Extraction of bacteria from soil: Efficiency of shaking or ultrasonication as indicated by direct counts and autoradiography

Abstract

Blending, shaking with a mechanical wrist-action shaker or ultrasonication were compared for removing bacteria from a sandy soil and a silt loam soil. Bacteria were counted microscopically after staining with acridine orange. There was no significant difference in the numbers of bacteria released from the sandy soil by the different treatments but ultrasonication extracted more bacteria from the silt loam. Highest recoveries of bacteria were obtained with a combination of Tris buffer as extractant, and 25% Ringer's solution as diluent. Ultrasonication of an irradiated sandy soil containing added Bacillus cereus removed 97% of the bacteria from the sand grains, but lysed some cells. Members of the indigenous soil population were more firmly held by the soil than added bacteria. Optimum conditions for recovering soil bacteria by ultrasonication were either 10.5 μm for 30s or 5 μm for 2 min. With the natural population, the numbers of bacteria and the proportion of metabolizing bacteria increased a little after ultrasonication as shown by labelling cells with 3H-glucose and detecting uptake by autoradiography. However, three isolates of bacteria tested showed varying amounts of injury after ultrasonication, with 5 μm for 2 min causing more damage than 10.5 μm for 30s. Ultrasonication was a more efficient means of extracting the natural bacterial population of soil than shaking but the time and amplitude must be adjusted to avoid injury to cultured bacterial cells.