Effect of micellar κ-casein dissociation on the formation of soluble protein complexes and acid gel properties

Accepted Manuscript Effect of micellar κ-casein dissociation on the formation of soluble protein complexes and acid gel properties Md. Sultan Mahomud, Nakako Katsuno, Takahisa Nishizu PII:

S0023-6438(17)30756-9

DOI:

10.1016/j.lwt.2017.10.018

Reference:

YFSTL 6581

To appear in:

LWT - Food Science and Technology

Received Date: 10 June 2017 Revised Date:

9 October 2017

Accepted Date: 9 October 2017

Please cite this article as: Mahomud, M.S., Katsuno, N., Nishizu, T., Effect of micellar κ-casein dissociation on the formation of soluble protein complexes and acid gel properties, LWT - Food Science and Technology (2017), doi: 10.1016/j.lwt.2017.10.018. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT 1

Effect of micellar κ-casein dissociation on the formation of soluble protein complexes

2

and acid gel properties

3 Md. Sultan Mahomud, Nakako Katsuno and Takahisa Nishizu*

5

Department of Applied Life Science, Gifu University, Yanagido1-1, Gifu 501-1193, Japan

6

*Corresponding author, Tel./Fax: +81 58 293 2888

7

E-mail address: [email protected]

AC C

EP

TE D

M AN U

SC

RI PT

4

1

ACCEPTED MANUSCRIPT 8

ABSTRACT

9 This study investigated the effect of micellar κ-casein dissociation, caused by cross-linking

11

agent glutaraldehyde, on the formation of soluble protein complexes and the texture of

12

resulting gels. Reconstituted skim milk containing different levels of added glutaraldehyde

13

(SM-GTA) and skim milk without glutaraldehyde (SM) were processed with or without

14

heating, and the structural properties of the prepared acid gel were studied. Acid gel prepared

15

from heated SM (without GTA) had significantly higher firmness (1.40 ± 0.02 N) and water

16

holding capacity (79.0 ± 3.0 %) compared to that made from heated SM treated with 0.1, 0.3,

17

0.5 mmol/L GTA, respectively. Higher storage modulus (G′) and denser microstructure were

18

observed in acid gel prepared from heated SM than those made from heated SM-GTA.

19

Electrophoretic analysis demonstrated that κ-casein levels in the soluble protein complexes of

20

SM was 9.60 ± 0.2 (%) which was decreased to 0.06 ± 0.01 (%) in SM treated with 0.5

21

mmol/L GTA. This was attributed to GTA decreasing micellar κ-casein dissociation and,

22

therefore, reducing the formation of soluble protein complexes. This reduction in soluble

23

protein complexes in SM-GTA resulted in weaker acid gels compared with those prepared

24

from SM without GTA.

AC C

EP

TE D

M AN U

SC

RI PT

10

25 26

Keywords: Cross-linking, dissociation, protein complexes, particle size, rheology, gel texture. 2

ACCEPTED MANUSCRIPT 27 Chemical compounds studied in this article:

29

Acridine orange (PubChem CID: 62344); β-mercaptoethanol (PubChem CID: 1567); calcium

30

chloride (PubChem CID: 24854); glucono delta-lactone (PubChem CID: 736); glutaraldehyde

31

(PubChem CID: 3485); imidazole (PubChem CID: 795); sodium chloride (PubChem CID:

32

5234).

SC

RI PT

28

M AN U

33

AC C

EP

TE D

34

3

ACCEPTED MANUSCRIPT 35

1. INTRODUCTION

36 The texture characteristics of acid milk gels are among the most important attributes for

38

determining their sensory evaluation and consumer acceptability. Furthermore, the most

39

common defects in acid milk gel are reduced firmness, high syneresis, and low viscosity (Xu,

40

He, Ma, Zhang, & Wang, 2015). Firmness is increased and syneresis notably decreased when

41

acid milk gels are prepared from heated milk. Heat treating milk results in the denaturation of

42

whey proteins, which allows them to interact with κ-casein via thiol–disulfide exchange

43

reactions to form whey protein/κ-casein aggregates (Vasbinder, Alting, Visschers, & de Kruif,

44

2003; Ozcan, Horne, & Lucey, 2015; Mahomud, Katsuno, Zhang, & Nishizu, 2017a). These

45

aggregates increase the firmness of acid gels. Moreover, our previous study (Mahomud,

46

Katsuno, & Nishizu, 2017b) demonstrated that the formation of soluble protein complexes as

47

new significant complexes in the heated milk and whey protein-enriched milk might increase

48

network connectivity and promote the number and strength of contact points, resulting in

49

firmer gels.

AC C

EP

TE D

M AN U

SC

RI PT

37

50

Milk is a colloidal suspension containing casein micelles, whey proteins, lipids, lactose,

51

and salts (White & Davies, 1958). Casein micelles in milk account for 80% of the total milk

52

protein, and are composed of a mixture of four common caseins: αS1-casein, αS2-casein,

53

β-casein, and κ-casein (Walstra, 1990). Whey proteins account for the remaining 20% of total 4

ACCEPTED MANUSCRIPT milk proteins, and consist mainly of β-lactoglobulin (β-lg) and α-lactalbumin (Mottar, Bassier,

55

Joniau, & Baert, 1989). The application of milk to produce diversified products largely

56

depends on the integrity of the casein micellar system. The surfaces of casein micelles are

57

covered with κ-casein, which appears to form an extended hairy layer that provides steric and

58

electrostatic stabilization to the micelles (Holt & Horne, 1996). Acid coagulation of casein

59

micelles using lactic acid bacteria or acidulants, such as glucono-δ-lactone (GDL),

60

destabilizes these steric conformations due to a reduced surface charge, resulting in

61

coagulation and the formation of acid curd at isoelectronic pH values of casein micelles near

62

4.6 (Lucey, Tamehana, Singh, & Munro, 1998).

M AN U

SC

RI PT

54

In addition to conformation, the interaction of the casein micellar system with whey

64

proteins also influences the application of milk in various products. β-lactoglobulin is a major

65

whey protein that contains two disulfide bridges and one free cysteine (Lucey et al., 1998).

66

Heating milk above 70 °C initiates thiol–disulfide exchange reactions to form disulfide bonds

67

between free cysteine groups of β-lactoglobulin and κ-casein, α-lactalbumin, or another

68

β-lactoglobulin, producing whey protein/κ-casein aggregates (Morand, Guyomarc’h, &

69

Famelart, 2011; Rodriguez & Dalgleish, 2006; Li, Dalgleish, & Corredig, 2015). These

70

aggregates are located partly in the serum phase as soluble protein complexes or on the

71

surface of casein micelles as micelle-bound complexes. However, the formation of soluble

72

protein complexes depends on the dissociation of κ-casein and its subsequent interaction with

AC C

EP

TE D

63

5

ACCEPTED MANUSCRIPT denatured β-lactoglobulin, which can be increased continuously by increasing the temperature

74

(Anema, Lowe, & Lee, 2004a; Anema, 2008). Another approach to altering the casein

75

micellar structure is adding a cross-linking agent, such as glutaraldehyde (GTA) or genipin, to

76

the milk, which severely inhibits micellar κ-casein dissociation and the soluble protein

77

complex formation. Although some studies (Silva, Sousa, Gübitz, & Cavaco-Paulo, 2004;

78

Casanova, Silva, Gaucheron, Nogueira, Teixeira, Perrone,… & Carvalho, 2017) have reported

79

the GTA-induced fixation of micellar κ-casein and a subsequent decrease in soluble protein

80

complex formation, the effect of these changes on the textural, rheological, and

81

microstructural attributes of the acid gel have not been well documented.

M AN U

SC

RI PT

73

Therefore, this study aimed to investigate the effect of micellar κ-casein dissociation by a

83

cross-linking agent (GTA) on the formation of soluble protein complexes and properties of

84

acid gels. Adding a small amount of GTA to skim milk (SM) can diminish micellar κ-casein

85

dissociation and decrease soluble protein complex formation. Furthermore, we attempted to

86

study the effect of increasing GTA concentration on micellar κ-casein dissociation and soluble

87

protein complex formation in heated and unheated SM using electrophoretic methods.

89

EP

AC C

88

TE D

82

2. MATERIALS AND METHODS

90 91

2.1. Preparation of milk samples 6

ACCEPTED MANUSCRIPT 92 Skim milk (SM) was prepared by adding skim milk powder (12 g; Difco Skim milk, Wako,

94

Osaka, Japan) to deionized water (100 g), affording a milk solution containing 100.0 g/kg

95

total solids, as described previously by Lucey et al. (1998). The SM pH was around 6.7

96

(unadjusted) at room temperature. The milk samples were stirred for at least 3 h at room

97

temperature and kept overnight at 4 °C to obtain complete hydration before further use.

SC

RI PT

93

99

M AN U

98 2.2. Addition of cross-linking agent to milk samples

100

The cross-linking agent, GTA (250 g/L in water, Nacalai Tesque, Kyoto, Japan), was added

102

to unheated SM at concentrations of 0.1, 0.3, and 0.5 mmol/L, respectively. The required

103

concentration was estimated as 0.4 mmol/L from the cross-linking efficiency of GTA

104

(Rodriguez & Dalgleish, 2006). Milk samples were shaken on a vortex mixer for 10 s and

105

then left for 1 h at room temperature to react with GTA. Milk samples with added GTA are

106

referred to as SM-GTA.

108

EP

AC C

107

TE D

101

2.3. Heat treatment of milk samples

109 110

The milk samples were subdivided into two heat treatment groups, with one subjected to 7

ACCEPTED MANUSCRIPT 111

heat treatment at 85 °C for 30 min in a thermostatically controlled water bath, followed by

112

rapid cooling in an ice bath, and the other part left untreated.

113 2.4. Centrifugation of milk samples

RI PT

114 115

To obtain serum protein (supernatant) from colloidal protein (pellet), heated and unheated

117

milk samples (1 mL) were placed in 1.5-mL plastic tubes for centrifugation at 20,100 ×g and

118

25 °C for 60 min (Hitachi Koki Centrifuge, Tokyo, Japan) as described previously (Nguyen,

119

Anema, Guyomarc, & Wong, 2015). After centrifugation, the clear supernatants were

120

carefully collected from the pellet and the protein compositions of the supernatants were

121

determined by electrophoresis.

124

M AN U

TE D

2.5. Polyacrylamide gel electrophoresis

EP

123

AC C

122

SC

116

125

Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was performed

126

using a Bio-Rad mini-gel slab electrophoresis system (Bio-Rad Laboratories, Richmond, CA,

127

USA). The supernatants of heated and unheated milk samples were diluted 1:20 with SDS

128

sample buffer. The protein profiles of the supernatants were measured after reduction by

129

adding β-mercaptoethanol (5 mL/L) and under non-reducing conditions. Gel preparation, 8

ACCEPTED MANUSCRIPT running, staining with Coomassie Brilliant Blue, and destaining were carried out as described

131

previously (Chevalier, Hirtz, Sommerer, & Kelly, 2009). Destained gels were scanned using

132

an image scanner (Brother MFC-9970CDW, Brother Co., Nagoya, Japan). The integrated

133

intensities and the percentage of the protein fractions were determined using ImageJ software

134

(Rasband, W. S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA)

135 2.6. Measurement of casein micelle size

M AN U

136

SC

RI PT

130

137

The sizes of the casein micelles in heated and unheated milk samples were analyzed by

139

dynamic light scattering (DLS) at 20 °C using a Malvern Zetasizer Nano-ZS (Malvern

140

Instruments Ltd., Malvern, UK) and following the method described by Meletharayil, Patel,

141

and Huppertz (2015). In brief, milk samples were diluted 50-fold with calcium–imidazole

142

buffer (5 mmol/L CaCl2, 20 mmol/L imidazole, 30 mmol/L NaCl, pH 7.0) and allowed to

143

stand for 10 min prior to measurement.

145

EP

AC C

144

TE D

138

2.7. Preparation of acid gels

146 147

In this study, acid milk gels were prepared from unheated SM/SM-GTA and heated SM/

148

SM-GTA. The milk samples were acidified with GDL (20.0 g/kg) at 30 °C until the milk pH 9

ACCEPTED MANUSCRIPT 149

reached 4.6, as described by Nguyen et al. (2015). After acidification, the acid gels were

150

cooled in ice water and stored at 4 °C until further analysis.

151 2.8. Water holding capacity of acid gels

RI PT

152 153

Water holding capacity (WHC) was measured using the procedure reported by Paramban

155

Rahila, Kumar, Mann, and Koli (2016) with modifications. Acid gel samples (AG, about 25

156

g) were centrifuged (KN-70, Kubota, Tokyo, Japan) at 1250 ×g for 25 min, and the whey

157

expelled (WE) was carefully separated and weighed. Measurements were carried out in

158

triplicate and the WHC was calculated as follows:

M AN U

SC

154

159

EP

161

2.9. Firmness of acid gels

AC C

160

AG − WE × 100 AG

TE D

WHC % =

162

The firmness of the acid gels was determined by a texture analyzer (Sun Rheometer

163

CR-200D, Sun Scientific Co., Tokyo, Japan) using a single-compression cycle test with a 2-N

164

load cell, according to a previous study (Sah, Vasiljevic, McKechnie, & Donkor, 2016) with

165

slight modifications. About 80 g acid gels were taken into 100-mL glass bottles keeping 50

166

mm sample height and placed under a 30-mm cylindrical probe. The probe penetrated 10

ACCEPTED MANUSCRIPT 167

vertically into the acid gels to a depth of 20 mm at 100 mm/min. The firmness (maximum

168

force of compression) was recorded from the force–time graph and expressed in Newtons (N).

169 2.10. Rheological measurement of acid gels

RI PT

170 171

The rheological properties of the acid gels during and after acidification were measured

173

using a controlled stress rheometer (AR2000, G2KG, TA Instruments, Newcastle, DE, USA)

174

with a cup-and-bob geometry consisting of two coaxial cylinders (diameters, 37.03 and 35.01

175

mm). After the addition of 20 g/kg GDL, each milk sample was stirred for 30 s and then a

176

15-mL aliquot was transferred to the rheometer cup. Gelation was monitored while the sample

177

was oscillated at 1.0 Hz with an applied strain 0.1. An applied strain of 0.1 was used, which

178

was in the linear viscoelastic range of acid gel. The storage modulus (G′) of the acid gel was

179

recorded every 3 min for 180 min at 30 °C during acidification. After acidification, the gel

180

was cooled from 30 to 5 °C at a rate of 1 °C min–1. A frequency sweep test was then carried

181

out by subjecting the same samples to a frequency ramp of 0.1–10 Hz (in log progression with

182

6 points per decade) at constant strain amplitude of 0.1. This procedure was slightly modified

183

from Chever, Guyomarc’h, Beaucher, and Famelart (2014).

AC C

EP

TE D

M AN U

SC

172

184 185

2.11. Confocal laser scanning microscopy 11

ACCEPTED MANUSCRIPT 186 The microstructure of the acid gels was monitored using confocal laser scanning

188

microscopy (CLSM; LSM710, Carl Zeiss microscopy GmbH, Jena, Germany), as described

189

previously (Ong, Dagastine, Kentish, & Gras, 2011). Acridine orange (Sigma Chemical Co.,

190

St Louis, MO, USA) was added to the selected milk samples (300 µL of acridine orange

191

solution (5 g/kg) per 100 g of milk). The milk sample with added dye and GDL was deposited

192

into the concave region of a glass petri dish (diameter, 35 mm) and covered with a cover slip

193

(Iwaki, Asahi Glass Co., Chiba, Japan), ensuring no air was trapped. The glass petri dish with

194

samples was incubated at 30 °C until the pH of the sample reached 4.6. After storing for 1 day

195

at 4 °C, the microstructure of the acid gels was analyzed using CLSM with excitation at 488

196

nm. Micrographs were captured using a 63 × objective with oil immersion (numerical

197

aperture = 1.4) and the emission band was collected at 515–530 nm.

200

SC

M AN U

TE D

EP

199

2.12. Statistical Analysis

AC C

198

RI PT

187

201

All experiments were performed three times with at least three replicates for each milk

202

sample. Any significant difference between the sample means (from three replicates) was

203

determined using Tukey’s honest significant difference test at a significance level of P < 0.05

204

using KaleidaGraph (Version 4.1.1, Synergy Software, Reading, PA, USA). 12

ACCEPTED MANUSCRIPT 205 206

3. RESULTS AND DISCUSSION

207 3.1. Effect of adding glutaraldehyde to skim milk on protein interactions and

209

distribution

RI PT

208

SC

210

The supernatants obtained from the heated and unheated milk samples with or without GTA

212

were analyzed using SDS-PAGE to investigate the effect of GTA on the dissociation of

213

micellar κ-casein from casein micelles and the formation of soluble protein complexes.

214

Reducing and non-reducing SDS-PAGE were used to observe the covalent aggregation of

215

denatured whey proteins with micellar κ-casein (Figs. 1 A & B). In non-reducing SDS-PAGE

216

analysis (Figure 1A), the band intensity for β-lactoglobulin was higher in unheated milk

217

samples (Fig. 1 A, lanes 1, 3, 5, and 7) than in heated samples (Fig. 1 A, lanes 2, 4, 6, and 8)

218

indicating heat-induced aggregation of β-lactoglobulin in the serum phase of heated samples.

219

In contrast, the κ-casein band in heated SM and heated SM+0.1 mmol/L GTA samples under

220

non-reducing conditions was almost disappeared, perhaps due to the heat-mediated

221

dissociation of micellar κ-casein and subsequent formation of covalent aggregates with

222

denatured β-lactoglobulin in the heated samples, which were treated as soluble protein

223

complexes (Singh & Creamer, 1991).

AC C

EP

TE D

M AN U

211

13

ACCEPTED MANUSCRIPT In reducing SDS-PAGE analysis, as shown in Figure 1 B, both heated and unheated

225

samples showed similar band patterns for β-lactoglobulin in heated milk samples (Fig. 1 B,

226

lanes 2, 4, 6, and 8) due to the cleavage of heat-induced aggregates by β-mercaptoethanol. In

227

contrast, κ-casein had a higher band intensity in heated SM (Fig. 1 B, lane 2) than in heated

228

SM-GTA (Fig. 1 B, lane 4, 6, 8), which was positively correlated with the GTA concentration

229

added. As the added GTA concentration increased, the level of κ-casein dissociation in serum

230

phases decreased, resulting in less κ-casein available to participate in heat-induced

231

aggregation with β-lactoglobulin. However, in unheated samples, GTA had no significant

232

influence on the formation of soluble protein complexes due to the lack of heat-mediated

233

β-lactoglobulin denaturation. Similarly, micellar κ-casein did not dissociated from casein

234

micelles to the serum phase in unheated SM-GTA, which reduced the formation of soluble

235

protein complexes.

TE D

M AN U

SC

RI PT

224

Therefore, the level of heat-induced aggregation in the serum phase between

237

β-lactoglobulin and κ-casein was higher in SM compared with SM-GTA, in agreement with

238

the findings of Rodriguez and Dalgleish (2006). Supernatants obtained from the SM-GTA

239

dispersion appeared to show less disulfide-linked aggregates of β-lactoglobulin and κ-casein

240

in the serum phase due to less κ-casein dissociation (Table 1). Differences in the levels of

241

aggregation between denatured β-lactoglobulin and κ-casein, either in casein micelles or in

242

the serum phase, might be influenced by the dissociation of micellar κ-casein from the casein

AC C

EP

236

14

ACCEPTED MANUSCRIPT micelles (Anema, Lee, Lowe, & Klostermeyer, 2004b). In this study, there was a lower level

244

of soluble κ-casein in SM-GTA. This may hinder the formation of soluble protein complexes

245

in SM-GTA. Therefore, the addition of GTA to SM had an antagonistic effect on κ-casein

246

dissociation and its interaction with β-lactoglobulin.

247

3.2. Effect of adding glutaraldehyde to skim milk on casein micelle particle size

SC

248

RI PT

243

M AN U

249

Dynamic light scattering was used to investigate the size of native and cross-linked

251

casein micelles as a function of the GTA concentration added to SM. The casein micelle

252

particle sizes in milk samples before and after heat treatment are presented in Table 1. In both

253

heated and unheated samples, the particle sizes in SM+0.5 mmol/L GTA samples were lower

254

than those in SM samples. As the concentration of GTA added was increased, the size of the

255

casein micelles decreased significantly (P < 0.05), in agreement with Nogueira Silva,

256

Saint-Jalmes, De Carvalho, and Gaucheron (2014), who reported that the addition of

257

cross-linking agents, such as genipin, could significantly collapse the hairy layer of micellar

258

κ-casein on the surface of casein micelles, resulting in decreased casein micelle size.

259

Furthermore, smoothing of the casein micelle surface as a function of genipin addition has

260

been reported using scanning electron microscopy. Similar findings have been reported for

261

casein micelles reported by Casanova et al. (2017). On the other hand, the size of the casein

AC C

EP

TE D

250

15

ACCEPTED MANUSCRIPT micelles was lower in the heated SM than those of unheated SM due to the dissociation of

263

micellar κ-casein and their interaction with β-lactoglobulin in the serum phase. Therefore, the

264

results of this study demonstrated that the adverse changes in the casein micelle particle size

265

in heated or unheated milk samples were strongly dependent on the cross-linking reagent

266

concentration. In serum phase, the particle size obtained from both heated and unheated SM

267

was significantly higher (P < 0.05) than those of the SM treated with 0.1, 0.3, and 0.5 mmol/L

268

GTA (Table 1), indicating less formation of soluble protein complexes due to the reduction of

269

micellar κ-casein dissociation.

270

M AN U

SC

RI PT

262

3.3. Water holding capacity of acid gels prepared from skim milk with/without added

272

glutaraldehyde

273

TE D

271

The water holding capacities (WHCs) of acid gels made from SM and SM-GTA are shown

275

in Table 2. The WHCs of acid gels prepared from heated SM were higher than that prepared

276

from unheated SM. A significant increase (P < 0.05) in WHC was observed for acid gels

277

prepared from heated SM compared with those prepared from heated SM-GTA. For heated

278

milk samples, the WHC of the acid gels decreased with increasing GTA concentration. In

279

contrast, the WHC of acid gels did not change in either unheated SM or unheated SM-GTA,

280

showing that GTA concentration had little or no direct effect on the WHC.

AC C

EP

274

16

ACCEPTED MANUSCRIPT The increase in the WHC of acid gels made from heated milk samples indicated that it was

282

influenced by heat treatment. This was due to the denaturation of whey proteins and their

283

interaction with micellar κ-casein. According to Lee and Lucey (2003), the denaturation of

284

whey proteins and formation of soluble protein complexes creates a homogeneous porous

285

structure into which free water is immobilized and entrapped. Therefore, acid gels can hold

286

huge amounts of water in their protein network, resulting in an increased WHC. However, SM

287

treated with GTA and heat had an antagonistic effect on the WHC of acid gels. The addition

288

of GTA to SM has been suggested to prevent the dissociation of micellar κ-casein (Migneault,

289

Dartiguenave, Bertrand, & Waldron, 2004), which would reduce the formation of soluble

290

protein complexes. Consequently, the WHC of acid gels decreased with a decreasing amount

291

of soluble protein complexes. Results from other reports (Xu et al., 2015; Schorsch, Wilkins,

292

Jonest, & Norton, 2001) have shown that the amount of soluble protein complexes played a

293

predominant role in improving the WHC of acid gels.

295 296

SC

M AN U

TE D

EP

AC C

294

RI PT

281

3.4. Firmness of acid gels prepared from skim milk with/without added glutaraldehyde

297

The effect of adding GTA to SM on the firmness of set acid gels is shown in Table 2. The

298

acid gel made from heated SM showed significantly higher firmness (P < 0.05) compared

299

with those made from heated SM-GTA. Adding GTA to heated SM afforded significantly 17

ACCEPTED MANUSCRIPT lower firmness in the resultant set acid gels compared to heated SM without GTA. For heated

301

milk samples, firmness of the acid gels decreased with increasing GTA concentration. In

302

contrast, the firmness of acid gels made from unheated SM-GTA was low due to an

303

incompatibility between undenatured whey proteins and κ-casein.

RI PT

300

The increase in the firmness of acid gels made from heated SM might be due to interactions

305

between denatured whey proteins and κ-caseins in the casein micelle to form micelle-bound

306

complexes, and between denatured whey proteins with κ-caseins in the serum phase to form

307

soluble protein complexes. Our previous study (Mahomud et al., 2017b) showed that soluble

308

protein complexes were the predominant factor in improving the firmness of yoghurt gel

309

rather than micelle-bound complexes. This result agreed with that of Donato, Alexander, and

310

Dalgleish (2007) and Guyomarc’h, Jemin, Le Tilly, Madec, and Famelart (2009) who reported

311

that soluble protein complexes might be important for increasing the stiffness of acid gels.

312

The amount of soluble protein complexes decreased with increasing GTA concentration in

313

SM because GTA prevented the dissociation of micellar κ-caseins from casein micelles.

M AN U

TE D

EP

AC C

314

SC

304

315

3.5. Rheological properties of acid gels prepared from skim milk with/without added

316

glutaraldehyde

317 318

Changes in the storage modulus (G′) of SM and SM-GTA samples as the pH decreased 18

ACCEPTED MANUSCRIPT from 6.7 to 4.6 with GDL treatment are presented in Fig. 2. The effect of adding a low GTA

320

concentration to SM on the rheological properties of the acid gels was monitored to determine

321

G′ every 3 min during acidification. Acid gels prepared from heated SM without GTA showed

322

higher final G′ values than those made from heated SM with GTA added (Fig. 2A). In

323

heat-treated samples, the G′ value was highly influenced by the amount of GTA added,

324

decreasing with increasing levels of GTA. In contrast, the G′ profiles obtained for unheated

325

SM were similar to those obtained from unheated SM-GTA samples, with GTA concentration

326

showing little or no effect on the gel properties. The value of tan δ was higher in acid gel

327

formed with unheated SM and unheated SM-GTA than those formed with heated SM and

328

heated SM-GTA (Fig. 2B). In the frequency sweep test, the log G′ of acid gels made from

329

heated SM with and/or without GTA increased linearly as log frequency increased (Fig. 2C).

330

Furthermore, the G′ value of acid gel made from heated SM was higher than for the gel made

331

from heated SM-GTA. On the other hand, the G′ value of acid gel made from unheated SM

332

was almost similar to those made from unheated SM-GTA.

AC C

EP

TE D

M AN U

SC

RI PT

319

333

Heating milk at a higher temperature accelerates the denaturation of β-lactoglubolin, which

334

forms complexes with κ-casein either in the serum phase or casein micelles (Singh & Creamer,

335

1991; Krzeminski, Großhable, & Hinrichs, 2011). The active participation of denatured whey

336

proteins in the gel network through the heat-induced aggregation of β-lactoglubolin and

337

κ-casein increases the number and strength of bonds between protein complexes, resulting in 19

ACCEPTED MANUSCRIPT an increase in acid gel storage modulus (Lucey et al., 1998; Damin, Alcântara, Nunes, &

339

Oliveira, 2009; Ozcan et al., 2015). Furthermore, acid gels prepared from heated SM without

340

GTA were found to have higher G′ values compared with those prepared from heated

341

SM-GTA. This phenomenon has been attributed to the formation of soluble protein complexes

342

by the interaction of β-lactoglubolin with micellar κ-casein in the serum phase of heated milk

343

(Anema, 2007). The dissociation of micellar κ-casein from casein micelles depends on GTA

344

addition to SM, which regulates the distribution of heat-induced complexes between the

345

serum and micellar phases. Similarly, a decrease in the dissociation of micellar κ-casein from

346

the casein micelles with increasing GTA concentration has been attributed to hinder in the

347

formation of heat-induced aggregates in the serum phases of heated SM-GTA.

M AN U

SC

RI PT

338

In the current study, acid gels were prepared from heated SM-GTA typically had lower G′

349

values due to fewer soluble protein complexes, which resulted in less covalent bonding in the

350

protein networks. Our previous work (Mahomud et al., 2017b) showed that an increase in

351

soluble protein complexes leads to stronger acid gels and an increase in G′ during

352

acidification. Acid gels made from heated SM-GTA had lower G′ values, in agreement with

353

Xu et al. (2015) and Vasbinder et al. (2003), who reported that a decrease in soluble protein

354

complexes in heated SM led to the formation of a weaker gel (lower G′). Other studies

355

(Guyomarc’h et al., 2009) reported that the increased levels of soluble protein complexes in

356

heated SM increased the G′ values of the acid gels. Therefore, it can be concluded that the

AC C

EP

TE D

348

20

ACCEPTED MANUSCRIPT dissociation of micellar κ-casein stimulates soluble protein complex formation, which is

358

responsible for increasing the G′ value of acid gels. Similarly, a decrease in soluble protein

359

complexes in SM-GTA reduced the number and strength of contact points, resulting in weaker

360

gels.

RI PT

357

361

3.6. Microstructure of acid gels prepared from skim milk with/without added

363

glutaraldehyde

M AN U

SC

362

364

The effect of adding cross-linking reagents to SM on the microstructure of acid gels was

366

analyzed using CLSM, as shown in Fig. 3. The microstructure of acid gels was captured to

367

observe the protein network, with a remarkable difference observed between the

368

microstructures of acid gels made from unheated and heated SM with different GTA

369

concentrations. Acid gels prepared from unheated SM with or without added GTA appeared to

370

contain a coarse network with little interconnectivity and more open networks with more

371

porous space (Fig. 3 A–D). In contrast, acid gels prepared from heated SM with or without

372

added GTA had more branched and compact protein networks with less porous space (Fig. 3

373

E–H). However, a more compact microstructure was observed in acid gels made from heated

374

SM without GTA compared with those made from heated SM with added GTA. A possible

375

reason for increasing network compactness is the presence of more aggregated protein

AC C

EP

TE D

365

21

ACCEPTED MANUSCRIPT 376

complexes of denatured β-lactoglubolin and κ-caseins in the serum phase (Schorsch et al.,

377

2001). Furthermore, the association of denatured β-lactoglubolin with either casein micelles or

379

serum κ-casein increased the number of strands by forming disulfide bonds, which enhanced

380

compactness in the protein network of acid gels made from heated SM (Lowe et al., 2004;

381

Jørgensen, Abrahamsen, Rukke, Johansen, & Skeie, 2016). In heated SM without GTA, a

382

large amount of denatured β-lactoglubolin was aggregated with dissociated κ-casein to form

383

soluble protein complexes, while β-lactoglubolin was aggregated with micellar κ-casein to

384

form micelle-bound complexes (Lucey et al., 1998). Consequently, the total number of bonds

385

and amount of protein involved in the protein networks would be higher in acid gels made

386

from heated SM than those made from SM-GTA. Nonetheless, the reduced dissociation of

387

micellar κ-casein from casein micelles under the action of GTA, resulting in the formation of

388

fewer soluble protein complexes, ultimately created a coarse and porous microstructure in the

389

acid gels.

391

SC

M AN U

TE D

EP

AC C

390

RI PT

378

4. CONCLUSIONS

392 393

The addition of different GTA concentrations to SM, followed by heat treatment, influenced

394

the gelation properties of the resulting acid gels. The dissociation of micellar κ-casein from 22

ACCEPTED MANUSCRIPT casein micelles after heat treatment markedly decreased with increasing GTA concentration in

396

SM. Therefore, the formation of soluble protein complexes, caused by the dissociation of

397

micellar κ-casein, was limited by GTA addition to SM. The gelation properties of the resultant

398

acid gels were influenced by the addition of small GTA concentrations to SM. In this study,

399

acid gels prepared from heated SM showed a higher WHC, higher firmness, and denser

400

microstructure than those prepared from heated SM-GTA. A decrease pattern of yogurt texture

401

made from heated SM-GTA was associated with the reduction of κ-casein dissociation

402

resulting lowered the formation of soluble protein complexes. Heating SM contributed to the

403

creation of complex gel networks with numerous aggregated particles, such as soluble protein

404

complexes and micelle-bound complexes, which increased the number and strength of contact

405

points in the protein network. The results suggested that the formation of soluble protein

406

complex was important for improving gelation properties and preparing a firmer acid gel.

407

Further studies on the destabilization of casein micelles and storage behavior of acid gels are

408

required to evaluate the effect of soluble protein complexes on textural properties of acid gels

409

as a function of cross-linking reagent concentration in SM.

411

SC

M AN U

TE D

EP

AC C

410

RI PT

395

ACKNOWLEDGEMENTS

412 413

The authors wish to thank to the Japanese Government for granting funding to the PhD 23

ACCEPTED MANUSCRIPT scholar through the MEXT scholarship. We thank Prof. Oumi Yasunori at the Life Science

415

Research Center in Gifu University for his support in using the rheometer and Zetasizer. We

416

also acknowledge Assistant Professor Shigeo Takashima at the Life Science Research Center

417

in Gifu University for his assistance with the confocal microscope. This research did not

418

receive any specific grants from funding agencies in the public, commercial, or not-for-profit

419

sectors.

SC

RI PT

414

M AN U

420 Compliance with ethical standards

422

Conflict of interest: The authors do not have any conflicts of interest.

423

Compliance with ethics requirements: This article does not contain any studies with human

424

or animal subjects.

427

EP

426

REFERENCES

AC C

425

TE D

421

428

Anema, S. G., Lowe, E. K. & Lee, S. K. (2004a). Effect of pH at heating on the acid-induced

429

aggregation of casein micelles in reconstituted skim milk. LWT - Food Science and

430

Technology. 37(7), 779–787.

431

Anema, S. G., Lee, S. K., Lowe, E. K., & Klostermeyer, H. (2004b). Rheological properties of

432

acid gels prepared from heated pH-adjusted skim milk. Journal of Agricultural and Food 24

ACCEPTED MANUSCRIPT 433

Chemistry. 52, 337–343. Anema, S. G. (2007). Role of κ-casein in the association of denatured whey proteins with

435

casein micelles in heated reconstituted skim milk. Journal of Agricultural and Food

436

Chemistry. 55(9), 3635–3642.

RI PT

434

Anema, S. G. (2008). On heating milk, the dissociation of kappa-casein from the casein

438

micelles can precede interactions with the denatured whey proteins. Journal of Dairy

439

Research, 75(4), 415–421.

M AN U

SC

437

Casanova, F., Silva, N. F. N., Gaucheron, F., Nogueira, M. H., Teixeira, A. V. N. C., Perrone,

441

I. T., Alves, M. P., Fidelis, P. C., & Carvalho, A. F. D. (2017) Stability of casein

442

micelles crosslinked with genipin: a physicochemical study as a function of pH,

443

International Dairy Journal. doi: 10.1016/j.idairyj.2016.12.006.

TE D

440

Chevalier, F., Hirtz, C., Sommerer, N., & Kelly, A. L. (2009). Use of reducing/nonreducing

445

two-dimensional electrophoresis for the study of disulfide-mediated interactions between

446

proteins in raw and heated bovine milk. Journal of Agricultural and Food Chemistry,

447

57(13), 5948–5955.

AC C

EP

444

448

Chever, S., Guyomarc’h, F., Beaucher, E., & Famelart, M. H. (2014). High-protein fat-free

449

acid milk gels: control of protein composition and heat treatment. International Dairy

450

Journal. 37(2), 95–103.

25

ACCEPTED MANUSCRIPT Damin, M. R., Alcântara, M. R., Nunes, A. P., & Oliveira, M. N. (2009). Effects of milk

452

supplementation with skim milk powder, whey protein concentrate and sodium caseinate

453

on acidification kinetics, rheological properties and structure of nonfat stirred yogurt.

454

LWT - Food Science and Technology. 42(10), 1744–1750.

RI PT

451

Donato, L., Alexander, M., & Dalgleish, D. G. (2007). Acid gelation in heated and unheated

456

milks: interactions between serum protein complexes and the surfaces of casein micelles.

457

Journal of Agricultural and Food Chemistry. 55, 4160–4168.

M AN U

SC

455

Guyomarc’h, F., Jemin, M., Le Tilly, V., Madec, M. N., & Famelart, M. H. (2009). Role of the

459

heat-induced whey protein/κ-casein complexes in the formation of acid milk gels: a

460

kinetic study using rheology and confocal microscopy. Journal of Agricultural and Food

461

Chemistry. 57, 5910–5917.

TE D

458

Holt, C., & Horne, D. S. (1996). The hairy casein micelle: Evolution of the concept and its

463

implications for dairy technology. Netherlands Milk and Dairy Journal. 50, 85-111.

464

Jørgensen, C. E., Abrahamsen, R. K., Rukke, E. O., Johansen, A. G., & Skeie, S. B. (2016).

465

Fractionation by microfiltration: effect of casein micelle size on composition and

466

rheology of high protein, low fat set yoghurt. International Dairy Journal, 1–9.

AC C

EP

462

467

Krzeminski, A., Großhable, K., & Hinrichs, J. (2011). Structural properties of stirred yoghurt

468

as influenced by whey proteins. LWT - Food Science and Technology. 44(10), 2134–

26

ACCEPTED MANUSCRIPT 469

2140.

Lee, W. J., & Lucey, J. A. (2003). Rheological properties, whey separation, and

471

microstructure in set-style yogurt: effects of heating temperature and incubation

472

temperature. Journal of Texture Studies. 34(5-6), 515-536.

RI PT

470

Li, Y., Dalgleish, D., & Corredig, M. (2015). Influence of heating treatment and membrane

474

concentration on the formation of soluble aggregates. Food Research International, 76,

475

309-316.

M AN U

SC

473

Lowe, E. K., Anema, S. K., Bienvenue, A., Boland, M. J., Creamer, L. K., & Jiménez-Flores,

477

R. (2004). Heat-induced redistribution of disulfide bonds in milk proteins. 2. Disulfide

478

bonding patterns between bovine β-lactoglobulin and κ-casein. Journal of Agricultural

479

and Food Chemistry. 52, 7669–7680.

TE D

476

Lucey, J. A., Tamehana, M., Singh, H., & Munro, P. A. (1998). Effect of interactions between

481

denatured whey proteins and casein micelles on the formation and rheological properties

482

of acid skim milk gels. Journal of Dairy Research. 65, 555–567.

AC C

EP

480

483

Mahomud, M. S., Katsuno, N., Zhang, L., & Nishizu, T. (2017a). Physical, rheological and

484

microstructural properties of whey protein enriched yoghurt influenced by the heating

485

milk at different pH values. Journal of Food Processing and Preservation. doi:

486

10.1111/jfpp.13236. 27

ACCEPTED MANUSCRIPT Mahomud, M. S., Katsuno, N., & Nishizu, T. (2017b). Formation of soluble protein

488

complexes and yoghurt properties influenced by the addition of whey protein concentrate.

489

Innovative Food Science and Emerging Technologies. doi: 10.1016/j.ifset.2017.05.010

RI PT

487

Meletharayil, G. H., Patel, H. A., & Huppertz, T. (2015). Rheological properties and

491

microstructure of high protein acid gels prepared from reconstituted milk protein

492

concentrate powders of different protein contents. International Dairy Journal, 47, 64–

493

71.

M AN U

SC

490

Migneault, I., Dartiguenave, C., Bertrand, M. J., & Waldron, K. C. (2004). Glutaraldehyde:

495

Behavior in aqueous solution, reaction with proteins, and application to enzyme

496

crosslinking. BioTechniques, 37(5), 790–802.

TE D

494

Mottar, J., Bassier, A., Joniau, M., & Baert, J. (1989). Effect of heat induced association of

498

whey proteins and casein micelles on yogurt texture. Journal of Dairy Science. 72,

499

2247–2256.

AC C

EP

497

500

Morand, M., Guyomarc’h, F., & Famelart, M. H. (2011). How to tailor heat-induced whey

501

protein/κ-casein complexes as a means to investigate the acid gelation of milk-A review.

502

Dairy Science and Technology. 91(2), 97–126.

503

Nguyen, N. H. A., Anema, S. G., Guyomarc, F., & Wong, M. (2015). The effect of the

504

addition of thiol reagents to heated milk on protein interactions and acid gelation 28

ACCEPTED MANUSCRIPT 505

properties. International Dairy Journal, 42, 42–50.

Nogueira Silva, N. F., Saint-Jalmes, A., De Carvalho, A. F. & Gaucheron, F. (2014).

507

Development of casein microgels from cross-linking of casein micelles by genipin.

508

Langmuir. 30, 10167–10175.

RI PT

506

Ong, L., Dagastine, R. R., Kentish, S. E., & Gras, S. L. (2011). Microstructure of milk gel and

510

cheese curd observed using cryo scanning electron microscopy and confocal microscopy.

511

LWT - Food Science and Technology. 44, 1291-1302.

513

M AN U

512

SC

509

Ozcan, T., Horne, D. S., & Lucey, J. A. (2015). Yogurt made from milk heated at different pH values. Journal of Dairy Science. 98, 1–10.

Paramban Rahila, M., Kumar, R., Mann, B., & Koli, P. S. (2016). Enzymatic modification of

515

milk proteins for the preparation of low fat dahi. Journal of Food Processing and

516

Preservation. doi: 10.1111/jfpp.12684.

EP

TE D

514

Rodriguez, C., & Dalgleish, D. G. (2006). Structures and some properties of soluble protein

518

complexes formed by the heating of reconstituted skim milk powder. Food Research

519

International. 39, 472–479.

AC C

517

520

Sah, B. N. P., Vasiljevic, T., McKechnie, S., & Donkor, O. N. (2016). Physicochemical,

521

textural and rheological properties of probiotic yogurt fortified with fibre-rich pineapple

522

peel powder during refrigerated storage. LWT - Food Science and Technology. 65, 978–

523

986. 29

ACCEPTED MANUSCRIPT Schorsch, C., Wilkins, D. K., Jonest, M. G., & Norton, I. T. (2001). Gelation of casein-whey

525

mixtures: effects of heating whey proteins alone or in the presence of casein micelles.

526

Journal of Dairy Research. 68(3), 471–481.

RI PT

524

Silva, C. J. S. M., Sousa, F., Gübitz, G., & Cavaco-Paulo, A. (2004). Chemical modifications

528

on proteins using glutaraldehyde. Food Technology and Biotechnology, 42(1), 51–56.

529

Singh. H., & Creamer, L. K. (1991). Aggregation and dissociation of milk protein complexes in heated reconstituted concentrated skim milk. Journal of Food Science. 56, 238–246.

M AN U

530

SC

527

Vasbinder, A. J., Alting, A. C., Visschers, R. W. & de Kruif, C. G. (2003). Texture of acid

532

milk gels: formation of disulfide cross-links during acidification. International Dairy

533

Journal. 13, 29–38.

TE D

531

White, J. C. D., & Davies, D. T. (1958). The relation between the chemical composition of

535

milk and the stability of the caseinate complex. I. General introduction, description of

536

samples, methods and chemical composition of samples. Journal of Dairy Research. 25,

537

236–255.

539

AC C

538

EP

534

Walstra, P. (1990). On the stability of casein micelles. Journal of Dairy Science. 73, 1965– 1979

540

Xu, W., He, S., Ma, Y., Zhang, Y., & Wang, R. (2015). Effect of the heat-induced whey

541

proteins/κ-casein complex on the acid gelation of yak milk. RSC Advances., 5(12), 8952–

542

8956.

543

30

ACCEPTED MANUSCRIPT 544

FIGURE CAPTIONS

545 Fig. 1. SDS-PAGE electrophoretograms of supernatants obtained from SM and SM-GTA. Gel

547

A is under non-reducing conditions and gel B is under reducing conditions. Lanes 1, 3, 5, and

548

7 are supernatants collected from unheated dispersions of SM, SM+0.1 mmol/L GTA,

549

SM+0.3 mmol/L GTA, and SM+0.5 mmol/L GTA, respectively, while lanes 2, 4, 6, and 8 are

550

supernatants collected from heated (85 °C for 30 min) dispersions of SM, SM+0.1 mmol/L

551

GTA, SM+0.3 mmol/L GTA, and SM+0.5 mmol/L GTA, respectively.

M AN U

SC

RI PT

546

552

Fig. 2. (A) Storage modulus (G′) and (B) tan δ of acid gels measured during acidification with

554

GDL at 30 °C, and (C) storage modulus (G′) as a function of frequency sweep measured after

555

acidification at 5 °C. Acid gels were made from unheated milk (dotted lines) containing SM

556

(□), SM+0.1 mmol/L GTA (◇), SM+0.3 mmol/L GTA (△), and SM+0.5 mmol/L GTA ( ),

557

respectively, and heated milk (solid lines) containing SM (■), SM+0.1 mmol/L GTA (◆),

558

SM+0.3 mmol/L GTA (▲), and SM+0.5 mmol/L GTA ( ), respectively.

EP

AC C

559

TE D

553

560

Fig. 3. Confocal laser scanning microscopy images of acid gels made from (A–D) unheated

561

dispersions of SM, SM+0.1 mmol/L GTA, SM+0.3 mmol/L GTA, and SM+0.5 mmol/L GTA,

562

respectively, and (E–H) heated dispersions of SM, SM+0.1 mmol/L GTA, SM+0.3 mmol/L 31

ACCEPTED MANUSCRIPT 563

GTA, and SM+0.5 mmol/L GTA, respectively. The protein matrix is white and the pores are

564

dark. Scale bar represents 10 µm.

565 Supplementary Fig. pH profile as a function of time for acid gels made with 20 g/kg GDL at

567

30 °C from unheated milk (dotted lines) containing SM (□), SM+0.1 mmol/L GTA (◇),

568

SM+0.3 mmol/L GTA (△), and SM+0.5 mmol/L GTA ( ), respectively, and heated milk

569

(solid lines) containing SM (■), SM+0.1 mmol/L GTA (◆), SM+0.3 mmol/L GTA (▲), and

570

SM+0.5 mmol/L GTA ( ), respectively.

AC C

EP

TE D

M AN U

SC

RI PT

566

32

ACCEPTED MANUSCRIPT Table 1 Z-average particle diameter of micellar and serum phases obtained from unheated and heated milk samples treated with 0, 0.1, 0.3, and 0.5 mmol/L glutaradehyde (GTA), respectively and protein fraction in soluble protein complexes of heated samples measured from SDS-PAGE.

Unheated

Particle diameter in serum phases (nm)

Heated

Unheated

SM SM+0.1 mmol/L GTA SM+0.3 mmol/L GTA

a

205 ± 2 200 ± 4ab 196 ± 2ab

a

a

193 ± 3 188 ± 2ab 187 ± 2ab

96 ± 2 90 ± 1b 85 ± 2c

SM+0.5 mmol/L GTA

193 ± 5b

186 ± 2b

82 ± 1c

Protein fraction in soluble protein complexes (%)

RI PT

Particle diameter in micellar phases (nm)

Heated

a

κ-casein

a

β-lg

131 ± 2 115 ± 2b 110 ± 1c

9.60 ± 0.2 5.30 ± 0.2b 0.20 ± 0.02c

12.50 ± 0.5a 8.50 ± 0.4b 5.70 ± 0.3c

99 ± 1d

0.06 ± 0.01c

3.70 ± 0.3d

SC

Sample

AC C

EP

TE D

M AN U

Results are expressed as means of three trials. a, b, c, d Different lowercase superscript letters in the same column are significantly different (P < 0.05).

ACCEPTED MANUSCRIPT Table 2 Water holding capacity and firmness of acid gel made from unheated and heated milk samples treated with 0, 0.1, 0.3, and 0.5 mmol/L glutaradehyde (GTA), respectively. Sample

Water holding capacity (%)

Firmness (N)

Heated

Unheated

Heated

SM SM+0.1 mmol/L GTA SM+0.3 mmol/L GTA

53 ± 3a 52 ± 3a 52 ± 2a

79 ± 3a 73 ± 2b 65 ± 2c

0.52 ± 0.05a 0.52 ± 0.04a 0.52 ± 0.02a

1.40 ± 0.02a 1.20 ± 0.03b 0.80 ± 0.04c

SM+0.5 mmol/L GTA

51 ± 2a

54 ± 3d

0.51 ± 0.08a

0.50 ± 0.03d

SC

RI PT

Unheated

AC C

EP

TE D

M AN U

Results are expressed as means of three trials. a, b, c, d Different lowercase superscript letters in the same column are significantly different (P < 0.05).

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 1

Highlights

2 Dissociation of micellar κ-casein occurred in heated milk.

4

Formation of soluble protein complexes was dependent on κ-casein dissociation.

5

Glutaraldehyde inhibited dissociation of micellar κ-casein.

6

Dissociation declined with the formation of soluble protein complexes.

7

Decrease in soluble protein complexes resulted in weaker acid gels

SC

RI PT

3

AC C

EP

TE D

M AN U

8

1