The inactivation of lactobacillus plantarum during drying

Cultures of lactic acid bacteria play an important role in the production of food and feed. The most important application is the use as a starter culture. The main role of a starter culture is to ferment a sugar in the raw material to lactic acid and/or acetic acid. In the food industry, the formation of small amounts of other flavour components can also be important. The flavour components give the fermented product the unique taste whereas the lowered pH, due to acid production, also acts as a food preservation method.

It is preferable, that formulations used contain enough bacteria to inoculate the process material directly. Frozen as well as freeze-dried formulations are used for this purpose. The main advantage of (freeze-)dried over frozen bacteria is the lower transport and storage costs of the cultures. For industrial use, large quantities of active bacteria are required. Freeze-drying is generally not considered an attractive method for the preparation of these large quantities, due to its high cost and inherent complexity. Application of convective drying methods, such as spray drying, fluidized bed drying and spray granulation, is much more economical. However, the main disadvantage of the convective drying processes is that a considerable part of the bacteria is inactivated during drying.

To optimize an industrial scale convective drying process for lactic acid bacteria, it is essential that an insight into the correlation of parameters of the drying process with the inactivation rate of the bacteria is developed. The main objective of this study was to develop and verify a mathematical model that is able to predict the inactivation of bacteria during drying. Lactobacillus plantarum was used as a model lactic acid bacterium and fluidized bed drying was used as the drying method. With an appropriate mathematical model, more insight can be obtained about those parameters of the drying process that are important for the inactivation of L. plantarum .

To define a criterion to quantify survival of L. plantarum after drying, the pH decrease in a phosphate buffer due to fermentation of glucose to lactic acid by non- growing L. plantarum cells was studied. The method used offers a rapid and reproducible means of measuring the glucose-fermenting activity of L. plantarum . The maximum observed velocity of pH decrease is defined as the activity of the cell suspension. The residual activity is defined as the ratio of the activity after and at the start of the drying process.

In order to describe the inactivation of L. plantarum during a fluidized bed drying process, the drying kinetics of particles with L. plantarum cells immobilized in potato starch was studied. Experimental drying data were described using the short-cut drying theory. This theory was also used for the calculation of moisture concentration profiles inside the particles. Temperature of the drying particles in the course of the drying process was calculated with a heat balance in which the heat transfer coefficient was derived from the Nusselt-number in a fluidized bed. The mathematical model of the drying kinetics is straightforward and can be implemented easily in an overall model which includes the inactivation of L. plantarum during drying.

To describe the thermal inactivation of L. plantarum during drying, the thermal inactivation kinetics of L. plantarum cells immobilized in potato starch were measured at different moisture concentrations. The measured temperature and moisture dependency of the inactivation rate was modelled with first-order kinetics. The thermal inactivation model was coupled to the drying kinetics model. Although this concept has been successfully applied in literature studies, the model significantly underestimates the measured inactivation during drying.

It was concluded that inactivation of L. plantarum during drying is caused by two mechanisms: thermal inactivation and inactivation due to dehydration. In the final simulation model these inactivation mechanisms were described as two processes occurring simultaneously during drying. The dehydration inactivation was quantified by fitting an arbitrary set of equations to the measured inactivation data obtained at a drying temperature at which the thermal inactivation can be neglected. Both inactivation models were coupled to the drying kinetics model. The overall model predicts the measured inactivation of L. plantarum during a fluidized-bed drying process, up to a drying temperature of 55 °C.

From experiments, it was concluded that the influence of the drying rate on the residual activity of L. plantarum can be neglected and that the moisture concentration rather than the water activity is the essential parameter in dehydration inactivation. With the help of a newly developed DNA/DNase method (in cooperation with the Laboratory of Dairying and Food Physics), it was shown that dehydration, unlike thermal inactivation, causes damage to the cell membrane/cell wall. In order to optimize the residual activity after drying, the influence of the harvesting time on the dehydration resistance of L. plantarum was studied. Cells harvested after the early stationary growth phase showed the highest dehydration resistance.

It was concluded that during drying dehydration -inactivation of cells is inevitable. This dehydration-inactivation will not be influenced by the choice of the drying process but will be determined by other factors.

The research project was financially supported by Gist-brocades NV, Delft, The Netherlands.