Modelling pseudomonad growth in milk and milk-based products

Predictive microbiology is a method by which the growth responses of microorganisms of concern are modelled mathematically in respect of the major factors affecting growth. A model for the growth of Pseudomonas (the main organism of concern in refrigerated milk and some milk products) was developed and validated. Psychrotrophic psuedomonads were isolated from various modified and whole milks, using both Victorian (obtained from CSIRO Dairy Research Laboratories) and Tasmanian milks. Growth rates in artificial broth media were determined using turbidimetric methods. By fitting a sigmoidal curve (using the Gompertz function) to the data, generation times can be calculated and growth rates (by turbidimetric or viable count methods) determined. All strains were grown in artificial broth media at 10°C and growth rates compared. The fastest growing strain (Pseudomonas putida 1442) was used to develop the model i.e., a worst case approach was adopted. A square root model (Ratkowsky et al, 1982) was developed for P.putida 1442 in artificial broth media by determining the generation time at 0.5°C intervals from 0 to 50 °C. A similar procedure was used to model the effect of water activity (using sodium chloride as the solute) on Pseudomonas spp. Generation times calculated by turbidimetric and viable counts were found to differ. This difference was found to be constant with respect to temperature and was incorporated into the modelling process so that all models expressed generation times equivalent to those calculated by viable counts, the standard method for enumerating microorganisms in food products. A literature search comparing the notional minimum temperature (T min) for growth of psychrotrophic pseudomonads found that the Tmin was the same (265.4K ± 0.7) despite the source of the organism. This implies that only one model for temperature dependence is needed for psychrotrophicPseudomonas in dairy, meat and poultry products. The model was validated both in the laboratory and in industry using various milk and milk based products. The validation process involved monitoring the growth of pseudomonads at various temperatures and comparing the observed generation times to those predicted using bias and accuracy factors (Ross, 1993). The temperature model has been incorporated into prototype computer software and trialled in the dairy industry. The validation process showed the psychrotrophic pseudmonad model to accurately predict the growth of pseudomonads in the products tested. In some instances in industry, further information in the form of the duration of lag phase, is required to maximise the accuracy of the model. The capacity to input this information was included in the prototype software