The effect of heat treatment on lysine availability and dye binding capacity of skim milk : a thesis presented in partial fulfilment of the requirements for the degree of Master of Technology at Massey University
The reported work on changes in lysine content in milk and dried milk is examined. The cause of these losses, the Maillard reaction, and the methods of lysine determination are discussed. All methods have recognised faults. Little information is available to the food processor regarding the kinetics of these losses, and the methods of their determination are not simple enough for routine quality control application. Although the lysine content of milk products determined after acid hydrolysis is known to be higher than nutritional studies indicate the causes of this are being established. Therefore acid hydrolysis in conjunction with a GLC method of amino acid analysis was adopted after some modification. (It was found that dialysis of the milk prior to hydrolysis resulted in cleaner chromatograms and that as the recovery of several amino acids, such a proline, leucine, and isoleucine, was not affected by heat treatment then these were used as internal 'internal standard'.) No simple rate expression could be found to fit the kinetics of the loss of acid released lysine. A first order model requiring the losses to be increased by a factor of 3.43 was devised and this could be used to satisfactorily predict values for acid available lysine in the heat treated milk. The possibility of the 3.43 factor being due to the regeneration of lysine by acid from Maillard intermediates, although requiring assumptions, was found to be not unreasonable. The energy of activation of the reaction leading to a loss in acid released lysine at 31.5 Kcal/mole is similar to literature values while the model value of 37.2 Kcal/mole is rather higher. The literature findings of little or no loss of lysine during pasteruization, evaporation, and sterilization of milk are supported. The technique of protein determination by dye binding was examined and applied to following changes in lysine in heat treated milk. The inconsistencies in reported work on dye binding is of little consequence as relative changes only are required. Changes in dye binding using amido black did not follow simple order kinetics, even when allowance was made for the constant binding by arginine and histidine. A first order model requiring the changes to be increased by a factor of 3.68 was developed. About 46% of this factor can be explained by assuming constant binding by arginine and histidine, the remainder of the factor possibly being due to Maillard intermediates binding dye, and/or a change in binding stoichiometry occurring. From the model it is possible to predict the observed changes in dye binding. Literature findings were supported. The energy of activation for the dye binding changes is 28.6 Kcal/mole, and for the model, 30.8 Kcal/mole. Ancillary investigations showed that the concurrent colour changes due to heat treatment have an energy of activation of about 30 Kcal/mole, and that there is a relationship between colour and dye binding capacity in heat treated milk. The relationship between the Pro-Milk and a typical absorbance spectrophotometer was determined, and an expression found which would enable a spectrophotometer to be used for protein determination.