Probability-of-growth modelling to optimize the use of hurdle technology to achieve microbiological stability of high moisture processed cheese : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Food Technology at Massey University, Palmerston North, New Zealand

This thesis reports a study of the application of hurdle technology to high moisture, low acid ambient shelf-stable hot packed processed cheese analogue (PCA). Hurdle technology makes use of a combination of mild stress factors. A combination of these stress factors can be more effective in inhibiting or inactivating the growth of micro-organism than individual stress factors. The current study focused on the application of hurdle technology to inhibit the growth of the food pathogen, Clostridium botulinum (C. botulinum). This micro-organism poses a hazard for consumers and is capable of growth in low acid food (pH> 4.5). As there are difficulties in working with C. botulinum in laboratory trails, Clostridium sporogenes (C. sporogenes) was used as an analogue of C. botulinum. C. sporogenes is very similar to C. botulinum in growth characteristics but is not dangerous. The effectiveness of selected preservatives on the growth of the target micro-organism was expressed as the probability of growth and was modelled as function of the concentrations of the selected preservatives in nutrient broth. Nutrient broth was initially used as it can be easily and accurately adjusted and controlled in terms of composition, and allows more rapid growth than is observed in PCA. A combination of salt (sodium chloride), sorbic acid (in the form of potassium sorbate), nisin and lysozyme was selected as stress factor. The inhibitory effect of these preservatives was then observed in the high-moisture nutrient broth at pH 7 (the optimum condition for spore of C. sporogenes to germinate) at 37ºC for eight weeks. It was found that lysozyme did not have a significant inhibitory effect on C. sporogenes in combination with salt, potassium sorbate and nisin. Therefore, the inhibitory effect of salt, sorbic acid and nisin at two different pHs (5.5 and 7) were subsequently evaluated in the nutrient broth at 37ºC for eight weeks. The probability of growth of C. sporogenes was modelled as a function of the concentrations of these selected preservatives at each pH. The results demonstrated that a combination of salt, nisin and potassium sorbate at relatively low concentrations can be used to inhibit growth. The inhibitory effects of the preservatives were pH dependent and their inhibitory effect is higher at pH 5.5. The developed models were validated using a fresh data set. Finally, the applicability of the developed model was checked in high moisture PCA. The results showed that the developed broth model underestimated the probability of growth in the PCA. Therefore, a specific probability of growth model was developed for the PCA using the PCA instead of nutrient broth as the growth medium. This model accurately predicted the probability of growth of C. sporogenes in the PCA for given combinations of preservative concentrations. The model developed for PCA allows the relative levels of preservatives to be easily quantified without the need for time consuming and expensive experimental work. The model would have limitations in the case of strongly varying formulations, since minor changes in processed cheese formulation or its production, could significantly alter its ability to support toxin production. Therefore, the model is applicable only to PCAs that have formulations similar to that used in this study. The general approach described in this thesis could be applied in the development of other high moisture, low acid foods