The frequency distribution of aggregates of casein micelles was determined inelectron micrographs of rennet-treated skim-milk. The average degree of aggregation was constant until after 60% of the coagulation time, then increased linearly to 100% of the coagulation time and subsequently rose more slowly. The observations indicated that the rate of aggregation of casein micelles was determined by the random collision of particles, with no preference for the formation or reaction of aggregates of any particular size. The linking together of micelles was not rate-determining.

Use of a model of the renneting reaction has led to the definition of the amounts of soluble casein present in milks of different concentrations, at the time of clotting. It is shown that, although the clotting time (RCT) is relatively little affected by the concentration of milk by ultrafiltration, the proportion of casein which is soluble at the RCT is highly dependent upon the concentrations of both milk and rennet. This may affect the structure of the curd.

Positively charged materials, polyethyleneimine, poly-L-lysine and partly estcrified proteins (i.e. methyl bovine serum albumin, methyl casein and ethyl casein), in aqueous solution at the natural pH of milk, caused milk coagulation. This occurred at pH 6–6·7 at ≤ 30 °C using a defined amount of coagulant, which was quantitatively recovered in the curd. The protein and fat recovery yields were higher than with acid and rennet coagulation; the (Ca + Mg) partition between whey and curd was very similar to rennet milk clotting: the curd obtained showed high syneresis rate.

Mechanisms controlling exchange and partition of minerals and β-casein between micellar and soluble phases during cooling have been studied using suspensions of milk proteins of varying concentrations in the presence of aqueous phases having varying mineral compositions. In these studies the final equilibrium between Ca and phosphate was found to be regulated by formation of Ca phosphates of varying solubilities. The effect of the total casein level on mineral and protein equilibria was also studied. In milk, part of the β-casein content is bound to micelles by hydrophobic bonds; the proportion present in this form was increased by spontaneous micellar demineralization during cooling as well as by addition of a complexant to milk. At low temperatures, an equilibrium between the micellar and the monomeric states was reached which depends upon the total casein level and on the hydrophobic-bonded β-casein content. Not all of these hydrophobic bonds were broken when milk was cooled.

Different fractions of human κ-casein characterized by their sensitivity to chymosin were separated, after reduction, by affinity chromatography on thiol-Sepharose in a 0·1 m-Tris-HCl buffer, pH 7·0, containing 7 M-urea, 0·3 m-NaCl and 1 mm-EDTA. The fraction retained on the column and bearing the SH-groups was then chromatographed on hydroxyapatite in a 5 mm-phosphate buffer, pH 6·8, containing 0·2 m-KC1, 4·5 m-urea and 2 mm 2-mercaptoethanol. The fraction not retained on this column was purified by chromatography on DEAE-cellulose in a 6·6 m-urea, 0·02 m-imidazole, 0·02 m 2-mercaptoethanol buffer, pH 7·0. Two chymosinsensitive fractions were identified by starch-gel electrophoresis at pH 8·6; one of low electrophoretic mobility, the other a smeared band covering the whole distance of migration, in which several bands with mobilities greater than those of β-caseins could be detected.

By resuspending casein micelles in whey and dialysate it is shown that the role of whey proteins in the ethanol (EtOH)-induced coagulation of skim-milk is minimal. Experiments involving the interchange of milk sera indicated that the position of the EtOH stability/pH profile along the pH axis was governed by the diffusible constituents of the milk serum phase. The identities of those serum components governing the shape and position of the EtOH stability/pH profile were investigated. The addition of Ca2+ caused a shift in the entire profile to higher pH. Reduction of the available Ca2+ by addition of EDTA (up to 5 ran) shifted the profile to lower pH. The addition of phosphate (up to 5 mM) or citrate (up to 1 mM) had no effect on the profile, though higher concentrations of citrate (up to 5 mMi) caused slight shifts to lower pH. When equimolar amounts of Ca and phosphate were added, the system showed a shift in profile approximately equivalent to that of the free Ca introduced. Increasing the ionic strength of a milk by the addition of NaCl did not shift the profile, but decreased the maximum EtOH stability of the high pH arm of the sigmoidal profile. The EtOH stability/pH profile retained the same sigmoidal shape in all cases.

The isotherms for Ca2+ binding to bovine β-casein have been measured at 5 temperatures in the range 4–40 °C and at 4 different ionic strengths. The results are interpreted by an interactive-site binding model, and are compared with results previously obtained on αs1-casein. The affinity of β-cesein for the first Ca2+ to bind is similar to the affinity of αsl-casein for the same binding event: however, binding of subsequent Ca2+ to β-casein is weaker than the binding to αs1-casein. The results are discussed in terms of precipitability of the 2 caseins caused by the binding of Ca2+.

Whole casein preparations obtained from paired quarters (one healthy and one subclinically infected) of each of 6 individual cows were fractionated quantitatively by column chromatography on hydroxyapatite. Infection was associated with an apparent decrease in the αs- and β-casein fractions and increase in the γ- and κ-casein fractions. The increase in the κ-casein fraction was attributable mainly to enrichment with chymosin-resistant minor components of casein.

The profile produced by column chromatography on DEAE-cellulose of casein isolated from a healthy quarter of a single cow differed significantly from that from a subclinically infected quarter of the same cow. These differences were increased by incubation of the quarter milks at 37 °C for 8 h. Extensive fragmentation of [14C]methylated αs2- and β-casein, and less extensive fragmentation of [14C]methylated αs1-casein, occurred during incubation of these proteins in milk from an infected quarter. Fragmentation was almost entirely inhibited by soybean trypsin inhibitor. From a consideration of the chromatographic behaviour of the radiolabeled fragments on DEAE-cellulose, and other data presented, it was concluded that proteolysis of caseins by a plasmin-like enzyme was sufficient to account, in qualitative terms, for most of the compositional changes exhibited by casein from subclinically infected quarters.

Dialysing milk against phosphate-free sera showed that the transition in the ethanol (EtOH) stability/pH profile was associated with the soluble phosphate component of the milk serum. Sigmoidal behaviour similar to that of milk was reproduced when the EtOH stability of artificial mixtures of casein, Ca and phosphate was measured as a function of pH. A mechanism for the s coagulation of skim-milk is discussed.

A modegl is proposed for the equilibria between the components of the salts in milk. The model includes complex formation between calcium ions and the various ionized forms of citric, lactic and phosphoric acids, and makes allowance for the effect of ionic strength. A computer program has been written to calculate the pH of the milk salt solutions and the concentration of each of the complexes formed. Calculated pH values agree with observed values for solutions of known composition.

A method for determining chymosin and bovine pepsin A in commercial extracts of bovine veils, based on the use of the synthetic hexapeptide (Leu-Ser-Phe(NO2)-Nle-Ala-Leu-OMe) as reference substrate, is reported. Chymosin and bovine pepsin A were separated chromatographically from extracts and assayed for clotting activity on a reconstituted skim-milk standardized with reference chymosin and bovine pepsin A, themselves standardized in relation to the hexapeptide. The effect of pH on the absorbance difference between the hexapeptide and the Leu-Ser-Phe(NO2) tripeptide resulting from its hydrolysis was studied. It was found that the ‘optimal’ pH for determining the activities of the reference enzyme solutions was 4·7.

Six chymosin and 3 bovine pepsin A preparations were assayed on the hexapeptide to define the relation between the proteolytic activity and the amount of active enzyme. At pH 4·7 and 30 °C 1 mg chymosin and 1 mg bovine pepsin A hydrolysed 100 and 2700 μm-peptide/s respectively. The clotting activity of these preparations was assayed on a reconstituted skim-milk to standardize it and thus define the relation between the clotting time and the amount of active enzyme. Chymosin had a specific clotting activity twice that of bovine pepsin A. At equal clotting activities, bovine pepsin A was 55 times more active than chymosin on the hexapeptide at pH 4·7.

The thermal behaviour of β-lactoglobulin was studied by differential scanning calorimetry (DSC) in the temperature range 40–160 °C. The DSC curves revealed, in addition to the usually observed denaturation peak near 80 °C, a distinct endothermal peak between 130 and 150 °C. When the pH was increased from 6·5, the area under the peak near 80 °C (denaturation heat) decreased significantly, whereas the peak area near 140 °C increased. The temperature of maximum heat absorption in the peaks near both 80 and 140 °C gradually increased as the pH decreased. Addition of sugars and variation of the heating rate both caused a temperature shift of the endothermal heat effect at 140 °C, similar to that at 80 °C. No peak near 140 °C was observed when β-mercapto-ethanol was added to the β-lactoglobulin solution before scanning. The origin and nature of the high temperature denaturation peak is discussed in terms of conformational changes of the protein.

The computer program described by Wood, Reid & Elvin (1981) has been extended to include the complexes formed in alkaline solutions of Ca citrate and phosphate. Mg complexes are also included. Supersaturation with respect to various solid phases is tested for and the composition of the aqueous phase adjusted accordingly. Additional tests of the program have confirmed its accuracy and shown how the results depend on the values chosen for the relevant equilibrium constants.

The effect of physicochemical characteristics (pH, temperature, composition) of the aqueous phase on the mineral balance in milk and milk retentate has been studied. The ratio of colloidal Ca to total protein decreased with pH, but at any given pH the higher the protein concentration, the higher was the ratio of colloidal Ca to total protein. The solubilization of Ca during cooling and the decrease in soluble Ca during heating were approximately the same in retentates and in milk. Among the components of the aqueous phase, soluble Ca and citrate ions were related to the amount of colloidal Ca.

The main components of bovine whole casein were characterized by electrofocusing; pi values of αs1-, β-, κ-, γ1-, γ2- and γ3-caseins were determined. A further identification of casein components was achieved by a 2-dimensional electrophoresis study. 2-Dimensional patterns of γ-caseins obtained from a hydrolysate of β-cascin by bovine plasmin are in good agreement with those of γ-caseins naturally present in whole casein.

Quality characteristics of goat's milk yoghurt made from homogenized and non-homogenized concentrates prepared by vacuum evaporation, addition of dried skimmed goat's milk, reverse osmosis or ultrafiltration were observed. Yoghurts made from ultrafiltered milk gave the best flavour and viscosity, whilst homogenization of the concentrates also increased these qualities. Cow's milk yoghurts prepared in the same way were improved to an even greater extent by homogenization. Stirred goat's milk yoghurt had too low a viscosity to be acceptable, unless made from homogenized ultrafiltered milk.

Using polyacrylamide gel electrophoresis, 2-dimensional gel electrophoresis and electrofocusing, it has been shown that 4 basic fragments were produced by action of plasmin on buffalo β-casein. These fragments were present in whole buffalo casein prepared from bulk milk. Their mol. wts, 20000, 16000 and 11000 (2 components) were determined by SDS-gradient polyacrylamide gel electrophoresis. It is suggested that 3 of them are homologous to bovine γ1-, γ2- and γ3-caseins respectively, while the last one (mol. wt 16000), which has no counterpart in plasmin digest of bovine β-casein, could arise from a cleavage at bond 68–69.

Additions of potassium iodate to milk at 0·05 and 0·1 mM (10 and 20 ppm) before UHT treatment markedly reduced the rate at which pressure built up during processing. This permitted the use of longer processing times before unacceptable pressures were reached in the heat exchangers. Iodate reduced the amount of protein deposited, particularly in the higher temperature sections of the plant, but had no effect on the deposition of minerals. The more compact nature of the highly mineral deposits offered less resistance to the flow path. Reduction in the amount of protein deposited is likely to be caused by increased denaturation of β-lactoglobulin and oxidation of heat activated sulphydryl groups by the iodate, thus reducing the formation of high molecular weight polymers of sulphur-containing proteins at the heated surfaces. Increasing the level of sulphydryl groups in the milk through the addition of L-cysteine-HCl caused an increase in the amount of deposit formed during UHT treatment. Whilst little detrimental effect on the quality of the milk resulted from additions of iodate at 0·05 mM, milks with 01 mM-iodate became bitter during subsequent aseptic storage. Bitterness was a result of iodate-induced proteolysis of casein.

The interactions of 23 bacteriophages on strains of lactic streptococci at 30, 38 and 40 °C were studied in reconstituted skim-milk and broth media. In milk, 3 different temperature mediated responses were observed. Extensive phage replication at 30, 38 and 40 °C with inhibition of acid production by infected cultures was observed for 8 phages; extensive phage replication at 30 and 38 °C, but not at 40 °C, was observed for 4 phages, while 11 phages replicated at 30 °C but not at 38 or 40 °C. Studies in broth media revealed that failure of phage to replicate at 38 and 40 °C was not due to lack of host growth in most cases, but due to a temperature sensitivity on the part of the phage. In one phage–host system, phage replication continued after growth of the host had ceased. The significance of these findings to starter selection for cheese manufacture is discussed.

The action of gastric proteinases on bovine caseins was studied in vivo on rats fed skim-milk, or whole casein in water or whole casein in mineral solution, by analysing the gastric content after 30 min digestion. Clotting of the diet in the stomach greatly reduced the rate of gastric emptying. The proteolysis of caseins observed by gel electrophoresis appeared to follow a different pathway for the 3 different diets. The amino acid compositions of the trichloracetic acid sediments of the stomach contents did not differ between the 3 diets. The presence of free amino acids in the stomachs at significant levels (0·8–5 % of the total amino acid content) was observed.