Evaluation of sample preparation methods for the isolation of Salmonella from alfalfa and mung bean seeds with the Bacteriological Analytical Manual's Salmonella culture method.

Five pre-enrichment methods were evaluated for effectiveness with the U.S. Food and Drug Administration's Bacteriological Analytical Manual Salmonella culture method in recovering S. Stanley, S. Poona, and S. Muenchen from artificially contaminated alfalfa seeds, and S. Saintpaul, S. Anatum, and S. Infantis from artificially contaminated mung bean seeds. The methods included: (1) Soak.--Test portions were inoculated into pre-enrichment media; (2) Rinse.--Test portions were rinsed with pre-enrichment media, and the media was decanted from the test portions; (3) Rinsed seed.--Pre-enrichment media was added to the test portions that were rinsed in the rinse method; (4) Wet blend.--Test portions were blended with the pre-enrichment media; and (5) Dry blend.--Test portions were blended prior to pre-enrichment. The methods of pre-enrichment were also evaluated for effectiveness in recovering Pantoea agglomerans from alfalfa and mung bean seeds with a modified culture method for the recovery of Enterobacteriaceae from foods. The purpose of these studies was to provide a model for the recovery of Salmonella that may occur in seeds as a natural contaminant. The relative effectiveness of the soak method was consistently superior to the rinse method in isolating the selected Salmonella serovars from both seed types. Statistically, the rinsed seed method was as effective as the soak method in all trials, except 1 of 3, with S. Muenchen and alfalfa seeds (P > 0.05). The relative effectiveness of the methods in isolating P. agglomerans from alfalfa and mung bean seeds was similar to that observed with the artificially contaminated test portions. The soak method was consistently the most effective method and the rinse method was consistently the most ineffective method. The rinsed seed, wet blend, and dry blend methods were also as effective as the soak method in all 3 trials with each seed type (P > 0.05).