the complex interactions between the components. One of the most important properties of alginate gels, which make them highly versatile, is their capacity for controlled uptake, release and retention of molecules. This capacity is due in turn, to specific interactions of the macromolecular network with the diffusing or retained molecule. In this connection, hydrocolloids as alginate, unlike emulsifiers, generally stabilize emulsions by increasing the viscosity, which inhibits the rate of coalescence between the oil droplets. This compound is adsorbed at the oil/water interface to form an interfacial film and sterically stabilize emulsions against flocculation and coalescence (Huang, Kakuda, & Cui, 2001). In addition to functioning as gelling agents, they also
A.M. Herrero et al. / Food Hydrocolloids 36 (2014) 374e381
exhibit properties as emulsifiers, encapsulating agents, dispersing agents, foam stabilizers, film formers, and crystal inhibitors (Gaonkar, 1991). Alginate gel properties and their utility as oil bulking agents can be modulated by mixing with other biopolymers. Various researchers have investigated the properties of mixed biopolymer gelling systems for that purpose in some cases achieving substantial enhancements in gel strength (Evageliou, Tseliou Mandala, & Komaitis, 2010; Harrington & Morris, 2009; Zimeri & Kokini, 2003). Inulin is a polysaccharide which offers interesting technological and health-beneficial properties. The technological properties are related to the degree of polymerization of its chains, making it suitable for use as a low-calorie sweetener, fat replacer or texturizing agent (Tungland & Meyer, 2002). The ability of inulin to form gels, and the physicochemical properties of these gels, have been addressed by numerous researchers (Bot, Erle, Vreeker, & Agterof, 2004; Kim, Faqih, & Wang, 2001; Zimeri & Kokini, 2002). Some authors have reported that inulin gels resemble solid fats in texture and may be used as fat mimetic (Bot et al., 2004). Increasing the amount of inulin can result in an increase in the gel strength of gellan (Evageliou et al., 2010). Another polysaccharide of interest is dextrin, a low-molecularweight glucose polymer resulting from partial hydrolysis of starch. Dextrin is a widely used material with a variety of applications, particularly in the food industry for adhesives, and also in other applications where it offers considerable potential for hydrogel production (Carvalho, Gonçalves, Gil, & Gama, 2007; Garcia, Barros, Gonçalves, Gama, & Gil, 2008). The properties of these matrices with oil entrapped in polysaccharide gels are the net result of complex interactions among the components. Raman spectroscopy is a direct, non-invasive technique which has proven useful for providing structural information on the various components (lipids, polysaccharides, etc.) (Choi, Yuen, Phillips, & Ma, 2010; Dobson, 2001) involved in the preparation of these matrices. This spectroscopic technique has been used to identify selected seaweed polysaccharides (Pereira, Amado, Critchley, van de Velde, & Ribeiro-Claro, 2009) and the structural behaviour of oligosaccharides in water (Ka curáková & Mathlouthi, 1996). In particular, some authors have performed Raman spectroscopic analysis of alginate hydrogels (Dumitriu, Mitchell, & Vasile, 2011; Pielesz & Ba˛ k, 2008). In addition, the technique has been used to investigate the conformational behaviour of lipid bilayer systems (dipalmitoylphosphatidylcholine) perturbed by cholesterol and water (Bush, Adams, & Levin, 1980). By means of Raman spectral frequencies and intensities it is possible both to monitor intramolecular changes occurring within the three structurally distinct regions of the phospholipid molecule and to follow alterations in lattice order and packing characteristics. The acyl chain mobilities of bovine milk fat globule lipids and component triglycerides have also been determined by Raman spectroscopy as a function of temperature. This study showed that the CH stretching region is the most useful for lipid chain order analysis (Forrest, 1978). Also, Raman spectroscopy has proven the role of proteins and lipids in emulsion formation (Howell, Henryk, & Li-Chan, 2001; Meng, Chan, Rousseau, & Li-Chan, 2005). Moreover, Raman spectroscopy has been used to characterize the development of polymeric matrices based on chitosan/cashew gum for encapsulation of a natural essential oil (Abreu, Oliveira, Paula, & de Paula, 2012). These findings speak to the feasibility or the potential of this technique as a means to elucidate interactions between the components of matrices of oil molecules involved in polysaccharide gels. The aim of the present study was to develop olive oil bulking agents using mixed biopolymer systems of alginate with inulin or dextrin with optimal characteristics for use as animal fat replacers
375
in the development of healthier meat products. Two main aspects of these agents were considered: their technological characteristics, since these affect quality properties of the reformulated products in which they will be incorporated; and the other structural characteristics of the system and interactions between their components, in order to establish causal relationships between their structural and technological properties. 2. Materials and methods 2.1. Materials Ingredients used for preparation of polysaccharide gels as oil’s bulking agents included: olive oil (13% SFA, 79% MUFA and 8% PUFA) (Carbonell Virgen Extra, SOS Cuétara SA, Madrid, Spain); sodium alginate (SA) (90% carbohydrates content) (Tradissimo, TRADES S.A., Barcelona Spain); calcium sulfate (CS) (Panreac Química, S.A. Madrid, Spain), tetra-sodium pyrophosphate anhydrous PRS (SP) (Panreac Química, S.A. Madrid, Spain), inulin (I) consisting mainly of chicory inulin (>90% inulin) (TRADES S.A., Barcelona Spain) and white maize dextrin (D) (CARGILL S.L.U- CTS Rubi, Barcelona, Spain). 2.2. Preparation of olive oil bulking matrices based on polysaccharide gels Different concentrations and combinations of inulin, dextrin, and alginate and olive oil were assayed to determine the best conditions, from a technological standpoint (optimal fat content more in line with health recommendations, without loss of exudate and with a suitable texture for cutting, grinding, etc.), for production of an oil bulking agent that could use as fat replacer (DelgadoPando, Cofrades, Ruiz-Capillas, Solas, & Jiménez Colmenero, 2010; Herrero, Carmona, Pintado, Jiménez-Colmenero, & Ruíz-Capillas, 2011a, 2011b; Jiménez-Colmenero et al., 2012; Jiménez-Colmenero, Herrero, Pintado, Solas, & Ruiz-Capillas, 2010). Table 1 show the concentrations and combinations of inulin, dextrin, alginate and olive oil assayed. Among them, we chosen OM-A/D and OM-A/I due to they showed the highest content of olive oil, which could provide adequate intake levels, and textural properties (Table 1). It is reasonable to assume that OM-A/D and OM-A/I could be used as animal fat replacer. These matrices were prepared by mixing sodium alginate (1%), CaSO4 (1%), sodium pyrophosphate (0.75%) and dextrin (2.25%) or inulin (2.25%) with water (40%) in a homogenizer (Thermomix TM 31, Vorwerk España M.S.L., S.C, Madrid) to obtaining OM-A/D and OM-A/I matrices respectively. Ca2þ-salt and sodium pyrophosphate were used in order to slow the gelling procedure (Weiss, Scherze, & Muschiolik, 2005). The mixtures were prepared at 1500 rpm for 20 s. Olive oil was gradually added to this mixture with the homogenizer in operation (1500 rpm). Each type of sample was stuffed into metal moulds (similar to moulds used for cooked ham) of 2 kg capacity under pressure (which is apply manual as the maximum to permit the moulds employed) to compact them and prevent air bubbles, and stored in a chilling room at 3 C for 24 h until analysis. Each matrix was prepared in duplicate. Technological properties, except penetration test, and structural characteristics were determined only in OM-A/D and OM-A/I because these samples showed the highest olive oil content and penetration force values, which indicate that they could be used as fat replacer. The oil content (55%) in this oil bulking agents seems appropriate for use as fat replacer for example in a meat product, which provided adequate intake levels of olive oil. The same mixtures of polysaccharide in aqueous solution (without added olive oil) were prepared for use as references for
376
A.M. Herrero et al. / Food Hydrocolloids 36 (2014) 374e381
Table 1 Formulation (g/100 g) and penetration force (N) of various olive oil bulking agents based on polysaccharides matrices. Samplesa
Sodium alginate
CaSO4
Sodium pyrophosphate
Inulin
OM-A/D30 OM-A/I30 OM-A30 OM-D30 OM-I30 OM-0.5A/D30 OM-0.5A/I30 OM-A/D OM-A/I OM-A OM-D OM-I OM-0.5A/D OM-0.5A/I OM-0.5A
1 1 1
1 1 1
0.75 0.75 0.75
2.25
Dextrin
Water
Olive oil
Penetration force (N)b
2.25
65 65 65 65 65 67.5 67.5 40 40 42.25 42.75 42.75 42.5 42.5 43.6
30 30 30 30 30 30 30 55 55 55 55 55 55 55 55
0.05 0.08
2.25 2.25 0.5 0.5 1 1 1
0.5 0.5 1 1 1
0.375 0.375 0.75 0.75 0.75
1.125 1.125 2.25 2.25 2.25 2.25
0.5 0.5 0.5
0.5 0.5 0.5
0.375 0.375 0.375
1.125 1.125
0.01 0.02
0.03 0.04 0.01
0.02 0.01
a Olive oil bulking agents (OM-) formulated with 30% of olive oil: OM-A/D30, prepared with alginate and dextrin; OM-A/I30, prepared with alginate and inulin; OM-A30, prepared with alginate; OM-D30, prepared with dextrin; OM-I30, prepared with inulin; OM-0.5A/D30, prepared with half of alginate and dextrin; OM-0.5A/I30, prepared with half of alginate and inulin. Olive oil bulking agents (OM-) formulated with 55% of olive oil: OM-A/D, prepared with alginate and dextrin; OM-A/I, prepared with alginate and inulin; OM-A, prepared with alginate; OM-D, prepared with dextrin; OM-I: prepared with inulin; OM-0.5A/D, prepared with half of alginate and dextrin; OM-0.5A/I, prepared with half of alginate and inulin. b ld: lower than 0.02 N (measurement limit in this condition).
spectroscopic measurements. These polysaccharide gels were denominated A/D and A/I for dextrin and inulin respectively combined with alginate.
first compression (N s); chewiness (Cw) ¼ Hd Ch Sp (N mm). Measurement of samples was carried out at room temperature.
2.3. Technological properties
2.4. Structural properties
2.3.1. Penetration test Penetration test was determined on the different samples showed in Table 1 at room temperature (22 1 C) (JiménezColmenero et al., 2010). The test was performed using a 6 mm diameter cylindrical stainless steel plunger attached to a 5 kg cell connected to the crosshead of TA-XT plus Texture Analyzer (Texture Technologies Corp., Scarsdale, NY). The corresponding forcee penetration curves (at 0.8 mm/s crosshead speed) were plotted and analysed. Penetration force (N) was determined as force exerted at 10 mm or the point of gel rupture. In same samples penetration force was not detected (ld).
2.4.1. FT-Raman spectroscopic analysis The following Raman spectra were measured: a) spectra of the polysaccharides gels in aqueous solution used as reference (A/D and A/I); b) spectra of olive oil (O) used as reference; and c) spectra of the different types of olive oil bulking matrices (OM-A/D and OM-A/I). In addition Raman spectra were measured in alginate, dextrin or inulin in aqueous solution at concentrations ranging from 30 to 90% to determine carbohydrate-water hydrogen bonding. Portions of the different samples were transferred to quartz cuvettes (ST-1/Q/10) (TEKNOKROMA, Barcelona, Spain) to fill them to a length of 1 cm. For each sample 1500 scans were recorded. This procedure was carried out in triplicate, giving a total of 4500 scans per sample. Measurements were performed on three samples from each olive oil bulking matrix. Spectra were excited with the 1064 nm Nd: YAG laser line and recorded on a Bruker RFS 100/S FTspectrometer. The scattered radiation was collected at 180 to the source, and frequency-dependent scattering of the Raman spectra, produced by the spectrometer, was corrected by multiplying point by point with (mlaser/m)4. The influence of the optics on the spectrometer was eliminated by using the Raman correction command from the Opus 2.2 software (Bruker, Karlsruhe, Germany). Reported frequencies are accurate to 0.5 cm1, as deduced from frequency standards measured in the spectrometer. Raman spectra were resolved at 4 cm1 resolution with a liquid nitrogencooled Ge detector. The samples were illuminated by laser power at 300 mW. Raman spectra were processed using the Bruker Opus 2.2 and Grams/AI version 9.1 (Thermo Electron Corporation, USA) software.
2.3.2. Stability of olive oil bulking matrices based on polysaccharide gels Thermal stability, in terms of water and fat binding properties, of the samples OM-A/I and OM-A/D was determined (in triplicate). Sample was stuffed into tubes, which were hermetically sealed and heated in a water bath for 30 min at 70 C. They were then opened and left to stand upside down (for 50 min) to release the separated fat and water onto a plate. Matrix stability, as total fluid release, was expressed as % of initial sample weight. 2.3.3. Texture Profile Analysis Texture profile analysis (TPA) was performed in a TA.XTplus Texture analyzer (Texture Technologies Corp., Scarsdale, NY, USA) as described by Bourne (1978). Six cylindrical cores (diam ¼ 20 mm, height ¼ 20 mm) from each olive oil (OM-A/I and OM-A/D) bulking matrix were axially compressed to 40% of their original height. Force-time deformation curves were obtained with a 5 kg load cell, applied at a crosshead speed of 1 mm/s. Attributes were calculated using the Texture Expert program as follows: hardness (Hd) ¼ peak force (N) required for first compression; cohesiveness (Ch) ¼ ratio of active work done under the second compression curve to that done under the first compression curve (dimensionless); springiness (Sp) ¼ distance (mm) the sample recovers after the first compression; adhesiveness (Ad) ¼ area under the abscissa after the
2.5. Statistical analysis Analysis of variance (ANOVA one way) and Tukey’s multiple range test were carried out in order to evaluate the statistical significance (P < 0.05) of the effect of olive oil bulking matrix formulation. The normal distribution of samples was checked using
A.M. Herrero et al. / Food Hydrocolloids 36 (2014) 374e381
the Shapiro Wilks test. The KruskaleWallis test was used to test samples that did not fit the normal distribution. Statistical analysis was performed using Statgraphics Plus version 5.0. 3. Results and discussion 3.1. Technological properties 3.1.1. Stability of olive oil bulking matrices based on polysaccharide gels Binding (water and fat) properties of the matrices OM-A/D and OM-A/I were optimal, with no noticeable release of exudates after heating (total fluid release was <0.5% for OM-A/I and <1.5% for OMA/D). Similar results have been reported in oil-in-water emulsions formulated with olive oil (Herrero et al., 2011a, 2011b) or combinations of plant and fish oils (Delgado-Pando et al., 2010). The excellent binding properties observed in the olive oil bulking matrices studied in the present work suggest that they are an appropriate choice for use as ingredients in the formulation of foods such as healthy lipid meat products. 3.1.2. Textural properties: Texture Profile Analysis The textural behaviour of olive oil bulking matrices based on polysaccharide gels (OM-A/D and OM-A/I) differed depending on the polysaccharide combination used in their formulation (Fig. 1). The olive oil bulking matrix stabilized by alginate/inulin (OM-A/I) presented the highest (P < 0.05) hardness, adhesiveness, and chewiness. Alginate gel formation is produced by the ionic interaction between guluronic acid residues from two or more alginate chains and cations, yielding a three-dimensional network of alginate molecules well described by the “egg-box model” (Braccini & Perez, 2001). Several important properties of these gels have been reported, such as mechanical strength, porosity, gel uniformity, biocompatibility and influence on encapsulated cells (Simpson, Grant, Blackband, & Constantinidis,. 2003). The formation of alginate gel and its textural properties are influenced by factors such as the concentration of alginate, calcium ion and sequestering agents, pH, temperature and rate of solubility (Roopa & Bhattacharya, 2010) Some authors (Roopa & Bhattacharya, 2009) have determined texture parameters of alginate gels with similar alginate/Ca2þ concentration to the ones studied in the present work, and the textural properties of these were more like those to polysaccharide matrices with oil stabilized by alginate/dextrin (OM-A/D). The ability of inulin to form gels by shearing or heating/cooling an inulin suspension and the physicochemical properties of these gels have been addressed in several papers (Bot et al., 2004; Kim et al.,
5,5
OM-A/I OM-A/D
377
2001; Zimeri & Kokini, 2002). The texture of inulin gels resembles that of solid fats, and therefore inulin gel may be suitable for use as a fat mimetic (Bot et al., 2004). Furthermore, some research papers have also studied the influence of inulin with other hydrocolloids in water (Giannouli, Richardson, & Morris, 2004; Zimeri & Kokini, 2003). The effect of inulin on the gelation of acyl gellan has been investigated (Evageliou et al., 2010). Acyl gellan is a water soluble polysaccharide which, like alginate, form gels on cooling in presence of cations. It has been reported that the presence of inulin affected gellan strength, so that increasing inulin amounts resulted in increase in gel strength. This behaviour was attributed to segregative interactions between gellan and inulin. This fact, may also explain why textural properties, in particular hardness and chewiness, were strongest (P < 0.05) in polysaccharide matrices with oil stabilized by alginate/inulin (OM-A/I) (Fig. 1). Similar textural behaviour to OM-A/I and OM-A/D has been found in olive oil-in-water emulsions stabilized with proteins or combination of proteins and microbial transglutaminase (Herrero et al., 2011a, 2011b), which have also been used as animal fat replacers in frankfurters with optimal technological properties (Jiménez-Colmenero et al., 2010). Moreover, textural parameters of OM-A/I and OM-A/D are in the same range as those found in gels made with carbohydrates like konjac glucomanan, also used as fat replacers in the preparation of cooked and dry- fermented healthier meat products (Ruiz-Capillas et al., 2012; SalcedoSandoval et al., 2013; Triki et al., 2013). These konjac glucomanan-based fat analogues had similar mechanical properties to pork fats after grinding (Jiménez-Colmenero et al., 2012), indicating that this konjac gel is especially suitable as a pork backfat replacer in fat reduction processes. Given that OM-A/D and OM-A/I had similar textural properties to konjac gel, it seems reasonable to suppose that these matrices will also be suitable as animal fat replacers. Studies performed with other cold binding agents (Herrero, Cambero, Ordóñez, de la Hoz, & Carmona, 2009) containing a complex mixture of proteins, such as rehydrated plasma powder, showed that the resulting gels were less hard than the olive oil polysaccharide matrices formulated with alginate and inulin or dextrin. These results suggest that the olive oil bulking agents developed in the present work (OM-A/I and OM-A/D) could be used as animal fat replacers in the development of healthier meat products. 3.2. Structural properties The combination of inulin, dextrin and alginate seems to form a three-dimensional network that immobilizes oil to produce polysaccharides matrices of oil with optimal technological properties in terms of mechanical and binding (fat and water) properties. In pursuit of insights into the molecular interactions involved in stabilization of these matrices (OM-A/D and OM-A/I), we examined through the Raman spectra of these matrices as reported below.
4,0
2,5
1,0
-0,5 Hd (N)
Ch
Sp (mm)
Ad (N s)
Cw (N mm)
Fig. 1. Texture Profile Analysis (TPA) parameters [hardness (Hd), cohesiveness (Ch), springiness (Sp), adhesiveness (Ad) and chewiness (Cw)] of the oil bulking agents prepared with alginate and dextrin or inulin (OM-A/D and OM-A/I, respectively).
3.2.1. Lipid interactions One of the spectroscopic regions of interest in Raman studies of molecular systems containing lipids corresponds to the CeH stretching modes (2800e3050 cm1), which are analysed as follows. In order to eliminate any spectral influence of water and other components (inulin, dextrin or alginate) in olive oil bulking agent spectra, the corresponding spectra of the aqueous mixtures of these components (A/D and A/I) were accordingly subtracted from olive oil polysaccharide matrix spectra (OM-A/I and OM-A/D). For that purpose the 985 cm1 band of CaSO4 (Nakamoto, 1986) was eliminated using a subtraction factor so that their intensity peak was not visible. In this way, the influence of CH stretching bands attributed to inulin, dextrin or alginate was removed. The bending
378
A.M. Herrero et al. / Food Hydrocolloids 36 (2014) 374e381
]CH band located near 1267 cm1, attributed to olive oil (Zou et al., 2009), was then used as internal standard for subsequent normalization of the resulting spectra. These OM-A/I and OM-A/D and olive oil spectra (used as reference) in the 2800e3050 cm1 region are included in Fig. 2. The spectra showed bands that are characteristic of symmetric and asymmetric CeH stretching vibrations of methyl and methylene groups in aliphatic molecules (Socrates, 2001). Thus, the CH2 asymmetric stretching band appears at 2931 cm1, the CH3 symmetric stretching band at 2898 cm1 and the CH2 symmetric stretching motion close to 2853 cm1. Another Raman band at 3007 cm1, attributable to cis-olefinic group ]CeH stretching vibration is clearly visible. Fig. 2 shows a decrease in the Raman intensity of the CeH stretching modes in the 2800e3050 cm1 range as a consequence of formation of the polysaccharide matrix with olive oil. This intensity decrease may be attributable to various molecular factors. For instance, the Raman intensity of a bond stretching mode is proportional to the square of the bond polarizability derivatives, a0 , which can be expressed analytically in terms of the electronegativity of the atoms in the bond (c), the bond order (n), and the bond length (r), as follows (Fontal & Spiro, 1977; Long & Plane, 1965)
Shortening of CeH bond length can result from the shifting of the above equilibrium to the polar forms of oil ester molecules due to the presence of hydrogen bonding where a carbonyl group acts as acceptor. In this way, the methylene CeH bonds adjacent to the carbonyl group increase their sp2 character to some extent, which involves shortening of CeH bond length compared with the sp3 Ce H bond length in the neutral form (Gussoni & Castiglioni, 2000). Therefore, such a reduction of CeH stretching intensity (Fig. 2) is indicative of the interaction between oil ester molecules and water and/or carbohydrate molecules through their carbonyl groups when the polysaccharide gel matrices with olive oil incorporated (OM-A/D and OM-A/I) are formed. In the Raman study of lipids, less use was made of features from polar headgroup vibrations as opposed to the different CeH stretching modes mentioned above. This was determined to some extent by the rather low intensity of a number of bands associated with the headgroup vibrations (Muik, Lendl, Molina-Díaz, & AyoraCañada, 2005). For instance, there was a comparatively weak band of the C]O stretching mode near 1748 cm1 (Fig. 3A). This band frequency decreased by about 1 cm1 in going from olive oil to the OM-A/D and OM-A/I matrices, which also suggest the presence of hydrogen bonds between oil carbonyl groups and water and/or carbohydrate molecules (Razumas et al., 1996). Additionally, the
a0 ¼ ð0:1Þcr3 ðn=2Þ As the atoms in the methylene bond are not changed upon formation of these matrices, the above equation suggests that the decrease in the Raman intensity of the CeH stretching modes in the 2800e3050 cm1 region (Fig. 2) is likely the result of a slight decrease in the length of the methylene CeH bond. Such as decrease in these matrices may arise from hyperconjugation of methylene CH2 groups with the electrons of double bonds such as carbonyl bonds in oil esters (triglycerides) as described below
Fig. 2. Raman spectra in the 2800e3050 cm1 region from olive oil (A) and difference spectra of the oil bulking agents prepared with alginate and dextrin (B) or inulin (C) [OM-A/D and OM-A/I, respectively] obtained after subtracting the corresponding spectra of the aqueous mixture A/D and A/I.
Fig. 3. Raman spectra in the 1720e1770 cm1 (A) and 1630e1680 cm1 (B) regions from olive oil (black solid line, upper spectrum) and difference spectra of the oil bulking agents prepared with alginate and dextrin (grey solid line) or inulin (black dashed line) [OM-A/D and OM-A/I, respectively] obtained after subtracting the corresponding spectra of the aqueous mixture A/D and A/I.
A.M. Herrero et al. / Food Hydrocolloids 36 (2014) 374e381
3.2.2. Polysaccharide interaction In order to study the behaviour of carbohydrate molecules in the formation of the polysaccharides matrix with added olive oil (OMA/D and OM-A/I), the oil spectral contributions were first appropriately subtracted using a set of olive oil spectra recorded under the same conditions. The criteria used were based on elimination of the 1267 cm1 band, attributed to olive oil (Zou et al., 2009), using a subtraction factor so that the intensity peaks were not visible. The resulting spectra were normalized using the CaSO4 band located near 985 cm1 (Nakamoto, 1986).
Table 2 Area values of nCH bands in the 2800e3000 cm1 region and relative intensity ratio of the asymmetric methylene stretching band to the symmetric methylene stretching band [nasCH2 band versus nsCH2 (InasCH2/InsCH2)] of the Raman spectra from oil bulking agents prepared with alginate and dextrin or inulin (OM-A/D and OM-A/I respectively). Samples
Area (2800e3000 cm1)
InasCH2/InsCH2
Olive oil OM-A/D OM-A/I
220.6 0.7a 114.3 1.0b 115.5 1.3b
1.076 0.03a 0.965 0.05a 0.970 0.02a
Mean values resulting from measurement of three replicates from each sample standard deviation. Different superscript letters in the same column indicate significant differences (P < 0.05).
Fig. 4. Comparison of normalized Raman spectra of aqueous solution of a mixture of calcium alginate, CaSO4, sodium phosphate and dextrin (A/D) or inulin (A/D) with their difference spectra (lower) obtained after subtracting the olive oil spectrum of the oil bulking agents prepared with alginate and dextrin or inulin (OM-A/D and OM-A/I respectively) in the 2775e3050 cm1 region.
The difference spectra show negative bands in the 2800e 3000 cm1 range (Fig. 4) which can result from interactions of oil with water and -carbohydrate molecules interactions and from carbohydrate-water hydrogen bonding. The difference in intensity between OM-A/I and OM-A/D suggests that these interactions were more relevant in the presence of inulin. In fact, Fig. 5 shows clearly that the intensity of the CeH stretching bands decreases with increasing water content. This decrease was more pronounced in the case of inulin, possibly due to carbohydrate-water hydrogen bonding, which was stronger in the presence of inulin than of dextrin. Scanning electron microscopy has shown that inulin is highly soluble in water and absorbs water; this produce a change in its structure and morphology, resulting in completely disaggregated inulin granules (Manno et al., 2009). As we found in this case, some authors have reported the presence of carbohydratewater hydrogen bonding in other carbohydrates such as sucrose or maltodextrin (van Duynhoven, Kulik, Jonker, & Haverkamp, 1999; Ka curáková & Mathlouthi, 1996). In addition, on the basis of some literature studies (Gussoni & Castiglioni, 2000; Mizuno, Imafuji, Fujiwara, Ohta, & Tamiya, 2003) dealing with hydrogen bonding of water with various molecular species having CH bonds
8 Alginate
Intensity ν CH
intensity of the CH2 bending band located near 1655 cm1 increased upon formation of OM-A/D and OM-A/I (Fig. 3B); this spectral feature is similar to reports for monoolein emulsions having hydrogen-bonded carbonyl groups (Razumas et al., 1996). The intensity decrease of this band upon formation of these polysaccharide gel matrices of olive oil may be explained in terms of an increase in lateral pressure due to an order-induced increase in density near the methyl end (Wong & Mantsch, 1983). Table 2 shows the comparative areas values of the 2800e 3000 cm1 spectral profiles of olive oil, OM-A/I and OM-A/D. Significant differences were found between pure olive oil and polysaccharides gel matrices of olive oil (OM-A/I and OM-A/D), which shows significant interactions of oil acyl chains with water and/or carbohydrate molecules in the formation of the oil matrices. Similar area values (P < 0.05) were found for both OM-A/I and OM-A/D, which may thus play a similar role to these polysaccharides in the hydrogen-bonded carbonyl groups. Another aspect to be considered in these systems is the structural organization of lipid molecules interacting with water, namely the degree to which water molecules penetrate the oil micelles (Lindman, Wennerström, Gustavsson, Kamenka, & Brun, 1980). In this connection, if water molecules penetrate the lipid micelles, all vibrational modes of the methylene chains should shift to higher frequencies due to the low refractive index of water (Umemura, Cameron, & Mantsch, 1980). However, the present results strongly indicate that the penetration of water in these micelles was insignificant given that not all methylene chain modes shift to higher frequencies (Fig. 2). These results are thus consistent with the presence of lipid carbonyl groups which are hydrogen-bonded to other polar groups at the micelle surface. It is assumed that the ratio of the peak-height intensity of the asymmetric methylene stretching band to that of the symmetric methylene stretching band directly reflects interchain vibrational coupling and mobility of acyl chains (Mendelsohn, Dluhy, Taraschi, Cameron, & Mantsch, 1981). Given this intensity ratio dependence, the results of Table 2 indicate lower (P < 0.05) acyl chain mobility in olive oil polysaccharides gel matrices than in pure olive oil. This is to be expected from micelle formations where lipid acyl chains are relatively fixed through hydrogen bonding and intermolecular order.
379
Inulin
6
Dextrin 4
2
0 30
50
70
90
Water content (%) Fig. 5. Intensity of CH stretching bands of alginate, inulin and dextrin versus water content.
380
A.M. Herrero et al. / Food Hydrocolloids 36 (2014) 374e381
within the oil micelles seems insignificant. Dextrin and inulin seem to play a similar role when hydrogen-bonded to carbonyl groups, and intermolecular contacts between lipid carbonyl groups and water molecules occur at the level of the micelle surface. As regards carbohydrate-water interactions, some Raman spectral features indicate the presence of carbohydrate-water hydrogen bonding. In this connection, inulin seems to be more strongly bonded to water than dextrin. This difference in the structural behaviour of inulin and dextrin in terms of hydrogen bonding to water may explain the differences referred to in technological properties. Acknowledgements This research was supported by projects AGL2010-19515, AGL2011-29644-C02-01, and Consolider-Ingenio 2010:CARNISENUSA (CSD2007-00016). Fig. 6. Normalized Raman spectra in the 1500e1800 cm1 region of aqueous solution of a mixture of calcium alginate, CaSO4, sodium phosphate and dextrin (A/D) or inulin (A/I).
that are contiguous to oxygen atoms, the decrease of CH stretching intensity can be explained by the binding of a hydrogen bond donor to the lone electron pairs of the carbohydrate oxygen atoms. The contention that inulin-water interactions are stronger than dextrin-water is also supported by the frequency of the scissoring mode of water molecules (Falk, 1984). Thus the intensity maximum of this vibrational mode in inulin aqueous solution (A/I) appears at a higher frequency than in dextrin aqueous solution (A/D) due to carbohydrate-water hydrogen bonding (Fig. 6), which is stronger in inulin-water interactions. This difference behaviour of inulin and dextrin regarding their hydrogen bonding to water may explain the differences in some textural properties of OM-A/I and OM-A/D systems, particularly hardness, adhesiveness and chewiness (Fig. 1). Additionally, the difference in the strength of inulin-water and dextrin-water interactions could be related to lower total fluid release of OM-A/I as mentioned above (Section 3.1). 4. Conclusion Olive oil can be immobilized by combining either inulin or dextrin with alginate, resulting in a three-dimensional network. This appears to be a feasible strategy for producing olive oil bulking agents based on polysaccharide gel matrices with desirable nutritional and technological properties that make them suitable for use as fat replacers. In terms of lipid content (55% olive oil), these matrixes present a fatty acid profile more in line with health recommendations. As regards technological properties these oil bulking agents displayed desirable binding (water and fat) properties and textural behaviour which support their suitability as animal fat replacers. All these aspects are very important for choosing the oil bulking agent that is most suitable and effective for use as food ingredient, as part of a technological strategy for the development of healthier food (i.e. healthy meat product) and for consumer acceptance, since they affect quality properties of the reformulated products in which they will be incorporated. The stabilization of these polysaccharide gel matrices with olive oil incorporated can be the result of interactions of lipid phase with water and -carbohydrate molecules and carbohydrate-water interactions. The interactions involving the oil molecules consist of hydrogen bonding between oil carbonyl groups and water and/or carbohydrate molecules. Therefore, lipid acyl chains are relatively fixed through hydrogen bonding and intermolecular order upon micelle formation. In relation to structural organization of lipid molecules in terms of the penetration degree of water molecules
References Abreu, F. O. M. S., Oliveira, E. F., Paula, H. C. B., & de Paula, R. C. M. (2012). Chitosan/ cashew gum nanogels for essential oil encapsulation. Carbohydrate Polymers, 89, 1277e1282. Bot, A., Erle, U., Vreeker, R., & Agterof, W. G. M. (2004). Influence of crystallization conditions on the large deformation rheology of inulin gels. Food Hydrocolloids, 18, 547e556. Bourne, M. C. (1978). Texture profile analysis. Food Technology, 32, 62e65. Braccini, I., & Perez, S. (2001). Molecular basis of Ca2þ -induced gelation in alginates and pectins: the egg-box model revisited. Biomacromolecules, 2, 1089e1096. Bush, S. F., Adams, R. G., & Levin, I. W. (1980). Structural reorganizations in lipid bilayer systems: effect of hydration and sterol addition on Raman spectra of dipalmitoylphosphatidylcholine multilayers. Biochemistry, 19, 4429e4436. Carvalho, J., Gonçalves, C., Gil, A. M., & Gama, F. M. (2007). Production and characterization of a new dextrin based hydrogel. European Polymer Journal, 43, 3050e3059. Choi, S. M., Yuen, S. N., Phillips, D. L., & Ma, C. Y. (2010). Applications of vibrational spectroscopy to the analysis of polysaccharide and hydrocolloid ingredients. In E. C. Y. Li-Chan, P. R. Griffiths, & J. M. Chalmers (Eds.), Applications of vibrational spectroscopy in food science (Vol. II; pp. 577e592). West Sussex: John Wiley & Sons Ltd. Delgado-Pando, G., Cofrades, S., Ruiz-Capillas, C., Solas, M. T., & Jiménez Colmenero, F. (2010). Healthier lipid combination oil-in-water emulsions prepared with various protein systems. An approach for the development of functional meat product. European Journal of Lipid Science and Technology, 112, 791e801. Dobson, G. (2001). Spectroscopy and spectrometry of lipids e Part 1-Lipid analysis by vibrational spectroscopy. European Journal of Lipid Science and Technology, 103, 815e826. Dumitriu, R. P., Mitchell, G. R., & Vasile, C. (2011). Multi-responsive hydrogels based on N-isopropylacrylamide and sodium alginate. Polymer International, 60, 222e233. van Duynhoven, J. P. M., Kulik, A. S., Jonker, H. R. A., & Haverkamp, J. (1999). Solidlike components in carbohydrate gels probed by NMR spectroscopy. Carbohydrate Polymers, 40, 211e219. Evageliou, V., Tseliou, G., Mandala, I., & Komaitis, M. (2010). Effect of inulin on texture and clarity of gellan gels. Journal of Food Engineering, 101, 381e385. Falk, M. (1984). The frequency of the HeOeH bending fundamental in solids and liquids. Speclrochimica Acta, 4OA, 43e48. Fontal, B., & Spiro, T. G. (1977). Raman intensities and bond ordersdA simplified delta function equation. Spectrochimica Acta Part A: Molecular Spectroscopy, 33, 507e515. Forrest, G. (1978). Raman spectroscopy of the milk globule membrane and triglycerides. Chemistry and Physics of Lipids, 21, 237e252. Gaonkar, A. G. (1991). Surface and interfacial activities and emulsion characteristics of some food hydrocolloids. Food Hydrocolloids, 5, 329e337. Garcia, H., Barros, A. S., Gonçalves, C., Gama, F. M., & Gil, A. M. (2008). Characterization of dextrin hydrogels by FTIR spectroscopy and solid state NMR spectroscopy. European Polymer Journal, 44, 2318e2329. Giannouli, P., Richardson, R. K., & Morris, E. R. (2004). Effect of polymeric cosolutes on calcium pectinate gelation. Part 2. Dextrans and inulin. Carbohydrate Polymers, 55, 357e365. Gussoni, M., & Castiglioni, C. (2000). Infrared intensities. Use of the CH-stretching band intensity as a tool for evaluating the acidity of hydrogen atoms in hydrocarbons. Journal of Molecular Structure, 521, 1e18. Harrington, J. C., & Morris, E. R. (2009). Conformational ordering and gelation of gelatin in mixtures with soluble polysaccharides. Food Hydrocolloids, 23, 327e336. Herrero, A. M., Cambero, M. I., Ordóñez, J. A., de la Hoz, L., & Carmona, P. (2009). Plasma powder as cold-set binding agent for meat system: rheological and Raman spectroscopy study. Food Chemistry, 113, 493e499.
A.M. Herrero et al. / Food Hydrocolloids 36 (2014) 374e381 Herrero, A. M., Carmona, P., Pintado, T., Jiménez-Colmenero, F., & Ruíz-Capillas, C. (2011a). Olive oil-in-water emulsions stabilized with caseinate: elucidation of proteinelipid interactions by infrared spectroscopy. Food Hydrocolloids, 25, 12e18. Herrero, A. M., Carmona, P., Pintado, T., Jiménez-Colmenero, F., & Ruíz-Capillas, C. (2011b). Infrared spectroscopic analysis of structural features and interactions in olive oil-in-water emulsions stabilized with soy protein. Food Research International, 44, 360e366. Howell, N. K., Henryk, H., & Li-Chan, E. C. Y. (2001). Elucidation of proteinelipid interactions in a lysozymeecorn oil system by Fourier Transform Raman spectroscopy. Journal of Agricultural and Food Chemistry, 49, 1529e1533. Huang, X., Kakuda, Y., & Cui, W. (2001). Hydrocolloids in emulsions: particle size distribution and interfacial activity. Food Hydrocolloids, 15, 533e542. Jiménez-Colmenero, F. (2007). Healthier lipid formulation approaches in meatbased functional foods. Technological options for replacement of meat fats by non-meat fats. Trends in Food Science and Technology, 18, 567e578. Jiménez-Colmenero, F., Cofrades, S., Herrero, A. M., Fernández-Martín, F., RodríguezSalas, L., & Ruiz-Capillas, C. (2012). Konjac gel fat analogue for use in meat products: comparison with pork fats. Food Hydrocolloids, 26, 63e72. Jiménez-Colmenero, F., Herrero, A. M., Pintado, T., Solas, M. T., & Ruiz-Capillas, C. (2010). Influence of emulsified olive oil stabilizing system used for pork backfat replacement in frankfurters. Food Research International, 43, 2068e2076. Ka curáková, M., & Mathlouthi, M. (1996). FTIR and laser-Raman spectra of oligosaccharides in water: characterization of the glycosidic bond. Carbohydrate Research, 284, 145e157. Kim, Y., Faqih, M. N., & Wang, S. S. (2001). Factors affecting gel formation of inulin. Carbohydrate Polymers, 46, 135e145. Lindman, B., Wennerström, H., Gustavsson, H., Kamenka, N., & Brun, B. (1980). Some aspects on the hydration of surfactant micelles. Pure and Applied Chemistry, 52, 1307e1315. Long, T. V., & Plane, R. A. (1965). Calculation of absolute Raman intensities of A1 modes of molecules and ions from a delta-function potential model. Journal of Physical Chemistry, 43, 457e461. Manno, D., Filippo, E., Serra, A., Negro, C., De Bellis, L., & Miceli, A. (2009). The influence of inulin addition on the morphological and structural properties of durum wheat pasta. International Journal of Food Science & Technology, 44, 2218e2224. Mendelsohn, R., Dluhy, R., Taraschi, T., Cameron, D. G., & Mantsch, H. H. (1981). Raman and Fourier transform infrared spectroscopic studies of the interaction between glycophorin and dimyristoylphosphatidylcholine. Biochemistry, 20, 6699e6706. Meng, G., Chan, J. C. K., Rousseau, D., & Li-Chan, E. C. Y. (2005). Study of proteine lipid interactions at the bovine serum albumin/oil interface by Raman microspectroscopy. Journal of Agricultural and Food Chemistry, 53, 845e852. Mizuno, K., Imafuji, S., Fujiwara, T., Ohta, T., & Tamiya, Y. (2003). Hydration of the CH groups in 1,4-dioxane probed by NMR and IR: contribution of blue-shifting CH.OH2 hydrogen bonds. Journal of Physical Chemistry B, 107, 3972e3978. Muik, B., Lendl, B., Molina-Díaz, A., & Ayora-Cañada, M. J. (2005). Direct monitoring of lipid oxidation in edible oils by Fourier transform Raman spectroscopy. Chemistry and Physics of Lipids, 134, 173e182.
381
Nakamoto, K. (1986). Infrared and Raman spectra of inorganic and coordination compounds (4th ed.). New York: John Wiley and Sons Inc. Pereira, L., Amado, A. M., Critchley, A. T., van de Velde, F., & Ribeiro-Claro, P. J. A. (2009). Identification of selected seaweed polysaccharides (phycocolloids) by vibrational spectroscopy (FTIR-ATR and FT-Raman). Food Hydrocolloids, 23, 1903e1909. Pielesz, A., & Ba˛ k, M. K. K. (2008). Raman spectroscopy and WAXS method as a tool for analysing ion-exchange properties of alginate hydrogels. International Journal of Biological Macromolecules, 43, 438e443. Razumas, V., Talaikytedot, Z., Barauskasa, J., Larsson, K., Miezis, Y., & Nylander, T. (1996). Effects of distearoylphosphatidylglycerol and lysozyme on the structure of the monoolein-water cubic phase: X-ray diffraction and Raman scattering studies. Chemistry and Physics of Lipids, 84, 123e138. Roopa, B. S., & Bhattacharya, S. (2009). Alginate gels: rupture characteristics as a function of the conditions of gel formation. Journal of Food Engineering, 91, 448e454. Roopa, B. S., & Bhattacharya, S. (2010). Alginate gels: II. Stability at different processing conditions. Journal of Food Process Engineering, 33, 466e480. Ruiz-Capillas, C., Triki, M., Herrero, A. M., Rodríguez-Salas, L., & JiménezColmenero, F. (2012). Konjac gel as pork backfat replacer in dry fermented sausages: processing and quality characteristics. Meat Science, 92, 144e150. Salcedo-Sandoval, L., Cofrades, S., Ruiz-Capillas, C., Solas, M. T., & JiménezColmenero, F. (2013). Healthier oils stabilized in Konjac matrix as fat replacers in n-3 PUFA enriched Frankfurters. Meat Science, 93, 757e766. Simpson, N. E., Grant, S. C., Blackband, S. J., & Constantinidis, I. (2003). NMR properties of alginate microbeads. Biomaterials, 24, 4941e4948. Socrates, G. (2001). Infrared and Raman characteristic group frequencies: Tables and charts. England: John Willey & Sons, Ltd. Triki, M., Herrero, A. M., Jiménez-Colmenero, F., & Ruiz-Capillas, C. (2013). Effect of preformed konjac gels, with and without olive oil, on the technological attributes and storage stability of Merguez sausage. Meat Science, 93, 351e360. Tungland, B. C., & Meyer, D. (2002). Nondigestible oligo- and polysaccharides (dietary fiber): their physiology and role in human health and food. Comprehensive Reviews in Food Science and Food Safety, 1, 90e109. Umemura, J., Cameron, D. G., & Mantsch, H. H. (1980). FT-IR study of micelle formation in aqueous sodium n-hexanoate solutions. Journal of Physical Chemistry, 84, 2272e2277. Weiss, J., Scherze, I., & Muschiolik, G. (2005). Polysaccharide gel with multiple emulsion. Food Hydrocolloids, 19, 605e615. Wong, P. T. T., & Mantsch, H. H. (1983). Temperature-induced phase transition and structural changes in micellar solutions of sodium oleate observed by Raman scattering. Journal of Physical Chemistry, 87, 2436e2443. Zhang, J., Daubert, C. R., & Foegeding, E. (2005). Fracture analysis of alginate gels. Journal of Food Science, 70, 425e431. Zimeri, J. E., & Kokini, J. L. (2002). The effect of moisture content on the crystallinity and glass transition temperature of inulin. Carbohydrate Polymers, 48, 299e304. Zimeri, J. E., & Kokini, J. L. (2003). Rheological properties of inulinewaxy maize starch systems. Carbohydrate Polymers, 52, 67e85. Zou, M. Q., Zhang, X. F., Qi, X. H., Ma, H. L., Dong, Y., Liu, C. W., et al. (2009). Rapid authentication of olive oil adulteration by Raman spectrometry. Journal of Agricultural and Food Chemistry, 57, 6001e6006.
Contents lists available at ScienceDirect
Food Hydrocolloids journal homepage: www.elsevier.com/locate/foodhyd
Polysaccharide gels as oil bulking agents: Technological and structural properties A.M. Herrero a, *, P. Carmona b, F. Jiménez-Colmenero a, C. Ruiz-Capillas a a b
Instituto de Ciencia y Tecnología de Alimentos y Nutrición (ICTAN-CSIC), (Formerly, Instituto del Frío), Ciudad Universitaria, 28040 Madrid, Spain Instituto de Estructura de la Materia (CSIC), Serrano 121, 28006 Madrid, Spain
a r t i c l e i n f o
a b s t r a c t
Article history: Received 17 January 2013 Accepted 8 August 2013
Lipid and polysaccharide interactions in various polysaccharide gels prepared for use as oil bulking agents were investigated using Raman spectroscopy. Two different polysaccharide gel matrices containing olive oil were prepared using mixed biopolymer systems of alginate with inulin (OM-A/I) or dextrin (OM-A/D). Stability and textural properties were also evaluated in these matrices. Thermal stability of the different matrices was optimal. The textural behaviour of polysaccharide matrices with olive oil incorporated differed depending on the polysaccharide gels used in their formulation. OM-A/I presented the highest (P < 0.05) hardness, adhesiveness, and chewiness. These matrices were stabilized by hydrogen bonding between oil carbonyl groups and water and/or carbohydrate molecules. Lipid acyl chains are relatively fixed through hydrogen bonding and intermolecular order upon micelle formation. Raman spectroscopic results also showed carbohydrate-water hydrogen bonding, in which inulin seemed to be more strongly bonded to water than dextrin. This difference in the structural behaviour of inulin and dextrin in terms of hydrogen bonding to water may explain the differences in textural properties. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Polysaccharide Olive oil Texture Raman spectroscopy Structural characteristics
1. Introduction Improving the fat content of foods based on processing strategies (reformulation) is one of the most important current approaches to the development of potential functional foods, especially in the case of meat-based functional foods. Healthier lipid reformulation processes are generally based on replacement of the animal fat normally present in the product with another fat whose characteristics are more in line with health recommendations (Jiménez-Colmenero, 2007). A variety of non-meat fats of plant and marine origin have been used in product formulation (as non-meat ingredients) to partially replaces meat fats (mainly pork or beef). These oils have been incorporated in meat products in liquid and solid forms, encapsulated, or as oil-in-water emulsions (Jiménez-Colmenero, 2007). Strategies for incorporation of healthy oils in a gel-like matrix to form oil’s bulking agent (in which this new ingredient acts as an animal fat replacer) could offer new possibilities for improving the fat content of meat products. Polysaccharides, used either individually or in combination, can be used to create a variety of gel structures which may be suitable for immobilizing oil droplets and thus to acting as oil’s bulking agents.
* Corresponding author. Tel.: þ34 915616800; fax: þ34 91 5645557. E-mail address: [email protected] (A.M. Herrero). 0268-005X/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.foodhyd.2013.08.008
In this regard, healthy oils in a konjac matrix have been used to improve fat content in dry fermented sausage (Ruiz-Capillas, Triki, Herrero, Rodríguez-Salas, & Jiménez-Colmenero, 2012), fresh merguez sausages (Triki, Herrero, Jiménez-Colmenero, & Ruiz-Capillas, 2013) and frankfurters (Salcedo-Sandoval, Cofrades, Ruiz-Capillas, Solas, & Jiménez-Colmenero, 2013). Alginate gels may offer interesting possibilities as oil bulking agents. Alginates are of particular interest for their ability to form gels in the presence of calcium salt (Roopa & Bhattacharya, 2010; Zhang, Daubert, & Foegeding, 2005). These gels consist of polymeric molecules cross-linked to form a three-dimensional macromolecular network containing a large fraction of water within their structure which displaying mechanical rigidity. The properties of the gel are the net result of the complex interactions between the components. One of the most important properties of alginate gels, which make them highly versatile, is their capacity for controlled uptake, release and retention of molecules. This capacity is due in turn, to specific interactions of the macromolecular network with the diffusing or retained molecule. In this connection, hydrocolloids as alginate, unlike emulsifiers, generally stabilize emulsions by increasing the viscosity, which inhibits the rate of coalescence between the oil droplets. This compound is adsorbed at the oil/water interface to form an interfacial film and sterically stabilize emulsions against flocculation and coalescence (Huang, Kakuda, & Cui, 2001). In addition to functioning as gelling agents, they also
A.M. Herrero et al. / Food Hydrocolloids 36 (2014) 374e381
exhibit properties as emulsifiers, encapsulating agents, dispersing agents, foam stabilizers, film formers, and crystal inhibitors (Gaonkar, 1991). Alginate gel properties and their utility as oil bulking agents can be modulated by mixing with other biopolymers. Various researchers have investigated the properties of mixed biopolymer gelling systems for that purpose in some cases achieving substantial enhancements in gel strength (Evageliou, Tseliou Mandala, & Komaitis, 2010; Harrington & Morris, 2009; Zimeri & Kokini, 2003). Inulin is a polysaccharide which offers interesting technological and health-beneficial properties. The technological properties are related to the degree of polymerization of its chains, making it suitable for use as a low-calorie sweetener, fat replacer or texturizing agent (Tungland & Meyer, 2002). The ability of inulin to form gels, and the physicochemical properties of these gels, have been addressed by numerous researchers (Bot, Erle, Vreeker, & Agterof, 2004; Kim, Faqih, & Wang, 2001; Zimeri & Kokini, 2002). Some authors have reported that inulin gels resemble solid fats in texture and may be used as fat mimetic (Bot et al., 2004). Increasing the amount of inulin can result in an increase in the gel strength of gellan (Evageliou et al., 2010). Another polysaccharide of interest is dextrin, a low-molecularweight glucose polymer resulting from partial hydrolysis of starch. Dextrin is a widely used material with a variety of applications, particularly in the food industry for adhesives, and also in other applications where it offers considerable potential for hydrogel production (Carvalho, Gonçalves, Gil, & Gama, 2007; Garcia, Barros, Gonçalves, Gama, & Gil, 2008). The properties of these matrices with oil entrapped in polysaccharide gels are the net result of complex interactions among the components. Raman spectroscopy is a direct, non-invasive technique which has proven useful for providing structural information on the various components (lipids, polysaccharides, etc.) (Choi, Yuen, Phillips, & Ma, 2010; Dobson, 2001) involved in the preparation of these matrices. This spectroscopic technique has been used to identify selected seaweed polysaccharides (Pereira, Amado, Critchley, van de Velde, & Ribeiro-Claro, 2009) and the structural behaviour of oligosaccharides in water (Ka curáková & Mathlouthi, 1996). In particular, some authors have performed Raman spectroscopic analysis of alginate hydrogels (Dumitriu, Mitchell, & Vasile, 2011; Pielesz & Ba˛ k, 2008). In addition, the technique has been used to investigate the conformational behaviour of lipid bilayer systems (dipalmitoylphosphatidylcholine) perturbed by cholesterol and water (Bush, Adams, & Levin, 1980). By means of Raman spectral frequencies and intensities it is possible both to monitor intramolecular changes occurring within the three structurally distinct regions of the phospholipid molecule and to follow alterations in lattice order and packing characteristics. The acyl chain mobilities of bovine milk fat globule lipids and component triglycerides have also been determined by Raman spectroscopy as a function of temperature. This study showed that the CH stretching region is the most useful for lipid chain order analysis (Forrest, 1978). Also, Raman spectroscopy has proven the role of proteins and lipids in emulsion formation (Howell, Henryk, & Li-Chan, 2001; Meng, Chan, Rousseau, & Li-Chan, 2005). Moreover, Raman spectroscopy has been used to characterize the development of polymeric matrices based on chitosan/cashew gum for encapsulation of a natural essential oil (Abreu, Oliveira, Paula, & de Paula, 2012). These findings speak to the feasibility or the potential of this technique as a means to elucidate interactions between the components of matrices of oil molecules involved in polysaccharide gels. The aim of the present study was to develop olive oil bulking agents using mixed biopolymer systems of alginate with inulin or dextrin with optimal characteristics for use as animal fat replacers
375
in the development of healthier meat products. Two main aspects of these agents were considered: their technological characteristics, since these affect quality properties of the reformulated products in which they will be incorporated; and the other structural characteristics of the system and interactions between their components, in order to establish causal relationships between their structural and technological properties. 2. Materials and methods 2.1. Materials Ingredients used for preparation of polysaccharide gels as oil’s bulking agents included: olive oil (13% SFA, 79% MUFA and 8% PUFA) (Carbonell Virgen Extra, SOS Cuétara SA, Madrid, Spain); sodium alginate (SA) (90% carbohydrates content) (Tradissimo, TRADES S.A., Barcelona Spain); calcium sulfate (CS) (Panreac Química, S.A. Madrid, Spain), tetra-sodium pyrophosphate anhydrous PRS (SP) (Panreac Química, S.A. Madrid, Spain), inulin (I) consisting mainly of chicory inulin (>90% inulin) (TRADES S.A., Barcelona Spain) and white maize dextrin (D) (CARGILL S.L.U- CTS Rubi, Barcelona, Spain). 2.2. Preparation of olive oil bulking matrices based on polysaccharide gels Different concentrations and combinations of inulin, dextrin, and alginate and olive oil were assayed to determine the best conditions, from a technological standpoint (optimal fat content more in line with health recommendations, without loss of exudate and with a suitable texture for cutting, grinding, etc.), for production of an oil bulking agent that could use as fat replacer (DelgadoPando, Cofrades, Ruiz-Capillas, Solas, & Jiménez Colmenero, 2010; Herrero, Carmona, Pintado, Jiménez-Colmenero, & Ruíz-Capillas, 2011a, 2011b; Jiménez-Colmenero et al., 2012; Jiménez-Colmenero, Herrero, Pintado, Solas, & Ruiz-Capillas, 2010). Table 1 show the concentrations and combinations of inulin, dextrin, alginate and olive oil assayed. Among them, we chosen OM-A/D and OM-A/I due to they showed the highest content of olive oil, which could provide adequate intake levels, and textural properties (Table 1). It is reasonable to assume that OM-A/D and OM-A/I could be used as animal fat replacer. These matrices were prepared by mixing sodium alginate (1%), CaSO4 (1%), sodium pyrophosphate (0.75%) and dextrin (2.25%) or inulin (2.25%) with water (40%) in a homogenizer (Thermomix TM 31, Vorwerk España M.S.L., S.C, Madrid) to obtaining OM-A/D and OM-A/I matrices respectively. Ca2þ-salt and sodium pyrophosphate were used in order to slow the gelling procedure (Weiss, Scherze, & Muschiolik, 2005). The mixtures were prepared at 1500 rpm for 20 s. Olive oil was gradually added to this mixture with the homogenizer in operation (1500 rpm). Each type of sample was stuffed into metal moulds (similar to moulds used for cooked ham) of 2 kg capacity under pressure (which is apply manual as the maximum to permit the moulds employed) to compact them and prevent air bubbles, and stored in a chilling room at 3 C for 24 h until analysis. Each matrix was prepared in duplicate. Technological properties, except penetration test, and structural characteristics were determined only in OM-A/D and OM-A/I because these samples showed the highest olive oil content and penetration force values, which indicate that they could be used as fat replacer. The oil content (55%) in this oil bulking agents seems appropriate for use as fat replacer for example in a meat product, which provided adequate intake levels of olive oil. The same mixtures of polysaccharide in aqueous solution (without added olive oil) were prepared for use as references for
376
A.M. Herrero et al. / Food Hydrocolloids 36 (2014) 374e381
Table 1 Formulation (g/100 g) and penetration force (N) of various olive oil bulking agents based on polysaccharides matrices. Samplesa
Sodium alginate
CaSO4
Sodium pyrophosphate
Inulin
OM-A/D30 OM-A/I30 OM-A30 OM-D30 OM-I30 OM-0.5A/D30 OM-0.5A/I30 OM-A/D OM-A/I OM-A OM-D OM-I OM-0.5A/D OM-0.5A/I OM-0.5A
1 1 1
1 1 1
0.75 0.75 0.75
2.25
Dextrin
Water
Olive oil
Penetration force (N)b
2.25
65 65 65 65 65 67.5 67.5 40 40 42.25 42.75 42.75 42.5 42.5 43.6
30 30 30 30 30 30 30 55 55 55 55 55 55 55 55
0.05 0.08
2.25 2.25 0.5 0.5 1 1 1
0.5 0.5 1 1 1
0.375 0.375 0.75 0.75 0.75
1.125 1.125 2.25 2.25 2.25 2.25
0.5 0.5 0.5
0.5 0.5 0.5
0.375 0.375 0.375
1.125 1.125
0.01 0.02
0.03 0.04 0.01
0.02 0.01
a Olive oil bulking agents (OM-) formulated with 30% of olive oil: OM-A/D30, prepared with alginate and dextrin; OM-A/I30, prepared with alginate and inulin; OM-A30, prepared with alginate; OM-D30, prepared with dextrin; OM-I30, prepared with inulin; OM-0.5A/D30, prepared with half of alginate and dextrin; OM-0.5A/I30, prepared with half of alginate and inulin. Olive oil bulking agents (OM-) formulated with 55% of olive oil: OM-A/D, prepared with alginate and dextrin; OM-A/I, prepared with alginate and inulin; OM-A, prepared with alginate; OM-D, prepared with dextrin; OM-I: prepared with inulin; OM-0.5A/D, prepared with half of alginate and dextrin; OM-0.5A/I, prepared with half of alginate and inulin. b ld: lower than 0.02 N (measurement limit in this condition).
spectroscopic measurements. These polysaccharide gels were denominated A/D and A/I for dextrin and inulin respectively combined with alginate.
first compression (N s); chewiness (Cw) ¼ Hd Ch Sp (N mm). Measurement of samples was carried out at room temperature.
2.3. Technological properties
2.4. Structural properties
2.3.1. Penetration test Penetration test was determined on the different samples showed in Table 1 at room temperature (22 1 C) (JiménezColmenero et al., 2010). The test was performed using a 6 mm diameter cylindrical stainless steel plunger attached to a 5 kg cell connected to the crosshead of TA-XT plus Texture Analyzer (Texture Technologies Corp., Scarsdale, NY). The corresponding forcee penetration curves (at 0.8 mm/s crosshead speed) were plotted and analysed. Penetration force (N) was determined as force exerted at 10 mm or the point of gel rupture. In same samples penetration force was not detected (ld).
2.4.1. FT-Raman spectroscopic analysis The following Raman spectra were measured: a) spectra of the polysaccharides gels in aqueous solution used as reference (A/D and A/I); b) spectra of olive oil (O) used as reference; and c) spectra of the different types of olive oil bulking matrices (OM-A/D and OM-A/I). In addition Raman spectra were measured in alginate, dextrin or inulin in aqueous solution at concentrations ranging from 30 to 90% to determine carbohydrate-water hydrogen bonding. Portions of the different samples were transferred to quartz cuvettes (ST-1/Q/10) (TEKNOKROMA, Barcelona, Spain) to fill them to a length of 1 cm. For each sample 1500 scans were recorded. This procedure was carried out in triplicate, giving a total of 4500 scans per sample. Measurements were performed on three samples from each olive oil bulking matrix. Spectra were excited with the 1064 nm Nd: YAG laser line and recorded on a Bruker RFS 100/S FTspectrometer. The scattered radiation was collected at 180 to the source, and frequency-dependent scattering of the Raman spectra, produced by the spectrometer, was corrected by multiplying point by point with (mlaser/m)4. The influence of the optics on the spectrometer was eliminated by using the Raman correction command from the Opus 2.2 software (Bruker, Karlsruhe, Germany). Reported frequencies are accurate to 0.5 cm1, as deduced from frequency standards measured in the spectrometer. Raman spectra were resolved at 4 cm1 resolution with a liquid nitrogencooled Ge detector. The samples were illuminated by laser power at 300 mW. Raman spectra were processed using the Bruker Opus 2.2 and Grams/AI version 9.1 (Thermo Electron Corporation, USA) software.
2.3.2. Stability of olive oil bulking matrices based on polysaccharide gels Thermal stability, in terms of water and fat binding properties, of the samples OM-A/I and OM-A/D was determined (in triplicate). Sample was stuffed into tubes, which were hermetically sealed and heated in a water bath for 30 min at 70 C. They were then opened and left to stand upside down (for 50 min) to release the separated fat and water onto a plate. Matrix stability, as total fluid release, was expressed as % of initial sample weight. 2.3.3. Texture Profile Analysis Texture profile analysis (TPA) was performed in a TA.XTplus Texture analyzer (Texture Technologies Corp., Scarsdale, NY, USA) as described by Bourne (1978). Six cylindrical cores (diam ¼ 20 mm, height ¼ 20 mm) from each olive oil (OM-A/I and OM-A/D) bulking matrix were axially compressed to 40% of their original height. Force-time deformation curves were obtained with a 5 kg load cell, applied at a crosshead speed of 1 mm/s. Attributes were calculated using the Texture Expert program as follows: hardness (Hd) ¼ peak force (N) required for first compression; cohesiveness (Ch) ¼ ratio of active work done under the second compression curve to that done under the first compression curve (dimensionless); springiness (Sp) ¼ distance (mm) the sample recovers after the first compression; adhesiveness (Ad) ¼ area under the abscissa after the
2.5. Statistical analysis Analysis of variance (ANOVA one way) and Tukey’s multiple range test were carried out in order to evaluate the statistical significance (P < 0.05) of the effect of olive oil bulking matrix formulation. The normal distribution of samples was checked using
A.M. Herrero et al. / Food Hydrocolloids 36 (2014) 374e381
the Shapiro Wilks test. The KruskaleWallis test was used to test samples that did not fit the normal distribution. Statistical analysis was performed using Statgraphics Plus version 5.0. 3. Results and discussion 3.1. Technological properties 3.1.1. Stability of olive oil bulking matrices based on polysaccharide gels Binding (water and fat) properties of the matrices OM-A/D and OM-A/I were optimal, with no noticeable release of exudates after heating (total fluid release was <0.5% for OM-A/I and <1.5% for OMA/D). Similar results have been reported in oil-in-water emulsions formulated with olive oil (Herrero et al., 2011a, 2011b) or combinations of plant and fish oils (Delgado-Pando et al., 2010). The excellent binding properties observed in the olive oil bulking matrices studied in the present work suggest that they are an appropriate choice for use as ingredients in the formulation of foods such as healthy lipid meat products. 3.1.2. Textural properties: Texture Profile Analysis The textural behaviour of olive oil bulking matrices based on polysaccharide gels (OM-A/D and OM-A/I) differed depending on the polysaccharide combination used in their formulation (Fig. 1). The olive oil bulking matrix stabilized by alginate/inulin (OM-A/I) presented the highest (P < 0.05) hardness, adhesiveness, and chewiness. Alginate gel formation is produced by the ionic interaction between guluronic acid residues from two or more alginate chains and cations, yielding a three-dimensional network of alginate molecules well described by the “egg-box model” (Braccini & Perez, 2001). Several important properties of these gels have been reported, such as mechanical strength, porosity, gel uniformity, biocompatibility and influence on encapsulated cells (Simpson, Grant, Blackband, & Constantinidis,. 2003). The formation of alginate gel and its textural properties are influenced by factors such as the concentration of alginate, calcium ion and sequestering agents, pH, temperature and rate of solubility (Roopa & Bhattacharya, 2010) Some authors (Roopa & Bhattacharya, 2009) have determined texture parameters of alginate gels with similar alginate/Ca2þ concentration to the ones studied in the present work, and the textural properties of these were more like those to polysaccharide matrices with oil stabilized by alginate/dextrin (OM-A/D). The ability of inulin to form gels by shearing or heating/cooling an inulin suspension and the physicochemical properties of these gels have been addressed in several papers (Bot et al., 2004; Kim et al.,
5,5
OM-A/I OM-A/D
377
2001; Zimeri & Kokini, 2002). The texture of inulin gels resembles that of solid fats, and therefore inulin gel may be suitable for use as a fat mimetic (Bot et al., 2004). Furthermore, some research papers have also studied the influence of inulin with other hydrocolloids in water (Giannouli, Richardson, & Morris, 2004; Zimeri & Kokini, 2003). The effect of inulin on the gelation of acyl gellan has been investigated (Evageliou et al., 2010). Acyl gellan is a water soluble polysaccharide which, like alginate, form gels on cooling in presence of cations. It has been reported that the presence of inulin affected gellan strength, so that increasing inulin amounts resulted in increase in gel strength. This behaviour was attributed to segregative interactions between gellan and inulin. This fact, may also explain why textural properties, in particular hardness and chewiness, were strongest (P < 0.05) in polysaccharide matrices with oil stabilized by alginate/inulin (OM-A/I) (Fig. 1). Similar textural behaviour to OM-A/I and OM-A/D has been found in olive oil-in-water emulsions stabilized with proteins or combination of proteins and microbial transglutaminase (Herrero et al., 2011a, 2011b), which have also been used as animal fat replacers in frankfurters with optimal technological properties (Jiménez-Colmenero et al., 2010). Moreover, textural parameters of OM-A/I and OM-A/D are in the same range as those found in gels made with carbohydrates like konjac glucomanan, also used as fat replacers in the preparation of cooked and dry- fermented healthier meat products (Ruiz-Capillas et al., 2012; SalcedoSandoval et al., 2013; Triki et al., 2013). These konjac glucomanan-based fat analogues had similar mechanical properties to pork fats after grinding (Jiménez-Colmenero et al., 2012), indicating that this konjac gel is especially suitable as a pork backfat replacer in fat reduction processes. Given that OM-A/D and OM-A/I had similar textural properties to konjac gel, it seems reasonable to suppose that these matrices will also be suitable as animal fat replacers. Studies performed with other cold binding agents (Herrero, Cambero, Ordóñez, de la Hoz, & Carmona, 2009) containing a complex mixture of proteins, such as rehydrated plasma powder, showed that the resulting gels were less hard than the olive oil polysaccharide matrices formulated with alginate and inulin or dextrin. These results suggest that the olive oil bulking agents developed in the present work (OM-A/I and OM-A/D) could be used as animal fat replacers in the development of healthier meat products. 3.2. Structural properties The combination of inulin, dextrin and alginate seems to form a three-dimensional network that immobilizes oil to produce polysaccharides matrices of oil with optimal technological properties in terms of mechanical and binding (fat and water) properties. In pursuit of insights into the molecular interactions involved in stabilization of these matrices (OM-A/D and OM-A/I), we examined through the Raman spectra of these matrices as reported below.
4,0
2,5
1,0
-0,5 Hd (N)
Ch
Sp (mm)
Ad (N s)
Cw (N mm)
Fig. 1. Texture Profile Analysis (TPA) parameters [hardness (Hd), cohesiveness (Ch), springiness (Sp), adhesiveness (Ad) and chewiness (Cw)] of the oil bulking agents prepared with alginate and dextrin or inulin (OM-A/D and OM-A/I, respectively).
3.2.1. Lipid interactions One of the spectroscopic regions of interest in Raman studies of molecular systems containing lipids corresponds to the CeH stretching modes (2800e3050 cm1), which are analysed as follows. In order to eliminate any spectral influence of water and other components (inulin, dextrin or alginate) in olive oil bulking agent spectra, the corresponding spectra of the aqueous mixtures of these components (A/D and A/I) were accordingly subtracted from olive oil polysaccharide matrix spectra (OM-A/I and OM-A/D). For that purpose the 985 cm1 band of CaSO4 (Nakamoto, 1986) was eliminated using a subtraction factor so that their intensity peak was not visible. In this way, the influence of CH stretching bands attributed to inulin, dextrin or alginate was removed. The bending
378
A.M. Herrero et al. / Food Hydrocolloids 36 (2014) 374e381
]CH band located near 1267 cm1, attributed to olive oil (Zou et al., 2009), was then used as internal standard for subsequent normalization of the resulting spectra. These OM-A/I and OM-A/D and olive oil spectra (used as reference) in the 2800e3050 cm1 region are included in Fig. 2. The spectra showed bands that are characteristic of symmetric and asymmetric CeH stretching vibrations of methyl and methylene groups in aliphatic molecules (Socrates, 2001). Thus, the CH2 asymmetric stretching band appears at 2931 cm1, the CH3 symmetric stretching band at 2898 cm1 and the CH2 symmetric stretching motion close to 2853 cm1. Another Raman band at 3007 cm1, attributable to cis-olefinic group ]CeH stretching vibration is clearly visible. Fig. 2 shows a decrease in the Raman intensity of the CeH stretching modes in the 2800e3050 cm1 range as a consequence of formation of the polysaccharide matrix with olive oil. This intensity decrease may be attributable to various molecular factors. For instance, the Raman intensity of a bond stretching mode is proportional to the square of the bond polarizability derivatives, a0 , which can be expressed analytically in terms of the electronegativity of the atoms in the bond (c), the bond order (n), and the bond length (r), as follows (Fontal & Spiro, 1977; Long & Plane, 1965)
Shortening of CeH bond length can result from the shifting of the above equilibrium to the polar forms of oil ester molecules due to the presence of hydrogen bonding where a carbonyl group acts as acceptor. In this way, the methylene CeH bonds adjacent to the carbonyl group increase their sp2 character to some extent, which involves shortening of CeH bond length compared with the sp3 Ce H bond length in the neutral form (Gussoni & Castiglioni, 2000). Therefore, such a reduction of CeH stretching intensity (Fig. 2) is indicative of the interaction between oil ester molecules and water and/or carbohydrate molecules through their carbonyl groups when the polysaccharide gel matrices with olive oil incorporated (OM-A/D and OM-A/I) are formed. In the Raman study of lipids, less use was made of features from polar headgroup vibrations as opposed to the different CeH stretching modes mentioned above. This was determined to some extent by the rather low intensity of a number of bands associated with the headgroup vibrations (Muik, Lendl, Molina-Díaz, & AyoraCañada, 2005). For instance, there was a comparatively weak band of the C]O stretching mode near 1748 cm1 (Fig. 3A). This band frequency decreased by about 1 cm1 in going from olive oil to the OM-A/D and OM-A/I matrices, which also suggest the presence of hydrogen bonds between oil carbonyl groups and water and/or carbohydrate molecules (Razumas et al., 1996). Additionally, the
a0 ¼ ð0:1Þcr3 ðn=2Þ As the atoms in the methylene bond are not changed upon formation of these matrices, the above equation suggests that the decrease in the Raman intensity of the CeH stretching modes in the 2800e3050 cm1 region (Fig. 2) is likely the result of a slight decrease in the length of the methylene CeH bond. Such as decrease in these matrices may arise from hyperconjugation of methylene CH2 groups with the electrons of double bonds such as carbonyl bonds in oil esters (triglycerides) as described below
Fig. 2. Raman spectra in the 2800e3050 cm1 region from olive oil (A) and difference spectra of the oil bulking agents prepared with alginate and dextrin (B) or inulin (C) [OM-A/D and OM-A/I, respectively] obtained after subtracting the corresponding spectra of the aqueous mixture A/D and A/I.
Fig. 3. Raman spectra in the 1720e1770 cm1 (A) and 1630e1680 cm1 (B) regions from olive oil (black solid line, upper spectrum) and difference spectra of the oil bulking agents prepared with alginate and dextrin (grey solid line) or inulin (black dashed line) [OM-A/D and OM-A/I, respectively] obtained after subtracting the corresponding spectra of the aqueous mixture A/D and A/I.
A.M. Herrero et al. / Food Hydrocolloids 36 (2014) 374e381
3.2.2. Polysaccharide interaction In order to study the behaviour of carbohydrate molecules in the formation of the polysaccharides matrix with added olive oil (OMA/D and OM-A/I), the oil spectral contributions were first appropriately subtracted using a set of olive oil spectra recorded under the same conditions. The criteria used were based on elimination of the 1267 cm1 band, attributed to olive oil (Zou et al., 2009), using a subtraction factor so that the intensity peaks were not visible. The resulting spectra were normalized using the CaSO4 band located near 985 cm1 (Nakamoto, 1986).
Table 2 Area values of nCH bands in the 2800e3000 cm1 region and relative intensity ratio of the asymmetric methylene stretching band to the symmetric methylene stretching band [nasCH2 band versus nsCH2 (InasCH2/InsCH2)] of the Raman spectra from oil bulking agents prepared with alginate and dextrin or inulin (OM-A/D and OM-A/I respectively). Samples
Area (2800e3000 cm1)
InasCH2/InsCH2
Olive oil OM-A/D OM-A/I
220.6 0.7a 114.3 1.0b 115.5 1.3b
1.076 0.03a 0.965 0.05a 0.970 0.02a
Mean values resulting from measurement of three replicates from each sample standard deviation. Different superscript letters in the same column indicate significant differences (P < 0.05).
Fig. 4. Comparison of normalized Raman spectra of aqueous solution of a mixture of calcium alginate, CaSO4, sodium phosphate and dextrin (A/D) or inulin (A/D) with their difference spectra (lower) obtained after subtracting the olive oil spectrum of the oil bulking agents prepared with alginate and dextrin or inulin (OM-A/D and OM-A/I respectively) in the 2775e3050 cm1 region.
The difference spectra show negative bands in the 2800e 3000 cm1 range (Fig. 4) which can result from interactions of oil with water and -carbohydrate molecules interactions and from carbohydrate-water hydrogen bonding. The difference in intensity between OM-A/I and OM-A/D suggests that these interactions were more relevant in the presence of inulin. In fact, Fig. 5 shows clearly that the intensity of the CeH stretching bands decreases with increasing water content. This decrease was more pronounced in the case of inulin, possibly due to carbohydrate-water hydrogen bonding, which was stronger in the presence of inulin than of dextrin. Scanning electron microscopy has shown that inulin is highly soluble in water and absorbs water; this produce a change in its structure and morphology, resulting in completely disaggregated inulin granules (Manno et al., 2009). As we found in this case, some authors have reported the presence of carbohydratewater hydrogen bonding in other carbohydrates such as sucrose or maltodextrin (van Duynhoven, Kulik, Jonker, & Haverkamp, 1999; Ka curáková & Mathlouthi, 1996). In addition, on the basis of some literature studies (Gussoni & Castiglioni, 2000; Mizuno, Imafuji, Fujiwara, Ohta, & Tamiya, 2003) dealing with hydrogen bonding of water with various molecular species having CH bonds
8 Alginate
Intensity ν CH
intensity of the CH2 bending band located near 1655 cm1 increased upon formation of OM-A/D and OM-A/I (Fig. 3B); this spectral feature is similar to reports for monoolein emulsions having hydrogen-bonded carbonyl groups (Razumas et al., 1996). The intensity decrease of this band upon formation of these polysaccharide gel matrices of olive oil may be explained in terms of an increase in lateral pressure due to an order-induced increase in density near the methyl end (Wong & Mantsch, 1983). Table 2 shows the comparative areas values of the 2800e 3000 cm1 spectral profiles of olive oil, OM-A/I and OM-A/D. Significant differences were found between pure olive oil and polysaccharides gel matrices of olive oil (OM-A/I and OM-A/D), which shows significant interactions of oil acyl chains with water and/or carbohydrate molecules in the formation of the oil matrices. Similar area values (P < 0.05) were found for both OM-A/I and OM-A/D, which may thus play a similar role to these polysaccharides in the hydrogen-bonded carbonyl groups. Another aspect to be considered in these systems is the structural organization of lipid molecules interacting with water, namely the degree to which water molecules penetrate the oil micelles (Lindman, Wennerström, Gustavsson, Kamenka, & Brun, 1980). In this connection, if water molecules penetrate the lipid micelles, all vibrational modes of the methylene chains should shift to higher frequencies due to the low refractive index of water (Umemura, Cameron, & Mantsch, 1980). However, the present results strongly indicate that the penetration of water in these micelles was insignificant given that not all methylene chain modes shift to higher frequencies (Fig. 2). These results are thus consistent with the presence of lipid carbonyl groups which are hydrogen-bonded to other polar groups at the micelle surface. It is assumed that the ratio of the peak-height intensity of the asymmetric methylene stretching band to that of the symmetric methylene stretching band directly reflects interchain vibrational coupling and mobility of acyl chains (Mendelsohn, Dluhy, Taraschi, Cameron, & Mantsch, 1981). Given this intensity ratio dependence, the results of Table 2 indicate lower (P < 0.05) acyl chain mobility in olive oil polysaccharides gel matrices than in pure olive oil. This is to be expected from micelle formations where lipid acyl chains are relatively fixed through hydrogen bonding and intermolecular order.
379
Inulin
6
Dextrin 4
2
0 30
50
70
90
Water content (%) Fig. 5. Intensity of CH stretching bands of alginate, inulin and dextrin versus water content.
380
A.M. Herrero et al. / Food Hydrocolloids 36 (2014) 374e381
within the oil micelles seems insignificant. Dextrin and inulin seem to play a similar role when hydrogen-bonded to carbonyl groups, and intermolecular contacts between lipid carbonyl groups and water molecules occur at the level of the micelle surface. As regards carbohydrate-water interactions, some Raman spectral features indicate the presence of carbohydrate-water hydrogen bonding. In this connection, inulin seems to be more strongly bonded to water than dextrin. This difference in the structural behaviour of inulin and dextrin in terms of hydrogen bonding to water may explain the differences referred to in technological properties. Acknowledgements This research was supported by projects AGL2010-19515, AGL2011-29644-C02-01, and Consolider-Ingenio 2010:CARNISENUSA (CSD2007-00016). Fig. 6. Normalized Raman spectra in the 1500e1800 cm1 region of aqueous solution of a mixture of calcium alginate, CaSO4, sodium phosphate and dextrin (A/D) or inulin (A/I).
that are contiguous to oxygen atoms, the decrease of CH stretching intensity can be explained by the binding of a hydrogen bond donor to the lone electron pairs of the carbohydrate oxygen atoms. The contention that inulin-water interactions are stronger than dextrin-water is also supported by the frequency of the scissoring mode of water molecules (Falk, 1984). Thus the intensity maximum of this vibrational mode in inulin aqueous solution (A/I) appears at a higher frequency than in dextrin aqueous solution (A/D) due to carbohydrate-water hydrogen bonding (Fig. 6), which is stronger in inulin-water interactions. This difference behaviour of inulin and dextrin regarding their hydrogen bonding to water may explain the differences in some textural properties of OM-A/I and OM-A/D systems, particularly hardness, adhesiveness and chewiness (Fig. 1). Additionally, the difference in the strength of inulin-water and dextrin-water interactions could be related to lower total fluid release of OM-A/I as mentioned above (Section 3.1). 4. Conclusion Olive oil can be immobilized by combining either inulin or dextrin with alginate, resulting in a three-dimensional network. This appears to be a feasible strategy for producing olive oil bulking agents based on polysaccharide gel matrices with desirable nutritional and technological properties that make them suitable for use as fat replacers. In terms of lipid content (55% olive oil), these matrixes present a fatty acid profile more in line with health recommendations. As regards technological properties these oil bulking agents displayed desirable binding (water and fat) properties and textural behaviour which support their suitability as animal fat replacers. All these aspects are very important for choosing the oil bulking agent that is most suitable and effective for use as food ingredient, as part of a technological strategy for the development of healthier food (i.e. healthy meat product) and for consumer acceptance, since they affect quality properties of the reformulated products in which they will be incorporated. The stabilization of these polysaccharide gel matrices with olive oil incorporated can be the result of interactions of lipid phase with water and -carbohydrate molecules and carbohydrate-water interactions. The interactions involving the oil molecules consist of hydrogen bonding between oil carbonyl groups and water and/or carbohydrate molecules. Therefore, lipid acyl chains are relatively fixed through hydrogen bonding and intermolecular order upon micelle formation. In relation to structural organization of lipid molecules in terms of the penetration degree of water molecules
References Abreu, F. O. M. S., Oliveira, E. F., Paula, H. C. B., & de Paula, R. C. M. (2012). Chitosan/ cashew gum nanogels for essential oil encapsulation. Carbohydrate Polymers, 89, 1277e1282. Bot, A., Erle, U., Vreeker, R., & Agterof, W. G. M. (2004). Influence of crystallization conditions on the large deformation rheology of inulin gels. Food Hydrocolloids, 18, 547e556. Bourne, M. C. (1978). Texture profile analysis. Food Technology, 32, 62e65. Braccini, I., & Perez, S. (2001). Molecular basis of Ca2þ -induced gelation in alginates and pectins: the egg-box model revisited. Biomacromolecules, 2, 1089e1096. Bush, S. F., Adams, R. G., & Levin, I. W. (1980). Structural reorganizations in lipid bilayer systems: effect of hydration and sterol addition on Raman spectra of dipalmitoylphosphatidylcholine multilayers. Biochemistry, 19, 4429e4436. Carvalho, J., Gonçalves, C., Gil, A. M., & Gama, F. M. (2007). Production and characterization of a new dextrin based hydrogel. European Polymer Journal, 43, 3050e3059. Choi, S. M., Yuen, S. N., Phillips, D. L., & Ma, C. Y. (2010). Applications of vibrational spectroscopy to the analysis of polysaccharide and hydrocolloid ingredients. In E. C. Y. Li-Chan, P. R. Griffiths, & J. M. Chalmers (Eds.), Applications of vibrational spectroscopy in food science (Vol. II; pp. 577e592). West Sussex: John Wiley & Sons Ltd. Delgado-Pando, G., Cofrades, S., Ruiz-Capillas, C., Solas, M. T., & Jiménez Colmenero, F. (2010). Healthier lipid combination oil-in-water emulsions prepared with various protein systems. An approach for the development of functional meat product. European Journal of Lipid Science and Technology, 112, 791e801. Dobson, G. (2001). Spectroscopy and spectrometry of lipids e Part 1-Lipid analysis by vibrational spectroscopy. European Journal of Lipid Science and Technology, 103, 815e826. Dumitriu, R. P., Mitchell, G. R., & Vasile, C. (2011). Multi-responsive hydrogels based on N-isopropylacrylamide and sodium alginate. Polymer International, 60, 222e233. van Duynhoven, J. P. M., Kulik, A. S., Jonker, H. R. A., & Haverkamp, J. (1999). Solidlike components in carbohydrate gels probed by NMR spectroscopy. Carbohydrate Polymers, 40, 211e219. Evageliou, V., Tseliou, G., Mandala, I., & Komaitis, M. (2010). Effect of inulin on texture and clarity of gellan gels. Journal of Food Engineering, 101, 381e385. Falk, M. (1984). The frequency of the HeOeH bending fundamental in solids and liquids. Speclrochimica Acta, 4OA, 43e48. Fontal, B., & Spiro, T. G. (1977). Raman intensities and bond ordersdA simplified delta function equation. Spectrochimica Acta Part A: Molecular Spectroscopy, 33, 507e515. Forrest, G. (1978). Raman spectroscopy of the milk globule membrane and triglycerides. Chemistry and Physics of Lipids, 21, 237e252. Gaonkar, A. G. (1991). Surface and interfacial activities and emulsion characteristics of some food hydrocolloids. Food Hydrocolloids, 5, 329e337. Garcia, H., Barros, A. S., Gonçalves, C., Gama, F. M., & Gil, A. M. (2008). Characterization of dextrin hydrogels by FTIR spectroscopy and solid state NMR spectroscopy. European Polymer Journal, 44, 2318e2329. Giannouli, P., Richardson, R. K., & Morris, E. R. (2004). Effect of polymeric cosolutes on calcium pectinate gelation. Part 2. Dextrans and inulin. Carbohydrate Polymers, 55, 357e365. Gussoni, M., & Castiglioni, C. (2000). Infrared intensities. Use of the CH-stretching band intensity as a tool for evaluating the acidity of hydrogen atoms in hydrocarbons. Journal of Molecular Structure, 521, 1e18. Harrington, J. C., & Morris, E. R. (2009). Conformational ordering and gelation of gelatin in mixtures with soluble polysaccharides. Food Hydrocolloids, 23, 327e336. Herrero, A. M., Cambero, M. I., Ordóñez, J. A., de la Hoz, L., & Carmona, P. (2009). Plasma powder as cold-set binding agent for meat system: rheological and Raman spectroscopy study. Food Chemistry, 113, 493e499.
A.M. Herrero et al. / Food Hydrocolloids 36 (2014) 374e381 Herrero, A. M., Carmona, P., Pintado, T., Jiménez-Colmenero, F., & Ruíz-Capillas, C. (2011a). Olive oil-in-water emulsions stabilized with caseinate: elucidation of proteinelipid interactions by infrared spectroscopy. Food Hydrocolloids, 25, 12e18. Herrero, A. M., Carmona, P., Pintado, T., Jiménez-Colmenero, F., & Ruíz-Capillas, C. (2011b). Infrared spectroscopic analysis of structural features and interactions in olive oil-in-water emulsions stabilized with soy protein. Food Research International, 44, 360e366. Howell, N. K., Henryk, H., & Li-Chan, E. C. Y. (2001). Elucidation of proteinelipid interactions in a lysozymeecorn oil system by Fourier Transform Raman spectroscopy. Journal of Agricultural and Food Chemistry, 49, 1529e1533. Huang, X., Kakuda, Y., & Cui, W. (2001). Hydrocolloids in emulsions: particle size distribution and interfacial activity. Food Hydrocolloids, 15, 533e542. Jiménez-Colmenero, F. (2007). Healthier lipid formulation approaches in meatbased functional foods. Technological options for replacement of meat fats by non-meat fats. Trends in Food Science and Technology, 18, 567e578. Jiménez-Colmenero, F., Cofrades, S., Herrero, A. M., Fernández-Martín, F., RodríguezSalas, L., & Ruiz-Capillas, C. (2012). Konjac gel fat analogue for use in meat products: comparison with pork fats. Food Hydrocolloids, 26, 63e72. Jiménez-Colmenero, F., Herrero, A. M., Pintado, T., Solas, M. T., & Ruiz-Capillas, C. (2010). Influence of emulsified olive oil stabilizing system used for pork backfat replacement in frankfurters. Food Research International, 43, 2068e2076. Ka curáková, M., & Mathlouthi, M. (1996). FTIR and laser-Raman spectra of oligosaccharides in water: characterization of the glycosidic bond. Carbohydrate Research, 284, 145e157. Kim, Y., Faqih, M. N., & Wang, S. S. (2001). Factors affecting gel formation of inulin. Carbohydrate Polymers, 46, 135e145. Lindman, B., Wennerström, H., Gustavsson, H., Kamenka, N., & Brun, B. (1980). Some aspects on the hydration of surfactant micelles. Pure and Applied Chemistry, 52, 1307e1315. Long, T. V., & Plane, R. A. (1965). Calculation of absolute Raman intensities of A1 modes of molecules and ions from a delta-function potential model. Journal of Physical Chemistry, 43, 457e461. Manno, D., Filippo, E., Serra, A., Negro, C., De Bellis, L., & Miceli, A. (2009). The influence of inulin addition on the morphological and structural properties of durum wheat pasta. International Journal of Food Science & Technology, 44, 2218e2224. Mendelsohn, R., Dluhy, R., Taraschi, T., Cameron, D. G., & Mantsch, H. H. (1981). Raman and Fourier transform infrared spectroscopic studies of the interaction between glycophorin and dimyristoylphosphatidylcholine. Biochemistry, 20, 6699e6706. Meng, G., Chan, J. C. K., Rousseau, D., & Li-Chan, E. C. Y. (2005). Study of proteine lipid interactions at the bovine serum albumin/oil interface by Raman microspectroscopy. Journal of Agricultural and Food Chemistry, 53, 845e852. Mizuno, K., Imafuji, S., Fujiwara, T., Ohta, T., & Tamiya, Y. (2003). Hydration of the CH groups in 1,4-dioxane probed by NMR and IR: contribution of blue-shifting CH.OH2 hydrogen bonds. Journal of Physical Chemistry B, 107, 3972e3978. Muik, B., Lendl, B., Molina-Díaz, A., & Ayora-Cañada, M. J. (2005). Direct monitoring of lipid oxidation in edible oils by Fourier transform Raman spectroscopy. Chemistry and Physics of Lipids, 134, 173e182.
381
Nakamoto, K. (1986). Infrared and Raman spectra of inorganic and coordination compounds (4th ed.). New York: John Wiley and Sons Inc. Pereira, L., Amado, A. M., Critchley, A. T., van de Velde, F., & Ribeiro-Claro, P. J. A. (2009). Identification of selected seaweed polysaccharides (phycocolloids) by vibrational spectroscopy (FTIR-ATR and FT-Raman). Food Hydrocolloids, 23, 1903e1909. Pielesz, A., & Ba˛ k, M. K. K. (2008). Raman spectroscopy and WAXS method as a tool for analysing ion-exchange properties of alginate hydrogels. International Journal of Biological Macromolecules, 43, 438e443. Razumas, V., Talaikytedot, Z., Barauskasa, J., Larsson, K., Miezis, Y., & Nylander, T. (1996). Effects of distearoylphosphatidylglycerol and lysozyme on the structure of the monoolein-water cubic phase: X-ray diffraction and Raman scattering studies. Chemistry and Physics of Lipids, 84, 123e138. Roopa, B. S., & Bhattacharya, S. (2009). Alginate gels: rupture characteristics as a function of the conditions of gel formation. Journal of Food Engineering, 91, 448e454. Roopa, B. S., & Bhattacharya, S. (2010). Alginate gels: II. Stability at different processing conditions. Journal of Food Process Engineering, 33, 466e480. Ruiz-Capillas, C., Triki, M., Herrero, A. M., Rodríguez-Salas, L., & JiménezColmenero, F. (2012). Konjac gel as pork backfat replacer in dry fermented sausages: processing and quality characteristics. Meat Science, 92, 144e150. Salcedo-Sandoval, L., Cofrades, S., Ruiz-Capillas, C., Solas, M. T., & JiménezColmenero, F. (2013). Healthier oils stabilized in Konjac matrix as fat replacers in n-3 PUFA enriched Frankfurters. Meat Science, 93, 757e766. Simpson, N. E., Grant, S. C., Blackband, S. J., & Constantinidis, I. (2003). NMR properties of alginate microbeads. Biomaterials, 24, 4941e4948. Socrates, G. (2001). Infrared and Raman characteristic group frequencies: Tables and charts. England: John Willey & Sons, Ltd. Triki, M., Herrero, A. M., Jiménez-Colmenero, F., & Ruiz-Capillas, C. (2013). Effect of preformed konjac gels, with and without olive oil, on the technological attributes and storage stability of Merguez sausage. Meat Science, 93, 351e360. Tungland, B. C., & Meyer, D. (2002). Nondigestible oligo- and polysaccharides (dietary fiber): their physiology and role in human health and food. Comprehensive Reviews in Food Science and Food Safety, 1, 90e109. Umemura, J., Cameron, D. G., & Mantsch, H. H. (1980). FT-IR study of micelle formation in aqueous sodium n-hexanoate solutions. Journal of Physical Chemistry, 84, 2272e2277. Weiss, J., Scherze, I., & Muschiolik, G. (2005). Polysaccharide gel with multiple emulsion. Food Hydrocolloids, 19, 605e615. Wong, P. T. T., & Mantsch, H. H. (1983). Temperature-induced phase transition and structural changes in micellar solutions of sodium oleate observed by Raman scattering. Journal of Physical Chemistry, 87, 2436e2443. Zhang, J., Daubert, C. R., & Foegeding, E. (2005). Fracture analysis of alginate gels. Journal of Food Science, 70, 425e431. Zimeri, J. E., & Kokini, J. L. (2002). The effect of moisture content on the crystallinity and glass transition temperature of inulin. Carbohydrate Polymers, 48, 299e304. Zimeri, J. E., & Kokini, J. L. (2003). Rheological properties of inulinewaxy maize starch systems. Carbohydrate Polymers, 52, 67e85. Zou, M. Q., Zhang, X. F., Qi, X. H., Ma, H. L., Dong, Y., Liu, C. W., et al. (2009). Rapid authentication of olive oil adulteration by Raman spectrometry. Journal of Agricultural and Food Chemistry, 57, 6001e6006.