Are Terrestrial Mammals the Source for Exposure of Polar Bear to Brucella spp. in Alaska?

Antibodies against Brucella spp. have been detected in polar bear (Ursus maritimus; Tryland et al., 2001; Rah et al., 2005). In the absence of isolates, it is important to determine if polar bears are exposed to Brucella spp. from terrestrial mammals, marine mammals, or both. This is the topic of the article by O’Hara et al. (2010). I disagree with the conclusions of this study for the reasons presented herein. O’Hara and coauthors used a Western blot (WB) test originally developed to identify specific antibodies directed against both Brucella A and M antigens (Edmonds et al., 1999). These antigens, associated with the O-polysaccharide (OPS), elicit antibodies of different specificity (Alton et al., 1988). The aim of the study by Edmonds et al. (1999) was to differentiate exposure to Brucella abortus biovar 1 (an A-dominant strain) from exposure to Brucella suis biovar 4 (which expresses both A and M antigens) in moose (Alces alces) in Alaska. An important feature of this WB is that it does not allow identification of infection with a given Brucella species. Brucella melitensis biovar 2 and B. suis biovar 4 are strains that express both A and M antigens and sera from animals exposed to either of these Brucella species would be visualized in the same way in the WB. Edmonds et al. (1999) concluded that moose were exposed to B. suis biovar 4 because of the known epidemiology of brucellosis in Alaska. In a recent study of 41 marine mammal Brucella strains (McDonald et al., 2006), 22 were A-dominant, two were M-dominant, and 17 expressed both A and M antigens. Thus, more than 40% of marine Brucella express both A and M antigens and sera from animals exposed to these Brucella species would be visualized in the same way as sera from animals exposed to B. suis biovar 4. Thus, O’Hara and coauthors cannot conclude that the likely origin of antibodies in polar bears is terrestrial. Additionally, because the Yersinia WB was not described in the original article (Edmonds et al., 1999), a description of this test is needed by O’Hara and coauthors to assess its value in detecting Yersinia enterocolitica O:9 cross-reactive antibodies. O’Hara et al. (2010) used a recently described (Meegan et al., 2010) brucellosis competitive ELISA (cELISA) with unknown specificity and the sensitivity; the description of this cELISA contains no mention of the detection of anti-Brucella antibodies in polar bear and the issue of differentiating exposure to Brucella spp. from terrestrial or marine mammals is not discussed (Meegan et al., 2010). I contend that the conclusion of O’Hara et al. (2010) that ‘‘the source or pathway for the exposure to Brucella spp. antigen appears to be terrestrial, not marine, based on the comparative analyses of various assay methods and antigens used in this study’’ is not substantiated and is misleading.