Although the practical importance of adding sugars before drying is well known, the mechanism of protection of bacteria by sugars is not clear. The response of the dehydrated micro-organisms to rehydration is analysed in terms of structural and functional changes, and correlated with their potentiality to grow in rich media. These aspects are related with the membrane integrity and the metabolic state of the rehydrated bacteria, measured by means of surface properties and permeability. To attain this objective, Lactobacillus delbrueckii subsp. bulgaricus was dehydrated in the presence and in the absence of sucrose and trehalose. The bacterial response upon rehydration was investigated by determining: (i) the lag time of the bacterial growing in rich media, (ii) the restoration of the surface properties and the cellular volume and (iii) the membrane integrity. Lactobacillus delbrueckii subsp. bulgaricus was grown in MRS at 37 degrees C overnight [De Man et al. (1960)J Appl Bacteriol 23, 130] and then dehydrated for 10, 20 and 30 min at 70 degrees C in a vacuum centrifuge. The lag time of micro-organisms was determined by optical density changes after rehydration. The surface properties were determined by measuring the zeta potential of the bacteria suspended in aqueous solution. The cellular volume recovery was measured, after stabilization in saline solution, by light scattering and by the haematocrit method [Alemohammad and Knowles (1974)J Gen Microbiol 82, 125]. Finally, the membrane integrity has been determined by using specific fluorescent probes [SYTO 9 and propidium iodide, (PI)] that bind differentially depending on the integrity of the bacterial membrane. The lag time of Lact. delbrueckii subsp bulgaricus, dehydrated by heat in the presence of sucrose or trehalose and after that rehydrated, was significantly shortened, when compared with that obtained for bacteria dried in the absence of sugars. In these conditions, trehalose and sucrose maintained the zeta potential and the cell volume close to the control (nondried) cells. However, the membrane integrity, measured with fluorescent probes, was maintained only when cells were dehydrated for 10 min in the presence of sugars. For larger times of dehydration, the membrane integrity was not preserved, even in the presence of sugars. When the micro-organisms are dehydrated in the absence of protectants, the membrane damage occurs with a decrease in the absolute value of the zeta potential and a decrease in the cellular volume recovered after rehydration. In contrast, when the zeta potential and the cellular volume are restored after rehydration to that corresponding to nondried cells, the micro-organisms are able to recover and grow with a reduced lag time. This can only be achieved when the dehydration is carried out in the presence of sugars. At short dehydration times, the response is associated with the preservation of the membrane integrity. However, for longer times of dehydration the zeta potential and volume recovery occurs in the presence of sugars in spite of a severe damage at membrane level. In this condition, cells are also recovered. In conclusion, to predict the ability of growing after dehydration, other bacterial structural parameters besides membrane integrity, such as zeta potential and cellular volume, should be taken into account. The correlation of the lag time with the surface and permeability properties is of practical importance because the correlation of these two parameters with cell viability, allow to determine the potential bacterial capacity to grow in a rich medium after the preservation procedure, without necessity of performing a kinetic curve of growth, which is certainly time-consuming.