This study examines decontamination processes that were developed on an emergency basis to eliminate Bacillus anthracis spores from deliberately contaminated buildings. The recommended steps include a survey with sampling, the removal of sensitive items, and HEPA vacuuming of all readily available surfaces, followed by biocide treatment and subsequent analyses for viable cells. There are several analytical challenges posed by this approach. These include the ability to discriminate the added strain from naturally occurring resident microbes, determining detection limits for anthrax spores in settled dusts, and detecting viable but nonculturable spores. There are also logistical issues relating to the various skill sets required from investigation to reconstruction. In the present study, a model office was constructed, and a strain of Bacillus pumilus was isolated from the carpet and reintroduced to the office in excess. The abundance of the B. pumilus strain was monitored in settled dust using a strain-specific, quantitative polymerase chain reaction (QPCR)-based detection method following repeated HEPA vacuum cleanings. The QPCR method had a limit of detection corresponding to < or = 10(2) colony forming units per gram of settled dust. QPCR results were compared with measures of dust recoveries and fungal glucan and endotoxin levels in the dust samples. The largest fraction (ca. 81%) of added spores was recovered during the first HEPA cleaning. Subsequent cleanings resulted in incrementally lower recoveries, with removal of 93% of the initial inoculum by the third HEPA vacuuming. HEPA vacuuming prior to removal of items such as office contents and furnishings would result in much less resuspension of dust and limiting the extent of contamination. This approach also ensures that residual contaminants are as low as can be reasonably achieved.

In this study, we show that noncoding sequences from amplified fragment length polymorphisms (AFLPs) can provide robust and sensitive genetic markers suitable for PCR-based discrimination of closely related strains of Bacillus and Paenibacillus, and quantitative PCR (qPCR)-based tracking of the strains in complex natural systems like soil. Quantitative PCR was accurate in the approximately 1 x 10(9) to approximately 1 x 10(4) colony forming units (CFU)/g soil range. The detection limit was improved to approximately 1 x 10(2) CFU/g when amplicons were analyzed by gel electrophoresis. Studies with laboratory-contained intact soil-core microcosms indicated that environmental persistence trends vary among different strains. For example, Bacillus circulans ATCC 9500, Bacillus amyloliquefaciens DSL 13563-0, Bacillus licheniformis ATCC 12713, Paenibacillus polymyxa NRRL B-4317, and 3 Bacillus subtilisstrains (ATCC 6051A, ATCC 55405, and NRRL B-941) died down to below the 1 x 10(2) CFU/g detection limit by days 28-105. In contrast, over a 105-day period, B. licheniformis ATCC 55406, Bacillus megaterium NRRL B-14308, and P. polymyxa strains ATCC 55407 and DSL 13540-4 died down but persisted at levels just above the detection limit, whereas Bacillus thuringiensis ATCC 13367 experienced a less than 10-fold decrease in cell numbers.

No abstract