PREDICTIVE MODELING OF LACTOPEROXIDASE AND GAMMA-GLUTAMYL-TRANSPEPTIDASE INACTIVATION IN A HIGH-TEMPERATURE SHORT-TIME PASTEURIZER

Linear models were derived to describe the thermal inactivation of gam ma-glutamyl transpeptidase (GGTP) and lactoperoxidase (LP) in bovine m ilk in a high-temperature short-time (HTST) pilot scale pasteurizer. I ntegrated lethal effect (pasteurization effect; PE) was related by sim ple linear functions to the log(10) of the % initial activity with a p ower transformation of the PE values to improve linear fit. The PE val ues required for a 1-log reduction in GGTP and LP activity, calculated from the equation parameters were 5.46 and 26.5, respectively, sugges ting that LP is considerably more heat stable than GGTP. The two linea r equations were validated using data from additional trials. Greater variation was found with the LP assay as compared to the GGTP suggesti ng that the assay method requires improvement. These linear models can be incorporated into user-friendly software to facilitate the estimat ion of lethal effect achieved with other pasteurization equipment.