Gelation of carrageenan: Effects of sugars and polyols

Gelation of carrageenan: Effects of sugars and polyols

Food Hydrocolloids 54 (2016) 284e292 Contents lists available at ScienceDirect Food Hydrocolloids journal homepage: www.elsevier.com/locate/foodhyd ...

578KB Sizes 5 Downloads 55 Views

Food Hydrocolloids 54 (2016) 284e292

Contents lists available at ScienceDirect

Food Hydrocolloids journal homepage: www.elsevier.com/locate/foodhyd

Gelation of carrageenan: Effects of sugars and polyols Richard Stenner a, Nobuyuki Matubayasi b, c, Seishi Shimizu a, * a

York Structural Biology Laboratory, Department of Chemistry, University of York, Heslington, York YO10 5DD, United Kingdom Division of Chemical Engineering, Graduate School of Engineering Science, Osaka University, Toyonaka, Osaka 560-8531, Japan c Elements Strategy Initiative for Catalysts and Batteries, Kyoto University, Katsura, Kyoto 615-8520, Japan b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 28 August 2015 Received in revised form 8 October 2015 Accepted 12 October 2015 Available online 19 October 2015

The gelation of k-carrageenan, a common polysaccharide for food gelling and thickening, is enhanced by the addition of sugars and polyols. Three different hypotheses have been proposed to explain this phenomenon: (i) the enhancement of the water structure around polysaccharides, and the concurrent change in polysaccharide hydration, induced by the cosolvent; (ii) exclusion of cosolvents from polysaccharide surfaces; (iii) binding between sugars/polyols and polysaccharides in the gel phase. To examine the above hypotheses, as well as to reveal the true driving forces responsible for gelation enhancement, we applied our recent statistical thermodynamic theory of gelation, derived from the rigorous KirkwoodeBuff theory. The hydration change is shown to be negligibly small relative to cosolvent exclusion and cosolvent-biopolymer binding, which can be rationalized by considering a sol egel equilibrium in which (1) the exclusion of sugars/polyols from k-carrageenan's surface in the sol phase, and (2) the binding of sugars and polyols in the gel phase, shift the solegel equilibrium of kcarrageenan to the gel state. This novel picture is consistent with a wealth of experimental evidence and provides a mechanistic insight into how polyol and sugar cosolvents influence the gelation of k-carrageenan on a molecular level. © 2015 Elsevier Ltd. All rights reserved.

Keywords: k-Carrageenan Gelation Statistical thermodynamics KirkwoodeBuff theory Sugar Polyol

1. Introduction Carrageenan, an assortment of sulphated polyelectrolyte heteropolysaccharides extracted from the red seaweed, has long been used in food (Burey, Bhandari, Howes, & Gidley, 2008; Necas & Bartosikova, 2013; Prajapati, Maheriya, Jani, & Solanki, 2014; Saha & Bhattacharya, 2010; Saltmarsh, 2015). One member of the carrageenan family, k-carrageenan (Fig. 1), has been employed ubiquitously as a food additive because of its thickening and gelling capabilities (Burey et al., 2008; Necas & Bartosikova, 2013; Prajapati et al., 2014; Saha & Bhattacharya, 2010; Saltmarsh, 2015). The gelation of k-carrageenan, like other polysaccharides, has long been known to be strongly affected by cosolvents, in particular such as sugars and polyols; such cosolvents have therefore been exploited widely to enhance k-carrageenan gelation and gel stability (Arakawa & Timasheff, 1982; Gekko, Mugishima, & Koga, 1987; Loret, Ribelles, & Lundin, 2009; Nishinari & Watase, 1992; Nishinari, Watase, Williams, & Philips, 1990; Oakenfull, 2000;

Abbreviations: KB, KirkwoodeBuff. * Corresponding author. E-mail address: [email protected] (S. Shimizu). http://dx.doi.org/10.1016/j.foodhyd.2015.10.007 0268-005X/© 2015 Elsevier Ltd. All rights reserved.

Ramakrishnan & Prud'homme, 2000). What sugar/polyol cosolvents do to the gelation of k-carrageenan is well known; however, how these cosolvents actually enhance gelation, at a molecular level, remains a mystery, and has been a subject of intense debate for decades (Arakawa & Timasheff, 1982; Gekko et al., 1987; Loret et al., 2009; Nishinari & Watase, 1992; Nishinari et al., 1990; Oakenfull, 2000; Ramakrishnan & Prud'homme, 2000). Hence the aim of this paper is to provide a clear explanation, based upon a statistical thermodynamic foundation, into how exactly sugar and polyol cosolvents influence the gelation of biopolymers by examining the gelation of k-carrageenan. A number of hypotheses have been proposed regarding how cosolvents affect the gelation of biopolymers; these hypotheses can be classified into the following three categories: (i) cosolventinduced changes in the water structure, and the consequent hydration change of the biopolymer, induced by the cosolvent (Arakawa & Timasheff, 1982; Back, Oakenfull, & Smith, 1979; Gekko et al., 1987; Nishinari & Watase, 1992; Oakenfull, 2000); (ii) preferential exclusion of the cosolvent from the biopolymer surface (Back et al., 1979; Shimizu & Boon, 2004; Shimizu & Matubayasi, 2014a, b; Timasheff, 2002a, b) and (iii) direct binding of the cosolvent to the biopolymer chain in the gel state (Kumar, Modig, &

R. Stenner et al. / Food Hydrocolloids 54 (2016) 284e292

285

above, should be examined critically in the light of statistical thermodynamics. 1.2. Preferential exclusion of cosolvents from polysaccharide chains

Fig. 1. Chemical structure of k-carrageenan.

Halle, 2003; Loret et al., 2009; Makhatadze & Privalov, 1992; Prakash, Loucheux, Scheufele, Gorbunoff, & Timasheff, 1981; Ramakrishnan & Prud'homme, 2000). We will begin by briefly examining each hypothesis.

1.1. Cosolvent-induced change of water structure and hydration change This hypothesis stems originally from the classical view of biomolecular folding and stability, in which the hydrophobic effect has long been believed to play a dominant role (Dill, 1990; Frank & Evans, 1945; Kauzmann, 1959; Tanford, 1968). According to this view, the hydrophobic effect is caused by the entropy penalty due to the enhancement of hydrogen bonds between water molecules adjacent to the hydrophobic solute (Dill, 1990; Frank & Evans, 1945; Kauzmann, 1959; Tanford, 1968). Hence cosolvent molecules that enhance the hydrogen bond network of the surrounding water molecules, known formally as kosmotropes, are considered to concurrently enhance the hydrogen bond network around the hydrophobic solutes, thereby strengthening the hydrophobic effect and, consequently, facilitating biomoleculeebiomolecule association (Dill, 1990; Frank & Evans, 1945; Kauzmann, 1959; Tanford, 1968). The enhancement of the water structure by the cosolvent is thus the cause of the hydration change. Attempts have therefore been made to measure, or obtain evidence for, the cosolvent-induced changes in hydration accompanying gelation (Dill, 1990; Frank & Evans, 1945; Gekko et al., 1987; Kauzmann, 1959; Oakenfull, 2000; Tanford, 1968; Uedaira, Ikura, & Uedaira, 1989). A bulk of evidence has been reported to support this view: (i) the introduction of cosolvents changes the gelation process from an enthapically driven to an entropically driven process, indicating a cosolvent-induced release of water (Gekko et al., 1987); (ii) the increase in the isothermal compressibility of k-carrageenan gels, indicative of water release (Gekko & Kasuya, 1985); (iii) the observed linear correlation between the “water structure parameter” (derived from the partial molar volume of the cosolvent) and the cosolvent-induced increase in thermal stability of the gel (Oakenfull, 2000); this is underscored further by a linear correlation between the “water structure parameter” and the number of equatorial-OH groups on the cosolvent (Nishinari & Watase, 1992; Oakenfull, 2000; Uedaira et al., 1989) which correlates also with the sugar-enhanced thermal stability of k-carrageenan gels (Nishinari & Watase, 1992). In contrast to all above, our recent statistical thermodynamic study on the gelation of agarose and gelatin has shown that the contribution of cosolvent-induced hydration change is negligible (Shimizu & Matubayasi, 2014b). However, the question still remains as to whether this is also the case for k-carrageenan gelation. Also the previous evidence for this hypothesis, as summarised

This hypothesis also originates from biomolecular hydration thermodynamics; the enhancement of biomolecular association in the presence of cosolvents can be rationalised by the preferential exclusion of cosolvents (or equivalently, preferential hydration) from the biomolecular surface (Timasheff, 2002a, b) which has been supported by experiments (Casassa & Eisenberg, 1964; Tanford, 1970; Timasheff, 1998). The notion of preferential cosolvent exclusion has sometimes been invoked to rationalise the cosolvent-induced enhancement of biopolymer properties (Back et al., 1979; Shimizu & Matubayasi, 2014a; Timasheff, 2002a). Shimizu and Matubayasi have recently extended the rigorous statistical thermodynamic theory of the cosolvent effect to gelation equilibria (Shimizu & Matubayasi, 2014b), which afforded a rationalisation for the sugar- and polyol-induced enhancement of gelatin and agarose gelation in terms of cosolvent preferential exclusion. What has emerged from this rigorous statistical thermodynamic theory is the clarification that preferential exclusion is actually caused not by the increased hydration but by the exclusion of cosolvents from biomolecular surfaces (Shimizu & Matubayasi, 2014b). Hence this hypothesis is actually about the exclusion of cosolvents. Whether this is the case also for k-carrageenan remains to be examined. 1.3. Binding of cosolvents on polysaccharide gels Sugars and polyols bind directly to k-carrageenan, increasing the number of junction zones, with shorter average length, per unit volume (Nishinari & Watase, 1992; Oakenfull, 2000). This is attributed to the stabilization of the characteristic intermolecular hydrogen bonding between individual k-carrageenan strands in a typical junction zone by the formation of intermolecular, crosslinking hydrogen bonds between the OH-groups of the sugar/polyol cosolvent and k-carrageenan (Nishinari & Watase, 1992; Oakenfull, 2000). Such a change in cross-linking came originally from rheological evidence (Loret et al., 2009; Nishinari & Watase, 1987), as well as the cosolvent-induced elevation of the gel denaturation temperature for agarose gels (Watase, Kohyama, & Nishinari, 1992) and kcarrageenan gels (Gekko et al., 1987; Nishinari & Watase, 1992; Nishinari et al., 1990). Such evidence, more specifically, includes: i) the shorter average length, and therefore increased number per unit volume, of junction zones for k-carrageenan gels in the presence of sugars/polyol cosolvents (Nishinari & Watase, 1992; Oakenfull, 2000); ii) the decreased free energy of junction zone formation; due to stabilisation of intermolecular cross-linking of individual k-carrageenan molecules in the junction zones by intermolecular hydrogen-bonding between cosolvent OH-groups and individual k-carrageenan strands (Oakenfull, 2000); iii) the increased setting and melting temperature of k-carrageenan gels in the presence of sugars (Nishinari & Watase, 1992; Nishinari et al., 1990; Ramakrishnan & Prud'homme, 2000); iv) the increased rigidity of k-carrageenan gels with increasing sugar/polyol concentration (Loret et al., 2009); v) the change in rheological properties of k-carrageenan gels above a critical sugar concentration (Loret et al., 2009); and vi) the positive correlation between the number of equatorial-OH groups on a sugar and the increased thermal stability of k-carrageenan gels as equatorial groups are suspect to form cross-linking hydrogen bonds between biopolymers (Nishinari & Watase, 1992; Oakenfull, 2000). Similar observation and explanations have been reported for

286

R. Stenner et al. / Food Hydrocolloids 54 (2016) 284e292

agarose gelation in the presence of sugars/polyol cosolvents (Nishinari & Watase, 1987; Russ, Zielbauer, & Vilgis, 2014; Watase et al., 1992). X-ray diffraction, small angle x-ray scattering and neutron scattering experiments also support the cosolvent-binding hypothesis (Sugiyama et al., 1998; Yuguchi, Urakawa, & Kajiwara, 2003). This hypothesis, however, has not been examined previously in the framework of the statistical thermodynamic theory. The co-existence of three major hypotheses clearly demonstrates that the mechanism responsible for sugar-/polyol-induced enhancement of k-carrageenan gelation is not clearly understood. This lack of understanding reflects the lack of consensus on the molecular level mechanism of cosolvent-induced gelation; fortunately, a molecular level perspective can be ascertained through the application of statistical thermodynamics. We have recently developed a statistical thermodynamic theory of cosolventinduced gelation (Shimizu & Matubayasi, 2014b) based upon the rigorous KirkwoodeBuff (KB) theory (Kirkwood & Buff, 1951; Newman, 1994; Shimizu, 2004; Shimizu & Matubayasi, 2014b). This theory enables one to determine, in a quantitative manner, how cosolvents actually influence the gelation equilibria. Experimental data concerning how the gel melting temperature is affected by cosolvent concentration and pressure, as well as the enthalpy of gelation (data which is readily available from the literature) is all that is required to achieve this (Shimizu & Matubayasi, 2014b). Only from such experimental data can one quantitatively compare the contributions from hydration change, cosolvent exclusion and cosolvent-biopolymer binding, in order to examine the validity of all the aforementioned hypotheses (Shimizu & Matubayasi, 2014b). Our KB theory of gelation, which has been applied to gelatine and agarose, examined the first two hypotheses, and shown that the effect of hydration (hypothesis (i)) contributes negligibly to the cosolvent-induced change of gelation. The theory then attributed the dominant contribution, cosolvent-biopolymer interaction, to the exclusion of cosolvents from biopolymers (hypothesis (ii)), without explicit reference to the cosolvent-biopolymer binding, in accordance to the well-established scenario of molecular crowding, which attributes the cosolvent-induced enhancement of protein folding and binding to the exclusion of cosolvents from biomolecular surfaces. The aim of this paper is, therefore, to establish a full and complete picture of the overall cosolvent effect on gelation, through the examination of all three of the major hypotheses. 2. A statistical thermodynamic analysis of cosolvent-induced change of gelation 2.1. A KirkwoodeBuff theory of cosolvent-induced change of gelation equilibrium Consider polysaccharide molecules (i ¼ u) in the mixture of water (i ¼ 1) and cosolvent (i ¼ 2) molecules. Following Minton's thermodynamic approach, we consider the following two states: isolated polymer and aggregated states, in which sol (a ¼ s) is the isolated polymer state, whereas gel (a ¼ g) is the aggregated state (Minton, 1974; Shimizu & Matubayasi, 2014b). Note that the polysaccharide molecules can make linked network structure in the gel state, yet are dilute enough so that polysaccharideepolysaccharide interactions other than the formation of link can be ignored, which is also an implicit assumption of the thermodynamic model (Shimizu & Matubayasi, 2014b). Thus we consider an isolated polymer chain for the sol state, and a dilute polysaccharide aggregate as the gel state. Based upon the above setup, Shimizu and Matubayasi (Shimizu & Matubayasi, 2014b) made it possible to determine the interaction between a polysaccharide polymer (u) in the sol (a ¼ s) or gel

(a ¼ g) states and water (i ¼ 1), as well as polysaccharide and cosolvent (i ¼ 2), defined in terms of the following KB parameter:

Z

ðaÞ

Gui ¼

h i ðaÞ dr 4pr 2 gui ðrÞ  1

(1)

Here g(uia) (r) is the radial distribution function (RDF) between u (at state a) and i. At the gel state (a ¼ g), this function is introduced per isolated polymer; see (Shimizu & Matubayasi, 2014b) for detailed definition. There are several different contributions to G(uia): (i) excluded volume effect, namely the steric repulsion between molecule i and a polysaccharide, makes G(uia) negative, because g(uia) (r) ¼ 0 at short distances (Shimizu & Boon, 2004); (ii) attractive interaction between a molecule i and a polysaccharide results in higher peaks of g(uia) (r), and contributes more positively to G(uia) (Shimizu, 2004; Shimizu & Boon, 2004; Shimizu & Matubayasi, 2014b; Smith, 2004); (iii) if there is depletion of the local density of i at certain range of r compared to that in the bulk, namely g(uia) (r) < 1, it contributes negatively to G(uia) (Shimizu, 2004; Shimizu & Boon, 2004; Shimizu & Matubayasi, 2014b; Smith, 2004). Now the change of G(uia) between gel and sol states, DGui ¼ G(s) ui  (g) Gui can be quantitatively determined solely from experimental data; to this end, the change of entropy that accompanies gel melting, DSg/s, as well as the dependence of the gel melting temperature dTg/s on the bulk density of cosolvent, n2, as well as on the hydrostatic pressure, P, are necessary (Shimizu & Matubayasi, 2014b). All the relevant data have been taken from Gekko (Gekko & Kasuya, 1985; Gekko et al., 1987). From such data, DGu1 and DGu2 can be calculated by solving the following simultaneous equations derived by the combination of the KB theory and Clausius-Clapeyron equation by Shimizu and Matubayasi (Shimizu & Matubayasi, 2014b)

DSg/s dTg/s ¼ DGu1  DGu2 RT dn2 DSg/s

dTg/s ¼ DVg/s ¼ DGu1 dP

(2)

(3)

where DVg/s is the volume change which accompanies gel / sol transition, R is the gas constant; Eqs. (2) and (3) are valid for dilute polysaccharide and cosolvents in the system. DVg/s, DSg/s, dTg/s/ dP, and dDTg/s/dn2 are thus the values at n2 / 0 and in particular, the first three are evaluated in bulk water. Eq. (2) was derived from the following equation, which relates the free energy change that accompanies gel to sol transition, Dm*u, to the KB parameters DGu1 and DGu2

1 RT



vDm*u vn2

 T;P;nu /0; n2 /0

¼ DGu1  DGu2

(4)

Thus the quantitative values of DGu1 and DGu2 will provide crucial information on what really drives the cosolvent-induced gelation enhancement. 2.2. Determination of KirkwoodeBuff integrals from experimental data How a cosolvent influences the gelation of k-carrageenan is encoded into dTg/s, the change in melting temperature. Some cosolvents (such as polyols and sugars), when added, stabilize kcarrageenan gelation and consequently an increase in Tg/s is observed (Gekko et al., 1987; Nishinari & Watase, 1992; Nishinari et al., 1990), whereas other cosolvents (such as urea and guanidine hydrochloride) destabilise k-carrageenan gelation, leading to a

R. Stenner et al. / Food Hydrocolloids 54 (2016) 284e292

287

Matubayasi, 2014b). The effect of KCl concentrations on the KB parameters DGu1 and DGu2 have been discussed in Appendix.

decrease in Tg/s (Nishinari et al., 1990). How the cosolvent influences gelation on a molecular level is encoded into the KB parameters, DGu1 and DGu2, which are connected to Tg/s by Eqs. (2) and (3). DGu1 and DGu2 have been calculated for k-carrageenan gelation in the presence of 15 different sugars and polyols, in the presence of dilute KCl, which provides counterions for k-carrageenan. In the literature, aqueous solutions of KCl at various concentrations, i.e., 0.0075, 0.015, 0.025 and 0.035 mol dm3 were reported. Accordingly, we have calculated the KB parameters at each concentration from the experimental values of DVg/s, DSg/s, dTg/s/dP, and dDSg/s/dn2 are taken from Gekko (Gekko & Kasuya, 1985; Gekko et al., 1987). We emphasise here that DVg/s, DSg/s, and dTg/s/dP were measured in the absence of cosolvents (Gekko & Kasuya, 1985), as required for Eq. (3). The experimental values used as input for our calculations, together with the resulting KB parameters, are tabulated for each KCl concentrations: Table 1 (0.0075 mol dm3), Table 2 (0.015 mol dm3), Table 3 (0.025 mol dm3) and Table 4 (0.035 mol dm3). From the KB parameters, DGu1 and DGu2, the merit of the three hypotheses, (i) hydration/water structure changes, (ii) preferential interactions and (iii) cosolvent-k-carrageenan binding, can be evaluated from a molecular perspective. Note that, in order to calculate DGu1 and DGu2 based upon Eqs. (2) and (3), the experimental data on DSg/s is crucial. In the experimental literature, there are two ways of reporting the entropy change of gel / sol transition: (i) entropy per gram of biopolymer; (ii) entropy per mole of cross-links (Shimizu & Matubayasi, 2014b). Additional information is necessary in order to convert data in each respective dimension: (i) average molecular weight of the biopolymer; (ii) number of cross-links per biopolymer. Even though the molecular weight and the number of cross-links are quite commonplace quantities, they are often not reported in the literature; on top of that, these quantities depend highly on the quality and preparation of the sample (Shimizu & Matubayasi, 2014b). In this paper, we report KB parameters per mole of cross-links, which can readily be converted to per solute mole upon the availability of the number of cross-links per solute in the precise experimental condition. Yet for our goal e to clarify quantitatively the importance of biopolymer-water versus biopolymerecosolvent interactions e the lack of this constant (number of cross-links) would not be a hindrance; although our formulation is done with per-biopolymer basis, it is related to the formulation per cross-link simply through the proportionality constant determined by the number of cross-links (Shimizu &

3. Mechanism of the cosolvent effects on k-carrageenan gelation 3.1. Water structure and hydration hypothesis The water structure hypothesis is based upon the traditional view on the origin of the hydrophobic effect, which is assumed to be the driving force for macromolecular association (Dill, 1990; Frank & Evans, 1945; Kauzmann, 1959; Tanford, 1968). In turn, the hydrophobic effect, namely the low solubility of non-polar solutes in water, is due to the entropic penalty arising from the enhancement of the hydrogen-bonded network of water around the hydrophobic group (Dill, 1990; Frank & Evans, 1945; Kauzmann, 1959; Tanford, 1968). Therefore, according to this hypothesis, kosmotropic (structure-making) cosolvents enhance the water structure around them, thereby promoting the hydrogen-bonded network of water around the hydrophobic solutes and strengthening the hydrophobic effect. Chaotropic (structure-breaking) cosolvents, on the other hand, break the hydrogen-bonded network of water, thereby weakening the hydrogen-bonded network of water around the hydrophobic solutes and weakening the hydrophobic effect (Dill, 1990; Frank & Evans, 1945; Kauzmann, 1959; Marcus, 2009; Politi, Sapir, & Harries, 2009; Tanford, 1968). It is the hydration of solutes that is affected by the kosmotropic and chaotropic cosolvents (Dill, 1990; Frank & Evans, 1945; Kauzmann, 1959; Marcus, 2009; Oakenfull, 2000; Politi et al., 2009; Tanford, 1968). Hence, in the language of the KB theory, the water structure hypothesis is equivalent to the dominance of DGu1 over DGu2, i.e., it is DGu1 that is the major driving force of cosolvent-induced changes in k-carrageenan gelation (Prakash et al., 1981). Tables 1e4 show the contrary: KB theory has demonstrated that jDGu2j >> jDGu1j; i.e. DGu2 is the major driving force behind polyol/sugar induced kcarrageenan gelation. This conclusion, that DGu1 contributes negligibly to sugar/ polyol-induced k-carrageenan gelation and therefore hydration changes cannot account for the cosolvent-induced change in gelation, is underscored by the study of (V2eV2w)/V2w (where V2 is the partial molar volume of the cosolvent and V2w is the van der Waals volume of the cosolvent), which reflects ‘structure-making’ ability of the cosolvent, and dTs/g/dn2, the thermal stability

Table 1 The effect of several cosolvents on the gelesol transition of k-carrageenan in a 0.0075 mol dm3 KCl aqueous solution. Note that mol1 here refers to per mol of intermolecular cross-links, following the convention adopted commonly in the literature. 1 dTg/s RT dn2

Lactose Maltose Sucrose Glucose Mannose Fructose Galactose Xylose Arabinose Ribose Maltitol Sorbitol Xylitol Erythritol Glycerol

K m kg1 s2

3.71  106 3.22  106 3.75  106 1.43  106 9.12  107 9.48  107 9.97  106 8.27  107 8.72  107 4.03  107 4.07  106 1.97  106 1.07  106 1.47  106 1.25  106

dTg/s dP

K m kg1 s2

6.62 6.62 6.62 6.62 6.62 6.62 6.62 6.62 6.62 6.62 6.62 6.62 6.62 6.62 6.62

              

108 108 108 108 108 108 108 108 108 108 108 108 108 108 108

1 RT





vDmu vn2 T;P;n ;n /0 u 2

482 419 487 186 118 123 129 107 113 52 530 256 221 186 162

cm3 mol1

DGu1 cm3 mol1

DGu2 cm3 mol1

8.6 8.6 8.6 8.6 8.6 8.6 8.6 8.6 8.6 8.6 8.6 8.6 8.6 8.6 8.6

474 410 478 177 110 114 121 99 104 43 521 247 212 178 154

288

R. Stenner et al. / Food Hydrocolloids 54 (2016) 284e292

Table 2 The effect of several cosolvents on gelesol transition of k-carrageenan in a 0.015 mol dm3 KCl aqueous solution. Note that mol1 here refers to per mol of intermolecular crosslinks, following the convention adopted commonly in the literature. 1 dTg/s RT dn2

Lactose Maltose Sucrose Glucose Mannose Fructose Galactose Xylose Arabinose Ribose Maltitol Sorbitol Xylitol Erythritol Glycerol

K m kg1 s2

3.71  106 3.22  106 3.75  106 1.43  106 9.12  107 9.48  107 9.97  106 8.27  107 8.72  107 4.03  107 4.07  106 1.97  106 1.07  106 1.47 106 1.25 106

dTg/s dP

K m kg1 s2

5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68 5.68

              

108 108 108 108 108 108 108 108 108 108 108 108 108 108 108

1 RT





vDmu vn2 T;P;n ;n /0 2 u

cm3 mol1

883 767 891 340 217 225 237 196 207 96 970 468 405 341 297

DGu1 cm3 mol1

DGu2 cm3 mol1

13.5 13.5 13.5 13.5 13.5 13.5 13.5 13.5 13.5 13.5 13.5 13.5 13.5 13.5 13.5

870 753 877 327 203 212 223 183 193 82 956 455 391 328 284

hydration hypothesis encapsulates the failure of classical thermodynamics to fuse an explicit connection between the molecular scale and gelation. The KB parameters have provided this molecular connection through jDGu2j >> jDGu1j, and have demonstrated the water structure/hydration hypothesis to be inadequate. The water

imparted to the gel by the cosolvent (Gekko et al., 1987). No correlation was observed between the two parameters, indicating that changes in hydration contribute negligibly to the cosolventinduced thermal stability of k-carrageenan gels (Gekko et al., 1987). The aforementioned evidence favouring the water structure/

Table 3 The effect of several cosolvents on gelesol transition of k-carrageenan in a 0.025 mol dm3 KCl aqueous solution. Note that mol1 here refers to per mol of intermolecular crosslinks, following the convention adopted commonly in the literature. 1 dTg/s RT dn2

Lactose Maltose Sucrose Glucose Mannose Fructose Galactose Xylose Arabinose Ribose Maltitol Sorbitol Xylitol Erythritol Glycerol

K m kg1 s2

3.71  106 3.22  106 3.75  106 1.43  106 9.12  107 9.48  107 9.97  106 8.27  107 8.72  107 4.03  107 4.07  106 1.97  106 1.07  106 1.47  106 1.25  106

dTg/s dP

K m kg1 s2

5.20 5.20 5.20 5.20 5.20 5.20 5.20 5.20 5.20 5.20 5.20 5.20 5.20 5.20 5.20

              

108 108 108 108 108 108 108 108 108 108 108 108 108 108 108

1 RT





vDmu vn2 T;P;n ;n /0 2 u

cm3 mol1

1692 1470 1707 653 415 432 454 377 397 184 1858 897 776 655 570

DGu1 cm3 mol1

DGu2 cm3 mol1

23.7 23.7 23.7 23.7 23.7 23.7 23.7 23.7 23.7 23.7 23.7 23.7 23.7 23.7 23.7

1669 1446 1683 629 392 408 430 353 373 160 1834 874 752 631 546

Table 4 The effect of several cosolvents on gelesol transition of k-carrageenan in a 0.035 mol dm3 KCl aqueous solution. Note that mol1 here refers to per mol of intermolecular crosslinks, following the convention adopted commonly in the literature. 1 dTg/s RT dn2

Lactose Maltose Sucrose Glucose Mannose Fructose Galactose Xylose Arabinose Ribose Maltitol Sorbitol Xylitol Erythritol Glycerol

K m kg1 s2

3.71  106 3.22  106 3.75  106 1.43  106 9.12  107 9.48  107 9.97  106 8.27  107 8.72  107 4.03  107 4.07  106 1.97  106 1.07  106 1.47  106 1.25  106

dTg/s dP

K m kg1 s2

5.86 5.86 5.86 5.86 5.86 5.86 5.86 5.86 5.86 5.86 5.86 5.86 5.86 5.86 5.86

              

108 108 108 108 108 108 108 108 108 108 108 108 108 108 108

1 RT





vDmu vn2 T;P;n ;n /0 u 2

2300 1997 2320 887 565 587 617 512 540 250 2525 1220 1055 890 775

cm3 mol1

DGu1 cm3 mol1

DGu2 cm3 mol1

36.3 36.3 36.3 36.3 36.3 36.3 36.3 36.3 36.3 36.3 36.3 36.3 36.3 36.3 36.3

2263 1961 2283 851 528 551 581 476 503 213 2488 1183 1018 853 738

R. Stenner et al. / Food Hydrocolloids 54 (2016) 284e292

3.2. Cosolvent exclusion hypothesis In the language of KB theory, the dominance of cosolvent exclusion signifies first of all the dominance of DGu2 over DGu1 (Shimizu & Matubayasi, 2014b), which has already been shown via Tables 1e4 and Eq. (2). In addition, the gel state is more compact than the sol, per biopolymer. Hence DGu2, which accompanies the gel / sol transition, has a large negative value, reflecting an expansion per biopolymer, which agrees with Tables 1e4 Taken together, the exclusion of cosolvents drives cosolvent-induced gelation. What is the molecular basis of cosolvent exclusion from biomolecular surfaces? To answer this question, let us note that the KB parameters for cosolvent-biopolymer interactions, DGu2, are larger for disaccharides than for monosaccharides. By the virtue of Eq. (2), it is possible to interpret the well-known correlation between the thermal stability of k-carrageenan gels and the average number of equatorial-OH groups (Nishinari & Watase, 1992; Oakenfull, 2000). A good correlation between DGu2 and number of equatorialOH groups on the cosolvent (Table 5 and Fig. 2) supports the preferential interactions hypothesis and can be interpreted from a molecular perspective. The degree of hydration of a sugar/polyol cosolvent is a function of its average number of equatorial-OH groups (Galema & Hoeiland, 1991; Gekko & Koga, 1983; Uedaira et al., 1989). A preferentially hydrated cosolvent should exhibit increased exclusion with increasing number of equatorial-OH groups, which accounts for the correlation in Fig. 2. This indicates that the hydration of the cosolvent may be the driving force for cosolvent exclusion due possibly to the cosolvent's favourable hydration in the bulk phase. The importance of Fig. 2, therefore, is the molecular insight into cosolvent exclusion it provides by connecting DGu2 to the molecular structure of the cosolvent, suggesting the equatorial-OH groups on the cosolvent are important molecular features governing cosolvent-k-carrageenan interactions in the sol state. 3.3. Cosolvent-gel binding hypothesis In the language of KB theory, this hypothesis is equivalent to a positive G(g) ui , which makes DGui more negative for the gelesol (g) transition as DGui ¼ G(s) ui  Gui and equates to binding of the cosolvent to the polysaccharide in the gel state. From Tables 1e4, it

2500

2000

-∆Gu2 (cm3 mol-1)

structure hypothesis here was indeed a part of a more general paradigm which attempted to rationalise the cosolvent effect on biomolecules based upon the change of water structure (Gekko et al., 1987; Nishinari & Watase, 1992; Oakenfull, 2000). Yet accumulating evidence in protein folding, binding and macromolecular assemblies suggest strongly that hydration changes cannot account for these processes (Shimizu & Matubayasi, 2014b).

289

1500

1000

500

0

0

1

2

3

4

5

6

7

8

# Equatorial OH groups on cosolvent Fig. 2. DGu2 for k-carrageenan-cosolvent at 0.0075 mol dm3(black circles) and 0.035 mol dm3 (yellow triangles) KCl against average number of equatorial-OH groups per cosolvent (Galema & Hoeiland, 1991; Gekko & Koga; 1983; Uedaira et al., 1989). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

can be readily appreciated that all 15 cosolvents exhibited negative (g) DGu2 (DGu2 ¼ G(s) u2  Gu2 < 0) for the gelesol transition; cosolvent-k-

carrageenan binding, which would make G(g) u2 positive, may be an additional factor in making DGu2 so highly negative for the gelesol transition. This postulation is supported by the aforementioned differential scanning calorimetry and rheological experimental measurements (Loret et al., 2009; Nishinari & Watase, 1987; Watase et al., 1992). The KB parameters and experimental evidence (Loret et al., 2009; Nishinari & Watase, 1987; Watase et al., 1992) suggest kcarrageenan-cosolvent binding, in the gel state, is an important factor governing cosolvent-induced k-carrageenan gelation. The experimental evidence indicates the cosolvent's OH-groups are important molecular features which govern the interactions between the cosolvent and k-carrageenan in both the sol (via preferential hydration) and gel (via intermolecular hydrogen-bonding) states, a conclusion corroborated by the observed correlation in Fig. 2. The enhanced thermal stability of k-carrageenan gels with increasing average number of equatorial-OH groups on the cosolvent (Nishinari & Watase, 1992) is consistent with k-carrageenancosolvent binding in the gel state, as an increased average number of equatorial-OH groups on the cosolvent will result in increase in intermolecular hydrogen-bonding between the cosolvent and individual k-carrageenan strands in the gel state (Nishinari & Watase, 1992; Oakenfull, 2000). Nishinari concluded it was impossible to determine, from DSC experiments, whether a change in the structure of water or direct biomolecule-cosolvent binding was the

Table 5 DGu2 at 0.0075 and 0.035 mol dm3 KCl and the average number of equatorial-OH groups on the cosolvent (Galema & Hoeiland, 1991; Gekko & Koga, 1983; Uedaira et al., 1989). Note, only 10 of the studied 15 cosolvents are reported here due to limitations in available data in the literature. Cosolvent

# Equatorial OH

DGu2 cm3 mol1 in 0.0075 mol dm3 KCl

DGu2 cm3 mol1 in 0.035 mol dm3 KCl

Lactose Maltose Sucrose Glucose Mannose Fructose Galactose Xylose Arabinose Ribose

6.67 7.62 6.5 4.66 3.3 3.1 3.7 3.63 2.37 2.1

474 410 478 177 110 114 121 99 104 43

2264 1961 2284 851 528 551 581 476 503 213

290

R. Stenner et al. / Food Hydrocolloids 54 (2016) 284e292

dominant driving force behind cosolvent-induced changes in kcarrageenan gelation (Nishinari & Watase, 1992); in contrast, from KB theory, jDGu2j >> jDGu1j indicates cosolvent-biomolecule binding, and not changes in the water structure, is the case. 3.4. Mechanism for cosolvent-induced k-carrageenan gelation in light of KirkwoodeBuff theory From KB theory, it has been demonstrated that i) cosolvent exclusion in the sol state (G(s) u2 < 0) (Shimizu & Matubayasi, 2014b) and ii) cosolvent-k-carrageenan binding in the gel state (G(g) u2 > 0) (Shimizu & Matubayasi, 2014b) are the driving forces behind cosolvent-induced k-carrageenan gelation; changes in the structure of water/hydration contribute negligibly (jDGu2j >> jDGu1j; Tables 1e4), a conclusion supported by the absence of correlation between DG21 and Tm (Gekko et al., 1987). This paints an intuitive picture of the molecular scale process governing sugar/polyol cosolvent-induced changes in k-carrageenan gelation:  The strongly hydrated polyol/sugar cosolvent is preferentially excluded from the polymer surface in the presence of k-carrageenan (Back et al., 1979; Casassa & Eisenberg, 1964; Shimizu & Matubayasi, 2014a, b; Tanford, 1970; Timasheff, 1998, 2002a, b).  Exclusion of the cosolvent reduces the average water molecule population in the vicinity of k-carrageenan strands (Oakenfull, 2000).  The decreased water population weakens k-carrageenanewater interactions, causing k-carrageenan strands to aggregate towards neighbouring strands (Dill, 1990; Frank & Evans, 1945; Kauzmann, 1959; Oakenfull, 2000; Tanford, 1968)  The polyol/sugar cosolvent, through equatorial-OH groups, form cross-linking hydrogen bonds with k-carrageenan strands, facilitating junction zone formation and gelation (Nishinari & Watase, 1987, 1992; Oakenfull, 2000; Ramakrishnan & Prud'homme, 2000) To summarise, cosolvent exclusion in the sol state perturbs the solegel equilibrium in favour of the gel, where the gel is stabilised by favourable k-carrageenan-cosolvent interactions as a result of biomolecule-cosolvent binding (Fig. 3). The magnitude of the kcarrageenan-cosolvent interactions is a function of the number of OH-groups on the cosolvent, indicating OH-groups are a significant molecular feature governing sugar/polyol cosolvent-induced changes in k-carrageenan gelation (Nishinari & Watase, 1987, 1992; Oakenfull, 2000; Ramakrishnan & Prud'homme, 2000).

4. Conclusion Sugars and polyols enhance the gelation of carrageenan (Arakawa & Timasheff, 1982; Gekko et al., 1987; Loret et al., 2009; Nishinari et al., 1990; Nishinari & Watase, 1992; Oakenfull, 2000; Ramakrishnan & Prud'homme, 2000). Three conflicting hypotheses have been proposed to explain this simple observation: (i) the change of saccharide hydration caused by the enhancement of water structure around the cosolvent (Gekko et al., 1987; Nishinari & Watase, 1992; Oakenfull, 2000; Ramakrishnan & Prud'homme, 2000); (ii) the exclusion of cosolvents from polysaccharide surfaces (Shimizu & Matubayasi, 2014b); (iii) binding of cosolvents on carrageenan gel (Nishinari & Watase, 1987, 1992; Oakenfull, 2000). Only statistical thermodynamics, which is the very branch of science to provide a microscopic-macroscopic link, can answer this question. Thus we applied our recent statistical thermodynamic theory of gelation based upon our rigorous KB theory of cosolvency (Shimizu & Matubayasi, 2014b). Polysaccharide-water and polysaccharide-cosolvent interactions can be evaluated solely from thermodynamic data on the dependence of transition temperature on cosolvent concentration, hydrostatic pressure and the entropy of gel / sol transition (Shimizu & Matubayasi, 2014b). We identified the two driving forces (i) the exclusion of sugars and polyols from k-carrageenan surfaces in the sol state and (ii) the binding of sugars and polyols to k-carrageenan in the gel. Both (i) and (ii) contribute to make the cosolvents more excluded from the sol state than from the gel states. This is supported by experiments (Gekko et al., 1987; Loret et al., 2009; Nishinari et al., 1990; Nishinari & Watase, 1987, 1992; Oakenfull, 2000; Ramakrishnan & Prud'homme, 2000; Russ et al., 2014; Sugiyama et al., 1998; Uedaira et al., 1989; Watase et al., 1992; Yuguchi et al., 2003). Polysaccharide hydration, in contrast, contributes only negligibly to the thermodynamics of gelation (Shimizu & Matubayasi, 2014b), in direct contrast to the classical hypothesis that cosolvent-induced hydration change is the driving force of gel stabilisation (Gekko et al., 1987; Nishinari & Watase, 1992; Oakenfull, 2000; Ramakrishnan & Prud'homme, 2000). The contribution of cosolvent binding to the gel phase as one of the driving forces of the stabilization of the gel phase marks a departure from the traditional molecular crowding scenario, namely the shift of equilibrium to the direction of the reduced cosolvent excluded volume, which applies universally to protein folding, protein-ligand binding, aggregation of helices, and association of small molecular species, and is considered to be the dominant driving force of these seemingly different processes (Minton, 1974;

Fig. 3. The gelation of k-carrageenan by the addition of a polyol/sugar cosolvent, based upon Shimizu and Matubayasi (2014b). The unfavourable interactions between the kcarrageenan strands and the cosolvent in the sol state shift the solegel equilibrium to the gel state, where intermolecular hydrogen bonds between the cosolvent and k-carrageenan stabilise the gel. Blue dots ¼ Water; Green dots ¼ cosolvent; Red lines ¼ k-carrageenan polymers. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

R. Stenner et al. / Food Hydrocolloids 54 (2016) 284e292

Shimizu, 2004; Shimizu & Boon, 2004; Shimizu & Matubayasi, 2014a, 2014b). In gelation, there is an additional factor, cosolventgel binding, which contributes to the stabilization of the gel phase (Gekko et al., 1987; Nishinari & Watase, 1987, 1992; Oakenfull, 2000; Ramakrishnan & Prud'homme, 2000; Russ et al., 2014; Sugiyama et al., 1998; Watase et al., 1992; Yuguchi et al., 2003). Which of these two is the dominant factor? In the present theoretical framework, answering this question requires one to clarify the solvation of sol and gel states individually. Experimentally, this would require the transfer of sol and gel from vacuum to aqueous cosolvent solutions, which is impossible. Computer simulation would therefore be crucial in distinguishing the contributions from the two states.

Acknowledgements This work is supported by the Grants-in-Aid for Scientific Research (Nos. 15K13550, 23651202, and 26240045) from the Japan Society for the Promotion of Science, and by the Elements Strategy Initiative for Catalysts and Batteries from the Ministry of Education, Culture, Sports, Science, and Technology, and by Computational Materials Science Initiative, Theoretical and Computational Chemistry Initiative, and HPCI System Research Project (Project IDs: hp140156, hp140214, hp150131, hp150137, and hp150231) of the Next-Generation Supercomputing Project.

Appendix In addition to carrageenan (i ¼ u), water (i ¼ 1) and cosolvent (i ¼ 2) molecules, let us explicitly consider here ionic species: Kþ (i ¼ 3) and Cl (i ¼ 4). Following our previous works, the pair of GibbseDuhem equations under constant T and P in the presence and absence of carrageenan can be written as

n*u dm*u þ n*1 dm1 þ n*2 dm2 þ n*3 dm3 þ n*4 dm4 ¼ 0

(A1)

n1 dm1 þ n2 dm2 þ n3 dm3 þ n4 dm4 ¼ 0

(A2)

Combining Eqs. (A1) and (A2), and introducing the KB parameter

Gui ¼

n*i  ni nu ni

(A3)

one obtains

 vm* n  u ¼ n1 Gu1 þ 3 vm1 n1  n  n1 1 þ 3 n1

vm3 n vm4 G þ 4 G vm1 u3 n1 vm1 u4  vm3 n4 vm4 G þ vm1 n1 vm1 u2

 (A4)

In the experiments, species 3 and 4 are used as counterions, both of which are far more dilute than water, such that n3/n1 < < 1 and n4/n1 < < 1. Moreover, when the species 2 is added to change m1, vm4 3 it is safe to assume that vm vm1 and vm1 do not change drastically. Hence the first and second terms respectively become Gu1 and Gu2. Although the terms with Gu3 and Gu4 are considered to be negligible in Eq. (A4), the effect of KCl is seen in Tables 1e4. Actually, Gu1 and Gu2 are determined by the interactions of carrageenan with water and cosolvent, respectively, which are both mediated in turn by water, cosolvent, and KCl. Tables 1e4 show that Gu1 and Gu2 decrease with the KCl concentration. The effective interactions of carrageenan and water with water and cosolvent becomes more repulsive with the addition of KCl.

291

References Arakawa, T., & Timasheff, S. N. (1982). Stabilization of protein structure by sugars. Biochemistry, 21, 6536e6544. Back, J. F., Oakenfull, D., & Smith, M. B. (1979). Increased thermal stability of proteins in the presence of sugars and polyols. Biochemistry, 18, 5191e5196. Burey, P., Bhandari, B. R., Howes, T., & Gidley, M. J. (2008). Hydrocolloid gel particles: formation, characterization, and application. Critical Reviews in Food Science and Nutrition, 48, 361e377. Casassa, E. F., & Eisenberg, H. (1964). Thermodynamic analysis of multicomponent solutions. Advances in Protein Chemistry, 19, 287e394. Dill, K. A. (1990). Dominant forces in protein interactions. Biochemistry, 29, 7133e7155. Frank, S. H., & Evans, M. W. (1945). Free volume and entropy in condensed systems III. Entropy in binary liquid mixtures; partial molal entropy in dilute solutions; structure and thermodynamics in aqueous electrolytes. Journal of Chemical Physics, 13, 507e532. Galema, S. A., & Hoeiland, H. (1991). Stereochemical aspects of hydration of carbohydrates in aqueous solutions. 3. Density and ultrasound measurements. Journal of Physical Chemistry, 95, 5321e5326. Gekko, K., & Kasuya, K. (1985). Effect of pressure on the sol-gel transition of carrageenans. International Journal of Biological Macromolecles, 7, 299e306. Gekko, K., & Koga, S. (1983). Increased thermal stability of collagen in the presence of sugars and polyols. Journal of Biochemistry, 94, 199e205. Gekko, K., Mugishima, H., & Koga, S. (1987). Effects of sugars and polyols on the solegel transition of k-carrageenan: calorimetric study. International Journal of Biological Macromolecles, 9, 146e152. Kauzmann, W. (1959). Some factors in the interpretation of protein denaturation. Advances in Protein Chemistry, 14, 1e63. Kirkwood, J. G., & Buff, F. P. (1951). The statistical mechanical theory of solution. Journal of Chemical Physics, 19, 774e777. Kumar, S., Modig, K., & Halle, B. (2003). Trifluoroethanol-induced b/a transition in b-lactoglobulin: hydration and cosolvent binding studied by 2H, 17O, and 19F magnetic relaxation dispersion. Biochemistry, 42, 13708e13716. Loret, C., Ribelles, P., & Lundin, L. (2009). Mechanical properties of k-carrageenan in high concentration of sugar solutions. Food Hydrocolloids, 23, 823e832. Makhatadze, G. I., & Privalov, P. L. (1992). Protein interactions with urea and guanidinium chloride. Journal of Molecular Biology, 226, 491e505. Marcus, Y. (2009). Effect of ions on the structure of water: structure making and breaking. Chemical Reviews, 109, 1346e2137. Minton, A. P. (1974). A thermodynamic model for gelation of sickle-cell hemoglobin. Journal of Molecular Biology, 82, 483e498. Necas, J., & Bartosikova, L. (2013). Carrageenan: a review. Veterinarni Medicina, 58, 187e205. Newman, K. E. (1994). KirkwoodeBuff solution theory: derivation and applications. Chemical Society Reviews, 23, 31e40. Nishinari, K., & Watase, M. (1987). Effects of polyhydric alcohols on thermal and rheological properties of polysaccharide gels. Agricultural and Biological Chemistry, 51, 3231e3238. Nishinari, K., & Watase, M. (1992). Effects of sugars and polyols on the gel-sol transition of kappa-carrageenan gels. Thermochimica Acta, 206, 149e162. Nishinari, K., Watase, M., Williams, P. A., & Philips, G. O. (1990). k-Carrageenan gels: effect of sucrose, glucose, urea and guanidine hydrochloride on the rheological and thermal properties. Journal of Agricultural and Food Chemistry, 38, 1188e1193. Oakenfull, D. (2000). Solvent structure and gelation of polysaccharides in concentrated solutions of simple sugars. Gums and Stabilisers for the Food Industry, 10, 277e284. Politi, R., Sapir, L., & Harries, D. (2009). The impact of polyols on water structure in solution: a computational study. Journal of Physical Chemistry A, 113, 7548e7555. Prajapati, V. D., Maheriya, P. M., Jani, G. K., & Solanki, H. K. (2014). Carrageenan: a natural seaweed polysaccharide and its applications. Carbohydrate Polymers, 105, 97e112. Prakash, V., Loucheux, C., Scheufele, S., Gorbunoff, M. J., & Timasheff, S. N. (1981). Interactions of proteins with solvent components in 8M urea. Archives of Biochemistry and Biophysics, 210, 455e464. Ramakrishnan, S., & Prud’homme, R. K. (2000). Behaviour of k-carrageenan in glycerol and sorbitol solutions. Carbohydrate Polymers, 43, 327e332. Russ, N., Zielbauer, B. I., & Vilgis, T. A. (2014). Impact of sucrose and trehalose on different agarose-hydrocolloid systems. Food Hydrocolloids, 41, 44e52. Saha, D., & Bhattacharya, S. (2010). Hydrocolloids as thickening and gelling agents in food: a critical review. Journal of Food Science and Technology, 47, 587e597. Saltmarsh, M. (2015). Recent trends in the use of food additives in the United Kingdom. Journal of Science and Food Agriculture, 95, 649e652. Shimizu, S. (2004). Estimating hydration changes upon biomolecular reactions from osmotic stress, high pressure, and preferential hydration experiments. Proceedings of the National Academy of Sciences of USA, 101, 1195e1199. Shimizu, S., & Boon, C. L. (2004). The KirkwoodeBuff theory and the effect of cosolvents on biochemical reactions. Journal of Chemical Physics, 121, 9147e9155. Shimizu, S., & Matubayasi, N. (2014a). Preferential hydration: dividing surface vs excess numbers. Journal of Physical Chemistry B, 118, 3922e3930. Shimizu, S., & Matubayasi, N. (2014b). Gelation: the role of sugars and polyols on

292

R. Stenner et al. / Food Hydrocolloids 54 (2016) 284e292

gelatin and agarose. Journal of Physical Chemistry B, 118, 13210e13216. Smith, P. E. (2004). Cosolvent interactions with biomolecules: relating computer simulation data to experimental thermodynamic data. Journal of Physical Chemistry B., 108, 18716e18724. Sugiyama, M., Yuasa, C., Hara, K., Hiramatsu, N., Nakamura, A., Hayakawa, Y., et al. (1998). Structural change of k-carrageenan gel near sol-gel transition point. Physica B, 241e243, 999e1001. Tanford, C. (1968). Protein denaturation. Advances in Protein Chemistry, 23, 121e282. Tanford, C. (1970). Part C: theoretical models for the mechanism of denaturation. Advances in Protein Chemistry, 24, 2e95. Timasheff, S. N. (1998). In disperse solution, “osmotic stress” is a restricted case of preferential interactions. Proceedings of the National Academy of Sciences USA, 95, 7363e7367.

Timasheff, S. N. (2002a). Protein-solvent preferential interactions, protein hydration, and the modulation of biochemical reactions by solvent components. Proceedings of the National Academy of Sciences, 99, 9721e9726. Timasheff, S. N. (2002b). Protein hydration, thermodynamic binding and preferential hydration. Biochemistry, 41, 13473e13482. Uedaira, H., Ikura, M., & Uedaira, H. (1989). Natural-abundance oxygen-17 magnetic relaxation in aqueous solutions of carbohydrates. The Bulletin of the Chemical Society of Japan, 62, 1e4. Watase, M., Kohyama, K., & Nishinari, K. (1992). Effects of sugars and polyols on the gel-sol transition of agarose by differential scanning calorimetry. Thermochimica Acta, 206, 163e173. Yuguchi, Y., Urakawa, H., & Kajiwara, K. (2003). Structure characteristics of carrageenan gels: various types of counter ions. Food Hydrocolloids, 17, 481e485.