D.S. Horne

Casein structure, self-assembly and gelation

The literature on recent studies of casein structure and function is reviewed. Where appropriate, we try to reconcile conflicting views on the issue of secondary structure in these proteins, steering a middle course where possible. A suggestion is put forward that a coarser view, treating the caseins as block copolymers may be sufficient to rationalise much of the behaviour of these proteins in self-association, adsorption and micellar assembly.

Rennet-induced Coagulation of Milk

Abstract All aspects of rennet coagulation are included in this review, from the properties of the caseins relevant to micellar assembly and stability, through the coagulating enzymes and their activity, the methodology for monitoring aggregation and gelation, the various theoretical and empirical models adopted in the analysis of such results, including their successes and failures, as well as the effects of various processing pretreatments of the milk. Emphasis is placed on rheology as a means of studying and quantifying the curd formation process, including the monitoring of syneresis in longer-duration experiments. In the review of rennet-induced micellar aggregation, the influence of electrostatic interactions is highlighted as this has been exploited in explaining the anomalous temperature of rennet coagulation and the role of added ionic calcium in cheesemaking.

Impact of preacidification of milk and fermentation time on the properties of yogurt

Casein interactions play an important role in the textural properties of yogurt. The objective of this study was to investigate how the concentration of insoluble calcium phosphate (CCP) that is associated with casein particles and the length of fermentation time influence properties of yogurt gels. A central composite experimental design was used. The initial milk pH was varied by preacidification with glucono-delta-lactone (GDL), and fermentation time (time to reach pH 4.6 from the initial pH) was altered by varying the inoculum level. We hypothesized that by varying the initial milk pH value, the amount of CCP would be modified and that by varying the length of the fermentation time we would influence the rate and extent of solubilization of CCP during any subsequent gelation process. We believe that both of these factors could influence casein interactions and thereby alter gel properties. Milks were preacidified to pH values from 6.55 to 5.65 at 40 degrees C using GDL and equilibrated for 4 h before inoculation. Fermentation time was varied from 250 to 500 min by adding various amounts of culture at 40 degrees C. Gelation properties were monitored using dynamic oscillatory rheology, and microstructure was studied using fluorescence microscopy. Whey separation and permeability were analyzed at pH 4.6. The preacidification pH value significantly affected the solubilization of CCP. Storage modulus values at pH 4.6 were positively influenced by the preacidification pH value and negatively affected by fermentation time. The value for the loss tangent maximum during gelation was positively affected by the preacidification pH value. Fermentation time positively affected whey separation and significantly influenced the rate of CCP dissolution during fermentation, as CCP dissolution was a slow process. Longer fermentation times resulted in greater loss of CCP at the pH of gelation. At the end of fermentation (pH approximately 4.6), virtually all CCP was dissolved. Preacidification of milk increased the solubilization of CCP, increased the early loss of CCP crosslinks, and produced weak gels. Long fermentation times allowed more time for solubilization of CCP during the critical gelation stage of the process and increased the possibility of greater casein rearrangements; both could have contributed to the increase in whey separation.

Determination of molecular weight of a purified fraction of colloidal calcium phosphate derived from the casein micelles of bovine milk.

Colloidal calcium phosphate (CCP) plays a key role in the formation and integrity of casein (CN) micelles. However, limited information is available on the molecular weight (M(w)) of CCP. Recently, we theoretically derived the M(w) of CCP and the objectives of this study were to experimentally determine the M(w) of CCP. We used 2 methods to prepare CCP fractions: skim milk was enzymatically digested with either trypsin or a combination of papain and proteinase enzymes to remove most CN. The CN phosphopeptides are resistant to trypsin hydrolysis. Digestion was carried out in a membrane tube that was dialyzed against the same bulk milk used in sample preparation to remove small peptides and to minimize perturbation of CCP. After digestion, the protein contents of the enzyme-treated milks were 0.92 and 0.36% for the trypsin and papain-proteinase treatments, respectively. Size-exclusion chromatography, coupled with multi-angle laser light scattering, was used to separate the CCP-phosphopeptide fraction from the digested mixture. Simulated milk ultrafiltrate was used as a mobile phase during size-exclusion chromatography separation to try to preserve the integrity of CCP. Size-exclusion chromatography peaks, which had higher Ca and P contents than the baseline, were identified as the likely fractions containing the phosphopeptide-stabilized CCP; this peak eluted with retention times of 100 to approximately 110 min for trypsinated samples. The papain-proteinase treatment caused excessive loss of CN that were needed to stabilize CCP, which resulted in no obvious peak that had elevated Ca and P contents. Debye plots at these retention times indicated that the weight-average M(w) for the fraction prepared by trypsin was 17,450 g/mol. Attempts to estimate the M(w) of the phosphopeptides associated with CCP using sodium dodecyl sulfate-PAGE were not successful, as we did not observe any peptide bands in these gels, presumably because of their low concentration in the isolated, unconcentrated fraction. Assuming that 4 CN phosphopeptides stabilized each CCP and if the M(w) of each of these phosphopeptides was about 2,500 g/mol, then the M(w) of CCP would be around 7,450 g/mol. This experimental value was close to the theoretically-derived M(w) of 4,897 and 9,757 g/mol for tetrahedron and bi-pyramid shaped objects, respectively, when using the brushite form of calcium phosphate.

Effect of dextran and dextran sulfate on the structural and rheological properties of model acid milk gels

Various types of polysaccharides are widely used in cultured dairy products. However, the interaction mechanisms, between milk proteins and these polysaccharides, are not entirely clear. To explore the interactions between uncharged and charged polysaccharides and the caseins, we used a model acid-milk-gel system, which allowed acidification to occur separately from gelation. The effect of adding uncharged dextran (DX; molecular weight ~2.0×10(6) Da) and negatively charged dextran sulfate (DS; molecular weight ~1.4×10(6) Da) to model acid milk gels was studied. Two concentrations (0.075 and 0.5%, wt/wt) of DX or DS were added to cold milk (~0°C) that had been acidified to pH values 4.4, 4.6, 4.8, or 4.9. Acidified milks containing DX or DS were then quiescently heated at the rate of 0.5°C/min to 30°C, which induced gelation, and gels were then held at 30°C for 17 h to facilitate gel development. Dynamic small-amplitude-oscillation rheology and large-deformation (shear) tests were performed. Microstructure of gels was examined by fluorescence microscopy. Gels made with a high concentration of DX gelled at a lower temperature, but after 17 h at 30°C, these gels exhibited lower storage moduli and lower yield-stress values. At pH 4.8 or 4.9 (pH values greater than the isoelectric point of caseins), addition of 0.5% DS to acidified milk resulted in lower gelation temperature. At pH 4.4 (pH values less than the isoelectric point of caseins), addition of 0.5% DS to acidified milk resulted in gels with very high stiffness values. Gels made at pH 4.8 or 4.9 with both concentrations of DS had much lower stiffness and yield-stress values than control gels. Microstructural analysis indicated that gels made at pH 4.4 with the addition of 0.5% DX exhibited large protein strands and pores, whereas gels made with 0.075% DX or the control gels had a finer protein matrix. At higher pH values (>4.4), gels made with 0.5% DX had a finer structure. At all pH values, gels made with 0.5% DS exhibited larger pores than the control gels. This study demonstrated that low concentrations of uncharged DX did not significantly affect the rheological properties of model acid milk gels; high concentrations of DX resulted in earlier gelation, possibly caused by depletion-induced attractions between casein particles, which altered the microstructure and created weaker gels. At pH values <4.6, negatively charged DS produced stiff casein gels, which might be due to attractive crosslinking by electrostatic interactions between DS and caseins at pH values below the isoelectric pH of casein (i.e., positively charged casein regions interacted with negatively charged DS molecules).

Ethanol Stability and Milk Composition

This chapter reviews the history of the use of the ethanol stability test and surveys the milk compositional factors influencing its result. Developed originally as an indicator of fresh milk quality, it now finds use in some countries as a warning of the lack of suitability for high temperature heat treatment. Relevant factors influencing the heat stability of milk are introduced and mechanisms are proposed and compared for both instability reactions. It is concluded that a fairer option would be to develop a direct rapid heat stability method for use by field operatives.

Interactions between acidified dispersions of milk proteins and dextran or dextran sulfate.

Polysaccharides are often used to stabilize cultured milk products, although the nature of these interactions is not entirely clear. The objective of this study was to investigate phase behavior of milk protein dispersions with added dextran (DX; molecular weight = 2 × 10(6) Da) or dextran sulfate (DS; molecular weight = 1.4 × 10(6) Da) as examples of uncharged and charged polysaccharides, respectively. Reconstituted skim milk (5-20% milk solids, wt/wt) was acidified to pH 4.4, 4.6, 4.8, or 4.9 at approximately 0°C (to inhibit gelation) by addition of 3 N HCl. Dextran or DS was added to acidified milk samples to give concentrations of 0 to 2% (wt/wt) and 0 to 1% (wt/wt) polysaccharide, respectively. Milk samples were observed for possible phase separation after storage at 0°C for 1 and 24h. Possible gelation of these systems was determined by using dynamic oscillatory rheology. The type of interactions between caseins and DX or DS was probed by determining the total carbohydrate analysis of supernatants from phase-separated samples. At 5.0 to 7.5% milk solids, phase separation of milk samples occurred after 24h even without DX or DS addition, due to destabilization of caseins in these acidic conditions, and a stabilizing effect was observed when 0.7 or 1.0% DS was added. At higher milk solids content, phase separation was not observed without DX or DS addition. Similar results were observed at all pH levels. Gelation occurred in samples containing high milk solids (≥10%) with the addition of 1.0 to 2.0% DX or 0.4 to 1.0% DS. Based on carbohydrate analysis of supernatants, we believe that DX interacted with milk proteins through a type of depletion flocculation mechanism, whereas DS appeared to interact via electrostatic-type interactions with milk proteins. This study helps to explain how uncharged and charged stabilizers influence the texture of cultured dairy products.

A 100-Year Review: Progress on the chemistry of milk and its components

Understanding the chemistry of milk and its components is critical to the production of consistent, high-quality dairy products as well as the development of new dairy ingredients. Over the past 100 yr we have gone from believing that milk has only 3 protein fractions to identifying all the major and minor types of milk proteins as well as discovering that they have genetic variants. The structure and physical properties of most of the milk proteins have been extensively studied. The structure of the casein micelle has been the subject of many studies, and the initial views on submicelles have given way to the current model of the micelle as being assembled as a result of the concerted action of several types of interactions (including hydrophobic and the formation of calcium phosphate nanoclusters). The benefits of this improved knowledge of the type and nature of casein interactions include better control of the cheesemaking process, more functional milk powders, development of new products such as cream liqueurs, and expanded food applications. Increasing knowledge of proteins and minerals was paralleled by developments in the analysis of milk fat and its synthesis together with greater knowledge of its packaging in the milk fat globule membrane. Advances in analytical techniques have been essential to the isolation and characterization of milk components. Milk testing has progressed from gross compositional analyses of the fat and total solids content to the rapid analysis of milk for a wide range of components for various purposes, such as diagnostic issues related to animal health. Up to the 1950s, research on dairy chemistry was mostly focused on topics such as protein fractionation, heat stability, acid-base buffering, freezing point, and the nature of the calcium phosphate present in milk. Between the 1950s and 1970s, there was a major focus on identifying all the main protein types, their sequences, variants, association behavior, and other physical properties. During the 1970s and 1980s, one of the major emphases in dairy research was on protein functionality and fractionation processes. The negative cloud over dairy fat has lifted recently due to multiple reviews and meta-analyses showing no association with chronic issues such as cardiovascular disease, but changing consumer misconceptions will take time. More recently, there has been a great deal of interest in the biological and nutritional components in milk and how these materials were uniquely designed by the cow to achieve this type of purpose.

Letter to the Editor: Hydrophobic interactions in the caseins: Challenging their dismissal by Holt et al. (2013)

We are concerned about various assertions regarding the nature of casein interactions made in the recent review by Holt et al. (2013). Now, because this article is being cited by other researchers, we thought it would be helpful to raise these concerns to avoid misleading others. Previously (De Kruif and Holt, 2003), Holt tacitly accepted the notion of a dual-binding model (i.e., involving both attractive interactions and the formation of calcium phosphate nanoclusters as essential mechanisms for the assembly of casein micelles). But, in spite of the strong evidence in favor of their occurrence (e.g., dissociation by SDS or urea, temperaturedependent release of β-casein from micelles), Holt et al. (2013) now claim that the attractive components in casein micelle assembly are not hydrophobic interactions. They do this through a series of 3 statements. As we show below, the first statement is erroneous and also irrelevant. The second is correct but incomplete. The third misses the point of the mechanism it challenges. First, Holt et al. (2013) claim we must reconsider the view that caseins are among the most hydrophobic proteins known to science. This must have come as a surprise to Fox and Brodkorb (2008), one of their supportive citations, for they noted only that β-casein was the most hydrophobic of the caseins. Holt et al. (2013) also cite Bigelow (1967) in support of their assertion, but closer inspection of the Bigelow data places α-casein as little more than the average from his table of protein average hydrophobicities per residue. In reality, it matters little where the caseins rank in the spectrum of average hydrophobicities per residue. What matters for the development of hydrophobic interactions between these caseins is that their sequences possess clusters, or strings, of consecutive hydrophobic residues, as explained in the following. For a hydrophobic solute to exist in an aqueous environment, a cavity must be created for it, disrupting the hydrogen bond structure between the water molecules. For small solutes, the free energy penalty so incurred increases with volume; however, for larger solutes, such as clusters of hydrophobic side chains, it shifts to a dependence on the surface area of the cavity created. Bringing together 2 hydrophobic clusters of amino acids is thus energetically favorable (attractive) because the surface of the cavity accommodating the combination is reduced by the sum of the areas of their original opposing interfaces. Moreover, the now close proximity of the opposing side-chains allows a further attractive component to come into play in the form of short-range Van der Waals and other dispersion forces. Hydrophobic interactions are now recognized in the modern chemistry field as a complex concerted process involving modification or rearrangement of aqueous hydrogen bonds as well as dispersion forces between hydrophobic side-chains. Hydrophobic interactions are weak and therefore transient; however, it is their multiplicity and promiscuity that is so important in maintaining their aggregates or networks. Holt et al. (2013) make much of the example of the displacement of β-casein by denatured β-lactoglobulin. This result lies not so much in the relative strengths of individual hydrophobic interactions in these proteins but in the lateral disulfide cross-links, of which only β-lactoglobulin is capable, creating a β-lactoglobulin carpet, which is consequently more difficult to dislodge. Holt et al. (2013) also suggest that likening β-casein self-association to the formation of detergent micelles is a poor example because no compact and anhydrous domain results, but this misses the fundamental point of the analogy. The 2 structures, β-casein micelles and detergent micelles, are the operational outcome of the same physics. The participating molecules each have a charged head (repulsive) and an attractive tail. Their aggregates are in equilibrium and organize themselves so that their repulsive centers are as far away from each other as the attraction of their tails allows. The tails therefore form a central core to the micelle and the charged heads an outer halo. The core density, or packing, is a function of both the repulsive (mainly) and attractive components. For the detergent micelle, with singly charged head groups, the repulsive element is low, whereas the attraction is high, because the hydrophobic interaction of the identical hydrocarbon tails is compleLetter to the Editor: Hydrophobic interactions in the caseins: Challenging their dismissal by Holt et al. (2013)

Chapter 10:Casein Interactions: Does the Chemistry Really Matter?

‘Soft matter’ was a term first used in 1991 by Pierre-Gilles de Gennes in his acceptance address of the Nobel Prize for Physics. Since then the study of soft condensed matter has become a major growth area in physics, attracting significant funding worldwide. This sub-discipline encompasses the topi...