Effect of water activity on the heat resistance of Salmonella in "dry" materials.

It is well known that the availability of water has a pronounced influence on the heat resistance of microorganisms. The resistance of completely dry organisms is many times higher than the resistance of the same organisms in a dilute suspension which has a water activity close to 1.0. There is, however, very little information about the resistance of vegatative bacterial cells at the intermediate water activities found in "dry" foods and feeds, and in foods containing moderate concentrations of sugar. This problem has important practical implications in relation to the survival of salmonellae in "dry" foods and feeds. Banwart and Ayres (1) observed that the survival of salmonellae in dry egg white at 50 to 70 C decreased rapidly when the water content of the egg white was increased from 3 to 5% to 6 to 12%. However, the water activity of the products was not recorded. Rasmussen et al. (4) observed that wet Salmonella cells added to dry meat and bone meal were reduced by a factor of 108 by heating for 15 min at 68 C; however, heating for 1 hr at 82 C was required to kill salmonellae in naturally contaminated meal. The water activity was not recorded. Mossel and Koopman (3) found a rapid 105-fold reduction in viable cells when a liquid culture of Salmonella was mixed into dry meat and bone meal, but additional storage for 5 days caused only a 10-fold decrease in the number of survivors. The water activity of the meat and bone meal was 0.46. J. Taylor (personal communication) reported that salmonellae appear to survive indefinitely at room temperature in bone meals, meat meals, dried egg, etc., when the water content is below 10 to 12%. The aim of this study was to measure the heat resistance of salmonellae at water activities in the range from 0.60 to 1.0. Samples of meat and bone meal with adjusted water contents were used in most of the experiments. Salmonella typhimurium and Salmonella senftenberg 775W were grown in a heat-sterilized suspension of meat and bone meal (1 :5) in water. The cultures were centrifuged, the supernatant fluid was discarded, and the wet residue, which contained large numbers of Salmonella, was dried at 40 C under vacuum for 12 hr. It was then stored at room temperature over concentrated H2SO4 in a sealed container. This dry material, which contained approximately 108 viable salmonellae per gram, was mixed into meat and bone meal for heat resistance studies. Various amounts of distilled water were added to commercial meat and bone meal and the water activity was determined according to the method of Landrock and Procter (2) after equilibration for 2 weeks in sealed mason jars that were kept at room temperature. The water activity of the meat and bone meal at different moisture levels are shown in Table 1. Salmoneilae-containing dry residue (5 g) was mixed into 100-g samples of meat and bone meal conditioned at various water activities, and the inoculated samples were equilibrated for 1 to 2 weeks at room temperature. A 100-fold decrease in the numbers of viable Salmonellae was observed during the first 3 days of equilibration when the water content of the meal was high (13 to 16%). The counts remained constant after this initial decline. No multiplication of salmonellae was observed during storage of meal samples that contained up to 16% water. The high water activity in meal with 13 or 16% water would be expected to permit growth, but growth arrest may be caused by inhibitory solutes present in the meal. These hypothetical inhibitors are diluted toa noneffective level when a 1:5 suspension of meal is prepared in water. Heat resistance tests were carried out by heating 2-g samples of inoculated equilibrated meal in screw-cap tubes in a thermostatically controlled water bath. The heating times were corrected for temperature lag. The correction was based on heat transmission measurements. Five ml of cold phosphate buffer (M/15, pH 7.0) that contained 0.1 % peptone was added to each tube after heating and mixed with the meal on a Vortex Junior Mixer. Survivors were counted on Brilliant Green agar by the agar layer technique. Each plate was layered with 1 ml of semisolid Brilliant Green agar that contained a dilu-