The use of automated tubidimetric data for the construction of kinetic models

Abstract

An automated tubidimetric instrument (Bioscreen) was used to observe the growth response ofListeria monocytogenes to combinations of temperature (15–30°C), hydrogen-ion (0.1–21.9 μm) (equivalent pH 4.66–7.0) and NaCl concentration (0.5–9.5% w/v). Compared to traditional plate count techniques, the technique allowed many more data points to be captured and replicates to be used, with less expenditure of effort. Optical density curves were filtered (smoothed) to minimize the effect of signal noise and the mean signal from uninoculated wells was subtracted to minimize the effect of signal draft. A novel procedure for fitting growth curves to optical density data has been developed. The procedure involves the use of the logistic function and a calibration equation for fitting, in a single step, in the dimension of optical density. This approach allowed the four parameters of the logistic equation to be derived at each set of experimental conditions. A quadratic response surface was then fitted to the curve parameters using temperature, NaCl and hydrogen-ion concentration as three independent variables. Predicted time to 1000-fold increase in cell numbers compared well to predictions from predictive microbial growth equations generated in other laboratories using traditional plate counting. We propose that this technique should be further evaluated as a method for generating data for modeling the kinetics of microbial growth.