Abstract:
Dairy industries are actively working on the development of new dairy ingredients and products with novel functional and physical properties. Enzymatic modification via protein crosslinking using transglutaminase (TG) is an effective way to achieve such properties. Milk is commonly used in the manufacture of a variety of dairy products and it contains casein proteins which are good substrates for TG-induced crosslinking. The main aim of this thesis is to determine the effect of TG treatment on the changes in the physico-chemical properties of low heat skim milk under acidic (~pH 6.7 to 4) and alkaline (~pH 6.7 to 10) conditions, and with added ethylenediaminetetraacetic acid (EDTA) (0 to 50 mM). Reconstituted skim milks untreated and treated with TG (~pH 6.7; 2.5 to 30% w/w), and incubated at 30oC for 15 h, were used. The susceptibility of proteins to TG crosslinking and degree of crosslinking in casein micelles in milk were determined. The changes in physicochemical properties of milks with and without TG-induced crosslinks under acidic and alkaline conditions, and with added EDTA, were analysed by different analytical physical and chemical methods. Physical methods included monitoring the viscoelastic properties during acid gelation by rheology, determining the degree of syneresis of acid milk gels, measuring the viscosity by capillary viscometry and particle size by dynamic light scattering, analysing the casein micelle fine structures by small angle X-ray scattering, and examining the casein micelles by cryogenic-transmission electron microscopy and the acid milk gels’ microstructures by confocal laser scanning microscopy. Chemical methods included determining the distribution of proteins using reversed-phase high performance liquid chromatography, determining the distribution of minerals (Ca2+, Mg2+, K+, Na+ and inorganic phosphate (Pi)) using atomic absorption and emission spectroscopy, colorimetry and 31Pnuclear magnetic resonance (31P-NMR), and characterising the calcium phosphate phases using Raman spectroscopy. The ion speciation in milk was calculated using the modified Carl Holt model. TG treatment did not alter the pH profiles and solubilisation of colloidal calcium phosphate (CCP) that would normally occur during acidification at 30, 35, 40 and 45oC. However, compared to 10% (w/w) TG untreated milk, TG treated milk (10% w/w) when acidified at ≥35oC induced differences in the elastic modulus (Gʹ) versus time curves, particularly by inhibiting the formation of the shoulder (Gʹshoulder). The G'shoulder is here proposed to be the transition of the first increase in Gʹ due to the aggregation of soluble protein complexes at early stage of acidification, to the second increase in Gʹ due to further aggregation of casein micelles (and aggregated soluble complexes) as acidification progressed. TG treatment produced firmer milk gels which were less prone to breakage at large deformation. TG treatment in milk (10% w/w) prevented the dissociation of caseins from casein micelles due to micellar swelling between ~pH 6.7 and 8. Above ~pH 8, TG treatment resisted full disruption of casein micelles and reduced the extent of casein dissociation. TG treatment demonstrated the stabilisation of casein micellar network against alkaline-induced disruption. Although TG treatment in milk altered the distribution of caseins between serum and micellar phase during acidification and alkalinisation, the mineral equilibrium was independent of the presence of TG-mediated crosslinks. Acidification resulted in the solubilisation of CCP; whereas alkalinisation resulted in the precipitation of calcium phosphate for both types of milk (with and without TG). Raman spectroscopy demonstrated that alkalinisation did not result in the transformation of amorphous calcium phosphate into crystalline phases. TG treatment in milk (10% w/w) did not influence the solubilisation of CCP upon the addition of EDTA, but reduced the extent of micellar disruption when >15 mM EDTA was added. In addition to the use of colorimetry method to measure the soluble Pi concentration, 31P-NMR was shown to be an adequate and rapid method to measure Pi without the need to remove the milk proteins. The ion equilibria in milk systems with added EDTA were successfully predicted using the modified Carl Holt model on the speciation of milk minerals, taking into account the addition of EDTA. The findings of this thesis might be used in industrial applications. For example, TG crosslinking of caseins might be used in yogurt manufacture to minimise syneresis; and the addition of EDTA to dissociate caseins might be used to improve the heat stability of milkbased systems.