Physiology and modelling of Escherichia coli growth inhibition due to pH, organic acids temperature and water activity

The inhibition of bacterial growth in foods is necessary to prevel].t food spoilage and food poisoning due to the presence of pathogenic bacteria. Organic acids are inhibitory to bacteria and are present in many foods. In some low pH foods they are present in sufficient concentration to prevent the growth of bacterial pathogens while in others their concentrations merely inhibit their growth rate. Organic acids have yet to be satisfactorily modelled in a way that is simple to understand, covers a range of concentrations and yields meaningful estimates of the concentrations of acids that limit growth. This study aims to describe simply and specifically the inhibition of bacterial growth caused by organic acids in such a way that helps to elucidate the biochemical and metabolic mechanisms of inhibition at a cellular level. To achieve this aim in Chapter 2 a series of models is developed which describe organic acid inhibition of growth rate of various Escherichia coli strains. These models are fitted to datasets that contain data for pH alone, or pH and lactic acid, or finally for pH and acetic acid. The pH responses of both pathogenic and non pathogenic strains were determined and modelled. The square root type pH models were based on the following hypotheses : that growth rate is proportional to the concentration of hydrogen ions, that pH inhibition is separate to organic acid inhibition and that growth rate is also proportional to the concentration of undissociated organic acid and to the concentration of dissociated organic acid. Therefore organic acid inhibition was modelled using terms adapted from the Henderson-Hasselbalch equation which gives the concentration of each form of the acid. The original model type was adapted to describe the individual datasets better. For the lactic acid model it was found that the addition of a term for inhibition due to high pH significantly improved the fit of the model. New terms were developed that better described the inhibition by dissociated acetic acid and for pH inhibition of different pathogenic and nonpathogenic strains. The consequences of these changes and whether these models still support the hypotheses is discussed. However these models fulfil the aim of providing a good mathematical description of growth rate inhibition as shown by the data under the conditions tested. In Chapter 3 the modelling of organic acid growth inhibition of Escherichia coli is extended to describe the boundary between those conditions under which growth is possible and those conditions under which growth is not possible. The simplification of the required information to a binary result (growth/no growth) for each observation allows collection of a larger number of datapoints and hence a much wider range of environmental conditions to be examined. The models are fitted to datasets that contain data for a wide range of pH, water activity, temperature and lactic acid conditions. These models are based on the hypotheses described for Chapter 2. The new hypothesis was that the growth rate model's mathematical equation can be adapted to calculate the probability of growth. The success of the adaptation of the model, and the principles developed for the generation of the type of data necessary to create better models are discussed. These models fulfil the aim of providing a good mathematical description of growth rate inhibition under the severely limiting conditions at the growth/no growth boundary. In Chapter 4 a method to elucidate the underlying biochemistry and mechanisms of organic acid inhibition is described. The mechanism of inhibitory action of organic acids has been described as the lowering of the internal pH of bacteria due to the permeation of undissociated organic acid ions. Alternatively specific effects of organic acids on cell metabolism, cell membrane transport and other cell functions could be responsible for the inhibitory effect of organic acids on bacteria. Intracellular pH measurement using the fluorescent probe 5 (and 6-) -carboxyfluorescein succinimidyl ester was trialled. This technique could determine the relationship of intracellular pH to variation in pH and organic acid conditions for E.coli. The previous technique developed for Gram positive organisms was found to be ineffective for labelling the cells, so new adaptations of techniques for the transitory permeabilisation of E.coli to allow labelling were studied. Under particular sets of conditions the cells were found to take up label but then they did not respond to the additions of the assay protocol. This implies the bacteria were unable to regulate their intracellular pH. Further experimentation with many treatment variations were not effective in producing labelled physiologically active E.coli. This technique does not appear to be readily applicable to the measurement of intracellular pH of Gram-negative organisms and so cannot be used to explore the physiology of pH stress: However, greater knowledge of the physiology of bacteria would enable a better understanding of the basis of modelled responses and could lead to the development of new ways of controlling bacterial growth.