A Survey of Bacterial Respiratory Pathogens in Native and Introduced Mountain Goats (Oreamnos americanus)

At the time of European settlement of North America, mountain goats (Oreamnos americanus) were distributed from the mainland mountain ranges of southern Alaska, throughout the western Canadian Provinces and Territories, and into the northwestern US including Washington, Oregon, and western Idaho and Montana (Côté and Festa-Bianchet 2003). While mountain goats did not experience the level of overexploitation and reduction typical of other North American ungulates, wildlife management agencies included mountain goats in translocation programs during the early- to mid-1900s and expanded their distribution beyond known historical ranges (Côté and Festa-Bianchet 2003). Translocation efforts are considered a success, with introduced populations now established throughout the western US and Canada (Flesch et al. 2016). Nonetheless, in contrast to general range expansion through translocation, many populations show signs of decline. For example, today native populations in Montana (outside of Glacier National Park) are a third to a quarter of their estimated abundances in the 1940s (Smith and DeCesare 2017). Moreover, mountain goat abundance in British Columbia, Canada, and Washington state declined by at least half from the 1950s and 60s to the early 2000s (Mountain Goat Management Team 2010; Rice and Gay 2010).

The difficulty in studying mountain goats presents a number of challenges in documenting cause-specific declines, which can include anthropogenic causes or predation (Côté and Festa-Bianchet 2003). While driving factors are regionally specific, recent documentation of pneumonia in adult and kid mountain goats in Nevada (Wolff et al. 2014; Anderson et al. 2016) highlights the susceptibility of mountain goats to pneumonia pathogens typically associated with bighorn sheep (Ovis canadensis) epizootics (Blanchong et al. 2018; Cassirer et al. 2018). Given the broad distribution of respiratory pathogens in bighorn sheep (Butler et al. 2017), susceptibility of mountain goats to epizootics may be a more-widespread cause for concern. Moreover, where mountain goats are sympatric with bighorn sheep there is strong niche overlap (Lowrey et al. 2018) and potential for interspecific disease transmission (Wolff et al. 2016; Blanchong et al. 2018). Resident pathogen communities of mountain goats are poorly characterized yet have important implications for management and conservation of both mountain ungulate species. We evaluated mountain goat respiratory samples collected through multiple monitoring efforts in Montana, Wyoming, Idaho, and Alaska to regionally characterize the resident pathogen community across a diverse range of mountain goat populations as an initial step to inform management efforts.

Sampling efforts were conducted within three regions of the Greater Yellowstone Area (GYA), with a smaller sampling effort in southeast Alaska (Fig. 1). Mountain goats are considered nonnative in the GYA and grew from 170 individuals introduced in the mid 1900s to roughly 1,650 individuals in 2016 (Flesch et al. 2016). Sampling efforts in the GYA targeted three regions: the northeast GYA, including portions of Yellowstone National Park, Grand Teton National Park (GTNP), and the southwest GYA, including the Snake River Range of Wyoming and Idaho (Fig. 1). Mountain goats are sympatric with bighorn sheep in the northern GYA and GTNP but are considered allopatric in the southwest GYA. Historically, all populations in the GYA were broadly sympatric with domestic livestock. We provided a comparison with native populations that have never shared ranges with domestic livestock by sampling four populations in southeast Alaska (Fig. 1). While the southwest GYA is the presumed source of mountain goats colonizing GTNP (Lowrey et al. 2017), there is no other suspected connectivity between the four regions.

We captured mountain goats via ground darting, helicopter net-gunning, or helicopter darting between 2010 and 2017. All captures were conducted in accordance with Montana State University Institutional Animal Care and Use Committee (2011-17, 2014-32) and agency permits. Following extensive studies in bighorn sheep (Besser et al. 2013; Butler et al. 2017), we focused on the presence of Pasteurellaceae organisms (specifically leukotoxigenic and beta-hemolytic Bibersteinia trehalosi, Mannheimia haemolytica, and Mannheimia spp. as well as Pasteurella multocida) from tonsil swabs and Mycoplasma ovipneumoniae from nasal swabs. Sampling protocols varied between regions, but generally adhered to Butler et al. (2017). The protocols varied with respect to the transport media for swabs, the type of diagnostic test to identify pathogens, and the diagnostic laboratory that conducted the analyses. All samples were collected using sterile, polyester-tipped applicators. Transport media included Port-A-Cul™ tubes (BD, Sparks, Maryland, USA), tryptic soy broth with 15% glycerol (TSB; Hardy Diagnostics, Santa Maria, California, USA) vials, and immediate inoculation of Columbia blood agar culture plates (Hardy Diagnostics). The inoculated TSB vials were frozen as soon as possible after each capture and later shipped overnight on dry ice to prevent freeze-thaw cycles. We assessed the presence of pathogens using a combination of PCR and culture techniques. We delivered samples to the Washington Animal Disease Diagnostic Laboratory or the Wyoming Game and Fish Department Wildlife Health Laboratory for detection and identification of respiratory pathogens. We did not account for detection probability (Butler et al. 2017) and present our results as a minimum resident pathogen community within each region.

We sampled a total of 98 individuals across the four study regions. Due to local logistics and funding availability, there was notable variation in the number of yearly samples collected (mean=9, SD=5.34, range=1–20). The PCR was the only diagnostic test used in all laboratory, year, and study area combinations for the detection of M. ovipneumoniae and had positive detections in the northeast and southwest GYA (Table 1). Mycoplasma ovipneumoniae was not detected in GTNP or southeast Alaska. With the exception of the southwest GYA, serology tests strongly agreed with the PCR results. The cause of discrepancy in the southwest GYA in 2013 was unknown, but could have been related to sample size, sample timing in relation to nasal shedding, or the PCR assay which had not been validated for mountain goats.

We detected B. trehalosi and Mannheimia spp. in all regions within the GYA. Mycoplasma haemolytica was detected in the northeast and southwest regions but not in GTNP. The southwest GYA was the only region where we detected P. multocida (Table 2). In contrast to the GYA, M. haemolytica was the only Pasteurellaceae species detected in southeast Alaska (Table 2). Because differentiating isolates or leukotoxins of M. haemolytica with Mannheimia glucosida is currently not possible, an unknown number of M. glucosida detections may have been misclassified as M. haemolytica. Lastly, there were no observed mountain goat die-offs in any region over the sampling period.

There was notable variation in pathogen detection within and among years (Tables 1, 2), which we attributed to low detection probability (Butler et al. 2017) and small annual sample sizes. Careful measures were taken to handle all samples according to protocols that optimized pathogen recovery and survival; however, mountain goats present challenging sampling conditions in rugged terrain at high elevations. Given the difficult sampling conditions, limited sample sizes, and imperfect detection, our failure to detect specific pathogens should be interpreted with caution. Moreover, recent sampling efforts suggest that nasal swabs are more effective than are swabs of tonsillar crypts for detecting P. multocida, an important bacterial respiratory pathogen (Weiser et al. 2003; Wood et al. 2017), but were not included in our sampling methods. Nonetheless, our results indicate a resident pathogen community containing both M. ovipneumoniae and Pasteurellaceae within the northeast and southwest GYA. Although multiple Pasteurellaceae species were detected in GTNP, we did not detect M. ovipneumoniae in the 14 animals sampled. Not surprisingly, southeast Alaska had the fewest detections of Pasteurellaceae and no detections of M. ovipneumoniae.

Little is known about diseases in mountain goats; however, recent findings in Nevada highlight their susceptibility to respiratory disease (Wolff et al. 2014; Blanchong et al. 2018). With a number of mountain goats populations in decline regionally, respiratory disease may serve as an important yet previously underappreciated cause of mortality. Moreover, because of the strong niche overlap of bighorn sheep and mountain goats (Lowrey et al. 2018), characterizing the respective pathogen communities should be an important prerequisite to establishing new sympatric populations or understanding the potential for interspecific transmission on existing sympatric ranges (Wolff et al. 2016). These dynamics highlight the challenges associated with managing sympatric mountain ungulates and the importance of characterizing local pathogen communities to inform the management and conservation of both species.

We would like to thank the many agency biologists, wildlife veterinarians, regional game wardens, and other personnel who facilitated capturing and sampling efforts. Financial support was provided by Wyoming Game and Fish Department, Idaho Fish and Game, Alaska Department of Fish and Game, Canon USA, Inc. (via Yellowstone Forever), Wyoming Governor's Big Game License Coalition, National Park Service, and the Rocky Mountain Goat Foundation. Leading Edge (Clarkston, Washington, USA) conducted many of the animal captures.