Comparison between structural changes of heat-treated and transglutaminase cross-linked beta-lactoglobulin and their effects on foaming properties

Food Hydrocolloids 25 (2011) 1758e1765

Contents lists available at ScienceDirect

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

Comparison between structural changes of heat-treated and transglutaminase cross-linked beta-lactoglobulin and their effects on foaming properties Germán D. Báez a, Andrea Moro a, Griselda A. Ballerini a, b, Pablo A. Busti a, Néstor J. Delorenzi a, * a

Área Fisicoquímica, Departamento de Química-Física, Facultad de Ciencias Bioquímicas y Farmacéuticas, Universidad Nacional de Rosario, Suipacha 531, S2002LKR Rosario, Argentina b Centro de Investigaciones y Desarrollo en Tecnología de los Alimentos, Universidad Tecnológica Nacional, Facultad Regional Rosario, Estanislao Zeballos 1341, 2000 Rosario, Argentina

a r t i c l e i n f o

a b s t r a c t

Article history: Received 3 December 2010 Accepted 23 February 2011

The main whey protein, b-lactoglobulin, was enzymatically modified by transglutaminase and analyzed for structural and conformational changes and their impact in protein foaming properties: foamability and foam stability. Solutions containing 25 mg mL
1 of b-lactoglobulin, 0.07 M cysteine in 20 mM sodium phosphate buffer pH 8.0 were incubated with transglutaminase, at a level of 1 U g
1 substrate, for different periods of time: 30, 60, 120 and 180 min. Protein structural characterization was discussed based on electrophoresis, fluorescence and viscosity studies. Comparison between the effects on foaming properties of this enzymatic treatment with those produced by heating, assayed in a previous work, was made. While 3 min was pointed as the critical time in heating treatment, 60 min was identified as the corresponding crucial time for transglutaminase treatment. The most significant conformational change, the greatest amount of dimers and trimers, and approximately the same proportion among protein species were verified at these times. Foamabilities were similar regardless of the treatment, but foam stability for heated b-lactoglobulin, measured through the change in foam volume with time is w250% higher than the same property for the protein enzymatically treated. Heating produces a higher degree of unfolding and index of surface hydrophobicity; less compact and more asymmetrical structures, with higher flexibility, which implies a greater capacity of rearrangement in the interface, producing a stiffer viscoelastic film, which slows down disproportionation through mechanisms that involve resisting compression and reducing gas transport. This improved film can be responsible for the higher foam volume stability. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Beta-lactoglobulin Heating treatment Transglutaminase Protein aggregates Foaming properties

1. Introduction The study of foam properties (foamability and foam stability) has a relevant application in food industry since many foods such as bread, meringue, ice cream or cake include foams as a vital component to improve their texture. Foamability can be understood as the capacity of achieving certain level of desired air phase volume and foam stability, as the foam endurance against destabilizing processes like mixing, cutting and heating or simply the aging process of the foam (Foegeding, Luck, & Davis, 2006). Proteins are frequently used as foaming agents in foods since they contribute not only to the formation but also to the stability of foams. Compared to low molecular weight surfactants, proteins are less effective to reduce the air/water interfacial tension but they

* Corresponding author. Tel./fax: þ54 341 4804598. E-mail address: [email protected] (N.J. Delorenzi). 0268-005X/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodhyd.2011.02.033

form an interfacial film exhibiting viscoelastic properties that is thought to improve the resistance of the foam under stress or aging conditions (Damodaran, 2005; Foegeding et al., 2006). Destabilization of quiescent foams is determined by processes such as film drainage, film rupture and disproportionation (Wilde & Clark, 1996), where proteins also play an important role, mainly through their conformation and state of aggregation. Among the known ways to produce protein aggregation, heating or enzymatic treatments of the protein sample are frequently used. In a recent work (Moro, Báez, Busti, Ballerini, & Delorenzi, 2010), we have shown that the time of previous heating is a crucial variable for the features of beta-lactoglobulin (b-LG) as a foaming agent, due to conformational changes and the different proportion of species in solution this treatment produces. In this cited work, 3 min was pointed as the critical time when 55 mg mL
1 b-LG solution was heated at 85  C, since the most significant conformational change and aggregation process occur at this time, producing non-native monomers and the greatest amount of

G.D. Báez et al. / Food Hydrocolloids 25 (2011) 1758e1765

dimers and trimers (monomer 51%, dimer 33% and trimer 16%). Heating treatment affects foamability and even more foam stability. Both foam properties are closely linked to structural changes of the protein. The increase in surface hydrophobicity (SH) is considered a decisive factor in the improved foamability, in spite of the presence of aggregates of higher molecular weight. Also, the best foam stabilization was achieved at 3 min of heating treatment, coincidentally with the occurrence of those conformational changes. Moreover, in spite of the variation in species composition that heating treatment produce in b-LG, the Kapp in Stern-Volmer plots remain almost constant after 3 min of heating treatment. Thus, understanding Kapp as a SH parameter (Moro, Gatti, & Delorenzi, 2001), it can be concluded that SH is almost constant for these conditions and therefore, in the species mixture, the exposure degree of the aminoacidic residues responsible for it (Trp) is also approximately the same in this experimental range: the resultant of all exposed Trp, in terms of accessibility of these residues, is approximately the same regardless the pattern of species involved. The enzyme transglutaminase (TG; protein-glutamine gglutamyltransferase; EC 2.3.2.13) catalyzes an acyl transfer reaction between the g-carboxyamide group of peptide-bound glutamine residues (acyl donors) and a variety of primary amines (acyl acceptors), including the 3-amino group of lysine residues in certain proteins. TG can modify proteins by means of amine incorporation, cross-linking and deamidation (Motoki & Seguro, 1998). When the enzyme acts on protein molecules, intra- and intermolecular 3-(g-glutamyl)lysine cross-links are formed. With the microbial TG discovery, its use by the food industry (Kuraishi, Yamazaki, & Susa, 2001) and the cross-linking produced by the enzyme have been extensively researched for some proteins as caseins, gluten, myosin, soy protein and milk protein. TG is used to improve the texture and to modify the functional properties of prepared foods in general like seafood, surimi products, noodles and dairy products (Gauche, Vieira, Ogliari, & Bordignon-Luiz, 2008). In view of the previous knowledge about the effects of heating treatment on foaming properties of b-LG, the aim of this work is comparing the effects on these functional properties when an alternative aggregation method, the enzymatic reaction with TG, is used on this milk protein. 2. Materials and methods 2.1. Materials

b-LG was purchased from Sigma Chemicals Co. (St. Louis, MO, USA) and used without further purification. A commercial preparation of microbial TG (ActivaÒ) was donated by Ajinomoto Co., Ltd. (Tokyo, Japan). Commercial TG is a mixture containing 99% maltodextin and 1% enzyme with 50 U g
1 of declared activity measured by hidroxamate method (Folk & Cole, 1966). The enzyme was used in the original form without any further purification. All other chemicals were of analytical grade. 2.2. TG cross-linking of b-LG Solutions containing 25 mg mL
1 of b-LG, 0.07 M cysteine (Cys) and TG at a level of 1 U g
1 substrate were prepared in 20 mM sodium phosphate buffer pH 8.0. The aggregation reaction was carried out at 40  C for different periods of time (30, 60, 120 and 180 min). Enzyme reaction was stopped at 60  C for 15 min (Rodriguez-Nogales, 2006; Walsh, Cleary, McCarthy, Murphy, & FitzGerald, 2003). These enzymatically treated samples were then cooled to 4  C in a water/ice bath, and then frozen and lyophilized (with a freeze dryer lyophilizer LIOTOP L 101, Liobras Ind Com e Serv de Liofilizadores Ltd., Brazil). For subsequent analyses these

1759

lyophilized samples were redissolved to the desired concentration with 20 mM sodium phosphate buffer pH 6.8. In a previous work (Eissa, Bisram, & Khan, 2004), it was noted that the b-LG molecule is a poor substrate for TG and that it must be partially or completely denatured to undergo enzymatic crosslinking. Reducing agents, like Cys, produce conformational changes on proteins and subsequent exposure of the enzymetargeted sites, because of the disruption of disulfide bonds in the substrate molecule. Fort, Carretero, Parés, Toldrá, and Saguer (2007), working with 0.02 M Cys as reducing agent, indicated that its presence does not improve TG activity when plasma proteins are used as substrate. This was probably due to the fact that Cys cannot increase the protein unfolding degree, hindering the cross-linking reactions among lysine and glutamine residues. However, Battaglin VillasBoas, Viera, Trevizan, de Lima Zollner, and Netto (2010) have shown the effectiveness of Cys to promote TG action, using higher concentrations, between 0.05 and 0.4 M. Therefore, considering that the effectiveness of Cys as reducing agent depends on its concentration, in this work Cys 0.07 M was used. 2.3. Heat treatment of b-LG A stock 55 mg mL
1 b-LG solution was prepared in 20 mM sodium phosphate buffer at pH 6.8. An aliquot from this solution was placed in small glass tube and heated during 3 min in a water bath at 85  C. This heated sample (HT-b-LG) was cooled to 4  C in a water/ice bath, and then frozen and lyophilized. For subsequent analyses these lyophilized samples were redissolved to the desired concentration with 20 mM sodium phosphate buffer pH 6.8. 2.4. Electrophoresis SDS-PAGE of cross-linked b-LG with TG was performed as described by Laemmli (1970) described, using a stacking gel of 10% acrylamide and a running gel of 15% acrylamide. The samples were previously reduced with b-mercaptoethanol (Battaglin Villas-Boas et al., 2010). After electrophoresis, gels were stained with Coomassie Brilliant Blue R250 and scanned using a HewlettePackard ScanJet 5p connected to a computer. To quantify the relative intensities of the stained protein bands, the pixel densities of digitized images were analyzed using software developed by our group (Palazolo, Rodriguez, Farruggia, Picó, & Delorenzi, 2000). The molecular weight of each protein band was matched to known standard proteins.

2.5. Fluorescence studies 2.5.1. Fluorescence quenching Conformational studies were carried out with the fluorescence quenching method of Moro et al. (2001) for different samples of b-LG: without any treatment (native), reduced by Cys, heated and treated with TG in the presence of 8 M urea. In this method, 3.0 mL of b-LG sample (10 mM in 20 mM sodium phosphate buffer at pH 6.8) was placed in the cell of a Jasco FP-770 spectrofluorometer and the fluorescence intensity (F0) was measured at 337 nm, using excitation at 295 nm. Aliquots of 5 M acrylamide, used as fluorescence quencher, were sequentially added in the cell content, and the new fluorescence intensities were measured (F). Acrylamide concentrations ranged from 0 to 0.2 M. The F0/F ratio was plotted versus the quencher concentration (SterneVolmer plot). In the range used, this plot was linear and the SterneVolmer equation can be expressed as:

1760

G.D. Báez et al. / Food Hydrocolloids 25 (2011) 1758e1765

F0 ¼ 1 þ Kapp ½acrylamide F

(1)

The initial slope of SterneVolmer plots, Kapp, is an index of protein conformation. Kapp is an apparent constant because b-LG has more than one tryptophanyl residue that can be quenched by acrylamide. 2.5.2. Urea unfolding profiles Equilibrium unfolding curves using urea is a convenient method to estimate the conformation stability of a protein (Shirley, 1995). Chemical unfolding profiles of b-LG species (native, reduced by Cys, heat treated and treated with TG) in the presence of urea were performed following the works of Moro et al. (2001) and Busti, Scarpeci, Gatti, and Delorenzi (2005). A series of urea solutions, from 0 to 8 M, were made from weighed quantities of urea in 20 mM sodium phosphate buffer at pH 6.8. An aliquot of pretreated b-LG solution was added to 3.0 mL of each of urea solution to give a final protein concentration of 0.13 mg mL
1. In order to determine the lmax for each sample, emission spectra were carried out, using an excitation wavelength of 295 nm. Assuming a two-state model (Pace, 1986, 1990; Tanford, 1968), in which a species of a protein possesses only two different conformational states with different values of an observed property (Y), the fraction of protein with a conformational change (FU), at any given urea concentration, can be determined from the following equation:

FU ¼

Y
YMIN YMAX
YMIN

2.6. Intrinsic viscosity Intrinsic viscosity was measured with an Ostwald capillary viscometer in a controlled temperature water bath (25  0.1  C). The water draining time was 89.44 s. To determine intrinsic viscosity of each sample, draining times of b-LG solutions of five different concentrations, between 10 and 60 mg mL
1, were measured. Two measurements for each concentration were performed. The viscosity increment, hi, was calculated from:

(3)

where hr is the relative viscosity, determined for:

hr ¼

hs r $ts ¼ s h0 r0 $t0

(4)

where h0, r0 and t0 are the viscosity, density and efflux time of solvent, while hs, rs and ts are the viscosity, density and efflux time of protein solution. On the basis of the well-known Huggins equation:

hi C

¼ ½h þ k$½h2 $C

  Ct VLF ¼ Vinit 1
Cinit

(5)

where C is the concentration of protein (g mL
1) and [h], the intrinsic viscosity, was determined from the extrapolation of the plot of hi/C versus C (Tanford, 1961). Density of each solution was measured using an Anton Paar DMA35N densimeter. 2.7. Foaming properties Foams were formed using a bubbling apparatus (Hagolle, Relkin, Popineau, & Bertrand, 2000; Loisel, Guéguen, & Popineau, 1993).

(6)

As it is reported previously (Croguennec, Renault, Bouhallab, & Pezennec, 2006; Fains, Bertrand, Baniel, & Popineau, 1997; Hagolle et al., 2000), foams are compared on the basis of (i) maximum foam density (FD), as a measure of foamability; and (ii) half-life time of drainage (T1/2) and (iii) volume variation with time, as measures of foam stability. FD is defined as the ratio between the maximal liquid incorporated into the foam (VLFmax) and the foam volume reached in the end of the sparging period (VFmax):

(2)

where YMIN and YMAx are the observed property values in 0 and 8 M urea, respectively. In the present work, lmax values were used as the observed property Y. [urea]1/2 is the urea concentration capable of changing the conformation of the 50% of protein molecules.

hi ¼ hr
1

Native and treated b-LG samples were dissolved to a final concentration of 0.1% (w/v) in 20 mM phosphate buffer pH 6.8. Determinations were made in a transparent acrylic tube (3.5 cm  20.0 cm) equipped with a pair of electrodes located at the base of the column and with a porous disk through which air, at a flow rate of 5 mL s
1, was passed and forced through the liquid (Vinit ¼ 10 mL), creating foam. Bubbling stopped when the foam reached a fixed volume of 115 mL (Vf). During the test, the conductivity and the volume of foam were recorded by a computer and a digital camera Olympus DS-580 4.0 mega pixel. Conductivity measurements at different times (Ct) and with reference to the initial conductivity (Cinit) were used to calculate the volume of liquid in the foam (VLF) (Chevalier, Chobert, Popineau, Nicolas, & Haertle, 2001; Loisel et al., 1993):

FD ¼

VLFmax VFmax

(7)

It has been noted that VFmax ¼ Vf
(Vinit
VLFmax). T1/2, the half-life time of drainage, is equal to:

T1=2 ¼ t1=2
t0

(8)

where t1/2 is the time when half of the maximum volume of liquid in the foam came back to the solution (VLFmax/2) and t0 is the time in the end of bubbling. On the other hand, in terms of volume variation with time, the longer it takes for the foam to collapse, the more stable the foam is (Wilde & Clark, 1996). This can be measured through the one quarter time of foam volume (T1/4), which is defined as:

T1=4 ¼ t1=4
t0

(9)

where t1/4 is the time required for VFmax to decay 25% and t0 is the time in the end of bubbling. Both times, T1/2 and T1/4, and FD were informed as relative ratios in reference to their values for the protein without any treatment: FD , T1/2 and T1/4 . All experiments were performed at 25  C. 3. Results and discussion 3.1. Protein structural characterization 3.1.1. Electrophoresis b-LG solutions non-treated and treated with TG for 30, 60, 120 and 180 min were analyzed by SDS-PAGE (reducing conditions) and the different b-LG species were quantified by gel densitometry. With increasing incubation time with TG, the amount of monomeric b-LG decreased and aggregates of several sizes were formed: dimers, trimers, oligomers and polymers (Fig. 1). For longer periods of incubation time, the quantities of dimers and trimers decreased, whereas the amount of aggregates larger than trimers increased abruptly. At 60 min of incubation time with TG, the

G.D. Báez et al. / Food Hydrocolloids 25 (2011) 1758e1765 Table 1 Fluorescence studies of different b-LG samples.

4.5 4.0

80

3.5 60

3.0 2.5

40

2.0 20

0

Treatment

Kapp (M
1)

[urea]1/2 (M)

b-LG b-LG-Cysb TG-CL-b-LG HT-b-LG b-LG þ ureac

1.97  0.18 2.48  0.15 3.35  0.12 3.74  0.10 12.05  0.10

4.88  0.10 4.66  0.15 4.30  0.05 4.09  0.08 e

a

K app (M -1)

Protien concentration ( )

100

1.5

0

30

60

120

180

1761

1.0

b-LG, b-LG-Cys, and b-LG þ urea were prepared in buffer phosphate 20 mM at pH 6.8. [urea]1/2 was determined from the chemical unfolding profiles of b-LG species in the presence of urea (0e8 M), by supposing a two-state model. Each value presented in this table is the average value obtained for triplicates. a b-LG without any treatment. b b-LG treated with 0.07 M Cys. c b-LG in 8 M urea.

Incubation time with TG (min) Fig. 1. Protein concentrations (%) of different species formed incubating b-LG solutions (25 mg mL
1) with TG in the presence of 0.07 M Cys, at 40  C, and for different periods of time: monomers (C), dimers (-), trimers (:) and aggregates not entering the running gel (oligomers and polymers) (A). Each point was the mean of two measurements. (B) Kapp (M
1) from the SterneVolmer graphs at different times of incubation with TG. Error bars were calculated from the standard error of three replicates.

largest amount of dimers and trimers was found, and oligomers and polymers appeared, representing approximately 50% of the species in solution at 180 min. From now on in this work, the sample enzymatically treated for 60 min, is called TG-CL-b-LG. Comparing both treatments, heating and enzymatic, it has been noted that the most significant conformational change was verified and the greatest amount of dimers and trimers was produced at 3 min in the former and at 60 min in the latter. Moreover, approximately the same proportion among protein species was present either in HT-b-LG sample (monomer 51%, dimer 33% and trimer 16%) (Moro et al., 2010) and in TG-CL-b-LG (monomer 53%, dimer 26%, trimer 18% and larger aggregates 3%) (Fig. 1). 3.1.2. Fluorescence quenching Protein unfolding causes a red shift on the lmax fluorescence emission due to the major exposition of Trp residues to the aqueous solvent, which also promotes an increase in the fluorescence quenching by acrylamide of denatured b-LG (Busti, Gatti, & Delorenzi, 2006; Moro et al., 2001; Palazolo et al., 2000). The more flexible the protein structure and the higher the unfolded species proportion, the more pronounced the slope in the SterneVolmer plot and then, the greater Kapp (Equation (1)). Besides, during the unfolding process, the protein surface becomes more and more hydrophobic due to the appearance of nonpolar amino acids, which were previously inside the protein structure. Thus, Kapp values are also used as a measure of SH (Moro et al., 2001). Fig. 1 shows the Kapp values from SterneVolmer plots for b-LG cross-linked by TG in the presence of Cys, for the cited different times. The initial Kapp value in this plot is higher than that obtained for b-LG without any treatment and in the absence of Cys (Table 1), as the presence of this reducing agent breaks disulfide bonds, making the opening of the protein structure easier. The effectiveness of the quenching process, measured through the Kapp values, increased with the length of enzymatic treatment, verifying a minimum value for the initial time, and a maximum after 60 min of incubation time with TG. The Kapp increment during the TG treatment can be ascribed to the presence of a percentage of nonnative monomers and an increasing percentage of irreversibly formed aggregates of low molecular weight of unfolded molecules. The slight decrease in Kapp value for 180 min of incubation time with TG is coherent with the presence of aggregates of high

molecular weight whose compact structure could reduce Trp accessibility. Consequently with the more important conformational changes, TG-CL-b-LG and HT-b-LG presented the maximum values of Kapp. However, Table 1 shows differences between Kapp values of treated b-LG, which are directly related to unfolding degree. The minimum Kapp value is verified for native b-LG without any treatment and the maximum, for b-LG chemically denatured with the highest concentration of urea, when the protein reaches its greater denaturation state. Between these extremes, it can be observed that Kapp decreased in the following order: HT-b-LG > TG-CLb-LG > b-LG-Cys. These results point at the presence of a residual structure that hinders the full Trp accessibility to the acrylamide in these species. 3.1.3. Urea unfolding profiles Urea promotes unfolding by both indirect and direct mechanisms. Direct urea interactions consist of hydrogen bonding to the polar moieties of the protein, particularly peptide groups, leading to screening of intramolecular hydrogen bonds. Solvation of the hydrophobic core proceeds via the influx of water and urea molecules. Urea also promotes protein unfolding in an indirect way by altering water structure and dynamics, as also occurs on the introduction of nonpolar groups to water, thereby diminishing the hydrophobic effect and facilitating the exposure of the hydrophobic core residues. Overall, urea-induced effects on water indirectly contribute to unfolding by encouraging hydrophobic solvation, whereas direct interactions provide the pathway (Bennion & Daggett, 2003; Rossky, 2008). Fig. 2 shows the FU (fraction of protein with a conformational change) for the different b-LG samples in the range of concentrations of urea assayed. The obtained [urea]1/2 values incremented in the following order: HT-b-LG < TG-CL-b-LG < b-LG-Cys < b-LG, as it is shown in Table 1. The higher the value of [urea]1/2, the more difficult the disruption on protein structure. In the case of b-LG-Cys, the observed conformational changes demonstrated that the rupture of disulfide bonds by the Cys as reducing agent, did not involve extensive unfolding of b-LG molecules. 3.1.4. Intrinsic viscosity Intrinsic viscosity is a measure of the hydrodynamic volume a molecule occupies. In the present work, native b-LG presented an intrinsic viscosity value of 3.7 mL g
1 (Table 2), which corresponds to a compact and globular spherical protein, regardless of their molecular weight (Ross-Murphy, 1994). A value 19% higher than this was presented by b-LG-Cys. This increment can be due to an increase of the molecular hydrodynamic volume of the protein, because of the loss of native structure caused by disruption of disulfide bonds (Tanford, 1968). Besides, this moderate increment is

1762

G.D. Báez et al. / Food Hydrocolloids 25 (2011) 1758e1765

1.0

heating treatment 0.8

HT- β -LG

FU

0.6 0.4

native β -LG

0.2

TG treatment

hydrophobic residue

0.0

TG-CL-β-LG 0

2

4

6

8

[urea] (M) Fig. 2. Urea unfolding profiles of different b-LG samples, determined through lmax values. (C) b-LG without any treatment, (-) b-LG-Cys, (B) HT-b-LG and (,) TG-CL-bLG. Errors bars were calculated from the standard error of three replicates.

coherent with the moderate conformational changes found in fluorescence studies. The highest value of intrinsic viscosity was obtained for HTb-LG, followed by a lower value for TG-CL-b-LG (Table 2). These results are consistent with the effects produced by heating (Vardhanabhuti & Foegeding, 1999): the formation of larger species, with larger effective volume fraction than native molecules and, more important in terms of viscosity, aggregates that are more asymmetric in shape. On the other hand, TG-CL-b-LG generates a more compact structure (Matsumura, Lee, & Mori, 2000; Tanimoto & Kinsella, 1988) and then, less asymmetric aggregates which lead to a lower intrinsic viscosity value. 3.1.5. Comparison between heating and enzymatic treatments De la Fuente, Singh, and Hemar (2002) have reviewed several different mechanisms that have been developed for the thermal denaturation/aggregation of b-LG. Regardless of the proposed mechanism, some events must be considered during the heating process. Firstly, a critical conformational change in the b-LG monomer exposes the free sulfhydryl group (Cys 121). This reactive group in the modified molecule can induce thiol/disulfide exchange reactions, leading to the formation of disulfide-linked aggregates. The disulfide linkage involved in the intermolecular interchange reaction would most likely be the Cys66eCys160, which is found in one of the external loops of b-LG. The other disulfide is buried in the inner parts of the protein and is less available for reaction (McKenzie, Ralston, & Shaw, 1972; Papiz et al., 1986). In addition to this aggregation by covalent intermolecular disulfide bonds, noncovalent interactions (ionic, van der Waals, hydrophobic) may also be involved. The contribution of non-covalent interactions will become of increasing importance with pH values closer to the isoelectric point and/or with higher salt concentrations, but these were not the environmental conditions assayed in the present Table 2 Intrinsic viscosity of different b-LG samples. Sample

[h] (mL g
1)

b-LG b-LG-Cys TG-CL-b-LG HT-b-LG

3.7  0.2 4.4  1.3 6.0  0.9 9.2  1.0

Each value presented in this table is the average value obtained for triplicates.

Fig. 3. Possible structures of different species of b-LG produced by heating treatment (HT-b-LG) and TG treatment (TG-CL-b-LG).

work. Therefore, in the first stage of heat-induced denaturation/ aggregation of b-LG, in the absence of added salt and near neutral pH, denatured monomers were linearly linked to form intermediate oligomers (di-, tri- and tetramers) via intermolecular disulfide-bond reactions (Bauer, Carrotta, Rischel, & Ogendal, 2000; Croguennec, O’Kennedy, & Mehra, 2004). The secondary structure of native b-LG consists of nine b-strands (w50%), a single a-helix (w15%), several turns (w20%), and random arrangements (w15%) (Boye, Ismail, & Alli, 1996). During thermal treatment, the hydrogen bonds which stabilized the native structure of b-LG are broken, causing a loss of a-helix and b-sheets structures, creating new intermolecular b-sheets arrangements and enhancing the exposition of hydrophobic residues to the solvent (Eissa, Puhl, Kadla, & Khan, 2006; Kim, Cornec, & Narsimhan, 2005; Moro et al., 2010). The treatment with TG promotes 3-(g-glutamyl)lysine bonds and causes aggregation of the protein, leading to high molecular weight species. In the case of subsequently applying both treatments, heating and then enzymatic TG cross-linking, Eissa et al. (2006) suggested that a low number of bonds had been created, in spite of the potential number of bonds which could be formed according to the glutamine and lysine residues present in the backbone of b-LG (glutamine and lysine residues per b-LG molecule are 9 and 15, respectively). On one hand, large aggregates formed by heating b-LG at low ionic strength, the condition assayed in this work, seem to be open in contrast to structures formed at salt concentrations larger than 0.1 M that appear much denser (Bauer et al., 2000; Foegeding, Bowland, & Hardin, 1995). On the other hand, Tanimoto and Kinsella (1988) have shown that aggregates produced by TG treatment of b-LG in the presence of another denaturant, like dithiothreitol, contains intramolecular bonds that impede unfolding of the molecules upon heating, hence suggesting a compact nature of the polymerized b-LG molecules. Based on the results of the present work, it can be concluded that when comparing both treatments, HT-b-LG produces a higher degree of unfolding with a greater exposition of hydrophobic residues to the solvent leading to a higher index of SH, a less compact structure and a more asymmetrical form of the oligomers formed. The possible different structures are schematized in Fig. 3. 3.2. Foaming properties There are two distinct phases to protein foaming: (i) the effectiveness of gas encapsulation (foamability) and (ii) the lifetime of

G.D. Báez et al. / Food Hydrocolloids 25 (2011) 1758e1765

3.0

3.0

2.5

2.5

2.0

2.0

1.5

1.5

1.0

1.0 0

30

60

120

Relative foam density

Relative decay times

180

Incubation time with TG (min) Fig. 4. Relative foam density (FD/FD ) (:) and relative decay times: T1/2/T1/2 (-) and T1/4/T1/4 (C), with incubation time with TG. FD , T1/2 and T1/4 are the parameters for native b-LG, without any treatment. FD ¼ 0.0864  0.0069, T1/2 ¼ 23.0  1.8 s and T1/4 ¼ 11.0  0.9 min. Error bars were calculated from the standard error of three replicates.

8 7

FD/FDº T1/2 / T1/2º

6

Relative parameters

the foam (foam stability) (Foegeding et al., 2006; Wilde & Clark, 1996). The increment of foamability produced by an increment in protein SH has been studied extensively (Foegeding et al., 2006; Kato & Nakai, 1980; Moro et al., 2001, 2010; Townsend & Nakai, 1983). Disordered, smaller and flexible proteins are more efficient surface agents for foam formation than ordered, larger and rigid ones (Martin, Grolle, Bos, Cohen Stuart, & van Vliet, 2002). The higher the extent of denaturation a protein suffers, the higher SH. This leads to a greater affinity of the protein for the interface, which allows it to overcome the barrier against adsorption, which is developed at the interface while proteins are closely packed (Wilde & Clark, 1996). A rapid decrease in surface tension is promoted, leading to an increase in protein foamability. Fig. 4 shows the effects of TG treatment on b-LG foamability, measured as relative foam density (FD/FD ). It can be observed that foamability was improved w29%, for all assayed times of incubation with TG. Both electrophoresis and fluorescence quenching studies demonstrated that TG treatment produced b-LG aggregates of high molecular weight with higher exposure of hydrophobic patches than native b-LG. These two features have opposite effects on foamability: whereas the aggregates have a lower diffusion coefficient, hindering the foam formation, the exposed hydrophobic sites promote it. Considering the results, although TG treatment enhanced foamability, it was independent of the incubation time with the enzyme, since the same foamability improvement was achieved regardless of the time of TG treatment. It is likely that the described opposite effects will offset each other and thus, foamability will remain constant. During foam formation, gas bubbles are surrounded by an interfacial film of proteins assuring the protection of the foam against destabilization. The processes involved in foam destabilization are: liquid drainage, a close approach of adjacent bubble surfaces which leads to film rupture (coalescence) and gas diffusion into the continuous phase (Ostwald ripening or disproportionation) resulting in bubble coarsening. All these mechanisms occur simultaneously after the air bubbling stops. While drainage and coalescence prevail at the beginning, when the bubbles are mainly spherical, disproportionation is more important at advanced stages, when the cells are polyedric. In the end, the foam collapses. Foaming stability, estimated either through T1/2 or T1/4 values, increased in different degrees, in comparison with the native b-LG sample, for each studied incubation time with TG (Fig. 4). T1/2, the

1763

T1/4 / T1/4º

5 4 3 2 1 0 HT- ß-LG

TG-CL- ß-LG

Fig. 5. Variation of relative foam density (FD/FD ) and relative decay times (T1/2/T1/2 and T1/4/T1/4 ) for HT-b-LG and TG-CL-b-LG. Errors bars were calculated from the standard error of three replicates.

half-time of drainage, verified an increment of w35% all over the range of incubation time. This enhancement in stability can be properly explained through the presence of aggregates of denatured b-LG with high hydrodynamic size, which produces a major protein solution viscosity and then, the drainage rate slows down (Foegeding et al., 2006). Considering the observed results, although the enzymatic treatment led to an improved stability, when this foaming property was measured through T1/2, the time of incubation with TG did not affect it. On the other hand, T1/4, the time required for a 25% decay for VFmax, increased even more than T1/2 in the same range of incubation time with TG. Therefore, with enzymatic treatment up to 180 min, a peak at 60 min appeared and remained constant for further times of incubation. At 60 min, the increment was w250%, which coincided with the most significant conformational changes on b-LG, showing the close link with structural changes of the protein. Aggregates formed by TG treatment between 60 and 180 min led to more stabilized foams, slowing down disproportionation because of the formation of stiffer films that resist compression and may reduce gas transport (Moro et al., 2010). 3.2.1. Comparison between heating and enzymatic treatments Fig. 5 shows the comparison among relative parameters of foaming properties for both assayed treatments at certain times. It has already been discussed that there are some particular times in which the treatments led to similar proportion of molecular species: 3 min of heating (HT-b-LG) and 60 min of enzymatic treatment (TG-CL-b-LG). While FD/FD reached similar values regardless of the treatment, T1/2 and T1/4 for HT-b-LG are higher than those values for TG-CL-b-LG. Nevertheless, the increments are not equal: T1/2 for heated samples is w32% higher than T1/2 for enzymatically treated ones, while T1/4 for heated samples is w250% higher. As it has been previously mentioned, intrinsic viscosity of the protein has a direct influence in the viscosity of the solution, which is the most important factor in the destabilizing mechanism by drainage, mainly estimated through T1/2. Thus, an increment in viscosity makes the drainage decay, stabilizing the foam. This is the reason because higher foam stability was observed in heated samples, which present higher intrinsic viscosity in comparison with the enzymatically treated samples (Table 2). Besides, aggregates formed by heating produce a stiffer film, which lead to more stabilized foams, but in this case through the decay of another

1764

G.D. Báez et al. / Food Hydrocolloids 25 (2011) 1758e1765

mechanism of destabilization, the disproportionation, which affects T1/4 (Moro et al., 2010). Both effects lead to the same consequence, more stable foams. On the other hand, [urea]1/2 for TG-CL-b-LG was higher than [urea]1/2 for HT-b-LG. Thus, HT-b-LG appears to have the most labile chemical structure and, therefore, it would exhibit a higher flexibility and a greater capacity of rearrangement in the interface. Croguennec et al. (2006) suggested that the forces involved in the formation of the closely packed layer of adsorbed proteins could be hydrogen bonds, hydrophobic associations and electrostatic interactions rather than disulfide bonds formed by sulfhydryl-disulfide exchange reactions. Fluorescence studies in the present work showed a greater degree of SH (Table 1) for HT-b-LG than for TG-CLb-LG. In view of the results of this work, the greater foam volume stability achieved by heating treatment can be due to the interactions between the new exhibited hydrophobic sites which lead to the formation of a stiffer viscoelastic film. Therefore, the exposure of hydrophobic sites allowed hydrophobic interactions that became the most important ones, leading to a much higher increment of w250% in volume foam stability. It had been observed that SH remained constant after the critical time of heating (3 min) and foam stability, measured through T1/4, abruptly decayed after this time (Moro et al., 2010). At these times, oligomers and polymers of higher molecular weight reached a proportion of 85% in the species mixture, and produced a steric impediment which led to the formation of a more open and weaker interfacial film, responsible for the foam stability decay. When the sample was enzymatically treated with TG, SH also remains approximately constant after the corresponding critical time (60 min), and T1/4 for the different times of treatment were also constant. In this case, the proportion of oligomers and polymers of higher molecular weight only reached 48% among the present species, and they could not have affected the interfacial film as much as they did in heating treatment. Therefore, foam stability did not decay, prevailing SH as one of the most important factors determining foam stability. 4. Conclusions Comparison between the effects on foaming properties of TG treatment with the effects produced by heating was made in this work. While 3 min was pointed as the critical time in heating treatment, 60 min has been identified as the corresponding crucial time for TG treatment. The most significant conformational change, the greatest amount of dimers and trimers, and approximately the same proportion among protein species are verified at those times. Foamabilities are similar regardless of the treatment, but foam stability for heated b-LG, measured through the change in foam volume with time, is w250% higher than the same property for the protein enzymatically treated. Heating produces a higher degree of unfolding and index of SH; less compact and more asymmetrical structures, with higher flexibility, which implies a greater capacity of rearrangement in the interface, producing a stiffer viscoelastic film. This improved film can be responsible for the higher foam volume stability. Acknowledgements This work was supported by a grant from the Agencia Nacional de Promoción Científica y Tecnológica, Argentina (PICT 2006-1836). References Battaglin Villas-Boas, M., Viera, K. P., Trevizan, G., de Lima Zollner, R., & Netto, F. M. (2010). The effect of transglutaminase-induced polymerization in the presence

of cystein on b-lactoglobulin antigenicity. International Dairy Journal, 20(6), 386e392. Bauer, R., Carrotta, R., Rischel, C., & Ogendal, L. (2000). Characterization and isolation of intermediates in b-lactoglobulin heat aggregation at high pH. Biophysical Journal, 79(2), 1030e1038. Bennion, B. J., & Daggett, V. (2003). The molecular basis for the chemical denaturation of proteins by urea. Proceedings of the National Academy of Sciences of the United States of America, 100(9), 5142e5147. Boye, J. I., Ismail, A. A., & Alli, I. (1996). Effect of physico-chemical factors on the secondary structure of beta-lactoglobulin. Journal of Dairy Research, 63(1), 97e109. Busti, P. A., Gatti, C. A., & Delorenzi, N. J. (2006). Binding of alkylsulfonate ligands to bovine b-lactoglobulin: effects on protein thermal unfolding. Food Research International, 39(4), 503e509. Busti, P. A., Scarpeci, S., Gatti, C. A., & Delorenzi, N. J. (2005). Binding of alkylsulfonate ligands to bovine b-lactoglobulin: effects on protein denaturation by urea. Food Hydrocolloids, 19(2), 249e255. Chevalier, F., Chobert, J. M., Popineau, Y., Nicolas, M. G., & Haertle, T. (2001). Improvement of functional properties of b-lactoglobulin glycated through the Maillard reaction is related to the nature of the sugar. International Dairy Journal, 11(3), 145e152. Croguennec, T., O’Kennedy, B. T., & Mehra, R. (2004). Heat-induced denaturation/ aggregation of b-lactoglobulin A and B: kinetics of the first intermediates formed. International Dairy Journal, 14(5), 399e409. Croguennec, T., Renault, A., Bouhallab, S., & Pezennec, S. (2006). Interfacial properties and foaming properties of sulfhydryl-modified bovine b-lactoglobulin. Journal of Colloid and Interface Science, 302(1), 32e39. Damodaran, S. (2005). Protein stabilization of emulsions and foams. Journal of Food Science, 70(3), R54eR66. De la Fuente, M. A., Singh, H., & Hemar, Y. (2002). Recent advances in the characterization of heat-induced aggregates and intermediates of whey proteins. Trends in Food Science & Technology, 13(8), 262e274. Eissa, A. S., Bisram, S., & Khan, S. A. (2004). Polymerization and gelation of whey protein isolates at low pH using transglutaminase enzyme. Journal of Agricultural and Food Chemistry, 52(14), 4456e4464. Eissa, A. S., Puhl, C., Kadla, J. F., & Khan, S. A. (2006). Enzymatic cross-linking of b-lactoglobulin: conformational properties using FTIR spectroscopy. Biomacromolecules, 7(6), 1707e1713. Fains, A., Bertrand, D., Baniel, A., & Popineau, Y. (1997). Stability and texture of protein foams: a study by video image analysis. Food Hydrocolloids, 11(1), 63e69. Foegeding, E. A., Bowland, E. L., & Hardin, C. C. (1995). Factors that determine the fracture properties and microstructures of globular protein gels. Food Hydrocolloids, 9(4), 237e249. Foegeding, E. A., Luck, P. J., & Davis, J. P. (2006). Factors determining the physical properties of protein foams. Food Hydrocolloids, 20(2e3), 284e292. Folk, J. E., & Cole, P. W. (1966). Transglutaminase: mechanistic features of the active site as determined by kinetic and inhibitor studies. Biochimica et Biophysica Acta (BBA) e Enzymology and Biological Oxidation, 122(2), 244e264. Fort, N., Carretero, C., Parés, D., Toldrá, M., & Saguer, E. (2007). Combined treatment of porcine plasma with microbial transglutaminase and cysteine: effects on the heat-induced gel properties. Food Hydrocolloids, 21(3), 463e471. Gauche, C., Vieira, J. T. C., Ogliari, P. J., & Bordignon-Luiz, M. T. (2008). Crosslinking of milk whey proteins by transglutaminase. Process Biochemistry, 43(7), 788e794. Hagolle, N., Relkin, P., Popineau, Y., & Bertrand, D. (2000). Study of the stability of egg white protein-based foams: effect of heating protein solution. Journal of the Science of Food and Agriculture, 80(8), 1245e1252. Kato, A., & Nakai, S. (1980). Hydrophobicity determined by a fluorescent probe method and its correlation with surface properties of proteins. Biochimica et Biophysica Acta, 624(1), 13e20. Kim, D. A., Cornec, M., & Narsimhan, G. (2005). Effect of thermal treatment on interfacial properties of beta-lactoglobulin. Journal of Colloid and Interface Science, 285(1), 100e109. Kuraishi, C., Yamazaki, K., & Susa, Y. (2001). Transglutaminase: its utilization in the food industry. Food Reviews International, 17(2), 221e246. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680e685. Loisel, W., Guéguen, J., & Popineau, Y. (1993). A new apparatus for analyzing foaming properties of proteins. In K. D. Schwenke, & R. Mothes (Eds.), Food proteins: Structure and functionality (pp. 320e323). New York: VCH Publishers. Martin, A. H., Grolle, K., Bos, M. A., Cohen Stuart, M. A., & van Vliet, T. (2002). Network forming properties of various proteins adsorbed at the air/water interface in relation to foam stability. Journal of Colloid and Interface Science, 254(1), 175e183. Matsumura, Y., Lee, D.-S., & Mori, T. (2000). Molecular weight distributions of alactalbumin polymers formed by mammalian and microbial transglutaminases. Food Hydrocolloids, 14(1), 49e59. McKenzie, H. A., Ralston, G. B., & Shaw, D. C. (1972). Location of sulfhydryl and disulfide groups in bovine b-lactoglobulins and effects of urea. Biochemistry, 11(24), 4539e4547. Moro, A., Báez, G. D., Busti, P. A., Ballerini, G. A., & Delorenzi, N. J. (2010). Effects of heat-treated b-lactoglobulin and its aggregates on foaming properties. Food Hydrocolloids, 25(5), 1009e1015. Moro, A., Gatti, C. A., & Delorenzi, N. J. (2001). Hydrophobicity of whey protein concentrates measured by fluorescence quenching and its relation with surface

G.D. Báez et al. / Food Hydrocolloids 25 (2011) 1758e1765 functional properties. Journal of Agricultural and Food Chemistry, 49(10), 4784e4789. Motoki, M., & Seguro, K. (1998). Transglutaminase and its use for food processing. Trends in Food Science & Technology, 9(5), 204e210. Pace, N. C. (1986). Determination and analysis of urea guanidine hydrochloride denaturation curves. Methods in Enzymology, 131, 266e280. Pace, N. C. (1990). Conformational stability of globular proteins. Trends in Biochemical Sciences, 15(1), 14e17. Palazolo, G., Rodriguez, F., Farruggia, B., Picó, G., & Delorenzi, N. J. (2000). Heat treatment of b-lactoglobulin: structural changes studied by partitioning and fluorescence. Journal of Agricultural and Food Chemistry, 48(9), 3817e3822. Papiz, M. Z., Sawyer, L., Eliopoulos, E. E., North, A. C., Findlay, J. B., Sivaprasadarao, R., et al. (1986). The structure of beta-lactoglobulin and its similarity to plasma retinol-binding protein. Nature, 324(6095), 383e385. Rodriguez-Nogales, J. M. (2006). Effect of preheat treatment on the transglutaminase-catalyzed cross-linking of goat milk proteins. Process Biochemistry, 41(2), 430e437. Ross-Murphy, S. B. (1994). Rheological methods. In S. B. Ross-Murphy (Ed.), Physical techniques for the study of biopolymers (pp. 343e392). Glasgow: Blackie Academic and Professional. Rossky, P. J. (2008). Protein denaturation by urea. Proceedings of the National Academy of Sciences of the United States of America, 105(44), 16825e16826.

1765

Shirley, B. A. (1995). Urea and guanidine hydrochloride denaturation curves. In B. A. Shirley (Ed.), Protein stability and folding, theory and practice (pp. 177e190). New Jersey: Humana Press Inc. Tanford, C. (1961). Transport processes, viscosity. In C. Tanford (Ed.), Physical chemistry of macromolecules (pp. 318e338). New York: John Wiley & Sons, Inc. Tanford, C. (1968). Protein denaturation. Advances in Protein Chemistry, 23, 121e282. Tanimoto, S.-Y., & Kinsella, J. E. (1988). Enzymatic modification of proteins: effects of transglutaminase cross-linking on some physical properties of b-lactoglobulin. Journal of Agricultural and Food Chemistry, 36(2), 281e285. Townsend, A., & Nakai, S. (1983). Relationships between hydrophobicity and foaming characteristics of food proteins. Journal of Food Science, 48(2), 588e594. Vardhanabhuti, B., & Foegeding, E. A. (1999). Rheological properties and characterization of polymerized whey protein isolates. Journal of Agricultural and Food Chemistry, 47(9), 3649e3655. Walsh, D. J., Cleary, D., McCarthy, E., Murphy, S., & FitzGerald, R. J. (2003). Modification of the nitrogen solubility properties of soy protein isolate following proteolysis and transglutaminase cross-linking. Food Research International, 36(7), 677e683. Wilde, P. J., & Clark, D. C. (1996). Foam formation and stability. In G. M. Hall (Ed.), Methods of testing protein functionality (pp. 110e152). London: Blackie Academic & Professional.