Validation of diagnostic assays and development of molecular epidemiological tools for brucellosis

Brucellosis, caused by Brucella abortus, is endemic in free-ranging elk (Cervus elaphus nelsoni) and bison ( Bison bison) of the Greater Yellowstone Area (GYA) of Wyoming, Idaho and Montana, poses a continuing transmission risk to cattle sharing their range. Elk were captured as part of the Wyoming Game and Fish Department's (WGFD) Brucellosis Surveillance Program on the Grey's River and Dell Creek winter elk feedgrounds located in northwestern Wyoming in the winters of 2005 and 2006. All female elk 1.5 years or older were serologically tested for brucellosis. Animals were considered to be seropositive for brucellosis if positive on three federally approved brucellosis serological assays (card, SPT, FPA, Rivanol). Twenty-two elk met the criteria, were considered positive, and were killed and necropsied. Seventeen maternal tissue samples from each seropositive elk and three fetal tissues (if present) were cultured for Brucella abortus; the bacterium was isolated from 9/22 (40.9%) seropositive animals. Isolates of B. abortus were characterized and biotyped. B. abortus biovars 1 and 4 were identified among the isolates; both biovars were identified from the tissues of one naturally infected elk. An identical set of tissues to those cultured was examined for B. abortus by immunohistochemistry and B. abortus was not detected in any tissue sections. DNA was extracted from all homogenized tissue for use in a B. abortus specific polymerase chain reaction (PCR) assay. Samples from three lymph nodes of each animal were screened by PCR; B. abortus specific DNA was not detected in any tissue sample. The detection limit of the PCR assay was determined to be 1 x 10 4 organisms/mL of tissue homogenate and it appears that this assay was not sensitive enough to detect low levels of infection in the culture positive elk. Bacterial DNA also was processed for a molecular genotyping assay to differentiate isolates of B. abortus from elk and cattle of the GYA. A PCR based multiple-locus variable-number tandem repeats analysis (MLVA) was used to characterize isolates. Products from PCR were visualized using gel electrophoresis and base pair measurements were converted into a number of repeat units by comparing sizes to known repeat units for B. abortus strain 19 vaccine. Cluster analysis performed with the unweighted pair group method using arithmetic averages of the genotyping patterns revealed general groupings of isolates according to biotype. To date, no other diagnostic assay has surpassed bacteriologic culture as a more sensitive or specific test for identifying B. abortus infection in infected elk. This study represents the first description and differentiation of B. abortus isolates from elk and cattle of the GYA using MLVA. This type of molecular genotyping shows promise for epidemiological investigations and isolate source traceback, but further analysis and selection of appropriate molecular markers must be completed before the assay can be applied to localized geographic areas, such as the GYA.