Ordered structures as observed in bilayer fruit (bottom)/yoghurt (top) products are described. Systematic studies were performed on model systems, where the yoghurt layer was modelled by milk and the fruit layer by sugar solutions or sweetened agar gels (5% to 40% (w/w) sugar). The structures formed were regularly spaced and grew vertically from the interface, only upwards in agar gel/milk or fruit/yoghurt systems, but also downwards in sugar solutions/milk systems. The mechanism previously proposed for the development of similar structures in other multicomponent bilayer systems was used to explain this behaviour. The starting point for the development of the fingers was a water/sugar diffusive transport between the two phases, due to an osmotic pressure gradient. With protein present ill the upper phase, these mass transfers caused a density inversion at the boundary that led to a convective flow, the upper phase tending to sink and the lower phase tending to float. Once the instability appeared, the fingers started to grow and continued to do so, owing to exchange of water and sugar between the upward- and downward-growing fingers.
Enzymic hydrolysis of lactose by beta-galactosidase increased the heat stability of milk or concentrated milks, especially raw milk. The possibility that the participation of galactose and glucose (produced on hydrolysis pf lactose) in the Maillard reaction is responsible for the stabilizing effect is discussed.
Proteolysis during storage of a semi-hard cheese, made from milk fortified with a whey protein concentrate, was assessed by electrophoresis and isoelectric focusing. Casein degradation was not affected by the addition of denatured whey proteins, but was strongly influenced by the kind of clotting enzyme used for cheesemaking. Denatured beta-lactoglobulin was not hydrolysed up to the end of the ripening period.
Salt transfer in white cheese during initial short-term brining (1-3 days) was investigated at two concentrations of brine (150 and 200 g/kg) and at two temperatures (4 and 20 degrees C). Two mathematical models were developed for the semi-infinite cheese geometry by neglecting and considering convective flow. Diffusion was the governing transport mode in the process. Brine concentration and temperature had insignificant and significant effects, respectively, on salt diffusivity in the cheese. Salt diffusivity decreased with increasing salt content of the cheese. Effective salt diffusivity was estimated to be 0.37 . 10(-9), 0.25 . 10(-9) and 0.22 . 10(-9) m(2)/s at 4 degrees C with brining for 1, 2, and 3 d, respectively, and 0.51 . 10(-9), 0.42 . 10(-9) and 0.39 . 10(-9) m(2)/s at 20 degrees C for 1, 2, and 3 d, respectively. Semi-infinite modelling described the process well with the average deviation between experimental and theoretical data less than 5%.
Eight triazinic dyes were assayed as ligands for chromatographic affinity purification of lactoferrin from rennet whey, at pH 5.0, 7.0 and 9.0. Blue R-HE, Orange R-HE and Scarlet G-A did not adsorb lactoferrin at any pH, whereas Yellow FR adsorbed a small amount of lactoferrin and showed poor selectivity. Red HE-3B was most selective and lactoferrin was most adsorbed at pH 5.0 and 7.0. Blue F3-GA, Red HE-7B and Red F-5B adsorbed intermediate amounts of lactoferrin. At pH 9.0, Blue F3-GA, Red HE-3B and Red F-5B adsorbed intermediate amounts of lactoferrin. Red HE-7B maintained its low lactoferrin adsorption and good selectivity. At pH 5.0-7.0, lactoferrin shows high adsorption on Red HE-3B and at pH 7.0, selectivity was better than at pH 5.0, because lactoferrin was accompanied by only a small amount of beta-lactoglobulin. Red HE-3B was selected for tests on ionic-strength gradient elution. At pH 7.0, lactoferrin eluted with NaCl at about 1 mol/L, and the contaminating beta-lactoglobulin did so with NaCl at 0.3 mol/L. Treatment of immobilized Red HE-3B with sodium dithionite and sodium nitrite did not improve separation and considerably decreased recovery. A technique was developed for batch purification based on data from gradient elution. The procedure allowed extraction of 82% of the lactoferrin whey content with a purity of 98%.
Many current processes for the commercial manufacture of oil-continuous emulsions utilise a phase inversion step from an oil-in-water (o/w) to a water-in-oil (w/o) emulsion. Whilst the operators of such processes may have practical experience in the control of inversion, little knowledge is available regarding the precise mechanisms and the controlling factors. The aim of this study was to examine the complete phase inversion of coarse (> 10 mu m) o/w emulsions which were cooled and sheared simultaneously. Batch experiments were conducted in a small stirred and jacketed vessel in which an o/w emulsion was prepared. The emulsion was then cooled with continuous stirring. Measurements of torque and conductivity during the cooling stage allowed monitoring of the phase inversion process. Factors such as fat level and type,emulsifier level and type, and different cooling regimes, were investigated. In addition, we studied the factors controlling oil droplet coalescence to help determine the underlying mechanism in phase inversion and to help identify the rate-limiting step in the process. From these studies and previously published results a mechanism for phase inversion has been proposed in which the rate-limiting step is the entrapment of water within the oil phase. Entrapment results from multi-oil droplet collisions in which film drainage is controlled by the presence of crystals at the interface. A mechanism is proposed in which the fat crystals limit oil droplet deformation. The primary role of the emulsifier is to position the fat crystals at the interface. A secondary role of some emulsifiers is to impose a steric barrier to film drainage, resulting in longer phase inversion times.
NPN (Non Protein Nitrogen) has a pronounced influence on the seasonal pattern of the casein number (Casein/Apparent Protein) of milk. A similar but smaller seasonal pattern appeared however after eliminating the NPN effect by calculating the Cn/TP (casein/true protein) ratio. Three possible causes of the residual variation are examined. Variation of the contribution of cows having different genetic protein codes to the milk produce has a very minor effect. A shift of the Cn/TP ratio during lactation of certain cows may have some effect. The effect of rather extreme cattle feed compositions seems to be relative small.
Heat-induced changes in the proteins of recombined milk were studied using centrifugation, electrophoresis and electron microscopy techniques. The results showed that, during the initial stages of heating at pHs in the range 6.3 to 7.1 at 130 degrees C whey proteins denatured and associated with kappa-casein in the fat globule surface and casein micelles, the extent of association being lower at a higher pH of the milk at heating. The dissociation of kappa-casein from the fat globule surface and casein micelles increased with increasing pH of the milk al heating; the dissociation from the fat globule surface was more sensitive to pH changes in the range from 6.3 to 6.7. This was followed by a rapid increase in the percentage of total protein and fat sedimented at 18 000 g for 20 min; the effect was largely independent of the pH of the milk at heating in the range from 6.5 to 7.1. Continued heating of pH 6.7 milk at 130 degrees C caused a relatively small change in the aggregation of protein and fat particles until, just before coagulation, a rapid aggregation occurred. The caseins and whey proteins, both at the fat globule surface and in the serum, were modified through the formation of new non-disulphide covalent cross-links as indicated by SDS-electrophoresis, Electron microscopic examination of heated milks showed that, when recombined milk was heated at below pH 6.7, the surface of casein micelles, both the free micelles in the serum and those adsorbed on to the fat globule surface, developed many appendages (possibly aggregated whey proteins). However, the surface of adsorbed and free casein micelles in recombined milk heated at pH 7.1 was not intact and appeared to be free of whey proteins. Samples heated close to the coagulation time at pH 6.7 clearly showed the formation of chains consisting of fat globules and casein particles in which individual fat globules and casein particles largely retained their original shapes. The fat globules appeared to be linked in chains via the adsorbed casein micelles.
Changes in the size of casein micelles caused by the addition of cleaning agent components to milk were investigated. Detergents (two anionic and one nonionic), urea and sodium tripolyphosphate in concentrations up to 7 g/L were added to skim-milk. The method of measuring average micelle size was based on the wavelength dependence of turbidity. Small concentration-dependent changes in micelle size were noticed in the presence of added detergents. Addition of urea caused hardly any change. The largest increase in the size of micelles was observed when sodium tripolyphosphate was added to milk. At a sodium tripolyphosphate concentration of 5 g/L, average micelle radius was 70% more than in milk free of sodium tripolyphosphate.The dependence of size of micelles on pH in the absence and presence of an anionic detergent and sodium tripolyphosphate was also studied. An increase in pH caused an increase in the average radius of casein micelles. The presence of an anionic detergent retarded growth in size of micelles after raising the pH, whereas the presence of sodium tripolyphosphate accelerated this process.
A screening test for residues of antibiotics in milk at the Maximum Residue Limit (MRL) has been developed. By using capacitance measurement in a growth medium with different combinations of test organisms and pH, sufficient sensitivity was obtained for a range of antibiotics. Most of the antibiotics tested could be detected at the MRL.
The formation of a cheese-like flavour from amino acids and peptides, by the action of enzymes from Lactococcus lactis subsp. cremoris B78, a Gouda cheese starter organism, was investigated. Cell-free extract (CFE) was obtained from the organism after ultrasonic disruption of cells and subsequent removal of cell debris by centrifugation. The CFE was ultrafiltered to remove low-molecular-mass compounds (<500 Da). The retentate (CFE>500) was used as an enzyme source in incubation experiments with various mixtures of amino acids in water and with peptide solutions. These peptide solutions contained the alpha(sl)-casein peptides alpha(sl)-CN(f1-23) and alpha(sl)-CN(f24-199) or a mixture of peptides of molecular weight 500-5000 Da, isolated from Gouda cheese. The solutions were incubated aseptically for one week at 25 degrees C. Sensory analysis showed that a cheese-like flavour developed in amino acid mixtures containing free methionine or in mixtures of peptides containing methionine. In the latter case, methionine was released from the peptides by the action of proteolytic enzymes in the CFE>500. The formation of volatile sulphur compounds from methionine could be demonstrated by gas chromatography-mass spectrometry. The results indicated that non-proteolytic enzymes from mesophilic lactococci are important for flavour formation in cheese.
Small-deformation oscillatory shear measurements have been made at neutral pH on concentrated heat-set emulsion gels containing pure beta-lactoglobulin as the sole emulsifying agent and various amounts of small-molecule surfactant added after emulsification. Emulsion gels were prepared ill situ by a standardized thermal processing treatment (30 minutes at 90 degrees C). The emulsion gel storage modulus G' at 30 degrees C has been found to be very sensitive to thr type and amount of the added surfactant (expressed in terms of the surfactant/protein molar ratio R), and the overall rheological behaviour has been found to be qualitatively different from that reported in earlier literature studies where the surfactant was present during emulsion formation. With nonionic Tween 20 as added surfactant, we report a substantial increase in G' at R approximate to 1, a sharp reduction at R approximate to 2, and a substantial increase again at R greater than or equal to 4. With anionic DATEM or zwitterionic lecithin present, however, we observe that G' increases continuously with the surfactant concentration. Taken together with results from previous interfacial and emulsion stability studies in our laboratory, the rheology of these beta-lactoglobulin emulsion gels can be Successfully explained in terms of the different kinds of competitive adsorption behaviour and protein-surfactant interactions occurring in the various systems.
The bacterial flora of 148 ripened Norwegian and Swedish semi-hard cheeses of adequate and defective quality was phenotypically classified. From countable plates of brain-heart infusion agar (BHI), acetate agar (AcA), M17 agar, count agar sugar-free (SFA) and wort agar, 308 isolates were randomly picked. They were tentatively assigned by key tests into lactobacillus-like (76% of the isolates), Gram-positives other than lactobacillus-like (20% of the isolates) and Gram-negatives (4% of the isolates). Half of the non-lactobacillus group was assigned to species of Bacillus: B. subtilis, B. megaterium, B. coagulans, B. cereus and B. pumilus. The other isolates of this group were cocci, some of which were assigned to Lactococcus lactis ssp. cremoris, Lc. lactis ssp. lactis, Leuconostoc mesenteroides and other leuconostocs. The 235 isolates of the lactobacillus-like group, together with 51 type and 181 reference strains, were tested on API 50CH (API Systems, Montalieu Vercieu, FR) and subjected to numerical analysis with Jaccard (S-J) coefficients and UPGMA clustering. At the S-J similarity level of 80%, 30 clusters were formed. Cluster 21 was associated with cheese of adequate quality and Cluster 11 with cheese of defective quality. Most of the isolates (71%) were assigned to the Lactobacillus paracasei/casei complex.
The inactivation of the endogenous milk enzymes alpha-L-fucosidase (EC 3.2.1.51), phosphohexoseisomerase (EC 5.3.1.9), phosphodiesterase I (EC 3.1.4.1) and alpha-mannosidase (EC 3.2.1.24) during heating of milk follows a first-order reaction. Within times of up to 800 s, alpha-fucosidase is inactivated in the temperature range between 52 degrees C and 62 degrees C (activation energy, E(A) = 378 +/- 11 kJ/mol) and phosphohexoseisomerase between 56 degrees C and 66 degrees C (E(A) = 362 +/- 8 kJ/mol), which allows a characterisation of the thermisation process. Phosphodiesterase I is inactivated during pasteurisation (up to 1600 s at 65 degrees C to 78 degrees C; E(A) = 493 +/- 9 kJ/mol). alpha-Mannosidase activity is suitable to describe heating within the lower range of high-temperature pasteurisation (up to 1000 s at 76 degrees C and 86 degrees C; E(A) = 553 +/- 11 kJ/mol). The mean values of enzyme activities in raw milks of individual cows varied between 16 and 50%. Compared to this, only 1.8 to 4.8% variation could be detected in bulk milk collecting tanks. No enzyme activity could be measured in 21 commercial UHT milk samples, indicating that no reactivation occurs. In 18 pasteurised milks, up to 15% residual activity of gamma-glutamyltransferase and up to 30% residual activity of phosphodiesterase were detected, indicating a heat treatment at the upper limits.
In various rural regions in Europe the use of industrial starter cultures for preparation of fermented milk products is not yet practised. Mesophilic lactic acid bacteria, isolated from raw milk and products from these regions and from home-brewed kefir produced in Ireland, were genotypically typed on the basis of 16S rRNA sequences. Various phenotypic characteristics were also determined. Selected strains were studied with respect to case in hydrolysis, and to the production of flavour compounds in milk, in a cheese paste (Ch-easy) model and in real cheese. Virtually all Lactococcus lactis isolates had the phenotypic characteristics of the subspecies lactis. However, 30% of these had the subsp. cremoris genotype, indicating that they differ from lactococci present in commercial cheese starters. One strain appeared to have a cell wall protease variant which differed from the known P-type proteases of lactococci in both hydrolytic activity and amino acid composition. A number of typical flavour characteristics were observed that normally are not or only poorly detected in Gouda-type cheese. Flavour formation in the Ch-easy model strongly correlated with results obtained in cheese. A number of these flavour characteristics were also detected after growth of the same strains in milk. GC-MS analysis showed the presence of high levels of 3-methylbutan-1-ol, methylaldehydes and ethylesters of fatty acids with selected bacterial strains. The results indicate that strains with 'new' or unusual properties may be present among the microflora in natural fermentations of milk.
This paper gives an overview of milk composition in relation to mass transfer from milk to cheese in the manufacture of Gouda cheese. All possible mass transfer processes are discussed and quantified as far as possible. Use is made of published data in the literature, unpublished data and some newly collected data. Contrary to what is common in the literature, this paper relates cheese yield to 100 kg paracasein instead of 100 kg milk. This approach has a close relation to common practice of filling a curd processor with such an amount of milk that a constant amount of cheese will be obtained.It was concluded that the paracasein content of milk to be used for cheese making cannot be accurately calculated from an apparent protein estimation (including non-protein nitrogen) nor from a true protein estimation, but may be derived from infra-red true protein readings in milk and its corresponding rennet whey. The calculated ratio of minerals and acids to protein in cheese was not completely in balance with the estimated ratio. It was not possible to derive seasonal variation of the amount of colloidal minerals and acids per 100 kg paracasein from available information. It seems to be a variation of almost 1 kg per 100 kg paracasein. There are also some conflicting data about the amount of serum proteins in cheese. There are three main reasons why it is better to take the paracasein content as starting point: i) its determination resembles closely the actual cheese-making process, ii) there is no problem with the (unknown) content of carbohydrate in kappa-casein (because it is removed by rennet), iii) adsorption of peptides to paracasein (such as proteose-peptone) is included in the paracasein determination.It is suggested for future research to collect accurate figures regarding the seasonal pattern of the colloidal minerals and acids associated with (para)casein. Also, additional accurate cheese-making experiments are necessary to calculate the retention of individual minerals and acids in Gouda cheese.
Duplicate samples of mixed milk (n = 13 927) were collected to compare somatic cell counting with the Somascope and the Fossomatic 360. To estimate sampling and assay variability, somatic cell count (C) was measured with the Fossomatic 360 on 295 duplicate milk samples. The total variance was 1042 x 10(6). The contribution of the measuring instrument (Fossomatic 360) and the sampling procedure were 12.7 and 87.3%, respectively. The overall variance of the paired differences of the logarithmically transformed numerical value of C (expressed in mu L(-1)) (ln{C}) of the 295 duplicate milk samples all measured with the Fossomatic 360, was 0.08. Additionally, accuracy of the Somascope was evaluated on 9558 duplicate milk samples. In one of the duplicate milk samples, C was measured with the Fossomatic 360; in the other, with the Somascope. The variance of the paired differences of ln{C} between the Somascope and the Fossomatic 360 was 0.14. Comparison of the In{C} per class (class width = 1 on the log, scale) showed 77.4, 28.1 and 11.3% overestimation of In{C}, measured with the Somascope, in the classes less than or equal to 2,2-3 and 3-4, respectively, and 1.4, 2.6 and 4.6% underestimation in the classes 6-7, 7-8 and >8, respectively, indicating a lower accuracy of the Somascope in these classes. The Pearson correlation of ln{C} between the Fossomatic 360 and the Somascope was 0.96. The Pearson correlation of In{C} of the duplicate milk samples measured twice with the Fossomatic 360 was 0.97. Simple linear regression (Somascope against Fossomatic) showed a significant intercept, b(0) (+/-s.e.) = 1.13 (+/-0.011) and the slope, b(1) (+/-s.e.) = 0.803 (+/-0.0022), differed from the ideal value of 1 (P < 0.001). The overall residual variance of the linear regression analysis was 0.08. With a weighted non-linear regression analysis, a relationship X(som) = P-1 + X(fos)/(1 + P(2)X(fos)) was calculated, where P-1, is a constant that describes the overestimate of C (greatest at lowest values) with the Somascope, and P(2)X(fos) is a term that reflects the underestimate with the Somascope at higher values. The overall prevalence of cows with C higher than 250 x 10(3) mL(-1), based on the results of the Fossomatic 360 and on the Somascope, was 20.2 and 21.9%, respectively. Test agreement (K), sensitivity and specificity for detecting cows with C > 250 x 10(3) mL(-1) using the Somascope as compared to the Fossomatic 360, was 0.89, 0.95 and 0.97, respectively. The Somascope, when correctly calibrated, is an acceptable alternative for measuring C for field trial purposes and for dairy herd improvement programme.
Some general aspects of large deformation and fracture behaviour of semi-hard and hard cheeses are discussed with special emphasis on changes due to maturation. Fracture properties are shown to be much more dependent on large inhomogeneities than the resistance against small deformation (the modulus). The modulus increases during maturation, primarily due to the decrease in water content. Fracture properties are affected by the drying up of the cheese as well as by proteolysis. These factors not only affect fracture stress and strain directly, but also indirectly by influencing energy dissipation during the deformation and fracture process. The latter may cause the dependence of fracture strain on the rate of deformation to be different for various cheeses.
This paper describes the formation of lysylpyrraline, an advanced Maillard product, in milk-like model systems (glucose or lactose mixed with sodium caseinate in a milk salt solution) during controlled heating (110-150 degrees C, up to 30 min). A reversed-phase HPLC method (using a C-18 column with a binary gradient (methanol/water) and detection at 297 nm) was used to quantify lysylpyrraline in milk-like systems after a three-step enzymatic hydrolysis of the protein. The formation of lysylpyrraline was faster in the presence of glucose than lactose for the same sugar/protein molar ratio. The formation of lysylpyrraline could be described by zero-order kinetics. The temperature dependence of lysylpyrraline formation as measured by activation energy was 109.1 kJ/mol for the system lactose/casein and 138.1 kJ/mol for the system glucose/casein. The difference is probably due to higher reactivity of glucose at higher temperatures. Kinetic modelling was done to simulate a realistic chemical pathway for formation of lysylpyrraline where both the Maillard reaction and isomerization and degradation of the sugar were taken into account. It appeared very likely that 3-deoxyglucosone (a breakdown product of the sugar as well as a Maillard reaction product) is the key compound in lysylpyrraline formation. Kinetic modelling showed that formation of 3-deoxyglucosone from the sugar was much more important than its formation in the Maillard reaction. The experiments showed that lysylpyrraline could be a useful heat-induced marker to distinguish industrially processed milk as well as a powerful tool for studying the advanced Maillard reaction.
