Interactions of Casein Micelles with Calcium Phosphate Particles

Jun 4, 2014 - Riddet Institute, Massey University, Private Bag 11 222, Palmerston North 4442, New Zealand. ‡. Fonterra Research and Development Cent...
1 downloads 0 Views 3MB Size
Subscriber access provided by MEMORIAL UNIV

Article

Interactions of Casein Micelles with Calcium Phosphate Particles Lucile Tercinier, Aiqian Ye, Skelte G. Anema, Anne Singh, and Harjinder Singh J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/jf5018143 • Publication Date (Web): 04 Jun 2014 Downloaded from http://pubs.acs.org on June 5, 2014

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Agricultural and Food Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 41

Journal of Agricultural and Food Chemistry

1

Interactions of Casein Micelles with Calcium Phosphate Particles

2

3

Lucile Tercinier,a,b Aiqian Ye,*,a Skelte G. Anema,b Anne Singh,b and Harjinder Singha

4

5

a

Riddet Institute, Massey University, Private Bag 11 222, Palmerston North 4442, New

6

Zealand

7

b

8

New Zealand

Fonterra Research and Development Centre, Private Bag 11 029, Palmerston North 4442,

9

10

11

12

13

14

*

15

Dr A. Ye

16

Riddet Institute,

17

Massey University,

18

Private Bag 11 222,

19

Palmerston North 4442, New Zealand.

20

Tel.: +64-6-350-5072;

21

Fax: +64-6-350-5655.

22

E-mail address: [email protected].

Corresponding author.

23 1 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 2 of 41

24

Abstract

25

Insoluble calcium phosphate particles, such as hydroxyapatite (HA), are often used in

26

calcium-fortified milks as they are considered to be chemically unreactive. However, this

27

study showed that there was an interaction between the casein micelles in milk and HA

28

particles. The caseins in milk were shown to bind to the HA particles, with the relative

29

proportions of bound β-casein, αs-casein, and κ-casein different to the proportions of the

30

individual caseins present in milk. Transmission electron microscopy showed no evidence of

31

intact casein micelles on the surface of the HA particles, which suggested that the casein

32

micelles dissociated either before or during binding. The HA particles behaved as ion

33

chelators, with the ability to bind the ions contained in the milk serum phase. Consequently,

34

the depletion of the serum minerals disrupted the milk mineral equilibrium, resulting in

35

dissociation of the casein micelles in milk.

36

37

Keywords: Milk, Caseins, Casein micelle, Calcium phosphate, Hydroxyapatite, Adsorption.

38

39

2 ACS Paragon Plus Environment

Page 3 of 41

Journal of Agricultural and Food Chemistry

40

INTRODUCTION

41

Calcium fortification is a growing trend in the food industry, especially for milk and dairy

42

products. However, fortifying milk with calcium can be a challenge, especially if the product

43

has to be heat treated. When soluble calcium salts are added to milk, the ionic calcium

44

modifies the mineral equilibrium of the milk. Consequently, the physicochemical

45

characteristics of the casein micelles are modified and their heat stability decreases.1

46

Therefore, insoluble calcium phosphate particles such as hydroxyapatite (HA) or calcium

47

carbonate, rather than soluble salts, are often used for the calcium fortification of milk. When

48

HA particles were added to cow milk or soy milk, they were reported not to modify the ionic

49

calcium content or the pH of the milk, and therefore did not affect the heat stability of the

50

casein micelles when the milk was sterilized.2–4 They are therefore considered to be

51

unreactive when added to milk. However, it is known that proteins can interact with HA

52

particles, e.g., in chromatographic columns or nano-ceramic materials.5,6 HA particles are not

53

homogeneously charged. Proteins can adsorb to both positive and negative adsorbing sites

54

that are expressed at the surface of HA particles.

55

The negative sites are formed by oxygen atoms associated with phosphate and are called

56

P-sites. Positively charged groups of proteins, such as amino groups, can adsorb to P-sites.7

57

The positively charged sites, called C-sites, are formed by calcium ions in the crystal lattice.

58

C-sites can complex specifically with negatively charged groups of proteins, such as carboxyl

59

or phosphoserine groups.7 As milk proteins are negatively charged at the natural pH of milk,

60

they can adsorb to the C-sites of HA, as has been reported in several previous studies. For

61

example, Reynolds and Wong8 showed that, at pH 7, all the main milk proteins (αs1-casein,

62

β-casein, κ-casein, β-lactoglobulin, and α-lactalbumin) adsorb on to HA disks and cause them

63

to acquire an overall negative charge. In our previous work,9 we showed that caseins from

3 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 4 of 41

64

sodium caseinate (SC) adsorb on to HA particles, resulting in negatively- charged particles

65

and providing an electrosteric stabilization to the particles.

66

Whey proteins from whey protein isolate (WPI) are also able to bind to HA particles but to a

67

lesser extent than the caseins. 9 They are believed to bind to HA particles through their

68

carboxyl groups, whereas the caseins in SC are considered to bind more strongly on to HA

69

particles through both their carboxyl groups and their phosphoserine groups. SC is a mixture

70

of the natural caseins contained in milk, without the calcium and phosphate, which are

71

removed from the micelles during processing and are replaced by sodium.10 Therefore, in SC,

72

the casein molecules re-associate randomly and their phosphoserine groups are available for

73

binding to the calcium of HA particles. In milk, however, the phosphoserine groups of the

74

caseins are involved in the casein micelle structure and are already bound to the colloidal

75

calcium phosphate (CCP).11 Therefore, if exogenous HA particles are added to milk, it is not

76

certain whether the caseins in the micelles will be able to bind to the HA particles. Also

77

about 20% of the protein in milk is whey proteins, and these may also bind to HA particles.

78

The salt composition of milk is likely to affect the adsorption of proteins on to HA particles,

79

as the binding of proteins has been shown to depend on the mineral composition of the

80

suspending medium.12,13 Potential-determining ions can change the surface charge of the HA

81

particles, thereby affecting the amount of protein that can bind to them. For example, calcium

82

ions have been shown to give a positive charge to the surface of HA and this enhances

83

protein adsorption by increasing the number of sites available for protein binding through

84

their carboxyl and phosphoserine groups.14 Preliminary experiments (unpublished data)

85

showed that the adsorption of both caseins and whey proteins decreased in simulated milk

86

ultrafiltrate (SMUF), and this was attributed to the binding of phosphate or citrate ions from

87

the milk serum, which competed with the negatively charged groups of the proteins for

88

adsorption. As milk contains free calcium, phosphate, and citrate ions, it is likely that these 4 ACS Paragon Plus Environment

Page 5 of 41

Journal of Agricultural and Food Chemistry

89

ions will adsorb on to HA particles. If the ions in milk bind to the HA surface, the mineral

90

equilibrium of the milk serum phase could be disrupted, potentially affecting the integrity of

91

the casein micelles.15

92

As HA particles are often added to milk for calcium fortification, it is important to understand

93

the interactions that could take place between the milk proteins, the milk minerals, and the

94

HA particles, and whether they are likely to cause any instability in the products. Therefore,

95

the objective of this research was to investigate whether or not the casein micelles and whey

96

proteins contained in skim milk (SM) adsorbed on to HA particles when HA was added. The

97

binding of casein micelles on to HA particles was investigated using both SM and whey-

98

protein-depleted skim milk (WPD-SM). Transmission electron microscopy (TEM) was used

99

to check whether or not casein micelles could be observed at the surface of the HA particles.

100

To check whether the milk mineral equilibrium was affected by the addition of HA particles,

101

increasing levels of HA particles were added to SMUF and SM and the soluble minerals were

102

quantified. SM was also dialyzed against SMUF or against SMUF containing HA particles to

103

check whether the presence of HA particles caused disruption of the casein micelles when

104

casein micelles and HA particles were separated

105 106

EXPERIMENTAL SECTION

107

Materials. Fresh SM was obtained from a local supermarket and was used to prepare the

108

WPD-SM. Low heat skim milk powder (SMP) was obtained from Fonterra Co-operative

109

Group Ltd, New Zealand. The protein content of the SMP was 33% (w/w). Food grade HA

110

powder was purchased from Budenheim (TCP 53-83, Budenheim, Germany). The median

111

particle size d(0.5) was found to be 4.5 µm. The specific surface area of the powder was

112

determined by the BET (Brunauer–Emmett–Teller) method and was reported to be 65 m2/g.

5 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

113

SMUF was prepared according to Jenness and Koops,16 and the pH was adjusted to 6.67. All

114

the chemicals used were of analytical grade and were obtained from Sigma Aldrich unless

115

otherwise specified.

Page 6 of 41

116

117

Preparation of WPD-SM. WPD-SM was prepared using a bench-scale microfiltration (MF)

118

membrane unit (Amicon hollow fiber membrane, H1P30-43, pore size 0.1 µm). This

119

membrane unit allows the passage of the two main whey proteins (β-lactoglobulin and

120

α-lactalbumin), the lactose, and the serum-phase minerals to the MF permeate, but retains the

121

casein micelles. Fresh SM (1.3 L) was run through the membrane. The volume of the MF

122

permeate was monitored and an equivalent volume of SMUF containing 5% (w/w) lactose

123

was added back to the milk. Seven successive rinses with SMUF were necessary to obtain a

124

final WPD-SM containing only ~ 5% of the initial whey proteins. The WPD-SM was then

125

spray dried using a Yamato Pulvis spray drier (Model GB-22, Yamato Scientific Co., Tokyo,

126

Japan) and the powder obtained was stored in an air-tight container in a refrigerator until

127

needed. The protein content of the WPD-SM powder was measured using the Kjeldahl

128

method and was found to be 31% (w/w).

129

130

Adsorption Experiment. Stock solutions (10% (w/w) total solids (TS) on a powder basis,

131

corresponding to 3.3% (w/w) total protein) of WPD-SM and SM were reconstituted from

132

WPD-SM powder and SMP in Milli-Q water. The solutions were stirred for at least 1 h and

133

were left overnight at 4 °C to equilibrate.

134

The stock solutions were centrifuged at low speed (3000 g for 2 min) prior to the adsorption

135

experiment. A small pellet of casein micelles was centrifuged down at this speed (variable

6 ACS Paragon Plus Environment

Page 7 of 41

Journal of Agricultural and Food Chemistry

136

amount between 5 and 10% of the total casein micelles). The supernatants were carefully

137

poured out and used for further experiments.

138

Ranges of WPD-SM and SM solutions at different concentrations (0.15–2.8% (w/w) total

139

protein) were prepared by diluting appropriate volumes of the WPD-SM and SM stock

140

solutions in SMUF. As protein adsorption varies with ionic strength, pH, and mineral

141

composition, SMUF was used to prepare successive dilutions of the WPD-SM and SM

142

solutions so that the casein micelles were suspended in a solution of similar pH, ionic

143

strength, and mineral composition, for all samples. A constant weight of the WPD-SM or SM

144

solutions (0.95 g) of various protein concentrations (0.15–2.8% (w/w)) was added to

145

Eppendorf tubes containing the same amount of HA powder (50 mg).

146

The suspensions were stirred for 2 h at room temperature (approximately 20 °C), as

147

preliminary experiments had shown that equilibrium was reached after 2 h of adsorption. The

148

suspensions were then centrifuged (3000 g for 2 min) to separate the HA particles from the

149

WPD-SM and SM solutions. Preliminary work (unpublished data) had established that, under

150

these conditions, less that 5% (w/w) of the HA particles remained in suspension after

151

centrifugation; this did not affect the results. The supernatants (referred to in this paper as the

152

adsorption supernatants) were kept aside for further analysis. The adsorption experiments

153

were repeated at least twice.

154

155

Analysis of the Unabsorbed Protein Contained in the Adsorption Supernatant.

156

Determination of surface protein concentration and composition by SDS-PAGE. The

157

concentration of unadsorbed protein contained in the adsorption supernatants was determined

158

as a percentage of that in the initial WPD-SM or SM sample, using microfluidic chip sodium

159

dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), as described by Anema,17 7 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 8 of 41

160

and conventional SDS-PAGE, as described by Tercinier et al.9 The amount of adsorbed

161

protein (in mg of protein/m2 of HA surface available) was calculated by difference between

162

the amount of protein in the initial WPD-SM or SM solution and that in the supernatant, as

163

described by Tercinier et al.9 No correction was applied for the small portion of casein

164

micelles that were centrifuged down during the centrifugation of HA particles, as it could not

165

be measured precisely and was different for all samples prepared at different initial

166

concentrations. However, it was estimated to be always less than 5% of the total unadsorbed

167

caseins, and did not have a significant effect on the calculation of the amount of adsorbed

168

protein.

169

Each adsorption experiment was repeated at least twice. Variations in the amount of proteins

170

determined by SDS-PAGE were ~ 5% for total protein, ~ 2% for αs-casein (αs1- + αs2-), ~ 4%

171

for β-casein, ~ 9% for κ-casein, ~ 4% for β-lactoglobulin, and 9% for α-lactalbumin.

172

Non-micellar casein content. Non-micellar caseins were defined as those that did not

173

sediment from the adsorption supernatants during centrifugation at 21000 g and 25 °C for 60

174

min using a bench centrifuge (Centrifuge 5417R, Eppendorf AG, Hamburg, Germany). The

175

serum containing the non-micellar caseins was carefully removed from the pellet (containing

176

the casein micelles), and the protein content was determined by microfluidic chip SDS-PAGE

177

and was expressed as a percentage of the protein content in the adsorption supernatants.

178

Casein micelle diameter measurement. Dynamic light scattering (DLS) was performed

179

using a Malvern Zetasizer Nano ZS instrument (Malvern Instruments,Malvern,

180

Worcestershire, U.K.) to determine the hydrodynamic diameter (Dh) of the casein micelles.

181

The samples were diluted in SMUF (25µL of sample in 5 mL of SMUF) and 1 mL was

182

transferred to a plastic cuvette and placed in the instrument; the automatic attenuator setting

183

was used and the temperature of the cell was maintained at 20 °C for the duration of the

184

experiments Measurements of the dynamics of the scattered light were collected at a 8 ACS Paragon Plus Environment

Page 9 of 41

Journal of Agricultural and Food Chemistry

185

scattering angle of 173°. Average diffusion coefficients were determined by the method of

186

cumulants and were translated into average particle hydrodynamic diameters using the

187

Stokes–Einstein relationship for spheres.

188

189

Transmission Electron Microscopy. Three selected samples were prepared for TEM

190

observation: WPD-SM, HA particles suspended in water, and 50 mg of HA particles

191

suspended in 0.9 mL of WPD-SM. The general TEM method was similar to the method used

192

by McKenna et al.18 The WPD-SM solution was pipetted into agarose tubes and the ends of

193

the tubes were sealed with agarose. The suspensions of HA particles in water and in WPD-

194

SM were mixed with warm 3% low-temperature-gelling agarose at a ratio of 1:1. This

195

solution was poured on to a microscope slide, was allowed to set, and was chopped into 1

196

mm3 cubes. Agarose-embedded tubes and cubes were put into bijoux bottles containing 3%

197

glutaraldehyde in 0.2 M sodium cacodylate buffer and were left at 5 °C for 24 h. The

198

glutaraldehyde was rinsed twice with 0.2 M sodium cacodylate buffer over 2 h. The agarose-

199

embedded samples were then placed in 1% osmium tetroxide overnight at 5 °C. The samples

200

were placed in 1% uranyl acetate for 30 min and then rinsed twice with distilled water before

201

dehydration. The dehydration process consisted of successive rinses of the samples carried

202

out at 5 oC in 25% acetone for 15 min, 50, 70, and 90% acetone for 30 min each, followed by

203

100% acetone (three changes over 90 min). The acetone was then replaced by embedding

204

resin (Procure 812) and the bottles were placed on a rotator for 36 h. The cubes (for samples

205

containing HA particles) and the tubes (for samples containing only WPD-SM) were then

206

placed into an embedding capsule containing resin and baked at 60 oC for 48 h. The

207

embedded samples were then sectioned to a thickness of 90 nm using a Reichert Ultracut

208

microtome (Reichert-Jung, Heidelburg, Germany). These sections were mounted on 3 mm

209

copper grids and stained with lead citrate before examination in a Philips TEM 201C 9 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 10 of 41

210

transmission electron microscope (Philips, Eindhoven, The Netherlands) at an accelerating

211

voltage of 60 kV. Selected sections were also observed by light microscopy, using toluidine

212

blue to specifically stain protein material.

213

214

Dialysis Experiment. A dialysis experiment was carried out using 10% TS (w/w) SM. A 5%

215

TS (w/w) SM solution was prepared by diluting the 10% TS SM in SMUF. These SM

216

solutions are referred to as 10%-SM and 5%-SM in this paper. Samples of the 5%-SM and

217

10%-SM (10 mL) were placed in dialysis bags (MEMBRA-CEL MC25-16x100 CLR) and

218

dialyzed overnight at room temperature against SMUF (control) or SMUF containing 10%

219

HA particles. The SM samples were removed from the dialysis bags, their visual turbidity

220

was assessed by taking photographs, and the samples were analyzed for hydrodynamic

221

diameters of casein micelle and non-micellar casein content.

222

223

Mineral Concentrations. Total calcium was measured by inductively coupled plasma-

224

atomic emission spectrometry. The ionic calcium concentration was measured using a

225

Radiometer (Copenhagen, Denmark) type F2112 Ca calcium-specific electrode paired with

226

an Orion (Beverly, MA) Sure-Flow model 90-02 Ag/AgCl double junction electrode with

227

0.08 M KCl in the outer electrode chamber. The standards used were CaCl2 solutions (1–5

228

mM) in 0.08 M KCl. Inorganic phosphate (PO43–) was determined by automated colorimetric

229

analysis using a Technicon Auto-Analyzer. The samples were pumped through a segmented

230

flow injection system and inorganic phosphate was determined by a procedure involving

231

measurement of the absorption at 420 nm of a complex formed by orthophosphate ions and a

232

phosphate color reagent, containing ammonium molybdate and ammonium metavanadate.

233

Citrate was measured by an enzymatic method, involving successive enzymatic reactions 10 ACS Paragon Plus Environment

Page 11 of 41

Journal of Agricultural and Food Chemistry

234

with citrate lyase, malate dehydrogenase, lactate dehydrogenase, and NADH/H+. After the

235

enzymatic reactions, the concentration of NADH/H+, which was stoichiometric with the

236

concentration of citric acid, was determined by measuring the absorbance at 365 nm.

237

238

Statistical Analysis. T-tests and analysis of variance (ANOVA) tests were carried out using

239

Minitab software (Minitab version 16, Minitab, Inc., State College, PA) to determine the

240

significance of the differences. Significance was established at P < 0.05.

241

242

RESULTS AND DISCUSSION

243

Adsorption of Caseins and Whey Proteins. The adsorption studies were carried out by

244

adding a constant amount of HA to WPD-SM or SM solutions of different protein

245

concentration (0.15−2.8% (w/w)). After stirring for 2 h, the HA was separated from the

246

WPD-SM or SM solutions by centrifugation and the supernatants (referred to as the

247

adsorption supernatants) were analyzed for protein composition and concentration using

248

SDS-PAGE. The adsorbed protein concentration was calculated as the difference between the

249

total protein concentration and the unadsorbed protein concentration in each sample.

250

Figure 1 shows the unadsorbed protein concentration measured in the adsorption supernatants

251

and the amount of adsorbed proteins on the surface of the HA particles as a function of the

252

initial protein concentration for both WPD-SM (Figure 1A) and SM (Figure 1B). For both

253

WPD-SM and SM, at low initial protein concentration (< 0.5%), the amount of protein that

254

remained in the supernatants (i.e. the unadsorbed protein) was very low. As the initial protein

255

concentration increased, the amount of unadsorbed protein increased considerably. For both

256

systems, the amount of adsorbed caseins increased gradually until it reached a plateau value,

11 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 12 of 41

257

corresponding to a maximum protein coverage of the surface of the HA particles. For HA

258

particles mixed with WPD-SM, the surface casein concentration increased gradually with the

259

initial protein concentration up to about 1.5% (w/w, total protein) and reached a maximum

260

value of ~ 2.9 mg/m2. For HA particles mixed with SM, the amount of adsorbed casein

261

increased with the initial protein concentration up to about 2% (w/w, total protein) and

262

reached a maximum value of ~ 2.8 mg/m2 (Figure 1B). These results clearly showed that the

263

caseins from WPD-SM and SM adsorbed at the surface of HA particles, to similar maximum

264

amounts of ~ 2.9 mg/m2. In contrast, the maximum amount of whey proteins from SM that

265

bound to HA particles was ~ 0.1 mg/m2, which was negligible, as it represented only ~ 3% of

266

the total protein bound (Figure 1B).

267

Both caseins and whey proteins have been shown to bind to HA at pH ~ 7, when adsorption

268

studies are carried out in water or buffer.8,9 As both caseins and whey proteins carry a net

269

negative charge at the milk pH, they mostly adsorb on to HA particles via an interaction

270

between their negatively charged groups (carboxyl groups for the whey proteins, and both

271

carboxyl and phosphoserine groups for the caseins) and the calcium sites (C-sites) on HA

272

particles.5–7 However, as the interactions between the C-sites and phosphoserine groups are

273

stronger than those between the C-sites and carboxyl groups, caseins bind to HA to a greater

274

extent than whey proteins.9 As the phosphoserine groups of the caseins are grouped in

275

clusters, the caseins probably also carry more anchor points to bind to the C-sites on HA

276

particles than the whey proteins, which would partly explain the higher surface load for

277

caseins compared with whey proteins.19,20

278

Also, in a milk environment, there is competition for binding between the milk ions and the

279

milk proteins. Our previous work (unpublished data) on the binding of both caseins from SC

280

and whey proteins from WPI in SMUF showed that the whey proteins from WPI could bind

281

to HA particles only at a very low level (~ 0.5 mg/m2), as they had to compete for binding 12 ACS Paragon Plus Environment

Page 13 of 41

Journal of Agricultural and Food Chemistry

282

with the divalent milk anions, mostly phosphate and citrate present in SMUF. However, as

283

the interactions between caseins and HA are stronger, caseins could compete better with the

284

milk ions for binding and more casein was bound (~ 2.4 mg/m2). Negligible whey protein

285

binding was therefore expected when HA particles were added to SM, and this is what was

286

observed (Figure 1). Devold et al.21 showed that, at the milk pH, only caseins were able to

287

bind to the HA in tooth enamel and there was no whey protein binding.

288

The caseins in SC are in a non-micellar form and can be present as individual molecules or

289

small aggregates,22 whereas the caseins in WDP-SM and SM are associated together to form

290

casein micelles. The Langmuir model is often used in studies on the adsorption of protein on

291

to HA, to compare the adsorption characteristics of different types of protein.23–25 The

292

Langmuir model was therefore fitted to the adsorption data obtained in this study for WPD-

293

SM and SM and for SC in water and SMUF (unpublished data), to compare the adsorption

294

characteristics of the caseins from SC, WPD-SM, and SM. The Langmuir model quantifies

295

the amount of protein adsorbed on to HA as a function of the protein concentration once

296

equilibrium has been reached (in this case, the unadsorbed protein concentration) at a given

297

temperature. This model assumes that the proteins bind to a series of distinct empty sites on

298

the HA surface. The surface is assumed to be energetically homogeneous and the model

299

allows only a monolayer coverage. The Langmuir model is given by Eq. (1), as follows:

300



301

where mabs/S is the amount of protein bound to the surface expressed in mg/m2, [P] is the

302

concentration of protein at equilibrium (g/100 g), qm is the maximum monolayer surface

303

coverage (mg/m2), and K is the Langmuir affinity constant and is dimensionless.

304

The Langmuir isotherm curves for casein adsorption on to HA for the adsorption experiments

305

carried out with WPD-SM and SM are given in Figure 2. Table 1 gives the calculated

K[P] mabs = q m× S 1  KP

















(1)]

13 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 14 of 41

306

Langmuir constants for each of the isotherms, and for the adsorption data of SC on to HA, in

307

water and in SMUF. The maximum surface protein concentrations obtained for both WPD-

308

SM and SM experiments (Figure 1) (3.0 and 2.7 mg/m2, respectively) were not different from

309

the maximum adsorption value obtained for SC in water (2.9 mg/m2, P < 0.05) and were only

310

slightly higher than the maximum adsorption value for SC in SMUF (2.4 mg/m2).

311

In food emulsion systems, casein micelles have been shown to bind as intact micelles on to

312

oil droplets.26,27 The surface protein concentration for casein-micelle-coated droplets has been

313

reported to be much higher than that for SC-coated droplets,27,28 with usually two or three

314

times more adsorbed protein for casein micelles than for SC. As casein micelles are larger

315

than SC aggregates, they form a thicker layer at the surface of the oil droplets so that the

316

protein surface load is higher. The fact that the extent of binding of caseins in WDP-SM and

317

SM did not bind to HA was similar to that in SC could indicate that casein micelles do not

318

bind as intact micelles to HA particles. The affinity constant of the Langmuir model

319

characterizes the initial slope of the adsorption isotherms and indicates a high affinity of

320

protein for HA.24 There was a variability of the affinity constants in this study (Table 1), as a

321

small variation in the first points of the adsorption isotherms (Figure 2) can lead to a high

322

variation of the slope, and therefore of the affinity constant. However, it is interesting to note

323

that the mean affinity constants were similar for the casein adsorption experiments carried out

324

on SC in SMUF, WDP-SM, and SM binding to HA (Table 1). This suggests that the

325

adsorption mechanisms and the affinity of HA for micellar and non-micellar forms of caseins

326

in a milk environment are somewhat similar.

327

328

Transmission Electron Microscopy. TEM was carried out to examine the structure of the

329

adsorbed caseins at the surface of the HA particles. Figure 3A and Figure 3B show TEM

14 ACS Paragon Plus Environment

Page 15 of 41

Journal of Agricultural and Food Chemistry

330

images from a solution of WPD-SM. The casein micelles in WPD-SM appear to be dark

331

spherical aggregates, to be well dispersed, and to have diameters ranging from ~ 30 to 150

332

nm, consistent with the images of casein micelles observed in previous studies.29–31 Figures

333

3C and 3D show TEM images of the HA powder used in this study. It consists of particles of

334

~ 4 µm average size. However, these particles are made of elementary nanometer-sized rod-

335

like particles (Figure 3D) that aggregate together into micron-sized particles, as has been

336

observed previously.32,33

337

Figures 3E and 3F show images of the HA particles suspended in WPD-SM. As the particles

338

had been stirred for 2 h in the WPD-SM before preparation for TEM, the caseins had

339

adsorbed to the HA particles. The HA particles had a similar appearance to the HA particles

340

in the control (Figure 3C). No casein micelles could be seen on the surfaces of the particles

341

(Figures 3E and 3F). It was therefore hypothesised that the caseins did not adsorb as intact

342

micelles, but as small aggregates and/or individual molecules, which would be too small to be

343

observed with TEM. The unadsorbed casein micelles could still be seen in the suspending

344

phase around the particles, as shown by the arrows in Figure 3F. These unadsorbed micelles

345

had a hairy and disorganized appearance compared with the control (Figures 3A and 3B),

346

which may have been because they were partly dissociated by the presence of HA particles.

347

Analysis of Unadsorbed Caseins. The initial adsorption measurements and TEM

348

observations led to the hypothesis that casein micelles did not bind to HA particles as intact

349

micelles, possibly because the micelles dissociated either before or shortly after binding. To

350

verify this hypothesis, the supernatants containing the unadsorbed proteins were analyzed for

351

casein content, particle size, and protein composition.

352

The supernatants were centrifuged at 21000 g and 25 °C for 60 min; the pellet contained

353

micellar caseins whereas the non-micellar caseins stayed in the serum phase. This serum

15 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 16 of 41

354

phase was analyzed for protein content using microfluidic chip SDS-PAGE. Figure 4 shows

355

the distribution of the unadsorbed caseins between micellar and non-micellar forms (Figure

356

4A) and their size (Figure 4B), as a function of the initial protein concentration.

357

At low initial protein concentration, more than half of the unadsorbed caseins were in a non-

358

micellar form. For example, when the HA particles were prepared with WPD-SM containing

359

0.9% protein, 0.08% of the unadsorbed protein was non-micellar (Figure 4A), which

360

represented ~ 57% of the unadsorbed proteins (total 0.14%). As the initial protein

361

concentration increased, the proportion of non-micellar casein in the unadsorbed protein

362

decreased, reaching approximately 5% when HA particles were prepared with WPD-SM

363

containing 2.4% protein (0.08% out of 1.4% unadsorbed protein), which was close to the

364

level of non-micellar casein in the initial WPD-SM solutions (~ 3% non-micellar casein, data

365

not shown).

366

At low initial protein concentration, the average diameter of the unadsorbed casein particles

367

was about ~ 122 nm (Figure 4B). As the initial protein concentration increased, the diameter

368

increased, until it reached ~ 155 nm, which was close to the size of the casein micelle in the

369

initial WPD-SM solutions (160 nm).

370

Figure 5 shows the protein composition of the unadsorbed proteins as a function of the initial

371

protein concentration. For the adsorption experiments carried out with WPD-SM containing

372

less than 1% (w/w) protein, the ratio of unadsorbed κ-casein to unbound αs- and β-caseins

373

was considerably greater than in the original WPD-SM (the composition of the original

374

WPD-SM is indicated by the lines on Figure 5, and was αs-casein 50%, β-casein 38%, and κ-

375

casein 12%, based on the theoretical proportions of the individual proteins in SM10). For

376

example, when HA particles were mixed with WPD-SM containing 0.6% protein, ~ 60% of

377

the unadsorbed protein was κ-casein, which is substantially higher than the proportion found

16 ACS Paragon Plus Environment

Page 17 of 41

Journal of Agricultural and Food Chemistry

378

in milk (12%). The ratio of unadsorbed αs-casein to unadsorbed β-casein remained

379

approximately constant. As the initial protein concentration in the SM increased, the relative

380

proportions of the three unadsorbed individual caseins gradually approached the initial WPD-

381

SM composition as more and more unadsorbed protein accumulated in the supernatant.

382

Analysis of the unadsorbed proteins showed that the proportion of micellar casein in the

383

unadsorbed casein fraction (Figure 4A), the casein micelle size (Figure 4B), and the protein

384

composition (Figure 5) were not constant across the range of initial protein concentrations

385

studied. The unadsorbed caseins obtained at low initial protein concentrations contained a

386

higher proportion of κ-casein, contained a higher proportion of non-micellar casein, and were

387

of a smaller size, compared with the unadsorbed caseins obtained at high initial protein

388

concentration. This confirmed that the caseins were not bound to the HA particles as intact

389

micelles, and that the casein micelle structure was disrupted by the addition of HA particles.

390

However, it was not possible to determine whether the casein micelles dissociated first and

391

then the liberated caseins bound to HA or casein micelles adsorbed as intact micelles and

392

then dissociated allowing only the strongly bound caseins to remain on the HA particle

393

surface. However, as there was no change in the proportions of the different caseins adsorbed

394

on to the HA particles as a function of time, from 1 min to 3 h (data not shown), it is believed

395

that the caseins first dissociated from the micelles and then bound independently. However,

396

further experiments would be needed to confirm this hypothesis.

397

The amounts of the individual proteins adsorbed on to HA were calculated from the

398

composition of the unadsorbed protein and the original protein composition of the WPD-SM

399

(Figure 6). When the initial protein concentration was varied from 0 to 2.5%, the amounts of

400

adsorbed αs-casein, β-casein, and κ-casein increased gradually until they reached maxima of

401

~ 1.5, 1.3, and 0.15 mg/m2, respectively. The maximum amount of bound κ-casein

402

represented only ~ 5% of the total bound protein (0.15 mg/m2 out of 2.9 mg/m2), even though 17 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 18 of 41

403

the original WPD-SM contained ~ 12% κ-casein. This confirms that αs-casein and β-casein

404

are preferentially adsorbed over κ-casein.

405

The preference for αs- and β-caseins binding to HA has been shown previously and has been

406

attributed to the high content of phosphoserine residues in both αs- and β-caseins, compared

407

with κ-casein.9 Caseins adsorb mainly through their phosphoserine residues, which bind to

408

calcium on the HA particles. αs1-Casein, αs2-casein, and β-casein contain 7− 9, 10−13, and 5

409

phosphoserine residues, respectively, whereas κ-casein contains only 1 phosphoserine

410

residue.34 As αs- and β-caseins carry more residues that can bind to the C-sites on HA

411

particles, they can bind to a greater extent. As the casein micelles are probably disrupted by

412

the addition of HA particles, the phosphoserine groups of αs- and β-caseins must become

413

available to bind to HA particles.

414

It has been shown previously that both αs- and β-caseins contain specific phosphopeptides

415

with a high ability to bind to calcium phosphate.35–37 They contain similar calcium phosphate

416

binding motifs, made of a sequence of amino acids containing three phosphoserine residues

417

followed by two glutamic acid residues.38 These casein-derived phosphopeptides (CPP) have

418

been produced and purified by enzymatic digestion and have been bound to amorphous

419

calcium phosphate (ACP) to form complexes (CPP–ACP).39-40 It is probably the same motifs

420

of amino acids as those involved in the CPP–ACP complexes that are involved in the binding

421

of αs1-casein, αs2-casein, and β-casein to HA particles.

422

Mechanism of Casein Micelle Dissociation by HA Particles. This study has shown that,

423

the addition of HA particles to WPD-SM and SM causes dissociation of casein micelles into

424

individual casein or casein aggregates that then bind to the HA particles. However, it is not

425

known why the addition of HA particles causes the dissociation of the casein micelles. To

18 ACS Paragon Plus Environment

Page 19 of 41

Journal of Agricultural and Food Chemistry

426

further investigate the mechanism, a dialysis experiment and mineral quantification were

427

carried out.

428

Dialysis experiments . 10%-SM and 5%-SM were dialyzed overnight against SMUF

429

(controls) or SMUF containing HA particles to ascertain whether the presence of HA

430

particles caused the dissociation of the casein micelles when not in direct contact.

431

When removed from the dialysis bags, the 10%-SM and 5%-SM dialyzed against SMUF

432

containing HA particles (cuvettes B and D, respectively, in Figure 7) were less turbid than

433

their respective controls dialyzed against SMUF (cuvettes A and C, respectively, in Figure 7).

434

For the lower concentration of SM (5%-SM), the sample dialyzed against SMUF containing

435

HA particles was almost transparent, indicating that almost all the casein micelles had

436

dissociated. This shows that the casein micelles in the milks dialyzed against HA particles

437

dissociated during the dialysis process, even though they were not in direct contact with the

438

particles. This was confirmed by measurement of the non-micellar casein content and the

439

particle size of the dialyzed milks (Table 2). The casein micelle diameter of the control SM

440

was ~ 230 nm for both 10%-SM and 5%-SM and both milks contained less than 5% non-

441

micellar caseins, which was similar to the original non-dialyzed SM. This shows that the

442

milks dialyzed against SMUF were stable during dialysis. The 10%-SM and 5%-SM dialyzed

443

against SMUF containing HA particles contained 73 and 94% non-micellar caseins,

444

respectively. The casein micelle size of 10%-SM dialyzed against SMUF containing HA

445

particles was 165 nm; that of 5%-SM treated identically could not be measured by the

446

Zetasizer because the sample did not shown the required turbidity.

447

448

19 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 20 of 41

449

Mineral quantification. The observations that HA caused the casein micelles to dissociate

450

even when not in direct contact (Figure 7) led to the hypothesis that HA particles were able to

451

draw out some of the minerals contained in SMUF, which would cause the casein micelles to

452

dissociate. It has been shown previously that calcium, phosphate, and citrate ions can adsorb

453

to HA.12,41 The ions contained in SMUF must have adsorbed on to HA particles, disrupting

454

the salt balance in the SMUF. This hypothesis was tested by adding different amounts of HA

455

particles (from 0 to 18% (w/w)) into SMUF and SM and then measuring ionic calcium, total

456

calcium, and inorganic phosphate.

457

Figure 8 shows the variation in the ionic calcium concentration when HA particles at

458

different concentrations (from 0 to 18% (w/w)) were added to SMUF or to SM. The SM and

459

SMUF contained ~ 1.7 and ~ 1.45 mM ionic calcium, respectively, which was in accordance

460

with the values reported in the literature.42–44 As more HA particles were added to both

461

SMUF and SM, the ionic calcium concentration decreased, with the decrease being more

462

pronounced in SMUF. The addition of 10% (w/w) HA particles led to ionic calcium

463

concentrations of ~ 0.1 mM in SMUF and ~ 0.55 mM in SM, which represented losses of 93

464

and 68% of the initial ionic calcium, respectively. The decrease in ionic calcium on the

465

addition of HA particles indicates that ionic calcium binds to these particles. As ionic calcium

466

binds to HA particles, CCP must be released from the casein micelles to compensate for the

467

loss of ionic calcium in the serum, thereby explaining the smaller decrease in ionic calcium

468

for SM than for SMUF. The binding of ionic calcium on to the HA particles was a relatively

469

fast process;the ionic calcium concentration reached a stable level in less than 5 min (data not

470

shown).

471

A mineral analysis was also carried out on SMUF and SM samples after the HA particles had

472

been removed from the solution by centrifugation. As HA particles were added to SM at

473

increasing concentration from 0 to 18% (w/w), the concentrations of total calcium, citrate, 20 ACS Paragon Plus Environment

Page 21 of 41

Journal of Agricultural and Food Chemistry

474

and inorganic phosphate decreased (Figure 9A). The initial SM contained ~ 21.5 mmol/kg of

475

inorganic phosphate, ~ 28.5 mmol/kg of calcium, and ~ 11 mmol/kg of citrate, which is

476

consistent with the values reported previously for SM.15 On the addition of 18% HA particles

477

to SM, the calcium concentration in the SM decreased by ~ 93%, from 28.5 to 2.2 mmol/kg,

478

the inorganic phosphate concentration decreased by ~ 68%, from 21.6 to 6.8 mmol/kg, and

479

the citrate concentration decreased by ~74%, from 11 to 2.85 mmol/kg. Part of the decrease

480

was probably because the calcium, phosphate, and citrate that were contained in the casein

481

micelles before they dissociated and adsorbed on the HA particles sedimented out with the

482

HA particles during centrifugation. However, CCP contains only ~ 50% of the total inorganic

483

phosphate, ~ 70% of the total calcium, and ~ 20% of the total citrate contained in milk. As

484

the percentage decreases in calcium, phosphate, and citrate on the addition of 18% HA

485

particles to SM were higher than the percentage of the corresponding mineral in CCP,

486

calcium, phosphate, and citrate ions must have bound to HA particles, depleting the milk

487

serum in these ions.

488

This was confirmed by the mineral analysis results when HA particles were added to SMUF

489

(Figure 9B). On the addition of 15% HA particles to SMUF, the total calcium concentration

490

decreased by ~ 95%, from 8.7 to 0.5 mmol/kg, the inorganic phosphate concentration

491

decreased by ~ 53%, from 11.6 to 5.5 mmol/kg, and the citrate concentration decreased by ~

492

77%, from 9.8 to 2.3 mmol/kg. The decrease in calcium was faster and more pronounced than

493

the decreases in phosphate and citrate, probably because phosphate and citrate have to

494

compete for the same binding sites on the HA surface (positive sites, C-sites), whereas only

495

calcium ions can bind to the negative sites (P-sites), and because more P-sites than C-sites are

496

expressed on the surface of HA at pH ~ 7, as shown by a negative zeta-potential of ~ –9 mV

497

(data not shown).

21 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 22 of 41

498

The disruption effect of HA on the integrity of the casein micelle and on the milk mineral

499

equilibrium can also be explained by taking into account the thermodynamic aspects of

500

calcium phosphate growth and sequestration. In milk, calcium and phosphate are sequestered

501

in the form of calcium phosphate nanoclusters inside the casein micelle.45 The phosphoserine

502

groups of the caseins are bound to the calcium phosphate nanoclusters inside the micelles,

503

keeping the calcium phosphate in a thermodynamically metastable state in which its

504

precipitation and further growth are prevented.46,47 However, when HA particles are added to

505

milk, it is possible that the most stable form of calcium phosphate (HA) grows at the expense

506

of the least stable form (calcium phosphate sequestered in the casein micelles). The calcium

507

phosphate inside the casein micelles would therefore be released to contribute to the growth

508

of the HA particles, leading to micelle dissociation. The process of calcium phosphate growth

509

is generally very slow. However, the HA particles used in this study may have provided a

510

very large surface of initially crystallised HA that would allow a fast crystallisation of

511

calcium phosphate from the minerals that were drawn out of the milk serum onto the HA

512

particles.48

513

In conclusion, this study showed the existence of interactions between casein micelles and

514

HA particles. The addition of HA particles to skim milk causes the dissociation of the casein

515

micelles and the binding of the individual caseins to the particles. This dissociation of the

516

casein micelles is a consequence of calcium, phosphate and probably citrate ions contained in

517

the milk serum binding to the surface of the HA particles. Consequently, CCP of casein

518

micelle is drawn out of the micelles and the structure of the casein micelles is disrupted.

519

Therefore, the phosphoserine groups of αS- and β-caseins become available to bind to HA

520

particles. Both αS- and β-caseins were preferentially bound compared with κ-casein, and this

521

was attributed to the higher content of phosphoserine residues of both αS- and β-caseins. The

522

results of this study indicate that the addition of hydroxyapatite particles to milk may 22 ACS Paragon Plus Environment

Page 23 of 41

Journal of Agricultural and Food Chemistry

523

compromise the stability of the milk by disrupting the equilibrium of the milk minerals and

524

dissociating the casein micelles.

525 526

ABBREVIATIONS

527

ACP: amorphous calcium phosphate

528

ANOVA: analysis of variance

529

CCP: colloidal calcium phosphate

530

CPP: casein-derived phosphopeptides

531

HA: hydroxyapatite

532

MF: microfiltration

533

SC: sodium caseinate

534

SDS-PAGE: sodium dodecyl sulfate polyacrylamide gel electrophoresis

535

SM: skim milk

536

SMP: skim milk powder

537

SMUF: simulated milk ultrafiltrate

538

TEM: transmission electron microscopy

539

TS: total solids

540

WDP-SM: whey protein-depleted skim milk

541

WPI: whey protein isolate

542

23 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 24 of 41

543

ACKNOWLEDGEMENTS

544

The authors gratefully acknowledge Michael Loh for his help with TEM, Claire Woodhall for

545

proofreading the manuscript and Fonterra Research and Development Centre for their

546

financial support.

547

548

549

550

24 ACS Paragon Plus Environment

Page 25 of 41

Journal of Agricultural and Food Chemistry

551

REFERENCES

552 553

(1)

Philippe, M.; Le Graët, Y.; Gaucheron, F. The effects of different cations on the physicochemical characteristics of casein micelles. Food Chem. 2005, 90, 673–683.

554 555 556

(2)

Chaiwanon, P.; Puwastien, P.; Nitithamyong, A.; Sirichakwal, P.P. Calcium Fortification in Soybean Milk and In Vitro Bioavailability. J. Food Comp. Anal. 2000, 13, 319–327.

557 558 559

(3)

Omoarukhe, E.D.; On-Nom, N.; Grandison, A.S.; Lewis, M.J. Effects of different calcium salts on properties of milk related to heat stability. Int. J. Dairy Technol. 2010, 63, 504–511.

560 561 562

(4)

Pathomrungsiyounggul, P.; Grandison, A.S.; Lewis, M.J. Effects of calcium-chelating agents and pasteurisation on certain properties of calcium-fortified soy milk. Int. J. Food Sci. Technol. 2010, 45, 2234–2240.

563 564

(5)

Bernardi, G.; Kawasaki, T. Chromatography of polypeptides and proteins on hydroxyapatite columns. Biochim. Biophys. Acta 1968, 160, 301–310.

565 566

(6)

Kawasaki, T. Hydroxyapatite as a liquid chromatographic packing. J. Chromatogr. A 1991, 544, 147–184.

567 568

(7)

Gorbunoff, M.J.; Timasheff, S.N. The interaction of proteins with hydroxyapatite: III. Mechanism. Anal. Biochem. 1984, 136, 440–445.

569 570

(8)

Reynolds, E.C.; Wong, A. Effect of adsorbed protein on hydroxyapatite zeta potential and Streptococcus mutans adherence. Infect. Immun. 1983, 39, 1285–1290.

571 572

(9)

Tercinier, L.; Ye, A.; Anema, S.; Singh, A.; Singh, H. Adsorption of milk proteins on to calcium phosphate particles. J. Colloid Interface Sci. 2013, 394, 458–466.

573 574

(10)

Wong, D.W.S.; Camirand, W.M.; Pavlath, A.E.; Parris, N.; Friedman, M. Structures and functionalities of milk proteins. Crit. Rev. Food Sci. Nutr. 1996, 36, 807–844.

575 576

(11)

Dalgleish, D.G. On the structural models of bovine casein micelles-review and possible improvements. Soft Matter 2011, 7, 2265–2272.

577 578

(12)

Yin, G.; Liu, Z.; Zhan, J.; Ding, F.; Yuan, N. Impacts of the surface charge property on protein adsorption on hydroxyapatite. Chem. Eng. J. 2002, 87, 181–186.

579 580 581

(13)

Zhu, X.D.; Fan, H.S.; Li, D.X.; Xiao, Y.M.; Zhang, X.D. Protein adsorption and zeta potentials of a biphasic calcium phosphate ceramic under various conditions. J. Biomed. Mater. Res., Part B 2007, 82, 65–73.

582 583

(14)

Wassell, D.T.H.; Hall, R.C.; Embery, G. Adsorption of bovine serum albumin onto hydroxyapatite. Biomaterials 1995, 16, 697–702.

584 585 586

(15)

Holt, C. The Milk Salts: Their Secretion, Concentrations and Physical Chemistry. In Developments in Dairy Chemistry—3, Fox, P.F., Ed.; Elsevier Applied Science Publishers: Amstersdam, Netherlands, 1985; pp. 143–181. 25 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 26 of 41

587 588

(16)

Jenness, R.; Koops, Preparation and properties of a salt solution which simulates milk ultrafiltrate. Neth. Milk Dairy J. 1962, 16, 153–164.

589 590

(17)

Anema, S.G. The use of “lab-on-a-chip” microfluidic SDS electrophoresis technology for the separation and quantification of milk proteins. Int. Dairy J. 2009, 19, 198–204.

591 592 593

(18)

McKenna, A.B.; Lloyd, R.J.; Munro, P.A.; Singh, H. Microstructure of whole milk powder and of insolubles detected by powder functional testing. Scanning 1999, 21, 305–315.

594 595

(19)

Johnsson, M.; Levine, M.J.; Nancollas, G.H. Hydroxyapatite binding domains in salivary proteins. Crit. Rev. Oral Biol. Med. 1993, 4, 371–378.

596 597

(20)

Wang, K.; Zhou, C.; Hong, Y.; Zhang, X. A review of protein adsorption on bioceramics. Interface Focus 2012, 2, 259–277.

598 599 600

(21)

Devold, T.G.; Rykke, M.; Isabey, D.; Sørensen, E.S.; Christensen, B.; Langsrud, T.; Svenning, C.; Borch-Iohnsen, B.; Karlsen, J.; Vegarud, G.E. In vitro studies of adsorption of milk proteins onto tooth enamel. Int. Dairy J. 2006, 16, 1013–1017.

601 602

(22)

Pitkowski, A.; Durand, D.; Nicolai, T. Structure and dynamical mechanical properties of suspensions of sodium caseinate. J. Colloid Interface Sci. 2008, 326, 96–102.

603 604 605

(23)

Iafisco, M.; Sabatino, P.; Lesci, I.G.; Prat, M.; Rimondini, L.; Roveri, N. Conformational modifications of serum albumins adsorbed on different kinds of biomimetic hydroxyapatite nanocrystals. Colloids Surf., B 2010, 81, 274–284.

606 607 608

(24)

Iafisco, M.; Di Foggia, M.; Bonora, S.; Prat, M.; Roveri, N. Adsorption and spectroscopic characterization of lactoferrin on hydroxyapatite nanocrystals. Dalton Trans. 2011, 40, 820–827.

609 610 611

(25)

Mohsen-Nia, M.; Massah Bidgoli, M.; Behrashi, M.; Mohsen Nia, A. Human Serum Protein Adsorption onto Synthesis Nano-Hydroxyapatite. Protein J. 2012, 31, 150– 157.

612 613 614

(26)

Euston, S.R.; Hirst, R.L. The Emulsifying Properties of Commercial Milk Protein Products in Simple Oil-in-Water Emulsions and in a Model Food System. J. Food Sci. 2000, 65, 934–940.

615 616

(27)

Ye, A. Functional properties of milk protein concentrates: Emulsifying properties, adsorption and stability of emulsions. Int. Dairy J. 2011, 21, 14–20.

617 618 619

(28)

Singh, H.; Ye, A. Interactions and functionality of milk proteins in food emulsions. In Milk Proteins. From Expression to Food; Thompson, A., Boland, M., Singh, H., Eds.; Academic Press: San Diego, CA, 2008; pp. 321–345.

620 621

(29)

Dalgleish, D.G. Food emulsions—their structures and structure-forming properties. Food Hydrocolloids 2006, 20, 415–422.

622 623

(30)

Schmidt, D.G.; Buchheim, W. The application of electron microscopy in dairy research. J. Microsc. 1992, 167, 105–121.

26 ACS Paragon Plus Environment

Page 27 of 41

Journal of Agricultural and Food Chemistry

624 625 626

(31)

Titapiccolo, G.I.; Alexander, M.; Corredig, M. Rennet-induced aggregation of homogenized milk: Impact of the presence of fat globules on the structure of casein gels. Dairy Sci. Technol. 2010, 90, 623–639.

627 628

(32)

Sun, R.-x.; Lu, Y.-p. Fabrication and characterization of porous hydroxyapatite microspheres by spray-drying method. Front. Mater. Sci. China 2008, 2, 95–98.

629 630 631

(33)

Xu, W.; Ganz, C.; Weber, U.; Adam, M.; Holzhüter, G.; Wolter, D.; Frerich, B.; Vollmar, B.; Gerber, T. Evaluation of injectable silica-embedded nanohydroxyapatite bone substitute in a rat tibia defect model. Int. J. Nanomed. 2011, 6, 1543–1552.

632 633

(34)

Walstra, P.; Jenness, R. Proteins. In Dairy Chemistry and Physics; John Wiley and Sons: New York, NY, 1984; pp. 98-122.

634 635 636

(35)

Cross, K.J.; Huq, N.L.; Palamara, J.E.; Perich, J.W.; Reynolds, E.C. Physicochemical characterization of casein phosphopeptide-amorphous calcium phosphate nanocomplexes. J. Biol. Chem. 2005, 280, 15362–15369.

637 638

(36)

Baum, F.; Ebner, J.; Pischetsrieder, M. Identification of multiphosphorylated peptides in milk. J. Agric. Food. Chem. 2013, 61, 9110–9117.

639 640 641

(37)

Reynolds, E.C.; Riley, P.F.; Adamson, N.J. A selective precipitation purification procedure for multiple phosphoseryl-containing peptides and methods for their identification. Anal. Biochem. 1994, 217, 277–284.

642 643

(38)

FitzGerald, R.J. Potential Uses of Caseinophosphopeptides Int. Dairy J. 1998, 8, 451– 457.

644 645

(39)

Cross, K.J.; Huq, N.L.; Reynolds, E.C. Casein Phosphopeptides in Oral Health Chemistry and Clinical Applications. Curr. Pharm. Des. 2007, 13, 793–800.

646 647

(40)

Reynolds, E.C. Anticariogenic complexes of amorphous calcium phosphate stabilized by casein phosphopeptides: A review. Spec. Care Dentist. 1998, 18, 8–16.

648 649 650

(41)

López-Macipe, A.; Gómez-Morales, J.; Rodríguez-Clemente, R. The Role of pH in the Adsorption of Citrate Ions on Hydroxyapatite. J. Colloid Interface Sci. 1998, 200, 114–120.

651 652 653 654 655

(42)

Gao, R.; Temminghoff, E.J.M.; van Leeuwen, H.P.; van Valenberg, H.J.F.; Eisner, M.D.; van Boekel, M.A.J.S. Simultaneous determination of free calcium, magnesium, sodium and potassium ion concentrations in simulated milk ultrafiltrate and reconstituted skim milk using the Donnan Membrane Technique. Int. Dairy J. 2009, 19, 431–436.

656 657

(43)

Lewis, M.J. The measurement and significance of ionic calcium in milk - A review Int. J. Dairy Technol. 2011, 64, 1–13.

658 659

(44)

Lin, M.-J.; Lewis, M.J.; Grandison, A.S. Measurement of ionic calcium in milk. Int. J. Dairy Technol. 2006, 59, 192–199.

660 661

(45)

Little, E.M.; Holt, An equilibrium thermodynamic model of the sequestration of calcium phosphate by casein phosphopeptides. C. Eur. Biophys. J. 2004, 33, 435–447. 27 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 28 of 41

662 663 664

(46)

Holt, C.; Wahlgren, N.M.; Drakenberg, T. Ability of a beta-casein phosphopeptide to modulate the precipitation of calcium phosphate by forming amorphous dicalcium phosphate nanoclusters. Biochem. J. 1996, 314, 1035–1039.

665 666 667 668

(47)

Holt, C.; Timmins, P.A.; Errington, N.; Leaver, J. A core-shell model of calcium phosphate nanoclusters stabilized by β-casein phosphopeptides, derived from sedimentation equilibrium and small-angle X-ray and neutron-scattering measurements. J. Eur. J. Biochem. 1998, 252, 73–78.

669 670 671

(48)

Nancollas, G.H.; Tomazic, B. Growth of calcium phosphate on hydroxyapatite crystals. Effect of supersaturation and ionic medium. J. Phys. Chem.1974, 78, 2218225

672 673

674

28 ACS Paragon Plus Environment

Page 29 of 41

675

Journal of Agricultural and Food Chemistry

FIGURE CAPTION

676

677

Figure 1. Amount of (●) caseins and (▲) whey proteins adsorbed on to HA particles

678

(mg/m2) and (○) unadsorbed protein concentration (% (w/w)) as a function of initial protein

679

concentration: (A) WPD-SM; (B) SM; 50 mg of HA particles were added in 0.95 mL of

680

WPD-SM and SM; the total surface area of HA was 3.25m2. Results are means of duplicates

681

or triplicates with error bars for the standard deviation.

682

Figure 2. Adsorption isotherms of caseins on to HA particles: (A) WPD-SM; (B) SM;

683

(●,▲,■) experimental data from different repeat experiments; (solid line) best-fit Langmuir

684

curve.

685

Figure 3. TEM micrographs obtained from: (A and B) WPD-SM solution reconstituted from

686

WPD-SMP in water (10% (w/w)); (C and D) suspension of HA particles in SMUF; (E and F)

687

suspension of HA particles in WPD-SM solution, corresponding to a stage of adsorption at

688

which the maximum amount of adsorbed protein was reached; the arrows show the remaining

689

unbound casein micelles in the suspending phase.

690

Figure 4. (A) Distribution of unadsorbed caseins between (white bar) micellar and (grey bar)

691

non-micellar forms and (B) diameters of unadsorbed caseins. Error bars represent standard

692

deviation.

693

Figure 5. Composition of the supernatants containing the unadsorbed protein after

694

adsorption of WPD-SM on to HA particles as a function of the initial protein concentration:

695

(●) αs-casein (αs1 + αs2); (▲) β-casein; (■) κ-casein; the lines indicate the composition of the

696

original WPD-SM in (dotted line) αs-casein (αs1 + αs2), (short dashed line) β-casein, and (long

697

dashed line) κ-casein. Error bars represent standard deviation.

29 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 30 of 41

698

Figure 6. Estimated amount of adsorbed caseins (mg/m2) on HA particles as a function of

699

initial protein concentration of WPD-SM: (●) αs-casein (αs1 + αs2); (▲) β-casein; (■)

700

κ-casein. Error bars represent standard deviation.

701

Figure 7. Visual turbidity of SM samples after dialysis against SMUF or against SMUF

702

containing 10% (w/w) HA particles: (A) 10%-SM dialyzed against SMUF; (B) 10%-SM

703

dialyzed against SMUF containing 10% (w/w) HA particles; (C) 5%-SM dialyzed against

704

SMUF; (D) 5%-SM dialyzed against SMUF containing 10% (w/w) HA particles.

705

Figure 8. Effect of addition of HA particles in (■) SMUF and in (▲) SM on ionic calcium

706

concentration. Error bars represent standard deviation.

707

Figure 9. Effect of addition of HA particles in (A) SM and (B) SMUF on (▼) total calcium,

708

(■) total citrate, and (●) inorganic phosphate concentrations; HA particles were removed

709

from the solution by centrifugation before the measurements.

30 ACS Paragon Plus Environment

Page 31 of 41

Journal of Agricultural and Food Chemistry

TABLES

Table 1. Parameters for the Adsorption of Caseins from WPD-SM, SM, and SC in Water and in SMUF, Calculated According to the Langmuir Model

casein source

affinity constant, K

WPD-SM SM SC in water SC in SMUF

24.5ab ± 9 31.5a ± 7.1 13.7b ± 4.8 31.5a ± 3.6

maximum surface coverage, qm (mg/m2) 3.0a ± 0.02 2.7ab ± 0.15 2.9a ± 0.2 2.4b ± 0.01

Means with different superscript letters differ significantly (P < 0.05; vertical comparison).

Table 2. Non-micellar Casein Content and Casein Micelle Size of the Dialyzed Milks sample

Non-micellar caseins (%)

casein micelle diameter (nm)

10%-SM dialyzed against SMUF dialyzed against HA in SMUF

3 73

230 165

5%-SM dialyzed against SMUF dialyzed against HA in SMUF

5 94

227 N/A

N/A: Not applicable

31 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 32 of 41

3.5

1.6

(A) 3.0

1.4

2.5

1.2 1.0

2.0

0.8 1.5 0.6 1.0

0.4

0.5

0.2

0.0

0.0 0.0

0.5

1.0

1.5

2.0

2.5

Unadsorbed protein concentration (%)

Amount of adsorbed protein (mg/m2)

FIGURE GRAPHICS

3.5

(B)

2.0

3.0 2.5

1.5

2.0 1.0

1.5 1.0

0.5 0.5 0.0

0.0 0.0

0.5

1.0

1.5

2.0

2.5

3.0

Unadsorbed protein concentration (%)

Amount of adsorbed protein (mg/m2)

Initial protein concentration (%)

Initial protein concentration (%)

Figure 1.

32 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Amount of adsorbed casein (mg / m2)

Page 33 of 41

3.5 3.0

(A)

2.5 2.0 1.5 1.0 0.5 0.0 0.0

0.5

1.0

1.5

2.0

Amount of adsorbed casein (mg / m2)

Casein concentration at equilibrium (%)

3.5 3.0

(B)

2.5 2.0 1.5 1.0 0.5 0.0 0.0

0.5

1.0

1.5

2.0

Casein concentration at equilibrium (%)

Figure 2.

33 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 34 of 41

Figure 3.

34 ACS Paragon Plus Environment

Page 35 of 41

Journal of Agricultural and Food Chemistry

Figure 4.

35 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 36 of 41

Figure 5.

36 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Amount of adsorbed protein (mg/m2)

Page 37 of 41

2.0

1.5

1.0

0.5

0.0 0.0

0.5

1.0

1.5

2.0

2.5

3.0

Initial protein concentration (%)

Figure 6.

37 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 38 of 41

Figure 7.

38 ACS Paragon Plus Environment

Page 39 of 41

Journal of Agricultural and Food Chemistry

1.8

Ionic calcium (mM)

1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 0

2

4

6

8

10 12 14 16 18 20

HA particles (%)

Figure 8.

39 ACS Paragon Plus Environment

30

(A) 25 20 15 10 5 0 0

2

4

6

8 10 12 14 16 18 20

HA particles concentration (%)

Mineral concentration (mmol/kg)

Mineral concentration (mmol/kg)

Journal of Agricultural and Food Chemistry

Page 40 of 41

14

(B)

12 10 8 6 4 2 0 0

2

4

6

8

10 12 14 16

HA particles concentration (%)

Figure 9.

40 ACS Paragon Plus Environment

Page 41 of 41

Journal of Agricultural and Food Chemistry

GRAPHICAL ABSTRACT

41 ACS Paragon Plus Environment