Bacillus cereus: emetic toxin production and gamma hypothesis for growth

Bacillus cereus is a food spoilage microorganism and a pathogen. Growth of B. cereus can

be prevented or delayed by adding growth limiting compounds to the food product or

by altered storage conditions. Combinations of growth limiting factors can show synergy,

or be multiplicative without synergy (gamma hypothesis). For food safety management,

it is important to understand if combinations are synergistic or not, to avoid making faildangerous

or overly fail-safe predictions. Therefore, the aim of this PhD project was to

validate the gamma hypothesis for specific combinations of hurdles commonly used in

food production. Since the relationship between growth and toxin production of B. cereus

is little understood, a second aim was to investigate the production of the emetic toxin

cereulide in more detail. Several new lines of research were set-up to deliver on these

aims. For growth data collection to quantify hurdle effects and to study combinations

of hurdles, the relative rate to detection method was found to be preferred over two

other methods evaluated. The gamma hypothesis was validated for combinations of pH

and undissociated acid. For combinations of pH and water activity lowering solutes, the

gamma hypothesis could neither be validated nor rejected. The validity of the gamma

hypothesis appeared to be dependent on the models chosen for the single hurdle effects,

which are subsequently combined into the gamma model. A systematic way of model

selection is therefore advocated. Investigating cells in the transition from lag phase (λ)

to exponential growth phase, it was found that trends in physiological processes could

be observed for different culture conditions, independent of the duration of λ. Esterase

activity and electron transport chain activity were found to be useful quantitative markers

for this transition phase. A new method to produce synthetic cereulide showing biological

activity was developed, allowing accurate quantification of cereulide in samples. The

use of valinomycin as a standard, the current procedure, underestimates the amount of

cereulide by approximately 10 %. Considering the onset of emetic toxin production, LCMS

analysis of B. cereus F4810/72 grown in BHI showed that cereulide production does

not start before cells are in mid to late stationary phase, although significant variation

was noted possibly related to variability in the growth parameters maximum specific

growth rate (μmax) and λ. Addition of salt to the growth medium delayed the production

of cereulide. This research has been able to deliver several new insights and tools that are

useful for food safety management of the emetic toxin producer B. cereus.