Response surface models for the growth kinetics of Escherichia coli O157:H7

Abstract

Growth curve data which had been fitted using the Gompertz function were submitted to response surface analysis to develop mathematical models for describing the effects of temperature (10-42°C), initial pH (4·5-8·5), sodium chloride concentration (5-50 g l -1), and oxygen availability (aerobic us anaerobic) on the growth of a three strain mixture of Escherichia coli O157:H7. Models were developed for the Gompertz B and M terms only after supplemental analyses indicated that the growth kinetics of the organism were independent of inoculum size and that the maximum population density achieved by E. coli were largely independent of the cultural variables. A total of 193 aerobic and 145 anaerobic growth curves representing 84 and 71 distinct variable combinations, respectively, were evaluated, with separate models being generated for the aerobic and anaerobic data. The data were analyzed using natural logarithm and square root transformations in combination with quadratic and cubic models. Subsequent evaluations indicated that the most effective response surface models were those based on a logarithmic transformation in combination with a quadratic model. These models provide a rapid means of estimating the effects of the four variables on the growth of E. coli O157:H7.