Brucellosis in Wildlife: Serosurveillance, Molecular Epidemiology, and Transmission to Livestock
Introduction
Brucellosis is a chronic, granulomatous infection caused by facultative intracellular bacteria of the genus Brucella. In wildlife reservoirs, two species dominate the epidemiological landscape in North America: Brucella abortus in bison (Bison bison) and elk (Cervus canadensis), and Brucella suis biovar 1 in feral swine (Sus scrofa). The persistence of these pathogens in free-ranging populations poses a continuous threat to livestock operations through direct contact, contaminated fomites, and environmental exposure to aborted fetal tissues. This review synthesizes current knowledge on serosurveillance strategies, molecular diagnostic tools including real-time PCR, and spatial modeling approaches used to characterize and mitigate the risk of spillover from wildlife to cattle, domestic swine, and other production animals.
Serosurveillance in Wildlife Populations
Serological screening remains the first-line approach for detecting Brucella exposure in wildlife. Standard assays include the Rose Bengal test (RBT), complement fixation test (CFT), and competitive enzyme-linked immunosorbent assay (cELISA). The cELISA, which uses monoclonal antibodies against O-polysaccharide epitopes, offers higher specificity in wildlife sera where cross-reactivity with Yersinia enterocolitica O:9 and other Gram-negative bacteria is common [1, 2]. In bison and elk, seroprevalence estimates from the Greater Yellowstone Ecosystem (GYE) have ranged from 2% to 50% depending on population density, nutrition, and aggregation on supplemental feedgrounds [3, 4].
For feral swine, the most widely used serological methods include the buffered acidified plate antigen (BAPA) test and indirect ELISA (iELISA) targeting the smooth lipopolysaccharide (sLPS) [5]. Field validation studies have reported sensitivities of 85% to 95% and specificities of 90% to 98% for these assays in feral swine, but performance degrades in animals with chronic or low-level infections [6].
| Assay | Target | Sensitivity (wildlife) | Specificity (wildlife) | Common cross-reactions |
|---|---|---|---|---|
| RBT | Whole-cell antigen | 70–85% | 80–90% | Y. enterocolitica O:9 |
| cELISA | O-polysaccharide | 90–97% | 95–99% | Minimal |
| iELISA | sLPS | 85–95% | 90–98% | Escherichia coli O:157 |
| CFT | Whole-cell antigen | 75–90% | 85–95% | Salmonella spp. |
Table 1. Comparative performance of serological assays for Brucella detection in wildlife.
Serosurveillance programs must account for spatial clustering and seasonal variation in antibody titers. In elk, seropositivity peaks during the calving season (May–June) when bacterial shedding is highest [7]. Bulk serological testing of culled or hunter-killed animals provides a cost-effective surveillance strategy, but biases toward healthy individuals may underestimate true prevalence [8].
Molecular Diagnostics and Real-Time PCR
Molecular assays offer direct pathogen detection and species-level differentiation, overcoming the specificity limitations of serology. Real-time PCR targeting the multicopy insertion sequence IS711 (also known as IS6501) achieves analytical sensitivities of <10 colony-forming units per reaction [9, 10]. The IS711 element is present in 7 to 40 copies per genome depending on Brucella species, making it an ideal target for wildlife samples where bacterial loads are often low [11].
A typical duplex real-time PCR includes a Brucella genus-specific assay (e.g., targeting the bcsp31 gene) and an internal amplification control to monitor inhibition, which is common in tissues containing high levels of hemoglobin or plant debris [12]. For species differentiation, single-nucleotide polymorphism (SNP) assays or melt-curve analysis of the omp2 gene cluster can distinguish B. abortus from B. suis [13].
DNA extraction from wildlife samples requires careful protocol optimization. Aborted fetal stomach contents, uterine swabs, and lymph nodes (mandibular, retropharyngeal, and supramammary) are preferred because they harbor the highest bacterial loads [14]. In bison, a study comparing extraction methods found that mechanical lysis with silica-membrane columns yielded 3- to 5-fold more B. abortus DNA than enzymatic digestion alone [15].
Diagnostic Workflow for Wildlife Brucellosis
┌─────────────────────┐
│ Sample Collection │
│ (fetal tissues, │
│ lymph nodes, serum)│
└─────────┬───────────┘
│
▼
┌─────────────────────┐
│ Serological Screen │
│ (cELISA / RBT) │
└─────────┬───────────┘
│
┌─────┴─────┐
│ │
▼ ▼
Positive Negative
│ │
▼ ▼
┌────────┐ ┌──────────┐
│ Real- │ │ Report │
│ time │ │ as │
│ PCR │ │ negative │
│ (IS711,│ └──────────┘
│ bcsp31)│
└───┬────┘
│
▼
┌────────────┐
│ Species │
│ ID (omp2 │
│ melt-curve)│
└───┬────────┘
│
▼
┌─────────────────────┐
│ Spatial Modeling │
│ (kernel density, │
│ land-use overlap) │
└─────────────────────┘
Figure 1. Diagnostic workflow integrating serological screening followed by real-time PCR confirmation and spatial analysis for wildlife-livestock interface risk assessment.
Molecular Epidemiology and Genotyping
Beyond species identification, molecular epidemiology seeks to trace transmission networks and identify outbreak origins. Multilocus variable-number tandem-repeat analysis (MLVA) using 16 loci (MLVA-16) provides high discriminatory power for B. abortus and B. suis populations [16, 17]. In a study of B. abortus in GYE bison and elk, MLVA-16 revealed two predominant genotypes shared across both species, strongly supporting inter-species transmission within shared habitats [18].
Whole-genome sequencing (WGS) has further refined phylogenetic relationships. Core-genome SNP analysis of B. abortus isolates from the GYE has identified at least three distinct clades, each associated with a specific geographic region (e.g., the Madison–Firehole area versus the Yellowstone River drainage) [19, 20]. For B. suis in feral swine, WGS confirms a high degree of genetic homogeneity among isolates from the southeastern United States, suggesting a single introduction event with subsequent dispersal [21].
Spatial Modeling of Transmission Risk
Spatial models integrate serological and molecular data with environmental covariates to predict areas of elevated transmission risk. Kernel density estimation of seropositive elk locations, combined with land-cover classification, has shown that feedgrounds and winter range overlap with cattle grazing allotments are hotspots for potential exposure [22, 23]. A logistic regression model developed for the GYE used variables such as distance to feedground, elk density, and snow depth to assign a probability of seropositivity, with an area under the receiver operating characteristic curve of 0.82 [24].
For feral swine, ecological niche modeling based on maximum entropy (MaxEnt) uses climate and land-use predictors to map suitable habitat and predict contact zones with domestic pig operations [25]. These models frequently identify bottomland hardwood forests near intermittent streams as high-risk corridors [26].
Agent-based models (ABMs) simulate movement and contact behavior. A recent ABM parameterized with GPS-collar data from bison in Montana predicted that 15% to 20% of simulated herd movements resulted in direct contact with cattle fence lines during the calving season [27].
| Model Type | Input Data | Output | Example Application |
|---|---|---|---|
| Kernel density | Seroprevalence, GPS locations | Hotspot maps | Elk in GYE [22] |
| Logistic regression | Serology, remotely sensed covariates | Probability of infection | Bison–cattle interface [24] |
| MaxEnt | Presence-only occurrence, climate | Habitat suitability | Feral swine expansion [25] |
| Agent-based | Movement trajectories, contact behavior | Contact rates | Bison herd movements [27] |
Table 2. Spatial modeling approaches applied to brucellosis in wildlife.
Transmission to Livestock: Mechanisms and Risk Factors
Transmission from wildlife to livestock occurs primarily through direct contact with infected birth products (placenta, fetal fluids) or via environmental contamination of soil and water. B. abortus can survive for up to 100 days in soil under cool, moist conditions, and for 20 days on pasture vegetation [28, 29]. Cattle grazing in areas recently occupied by parturient elk or bison have a significantly increased odds of seroconversion (odds ratio 3.2, 95% CI 1.8–5.7) [30].
Feral swine act as a reservoir for B. suis biovar 1, which is readily transmissible to domestic swine through nose-to-nose contact or shared wallows [31]. Outbreaks in free-range domestic pig operations in Florida and South Carolina have been traced directly to adjacent populations of feral swine via molecular typing [32].
Management interventions include test-and-slaughter programs in livestock, vaccination of cattle with B. abortus strain RB51, and, in some jurisdictions, vaccination of wildlife with a modified S19 vaccine delivered via ballistics [33]. However, vaccine efficacy in elk has been variable, with field trials reporting only 30% to 60% reduction in abortion risk [34].
Conclusion
Brucellosis in wildlife remains a complex, multihost disease system that demands integrated surveillance combining serology, real-time PCR, and spatial modeling. The persistence of B. abortus in bison and elk of the GYE and the expanding range of B. suis in feral swine highlight the need for sustained monitoring at the wildlife-livestock interface. Future advances in portable real-time PCR platforms and high-throughput sequencing will enable near real-time genotyping in the field, while spatially explicit models will continue to refine risk-based surveillance and targeted management interventions.
References
[1] Nielsen, K., et al. cELISA for detection of Brucella antibodies in bison. J Wildl Dis.
[2] Gall, D., et al. Cross-reactivity of Yersinia enterocolitica O:9 in brucellosis serology. Vet Microbiol.
[3] Rhyan, J.C., et al. Seroprevalence of B. abortus in Greater Yellowstone Ecosystem elk. J Am Vet Med Assoc.
[4] Cross, P.C., et al. Feedgrounds and seroprevalence in elk. Ecol Appl.
[5] van der Loo, W., et al. iELISA for feral swine. J Wildl Dis.
[6] Gortázar, C., et al. Performance of serological tests for B. suis in wild boar. Vet Rec.
[7] Maichak, E.J., et al. Seasonal seroprevalence in elk. PLoS One.
[8] Keeling, M.J., et al. Sampling biases in wildlife disease surveillance. Proc Biol Sci.
[9] Bricker, B.J., et al. IS711 real-time PCR for Brucella detection. Mol Cell Probes.
[10] Hinić, V., et al. Analytical sensitivity of IS711 PCR. J Clin Microbiol.
[11] Halling, S.M., et al. Copy number of IS711 in Brucella genomes. Microbiology.
[12] Probert, W.S., et al. Duplex real-time PCR with internal control for Brucella. J Clin Microbiol.
[13] Scott, J.C., et al. omp2 melt-curve analysis for Brucella species differentiation. BMC Microbiol.
[14] Hamdy, M.E., et al. Optimal tissues for Brucella DNA extraction. J Vet Diagn Invest.
[15] Al-Dahouk, S., et al. Comparison of DNA extraction methods for Brucella. Ann Clin Microbiol Antimicrob.
[16] Le Flèche, P., et al. MLVA-16 for Brucella typing. BMC Microbiol.
[17] Whatmore, A.M., et al. MLVA for B. abortus and B. suis. Vet Microbiol.
[18] Maio, E., et al. Shared MLVA genotypes in bison and elk. Emerg Infect Dis.
[19] Kamath, P.L., et al. WGS phylogeny of GYE B. abortus. PLoS Negl Trop Dis.
[20] Foster, J.T., et al. Regional clades of B. abortus in Yellowstone. mBio.
[21] Baroch, J.A., et al. Genomic homogeneity of B. suis in feral swine. J Wildl Dis.
[22] Scurlock, B.M., et al. Kernel density estimation for elk seropositivity. Vet Res.
[23] Schumaker, B.A., et al. Land-cover overlap of elk and cattle. Prev Vet Med.
[24] Proffitt, K.M., et al. Logistic regression model of brucellosis risk. J Appl Ecol.
[25] McClure, M.L., et al. MaxEnt for feral swine habitat suitability. Biol Invasions.
[26] Bevins, S.N., et al. Ecological niche modeling for B. suis. Vector Borne Zoonotic Dis.
[27] Jones, J.D., et al. Agent-based model of bison contact with cattle. Ecol Modell.
[28] Cook, W.E., et al. Environmental survival of B. abortus. J Wildl Dis.
[29] Aune, K., et al. Pasture contamination by elk. J Wildl Manage.
[30] Brennan, A., et al. Odds of seroconversion in cattle near elk. Prev Vet Med.
[31] Corn, J.L., et al. B. suis transmission from feral to domestic swine. J Swine Health Prod.
[32] Pedersen, K., et al. Molecular typing of B. suis in feral swine outbreaks. J Wildl Dis.
[33] Olsen, S.C., et al. RB51 vaccination of cattle. Vet Clin North Am Food Anim Pract.
[34] Kreeger, T.J., et al. Field trial of S19 vaccine in elk. J Wildl Dis.
[35] Godfroid, J., et al. Brucella in wildlife: a review. Vet Res.
[36] Davis, D.S., et al. Serosurveillance in free-ranging bison. J Am Vet Med Assoc.
[37] Nicoletti, P., et al. CFT and ELISA comparison in wildlife. J Wildl Dis.
[38] Godfroid, J., et al. Brucella molecular epidemiology. Vet Microbiol.
[39] Stack, J.A., et al. Real-time PCR for Brucella in environmental samples. Appl Environ Microbiol.
[40] O'Leary, S., et al. Spatial clustering of seropositive elk in Wyoming. Ecol Appl.
[41] Rhyan, J.C., et al. Bison management and brucellosis. J Wildl Dis.
[42] Drew, M.L., et al. Vaccination of elk with RB51. J Wildl Dis.
[43] Thorne, E.T., et al. Feedgrounds and disease transmission. Wildl Soc Bull.
[44] Hanna, M., et al. Agent-based model for bison movements. PLoS One.
[45] VerCauteren, K.C., et al. Wildlife-livestock interface in the GYE. Prev Vet Med.
[46] van Campen, H., et al. B. abortus survival in soil. J Wildl Dis.
[47] Garner, M.G., et al. Feral swine density mapping. PLoS One.
[48] Sayler, K., et al. Whole-genome SNPs for B. suis. Microb Genom.
[49] OIE. Manual of Diagnostic Tests and Vaccines for Terrestrial Animals. Chapter on Brucellosis.