The effect of hexametaphosphate addition during milk powder manufacture on the properties of reconstituted skim milk

The effect of hexametaphosphate addition during milk powder manufacture on the properties of reconstituted skim milk

International Dairy Journal 50 (2015) 58e65 Contents lists available at ScienceDirect International Dairy Journal journal homepage: www.elsevier.com...

438KB Sizes 1 Downloads 164 Views

International Dairy Journal 50 (2015) 58e65

Contents lists available at ScienceDirect

International Dairy Journal journal homepage: www.elsevier.com/locate/idairyj

The effect of hexametaphosphate addition during milk powder manufacture on the properties of reconstituted skim milk Skelte G. Anema* Fonterra Research and Development Centre, Private Bag 11029, Palmerston North, New Zealand

a r t i c l e i n f o

a b s t r a c t

Article history: Received 5 May 2015 Received in revised form 16 June 2015 Accepted 16 June 2015 Available online 6 July 2015

Skim milk powders with various levels of sodium hexametaphosphate (NaHMP) were prepared. Reconstituted skim milk samples were prepared from these powders. NaHMP slightly reduced the pH, markedly reduced the serum and ionic calcium and markedly increased the serum phase orthophosphate levels of the milks. This shift in the mineral equilibrium resulted in a drastic reduction in casein micelle integrity, with a marked dissociation of casein from the micelles. k-Casein was the predominant casein dissociated, although significant levels of aS-casein and b-casein were also transferred to the serum phase. This dissociation of the casein micelles caused a marked decrease in size and scattering properties of the casein micelles. In addition, a small decrease in the zeta potential of the casein micelles in the milk was observed. Heat treatment of the milks with added NaHMP induced further dissociation of k-casein, although much of the aS-casein and b-casein re-associated with the micelles. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction Sodium hexametaphosphate (NaHMP) is a food grade polyphosphate. NaHMP is recognised as a cyclic compound with the molecular formula of Na6P6O18; however, commercial food grade NaHMP may be a linear polyphosphate chain or possibly a mixture of the linear and cyclic compounds (Gao et al., 2010). NaHMP has the potential to bind up to 3 calcium atoms per molecule, and therefore is a very effective calcium chelator compared with simple monophosphates or citrates (de Kort, Minor, Snoeren, van Hooijdonk, & van der Linden, 2009; Odagiri & Nickerson, 1965). In addition to chelating calcium, NaHMP can bind to the casein micelles in milk. Vujicic, deMan, and Woodrow (1968) found that most of the NaHMP added to milk was bound to the casein micelles. Lower levels were found to be bound to the micelles in the studies of Odagiri and Nickerson (1965). de Kort, Minor, Snoeren, van Hooijdonk, and van der Linden (2011) proposed that, as NaHMP has six negative charges homogeneously distributed around the molecule, this charge distribution allowed the interaction of the NaHMP with both calcium ions and casein molecules. The NaHMP can bind to adjacent casein molecules either directly through the negative charges on the NaHMP interacting with the positive

* Tel.: þ64 6 350 4649. E-mail address: [email protected]. http://dx.doi.org/10.1016/j.idairyj.2015.06.006 0958-6946/© 2015 Elsevier Ltd. All rights reserved.

changes on the caseins or indirectly through the NaHMP and casein binding to the same calcium molecule. Thus this dual interaction of NaHMP can facilitate a cross-linking behaviour between proteins and mineral components. NaHMP and other polyphosphates or calcium chelators are often added to dairy systems to improve the processing ability of the dairy system or to impart desired functional properties to the derived products. For example, polyphosphates or other calcium chelators are routinely added during processed cheese manufacture to chelate calcium and aid emulsification of the cheese melt (Kapoor & Metzger, 2008). The fouling of heat exchanger surfaces during the manufacture of sterilised or UHT liquid milk products is markedly reduced on the addition of calcium chelators (Burdett, 1974; Prakash, Datta, Lewis, & Deeth, 2007), with polyphosphates such as NaHMP being more effective than the monophosphates (Burdett, 1974). The effect of NaHMP on the heat stability of milk is less clear cut. Abdulina (1975) and Abdulina, Kovalenko, and Alekseev (1970) reported improvements in the heat stability of milk of normal concentration on the addition of NaHMP, whereas Mittal, Hourigan, and Zadow (1990) reported that the addition of NaHMP to UHT recombined milk had a detrimental effect on heat stability, and this effect was more pronounced on the storage stability of the product. de Kort, Minor, Snoeren, van Hooijdonk, and van der Linden (2012) examined the effect of various calcium chelators on the heat stability of concentrated micellar casein solutions and found that

S.G. Anema / International Dairy Journal 50 (2015) 58e65

NaHMP was the least effective in improving the heat stability, whereas weaker calcium chelators had the greatest effect on heat stability. NaHMP addition is reported to improve the shelf life of UHT milk products by delaying the onset of age gelation. At least two mechanisms for age gelation exist (Datta & Deeth, 2001; Harwalkar, 1992). In one, there is an observed proteolysis during the storage of the UHT milk either by indigenous milk proteases such as plasmin, or through exogenous enzymes from bacterial contamination. In a second mechanism for age gelation, the gelation occurs on storage without obvious changes to proteins in the milk. McMahon (1996) has proposed a mechanism for this type of age gelation that involves the release of the heat-induced b-lactoglobulinek-casein complex from the casein micelles followed by a subsequent aggregation of this complex in the serum phase to form a gel network of cross-linked protein. NaHMP delays the onset of gelation by both mechanisms (Datta & Deeth, 2001; Harwalkar, 1992). Although the effects of NaHMP on the mineral equilibria in milk, and in particular the soluble and ionic calcium activity under ambient conditions, are reasonably well understood, its effect on the mineral distribution over a wider temperature range has not been studied, In addition, the effect of NaHMP addition to milk on the casein micelles, in particular the dissociation of casein micelles has not received much attention. Although milk powders do not generally contain polyphosphates, there is provision in the Codex standards for milk powders for the addition of low levels of polyphosphates (Codex Alimentarius Commission, 2011). Therefore in this study, NaHMP was added to milk during milk powder manufacture. The properties of the milks reconstituted from these milk powders were studied. The serum phase mineral composition at a range of temperatures was monitored, and various properties of the casein micelles were determined, including the levels of casein dissociated from the micelles. 2. Materials and methods 2.1. Milk powder and reconstituted skim milk samples Skim milk powders were manufactured with 0, 0.1, 0.5 and 1% added NaHMP (Innophos Cranbury, NJ, USA) on a dry basis. The appropriate quantity of NaHMP was added to chilled skim milk samples (~4  C) and dispersed by stirring with an overhead stirrer for ~15 min. The powders were prepared from the skim milk samples using the techniques and equipment described previously (Baldwin & Truong, 2007). In brief, the powders were manufactured on a pilot scale spray dryer (Anhydro, Copenhagen, Denmark, nominal evaporative capacity 80 kg h1) using nozzle atomisation. Skim milk was pre-heated at 90  C for 20 s by indirect heating and concentrated in a falling film vacuum evaporator with four effects to produce a concentrate of ~50% total solids. The concentrate was sprayed at ~125 bar pressure, with the inlet air temperature at 205  C and the outlet air temperature at ~85  C to produce a milk powder. Experimental powders were sealed in foil lined bags until use. Experimental milk samples of 10% (w/w) total solids were prepared by adding the appropriate quantity of the skim milk powder to water and allowing samples to stir for at least 2 h before use. The levels of NaHMP in the four reconstituted milk samples were, therefore, 0, 0.01, 0.05 and 0.1% (w/w), respectively. These milk samples will be referred to as reconstituted skim milk (RSM). 2.2. Additional heat treatments of skim milk samples Sub-samples (6 mL) of the reconstituted skim milk were placed in glass vials (8 mL total volume) and heated at 120  C for 6 min

59

(including the heat-up time of ~45 s) in a thermostatically controlled oil bath. After heating the samples were cooled in an ice/ water bath until the temperature was below 30  C. These milk samples will be referred to as heated reconstituted skim milk (HRSM). 2.3. Ultracentrifugation RSM and HRSM samples were transferred to tubes and centrifuged at 101,000  g for 1 h at 20  C in a Beckman L8-40 ultracentrifuge and the associated Ti80 rotor (Beckman Instruments Inc., Palo Alto, CA, USA). The supernatants were carefully removed and used for analysis while the pellets were discarded. 2.4. Polyacrylamide gel electrophoresis and laser densitometry The RSM and HRSM samples were analysed for native whey protein content and composition using native polyacrylamide gel electrophoresis techniques as has been described previously (Anema & McKenna, 1996). The skim milk samples and their respective ultracentrifugal supernatants were analysed for protein composition and concentration by reduced sodium dodecyl polyacrylamide gel electrophoresis (SDS-PAGE) as described in detail previously (Anema, 2009b). After staining and destaining procedures, the gels were scanned using a Molecular Dynamics scanning densitometer and the protein levels obtained by analysing the band intensities using the Imagequant integration software (Molecular Dynamics Inc., Sunnyvale, CA, USA). 2.5. Particle size measurements RSM and HRSM samples were dispersed in calciumeimidazole buffer (5 mM CaCl2, 20 mM imidazole and 30 mM NaCl at pH 7.0) and allowed to stand for 15 min. Casein micelle sizes were determined by photon correlation spectroscopy using a Malvern Zetasizer 4 instrument and the associated ZET5110 large bore sizing cell (Malvern Instruments Ltd, Malvern, UK). The correlation functions were collected at a scattering angle of 90 only. Each sample was measured 5 times and the average particle size was determined. In addition to the particle size, the intensity of the scattered radiation was recorded using the intensity function on the Zetasizer instrument. This involves the measurement of the scattered radiation, as photons per second, from the sample at an angle of 90 . 2.6. Zeta potential measurements RSM samples were dispersed in the calciumeimidazole buffer and allowed to stand for 15 min. Zeta potentials were determined by laser Doppler electrophoresis using the Malvern Zetasizer 4 instrument and the associated ZET5104 electrophoresis cell. An applied voltage of 60 V was used in all experiments, and the temperature was maintained at 20  C. A fresh sample was injected for each separate measurement as this provided more reproducible results. A total of 20 repeat measurements were made and the average determined. 2.7. Collection of permeate samples for determination of serum phase mineral levels RSM samples were warmed to temperatures between 20 and 60  C in a water bath and held for 1 h. Samples of milk ultrafiltrate were taken from the RSM at each temperature using an Amicon hollow fibre cartridge with a 10,000 Da cutoff membrane (Amicon, Inc., Beverly, MA, USA) as described previously (Anema, 2009a).

S.G. Anema / International Dairy Journal 50 (2015) 58e65

Total calcium, total phosphorus and total inorganic phosphorus were determined by an inductively coupled plasma emission spectroscopy method (Alkanani, Friel, Jackson, & Longerich, 1994). For the total calcium and total inorganic phosphate, the protein was removed by precipitation with trichloroacetic acid and for the total phosphorus the samples were dry ashed before analysis using the methods described previously (Sanders, 1931). The orthophosphate (orthoP) was determined using the molybdo-vanadate method (Allen, 1940). Ionic calcium levels were determined using a calcium specific electrode (Radiometer F2112Ca electrode, RadiometerCopenhagen, Copenhagen, Denmark) coupled with a reference electrode (Orion 90-02 double junction electrode, Beverly, Massachusetts, USA) as described previously (Anema & Li, 2003a). The pH of the milk samples was measured at a range of temperatures times using a combination glass electrode coupled with a temperature probe. The standard pH buffers were allowed to equilibrate for about 1 h at the desired temperature in a thermostatically controlled water bath and then used for calibration of the pH meter. This allowed the pH shift of buffers due to temperature differences to be taken into account during calibration. Milk samples were equilibrated at the desired temperatures for a minimum of 1 h in thermostatically controlled water bath. The pH of these samples was monitored at the desired temperature. 2.9. Experimental design The skim milk powders were analysed for total calcium, total phosphorus, total inorganic phosphorus and orthoP. Phosphorus distributions were determined from the total phosphorus, total inorganic phosphorus and orthoP levels. Total phosphorus is all the phosphorus in the milk powder and includes the natural inorganic phosphates of the milk, the organic phosphates esterified to the serine amino acids of the caseins and the added NaHMP. The total inorganic phosphorus is measured after caseins have been removed and therefore has the natural inorganic phosphates of the milk and the added NaHMP. The orthoP measures only the inorganic orthophosphate in the milk, and thus excludes the NaHMP and serine phosphates. Thus the difference between the total phosphorus and the inorganic phosphorus should give the organic phosphorus content in each sample, and the difference between the inorganic phosphorus and the orthoP should give the phosphorus attributable to the added NaHMP. For the RSM, the pH of the milk and total calcium, ionic calcium and orthoP in the serum phase were measured at a range of temperatures. The level of native whey proteins and the serum phase casein and whey proteins were also measured. In addition, the particles sizes, scattering intensities and zeta potentials of the particles in the milk were measured. For the HRSM, the level of native whey proteins, the serum phase casein and whey proteins, and the particles sizes and scattering intensities were measured. 2.10. Replication and statistical analysis All experiments were repeated at least twice. Statistical significance was determined using analysis of variance provided in the EZAnalyze statistical analysis program (Poynton, 2007). 3. Results and discussion 3.1. Phosphate distribution in milk samples The total calcium in the milk powders was found to be ~361 mmol kg1 and was similar for all four powders. In skim milk,

the calcium level is about 29e33 mM and the solids level is about 9% (Dalgleish & Law, 1989; Holt, 1985). This would convert to about 322e367 mmol kg1 on a dry basis, which is in the range found for these milk powders. The measured total phosphorus, inorganic phosphorus and orthoP in the powders and the calculated organic phosphorus and calculated phosphorus from the NaHMP are shown in Fig. 1A. As expected, the total phosphorus was greater than the inorganic phosphorus, and both increased with increasing levels of added NaHMP. The measured orthoP was essentially constant at ~250 mmol kg1 and very similar to the level of inorganic phosphorus for the sample with no added NaHMP. In natural skim milk, the level of phosphate is about 21 mM (Dalgleish & Law, 1989; Holt, 1985), and as the solids level is about 9%, this would equate to about 240 mM on a dry basis, which is close to the level found in these skim milk powders. The level of organic phosphorus was also essentially constant at about 90 mmol kg1. Skim milk has about 10 mM of phosphate esterified to casein, which would equate to about 100 mmol kg1 on a dry basis, which is reasonably consistent with the results for these milk powders. The calculated phosphorus attributable to the added NaHMP increased linearly with the level of added NaHMP, from about 0 mmol kg1 in the sample with no added NaHMP to about 90 mmol kg1 for the sample with 1% added NaHMP. Using a molecular weight of 612 g mol1 for NaHMP and assuming 6 phosphorus atoms per NaHMP, the level of phosphorus from the NaHMP was calculated (based on the addition levels) and compared with that determined experimentally (Fig. 1B). The calculated level was higher than that determined experimentally

500

A

400 300 200 -1

2.8. Mineral analysis and pH measurements

Phosphorus (mmol kg )

60

100 0 100

B

80 60 40 20 0 0.0

0.2

0.4

0.6

0.8

1.0

NaHMP in powder (% , w/w) Fig. 1. Panel A: phosphorus distribution in milk powder samples with added sodium hexametaphosphate (NaHMP). C, total phosphorus; B, total inorganic phosphorus; ;, orthophosphate; 7, organic phosphorus (calculated); -, NaHMP (calculated). Panel B: predicted NaHMP (B) and experimentally determined NaHMP (C). Error bars on experimental points represent the standard deviations for repeated measurements.

S.G. Anema / International Dairy Journal 50 (2015) 58e65

by about 10% at the highest addition level. This difference is probably expected as the NaHMP was added to a liquid milk stream before drying which may have somewhat different total solids than predicted, and the added NaHMP was on a dry basis and the powders had ~4% moisture. In addition, the NaHMP is unlikely to be absolutely pure and will have a lower NaHMP content, whereas the calculation assumed 100% purity. As can be seen in Fig. 1, skim milk powders with a range of NaHMP levels were produced, which would allow the effects of the NaHMP on the reconstituted milk properties to be determined. 3.2. Native whey protein levels in reconstituted milk samples As the milk powder had received a heat treatment of 90  C for 20 s during manufacture, some denaturation of whey protein was expected. Based on kinetic experiments (Anema & McKenna, 1996), this heat treatment was expected to denature approximately 40, 50 and 6% of the b-lactoglobulin A, b-lactoglobulin B and a-lactalbumin respectively. The levels of denaturation of the b-lactoglobulin A, b-lactoglobulin B and a-lactalbumin were approximately 42%, 60% and 10%, respectively, which is close to these predicted levels. The addition of NaHMP had no effect on the level of denaturation in the RSM samples. For the HRSM, the additional heat treatment of 120  C for 6 min was sufficient to fully denature the whey proteins in all samples (results not shown). 3.3. Serum phase mineral levels and pH of milk permeates RSM samples were equilibrated to temperatures in the range from 20 to 60  C, and the pHs were measured (Fig. 2A). The pH of the RSM samples decreased markedly and essentially linearly with an increase in temperature. The decrease was approximately 0.065 pH units per 10  C increase in temperature, which is very similar to that reported previously (Anema, 2009a; Pouliot, Boulet, & Paquin, 1989). There was a small decrease in pH with increasing NaHMP addition; however, this difference was less than 0.04 pH units between all samples at each temperature and although consistently observed this decrease was not significant (Fig. 2A). Mittal et al. (1990) and Vujicic et al. (1968) also reported a decrease in the pH of milk upon addition of NaHMP, consistent with this study. de Kort et al. (2009) showed a complex effect of NaHMP on the pH of calcium solutions with a pH decline at low addition levels (up to ~10 mM ¼ ~0.6%), followed by a pH increase at higher addition levels. In a modelling study on the addition of various components to simulated milk ultrafiltrate (SMUF), Gao et al. (2010) showed that NaHMP decreased the pH of the SMUF buffer; however, the effect was very small and was only about 0.2 pH units at an addition level of 0.5%. This equates to about 0.04 pH units per 0.1%, which is similar to that observed in Fig. 2A. The pH reduction is probably a consequence of the binding of calcium to the NaHMP, releasing bound protons (Gao et al., 2010; de Kort et al., 2009). The RSM samples equilibrated at the temperatures from 20 to 60  C were passed through an ultrafiltration membrane to collect permeate samples and the total calcium, ionic calcium and orthoP levels of the permeate samples were determined at each temperature (Fig. 2BeD respectively). The serum phase calcium level decreased markedly with increasing temperature (Fig. 2B). For the control sample, the decrease in serum phase calcium levels with temperature was approximately 0.6 mM per 10  C, which is consistent with literature reports (Anema, 2009a; Pouliot et al., 1989). At temperatures up to 40  C, increasing levels of added NaHMP markedly decreased the serum phase calcium level. At higher temperatures the effect was less pronounced, although the sample with 0.1% NaHMP maintained a lower serum phase calcium level at all temperatures (Fig. 2B).

61

As the analytical method measures the total calcium content of the serum phase, any calcium bound to NaHMP should still be measured. As the serum phase calcium level decreased with added NaHMP, these results indicate that at least some of the added NaHMP (along with its associated calcium) has bound to the colloidal phase. This is not unexpected as it is known that NaHMP, as well as some other polyphosphate salts, can bind to casein proteins, and can even cross-link the caseins (de Kort, 2012; de Kort et al., 2011; Vujicic et al., 1968). The observation that the differences are smaller at higher temperatures may be a consequence of less NaHMP/Ca binding to the colloidal phase at higher temperatures, possibly due to the pH decline at higher temperatures, thus making differences between control samples and those with added NaHMP similar at the higher temperatures. The ionic calcium levels of the permeate samples decreased with increasing temperature in a similar manner to the serum phase calcium. Increasing levels of added NaHMP progressively reduced the ionic calcium levels at each temperature investigated (Fig. 2C). As NaHMP binds calcium ions, a reduction in ionic calcium levels is expected and is consistent with previous studies (Gao et al., 2010; Holt, 1985; de Kort et al., 2011, 2012; Mittal et al., 1990; Odagiri & Nickerson, 1965; Tsioulpas, Koliandris, Grandison, & Lewis, 2010; Vujicic et al., 1968). The orthoP levels of the serum phases were determined (Fig. 2D). This is a measure of the natural phosphate levels of the milk and not the added NaHMP, and therefore indicates the change in orthoP equilibrium between the colloidal and serum phases. The orthoP level in the serum phase generally decreased with increasing temperature, consistent with earlier reports (Anema, 2009a; Pouliot et al., 1989); however the magnitude of the decrease was less pronounced as the level of added NaHMP was increased. In contrast to the serum phase calcium and ionic calcium levels, the level of serum phase orthoP increased as the level of added NaHMP was increased. The increase was somewhat more pronounced at the higher temperatures (Fig. 2D). Vujicic et al. (1968) also reported a marked increase in the serum phase orthoP level when NaHMP was added to milk. A solubility product type relationship exists between calcium and orthoP in milk (Chaplin, 1984; Holt, 1985). Addition of NaHMP chelates the serum phase calcium, thus the equilibrium between calcium and phosphate is disrupted and colloidal calcium phosphate will solubilise to restore the equilibrium to the concentrations dictated by the solubility product. A relationship between the serum phase calcium and serum phase phosphate or between ionic calcium and the serum phase phosphate would therefore be expected. Linear relationships between total serum calcium and serum phosphate (Fig. 2E) or between ionic calcium and serum phosphate (Fig. 2F) is observed at each temperature, with the slope being lower as the temperature increased. These major changes in serum phase calcium and orthoP levels indicate that the mineral equilibrium of the milk system was markedly altered by the addition of NaHMP. The NaHMP probably chelates the serum phase calcium and some of the NaHMP/calcium complex associates with the casein micelles. As the serum phase calcium and orthoP have a solubility product relationship with the colloidal calcium phosphate, some of the native colloidal calcium phosphate would solubilise to maintain solubility product concentrations of calcium phosphate in the serum phase. As both calcium and phosphate dissolve in the molecular ratio of the colloidal calcium phosphate, the result is a net increase in the serum phase orthoP level and a net decrease in the serum calcium level compared to the milk without added NaHMP, as is observed.

62

S.G. Anema / International Dairy Journal 50 (2015) 58e65

A

B

8.0

6.8

7.0

pH

6.6

6.5 6.4

6.0

Serum calcium -1 (mmol kg )

7.5

5.5

6.2

5.0

C

D

11.5

Ionic calcium (mM)

11.0 1.5

10.5 10.0

1.0

9.5 0.5

Serum orthophosphate -1 (mmol kg )

2.0

9.0 0.00

0.02

0.04

0.06

0.08

0.10

0.00

0.02

0.04

0.06

0.08

0.10

HMP in liquid milk (% )

8.0

E

F

2.0

7.0

1.5

6.5 1.0

6.0

Ionic calcium -1 (mmol kg )

Serum calcium -1 (mmol kg )

7.5

5.5 0.5

5.0 9.0

9.5

10.0

10.5

11.0

11.5

Serum orthophosphate (mmol kg-1)

9.0

9.5

10.0

10.5

11.0

11.5

Serum orthophosphate (mmol kg-1)

Fig. 2. Effect of sodium hexametaphosphate addition and temperature on A, pH; B, serum calcium; C, ionic calcium; D, serum orthophosphate levels in the reconstituted skim milk samples. Panels E and F: relationships between serum orthophosphate and serum calcium (E) and between serum orthophosphate and ionic calcium (F). Samples were measured at: C, 20  C; B, 30  C; ;, 40  C; , 50  C; -, 60  C. Error bars represent the standard deviations for repeated measurements.



3.4. Casein micelle zeta potential The results for the zeta potential determinations of the casein micelles in the RSM samples are given in Fig. 3. The zeta potential for the casein micelles in the RSM samples with no added NaHMP was found to be about 14.6 ± 0.2 mV which is consistent with that reported previously for milk samples dispersed in this buffer system (Anema & Klostermeyer, 1996; Dalgleish, 1984). The magnitude of the zeta potential decreased from 14.6 to 13.5 mV as the level of NaHMP in the RSM was increased from 0 to 0.1%. This is a small but significant decrease in the charge on the casein micelles and

suggests some surface modifications to the casein micelles, possibly involving the k-casein surface layers. de Kort et al. (2012) showed that NaHMP addition to concentrated casein micelle solutions resulted in an increase in the magnitude of the zeta potential; however, subsequent heating at 126  C caused a decrease in the magnitude of the zeta potential. Tsioulpas et al. (2010) also showed an increase in the magnitude of the zeta potential on NaHMP addition, and further increases on subsequent heating at 121  C for 15 min. However, in this latter study the NaHMP caused a significant increase in pH, which contrasts with that observed here (Fig. 2A). It is possible that the

S.G. Anema / International Dairy Journal 50 (2015) 58e65

220

Micelle diameter (nm)

z eta potential (-mV)

15.0 14.5 14.0 13.5

63

A

210 200 190 180

13.0 0.00

0.02

0.04

0.06

0.08

0.10

NaHMP in liquid milk (% ) Fig. 3. Effect of sodium hexametaphosphate on the zeta potential of the casein micelles in the reconstituted skim milk. Error bars represent the standard deviations for repeated measurements.

differences in behaviour in these different studies are a consequence of the nature of the NaHMP and the relative proportion of linear versus cyclic polyphosphate. The NaHMP used by de Kort et al. (2009) caused a substantial decrease in the pH of aqueous solutions suggesting that this may be predominantly a linear polyphosphate, as also suggested by Gao et al. (2010). Interestingly, Kocak and Zadow (1985) found the NaHMP from different suppliers had different physico-chemical properties and also differed in their abilities to control age gelation in UHT milks, confirming that could be composed of different levels and types of cyclic and linear polyphosphates. An alternative explanation for the differences observed between this study and those in the literature (de Kort et al., 2009; Mittal et al., 1990; Tsioulpas et al., 2010; Vujicic et al., 1968) is that the addition of NaHMP during the manufacture of skim milk powders has different effects to when the NaHMP is added to fresh liquid milk. The process of making milk powder involves preheating, evaporation and spray drying, and the powder is then reconstituted to form the RSM. The NaHMP may have different effects on the mineral components and the casein micelles during these processing and reconstitution steps that account for the differences in pH and zeta potential of the RSM samples when compared with the literature reports where the NaHMP was added to liquid milks. 3.5. Casein micelle size and scattering intensity The casein micelle size and scattering intensities of the RSM and HRSM samples were determined (Fig. 4). For the RSM, the average casein micelle size of the control sample was found to be about 203 nm (Fig. 4A), which is in the normal range for bovine micelles from reconstituted skim milk (Anema, Lowe, Lee, & Klostermeyer, 2014; Anema, Lowe, & Li, 2004). The average casein micelle size decreased markedly as the level of added NaHMP in the RSM was increased so that the average micelle size of the samples with 0.1% NaHMP was about 20 nm (10%) smaller than the control RSM samples (Fig. 4A). This decrease in micelle size was accompanied by a marked decrease in scattering intensity of the casein micelles (Fig. 4B). For the HRSM samples, the micelles sizes of the micelles increased by about 10e30 nm over the RSM samples, with a greater increase for the samples with higher levels of added NaHMP. There was a decrease in average micelle size as the level of NaHMP in the milk was increased, but the decrease was less pronounced than that

Scattering intensity -1 (kcounts sec )

260

B

240 220 200 180 160 140 120 0.00

0.02

0.04

0.06

0.08

0.10

NaHMP in liquid milk (% ) Fig. 4. Change in (A) casein micelle size and (B) scattering intensity for reconstituted skim milk samples (C) and heated reconstituted skim milk samples (B). Error bars represent the standard deviations for repeated measurements.

observed for the RSM samples (Fig. 4A). Similarly, the scattering intensity of the HRSM samples were markedly higher than the RSM samples, and showed a similar, although less pronounced, decrease with increasing NaHMP addition (Fig. 4B). Kalab and Harwalkar (1974), using electron microscopy, showed that the casein micelles in concentrated milk samples with added NaHMP were markedly disintegrated when compared with the control milk samples. In a very brief report, Abdulina et al. (1970) reported a decrease in casein micelle size and an increase in viscosity when NaHMP was added to milk that was supplemented with calcium and was either left unheated or heated. Interestingly Tsioulpas et al. (2010) showed an increase in casein micelle size for milks with added NaHMP and the size increased even more markedly after sterilisation of the milks that were subsequently sterilised (121  C for 15 min). 3.6. Serum phase proteins The marked changes in mineral equilibria and in particular the chelation of calcium by the NaHMP (Fig. 2) would be expected to have a detrimental effect on casein micelle integrity. This supposition is supported by the observation that there is a marked reduction in zeta potential (Fig. 3), size (Fig. 4A) and scattering intensity (Fig. 4B) of the colloidal particles with increasing NaHMP levels. In particular, the decrease in particle sizes and scattering intensities with increasing NaHMP levels (Fig. 4) for both the RSM and HRSM samples suggest a dissociation of the casein micelles with addition of NaHMP. Therefore the RSM and the HRSM samples were centrifuged and the supernatants analysed by SDS-PAGE. The integrated intensities of the major protein bands in the supernatants relative to that of the original milks are shown in Fig. 5A and B for the RSM and HRSM samples, respectively.

64

S.G. Anema / International Dairy Journal 50 (2015) 58e65

For the RSM, the SDS-PAGE analysis indicated significant levels of casein protein were dissociated from casein micelles on the addition of NaHMP (Fig. 5A). Although k-casein was the predominant protein dissociated, significant increases in the levels of serum phase as-casein and b-casein were induced by the addition of the NaHMP, particularly at a 0.1% addition rate. In the RSM samples, about 1.7% of the aS-casein, 4% of the b-casein and 13% of the kcasein was in the serum phase for the samples with no added NaHMP. On addition of 0.1% NaHMP these values increased to 17%, 28% and 50%, respectively. The level of these dissociated caseins (kcasein > b-casein > aS-casein) was inversely related to their content of phosphoserine residues (aS-casein > b casein > k-casein) indicating that the dissociation may be through solubilisation or modification of the native colloidal calcium phosphate. There was also an increase in the levels of serum phase a-lactalbumin and blactoglobulin with increasing NaHMP addition. However, this is probably due to the whey proteins that have associated with the kcasein, which dissociated from the micelles at increasing levels when the NaHMP levels were increased. The marked dissociation of casein from the micelles on addition of NaHMP would account for the reduced casein micelle size/ scattering intensity (Fig. 4). The high levels of dissociated k-casein through the addition of NaHMP may account for the reduced micellar zeta potential (Fig. 3) as k-casein is known to be the major milk protein responsible for the surface charge of the casein micelles (Anema & Klostermeyer, 1996; Dalgleish, 1984). Relative to the RSM samples, the HRSM samples had markedly increased levels of serum phase k-casein at all NaHMP additions whereas the levels of serum phase aS-casein and b-casein markedly

100

A

80

Serum phase protein (% of total)

60 40 20 0 100

B

80 60 40 20 0 0.00

0.02

0.04

0.06

0.08

0.10

decreased (Fig. 5B). This suggests that some of the dissociated aScasein and b-casein re-associated with the casein micelles on heating, and this may account for the increased casein micelle size on heating and the smaller changes in the casein micelle sizes on NaHMP addition when compared with the RSM samples (Fig. 4B). Some of the denatured whey protein associated with the casein micelles, and this could also account for some of the increase in casein micelle size when the RSM was heated (Anema & Li, 2003b, 2003c). When the RSM was heated, the level of calcium and phosphate in the serum decreased (Fig. 2). It is likely that this precipitated calcium and phosphate associated with the residual casein micelles and the aS-casein and b-casein via their serine phosphate moieties forming larger colloidal aggregates and depleting the serum of the aS-casein and b-casein. Interestingly, Anema and Li (2000) showed that aS-casein and b-casein that was dissociated on mild heating could re-associate with the micelles when heated at higher temperatures. Despite the high heat treatment, significant levels of a-lactalbumin and b-lactoglobulin remained in the serum phase, and these levels increased with increasing NaHMP addition. The severity of the heat treatment denatured virtually all of the whey proteins, which suggests that the serum phase whey proteins are complexed with the k-casein that had dissociated from the casein micelles (Anema, 2007, 2008). Therefore, the increasing levels of serum phase a-lactalbumin and b-lactoglobulin as the level of NaHMP is increased is likely to be due to the increased dissociation of the kcaseinewhey protein complexes from the casein micelles in the HRSM samples. This study has shown that the addition of NaHMP to milk caused major changes to the milk system, many of which would be expected to destabilise the milk to further processing such as a reduction in zeta potential and the dissociation of k-casein from the casein micelles. However, the addition of NaHMP generally has positive effects on the properties of the milks. The likely mechanism by which NaHMP addition modifies the properties of the milks is from the marked reduction in the serum phase calcium and in particular the ionic calcium levels. The serum phase calcium and/or ionic calcium levels are reported to markedly affect the properties of milk products (Deeth & Lewis, 2015; Lewis, 2010). In particular the fouling propensity, heat stability and subsequent storage stability of milk is intimately linked to the calcium content and distribution within the milk. The fouling of heated surfaces is increased with addition of calcium salts, whereas chelation of calcium with phosphates, including NaHMP markedly reduced fouling rates (Bansal & Chen, 2006; Daufin et al., 1987; de Jong, 1997; Prakash et al., 2007). Sedimentation levels in UHT milk are markedly increased when calcium salts are added to the milk. In contrast, sedimentation is markedly reduced when NaHMP or other calcium chelators are added to the milk (Lewis, Grandison, Lin, & Tsioulpas, 2011). It has been proposed that the addition of NaHMP retarded the age gelation of UHT milk by binding to the casein micelles and preventing dissociation of the k-casein/whey protein complexes (McMahon, 1996). Although NaHMP does bind to the casein micelles (de Kort et al., 2011; de Kort, 2012; Vujicic et al., 1968), it induces increased levels of dissociation of k-casein from the micelles (Fig. 5). Therefore this hypothesis that NaHMP prevents the dissociation of the k-casein/whey protein complexes from the micelles cannot be correct.

NaHMP in liquid milk (% ) Fig. 5. Effect of sodium hexametaphosphate on the serum phase proteins in supernatants from reconstituted skim milk (A) and heated reconstituted skim milk (B) samples: C, aS-casein; B, b-casein; ;, k-casein; , a-lactalbumin; -, b-lactoglobulin. Error bars represent the standard deviations for repeated measurements.



4. Conclusions The results of this study have shown that the addition of NaHMP to milk causes major changes to the chemical and physical

S.G. Anema / International Dairy Journal 50 (2015) 58e65

properties of the milk. These include a decrease in serum phase calcium and ionic calcium levels, increases in the serum phase orthoP levels, a decrease in the magnitude of the micellar zeta potential and particle size and a marked increase in the dissociation of casein, particularly k-casein, from the casein micelles. The dissociation of k-casein increased further on additional heating. Acknowledgements The author would like to thank David Newstead for manufacturing the milk powders and Bertram Fong for providing the samples of powder used in this study.

References Abdulina, T. N. (1975). The effect of some salts and of heat on structural changes in casein. Izvestiya Vysshikh Uchebnykh Zavedenii, Pishchevaya Tekhnologiya, 177e179. Abdulina, T. N., Kovalenko, M. S., & Alekseev, N. G. (1970). Changes in mean size of casein particles and viscosity of milk caused by addition of stabilizing salts. Industria del Latte, 31e34. Alkanani, T., Friel, J. K., Jackson, S. E., & Longerich, H. P. (1994). Comparison between digestion procedures for the multielemental analysis of milk by inductivelycoupled plasma-mass spectrometry. Journal of Agricultural and Food Chemistry, 42, 1965e1970. Allen, R. J. L. (1940). The estimation of phosphorus. Biochemical Journal, 34, 858e865. Anema, S. G. (2007). Role of k-casein in the association of denatured whey proteins with casein micelles in heated reconstituted skim milk. Journal of Agricultural and Food Chemistry, 55, 3635e3642. Anema, S. G. (2008). On heating milk, the dissociation of k-casein from the casein micelles can precede interactions with the denatured whey proteins. Journal of Dairy Research, 75, 415e421. Anema, S. G. (2009a). Effect of milk solids concentration on the pH, soluble calcium and soluble phosphate levels of milk during heating. Dairy Science & Technology, 89, 501e510. Anema, S. G. (2009b). The use of “lab-on-a-chip” microfluidic SDS electrophoresis technology for the separation and quantification of milk proteins. International Dairy Journal, 19, 198e204. Anema, S. G., & Klostermeyer, H. (1996). z-Potentials of casein micelles from reconstituted skim milk heated at 120  C. International Dairy Journal, 6, 673e687. Anema, S. G., & Li, Y. (2000). Further studies on the heat-induced, pH-dependent dissociation of casein from the micelles in reconstituted skim milk. Lebensmittel Wissenschaft und Technologie, 33, 335e343. Anema, S. G., & Li, Y. (2003a). Re-equilibration of the minerals in skim milk during reconstitution. Milchwissenschaft, 58, 174e178. Anema, S. G., & Li, Y. (2003b). Effect of pH on the association of denatured whey proteins with casein micelles in heated reconstituted skim milk. Journal of Agricultural and Food Chemistry, 51, 1640e1646. Anema, S. G., & Li, Y. (2003c). Association of denatured whey proteins with casein micelles in heated reconstituted skim milk and its effect on casein micelle size. Journal of Dairy Research, 70, 73e83. Anema, S. G., Lowe, E. K., Lee, S. K., & Klostermeyer, H. (2014). Effect of the pH of skim milk at heating on milk concentrate viscosity. International Dairy Journal, 39, 336e343. Anema, S. G., Lowe, E. K., & Li, Y. (2004). Effect of pH on the viscosity of heated reconstituted skim milk. International Dairy Journal, 14, 541e548. Anema, S. G., & McKenna, A. B. (1996). Reaction kinetics of thermal denaturation of whey proteins in heated reconstituted whole milk. Journal of Agricultural and Food Chemistry, 44, 422e428. Baldwin, A. J., & Truong, G. N. T. (2007). Development of insolubility in dehydration of dairy milk powders. Food and Bioproducts Processing, 85, 202e208. Bansal, B., & Chen, X. D. (2006). A critical review of milk fouling in heat exchangers. Comprehensive Reviews in Food Science and Food Safety, 5, 27e33. Burdett, M. (1974). Effect of phosphates in lowering amount of deposit formation during heat-treatment of milk. Journal of Dairy Research, 41, 123e129. Chaplin, L. C. (1984). Studies on micellar calcium-phosphate e composition and apparent solubility product in milk over a wide pH range. Journal of Dairy Research, 51, 251e257. Codex Alimentarius Commission. (2011). Milk and milk products (2nd ed.). Codex Alimentarius.

65

Dalgleish, D. G. (1984). Measurement of electrophoretic mobilities and z-potentials of particles from milk using laser Doppler electrophoresis. Journal of Dairy Research, 51, 425e438. Dalgleish, D. G., & Law, A. J. R. (1989). pH-induced dissociation of bovine casein micelles. II. Mineral solubilization and its relation to casein release. Journal of Dairy Research, 56, 727e735. Datta, N., & Deeth, H. C. (2001). Age gelation of UHT milk e a review. Food and Bioproducts Processing, 79, 197e210. Daufin, G., Labbe, J. P., Quemerais, A., Brule, G., Michel, F., Roignant, M., et al. (1987). Fouling of a heat-exchange surface by whey, milk and model fluids e an analytical study. Lait, 67, 339e364. Deeth, H. C., & Lewis, M. J. (2015). Practical consequences of calcium addition to and removal from milk and milk products. International Journal of Dairy Technology, 68, 1e10. Gao, R., van Halsema, F. E. D., Temminghoff, E. J. M., van Leeuwen, H. P., van Valenberg, H. J. F., Eisner, M. D., et al. (2010). Modelling ion composition in simulated milk ultrafiltrate (SMUF) II. Influence of pH, ionic strength and polyphosphates. Food Chemistry, 122, 710e715. Harwalkar, V. R. (1992). Age gelation of sterilized milks. In P. F. Fox (Ed.), Advanced dairy chemistry: Vol. 1. Proteins (pp. 691e734). London, UK: Elsevier. Holt, C. (1985). The milk salts: their secretion, concentrations and physical chemistry. In P. F. Fox (Ed.), Developments in dairy chemistry-3. Lactose and minor constituents (pp. 143e181). London, UK: Elsevier Applied Science Publishers. de Jong, P. (1997). Impact and control of fouling in milk processing. Trends in Food Science and Technology, 8, 401e405. Kalab, M., & Harwalkar, V. R. (1974). Milk gel structure. II. Relation between firmness and ultrastructure of heat-induced skim-milk gels containing 40e60% total solids. Journal of Dairy Research, 41, 131e135. Kapoor, R., & Metzger, L. E. (2008). Process cheese: scientific and technological aspects e a review. Comprehensive Reviews in Food Science and Food Safety, 7, 194e214. Kocak, H. R., & Zadow, J. G. (1985). Polyphosphate variability in the control of age gelation in UHT-milk. Australian Journal of Dairy Technology, 40, 65e68. de Kort, E. J. P. (2012). Influence of calcium chelators on concentrated micellar casein solutions: From micellar structure to viscosity and heat stability (PhD thesis). The Netherlands: Wageningen University. de Kort, E., Minor, M., Snoeren, T., van Hooijdonk, T., & van der Linden, E. (2009). Calcium-binding capacity of organic and inorganic ortho- and polyphosphates. Dairy Science and Technology, 89, 283e299. de Kort, E., Minor, M., Snoeren, T., van Hooijdonk, T., & van der Linden, E. (2011). Effect of calcium chelators on physical changes in casein micelles in concentrated micellar casein solutions. International Dairy Journal, 21, 907e913. de Kort, E., Minor, M., Snoeren, T., van Hooijdonk, T., & van der Linden, E. (2012). Effect of calcium chelators on heat coagulation and heat-induced changes of concentrated micellar casein solutions: the role of calcium-ion activity and micellar integrity. International Dairy Journal, 26, 112e119. Lewis, M. J. (2010). The measurement and significance of ionic calcium in milk e a review. International Journal of Dairy Technology, 63, 1e13. Lewis, M., Grandison, A., Lin, M.-J., & Tsioulpas, A. (2011). Ionic calcium and pH as predictors of stability of milk to UHT processing. Milchwissenschaft e Milk Science International, 66, 197e200. McMahon, D. J. (1996). Age-gelation of UHT milk: changes that occur during storage, their effect on shelf life and the mechanism by which age-gelation occurs. IDF special issue 9602: Heat treatments and alternative methods (pp. 315e325). Brussels, Belgium: International Dairy Federation. Mittal, S. B., Hourigan, J. A., & Zadow, J. G. (1990). Effect of added sodium hexametaphosphate on certain technological aspects of UHT recombined milk. Australian Journal of Dairy Technology, 45, 1e4. Odagiri, S., & Nickerson, T. A. (1965). Complexing of calcium by hexametaphosphate oxalate citrate and ethylenediamine-tetraacetate in milk. II. Dialysis of milk containing complexing agents. Journal of Dairy Science, 48, 19. Pouliot, Y., Boulet, M., & Paquin, P. (1989). Observations on the heat-induced salt balance changes in milk. 1. Effect of heating time between 4 and 90  C. Journal of Dairy Research, 56, 185e192. Poynton, T. A. (2007). EZAnalyze (version 3.0) [computer software and manual] (Vol. 2008). http://www.ezanalyze.com. Prakash, S., Datta, N., Lewis, M. J., & Deeth, H. C. (2007). Reducing fouling during UHT treatment of goat's milk. Milchwissenschaft e Milk Science International, 62, 16e19. Sanders, G. P. (1931). The determination of the calcium, magnesium, and acidsoluble phosphorus of milk by means of trichloroacetic acid filtrates. Journal of Biological Chemistry, 90, 747e756. Tsioulpas, A., Koliandris, A., Grandison, A. S., & Lewis, M. J. (2010). Effects of stabiliser addition and in-container sterilisation on selected properties of milk related to casein micelle stability. Food Chemistry, 122, 1027e1034. Vujicic, I., deMan, J. M., & Woodrow, I. L. (1968). Interaction of poly phosphates and citrate with skim milk proteins. Canadian Institute of Food Technology Journal, 1, 17e21.