ABSTRACT
Archived serum samples taken between 1997 and 2017 from 170 American black bears (Ursus americanus) in the Lake Tahoe area between California and Nevada, US, were tested for Toxoplasma antibodies to assess the seroprevalence of this agent. Samples were screened using a commercial porcine Toxoplasma (enzyme-linked immunosorbent assay [ELISA]) modified with Protein A/G peroxidase and compared to a traditional fluorescent antibody test. Results were analyzed to determine if there were differences in seroprevalence based on the test used, sex of bears, or habitat usage (urban-suburban vs. wildland). No significant differences in seroprevalence were attributable to any of these parameters. The overall seropositivity for bears was 36% (62/170), with urban-suburban bears scoring lower (31%; 37/119) than rural-wildland bears (40%; 18/45). Our results strongly support the use of a Protein A/G-modified ELISA for determining Toxoplasma exposure in black bears. We found somewhat lower levels of Toxoplasma antibodies in black bears from this region than in several reports from populations in the eastern US.
Toxoplasma gondii is a unicellular parasite with a worldwide distribution and infects a wide host range, including most mammals and some species of birds (Dubey et al. 2013). The only confirmed definitive hosts of T. gondii are members of the Family Felidae, and the domestic cat is capable of shedding more than 100 million oocysts after a single infection (Robert-Gangneux and Dardé 2012). An estimated 25–30% of the world's human population may be infected, with greater seroprevalence in populations with lower socioeconomic status, correlating with poor sanitation of food and water (Robert-Gangneux and Dardé 2012; Liu et al. 2015). Although environmental oocysts pose the greatest risk, the ingestion of contaminated meat, including bear meat, is a potential source of infection.
There is a paucity of economical, rapid, and large-scale Toxoplasma serologic tests in sylvatic species. Ideally, enzyme-linked immunosorbent assay (ELISA), indirect fluorescent antibody test (IFAT), and western blot tests require species-specific immunoglobulin conjugates for accurate results. The modified agglutination test has variable performance in different animal species but does perform well with ursid samples. The ELISA has the most promise for large-scale testing with minimal subjectivity, but the lack of commercially available species-specific anti-immunoglobulin G (IgG) conjugates for a wide range of species limits its utility (Robert-Gangneux and Dardé 2012).
The immunoglobulin species-specificity problem for bears has been circumvented by employing peroxidase-labeled Protein A and Protein G (Stöbel et al. 2002) or Protein A alone (Paillard et al. 2015) in ELISA tests. These proteins bind the fragment crystallizable region of IgG without interfering with the antigen binding sites of species-specific antibodies. In addition, comparison of Protein A/G performance when assessing sera from multiple species with known serologic status for T. gondii demonstrated excellent agreement with a commercial anti-T. gondii IgG ELISA developed for pigs (Al-Adhami and Gajadhar 2014). These studies indicate that Protein A/G may be a suitable method of testing for T. gondii in nondomestic species, including bears. The ability of this fusion protein to bind other types of ursid immunoglobulin, such as immunoglobin M and immunoglobulin A, is unclear, but such binding can occur in various other species.
Although clinical toxoplasmosis is rare in wild black bears, multiple studies have found that the seroprevalence of T. gondii in this species is among the highest of any animal in North America (Dubey et al. 1995; Nutter et al. 1998). The primary means by which black bears become infected is unknown but does occur predominately after birth rather than congenitally (Dubey et al. 2016). The high seroprevalence in this species may be due to an omnivorous diet, high trophic level within the ecosystem, and their long lifespan (Dubey et al. 2016).
Our study had three goals. First, to determine if a commercial T. gondii indirect ELISA test designed for pigs could be repurposed via incorporation of Protein A/G to accurately diagnose T. gondii exposure in black bears. Second, to determine the Toxoplasma seroprevalence in ursid serum samples in the Nevada Department of Wildlife archives. Third, to determine if there was a significant difference in the seropositivity of T. gondii in samples taken from bears in semiurban versus wildland environments. Bears may acquire T. gondii from consuming infected rats and mice, which are abundant in semiurban settings, or by ingesting oocysts from food, water, and human garbage contaminated by domestic cat feces, which are also more abundant in this environment. We hypothesized that bears in urban and suburban areas would have a higher seropositivity than bears from wildland areas.
Bear serum samples were mainly collected from the Lake Tahoe region between California and Nevada, US (Fig. 1). A total of 170 samples were taken between 1997 and 2017 and held frozen at –20 C. Circumstances of serum collection varied and included hunter-killed animals, relocation efforts, and humane euthanasia of injured or ill animals. Many samples were from animals euthanized due to repeated conflicts between the bear and humans. Data regarding circumstances and location of collection permitted classification of 164/170 bears as wildland or semiurban individuals. Classification was based on Beckmann and Berger (2003), which established criteria for the Nevada study area and demonstrated that black bears spend ≥90% of their time within specific environments. Data entry for sex was missing for 72 animals. Detailed data regarding age were generally not available and thus not analyzed, but the large majority of animals were adults. These samples were transferred to the Oregon Veterinary Diagnostic Laboratory (Corvallis, Oregon, USA) and held at –80 C. Samples were thawed to 22±3 C prior to use.
Area of western Nevada, USA where most of the 170 serum samples were collected from black bears (Ursus americanus) for serologic testing from 1997–2017. Samples were analyzed by both indirect fluorescent antibody and enzyme-linked immunosorbent assay methods for Toxoplasma exposure.
Area of western Nevada, USA where most of the 170 serum samples were collected from black bears (Ursus americanus) for serologic testing from 1997–2017. Samples were analyzed by both indirect fluorescent antibody and enzyme-linked immunosorbent assay methods for Toxoplasma exposure.
The IFAT served as the comparative gold standard to indicate the presence or absence of antibodies against T. gondii. The test kit included T. gondii IgG positive and negative controls, which were employed in each test run (VMRD T. gondii IgG FA Positive Control, catalog no. PC-IFA-TOXO-FEL-G; IgG FA Negative Control, catalog no. NC-IFA-TOXO-FEL; VMRD, Inc., Pullman, Washington, USA). The conjugate used was BioVision Protein A/G FITC (Biovision Inc., Milpitas, California, USA) diluted 1:40. Samples were diluted 1:25 prior to performing the test as per the kit instructions.
The PrioCHECK Toxoplasma Porcine Antibody Kit was used for the ELISA (PrioCHECK™ Porcine Toxoplasma Ab Kit, catalog no. 7610230, Prionics AG, Schlieren-Zurich, Switzerland). The kit was combined with the Pierce Recombinant A/G Peroxidase Conjugate to detect anti-T. gondii antibodies (Pierce Purified Recomb Protein A/G, Peroxidase Conjugated, product no. 32490; Pierce Biotechnology, Rockford, Illinois, USA). Bear serum samples and the supplied positive, weak positive, and negative control samples were used at a 1:10 dilution with fluid buffer solution. Quantitative measurement of anti-T. gondii IgG antibodies was done by optical density (OD) using a BioRad iMark plate reader (BioRad Laboratories, Hercules, California, USA). A sample was considered positive if the calculated percent positive was ≥20% (cutoff value). If the percent positive was <20%, the sample was classified as negative.
Statistical determination of overall prevalence agreement between IFAT and ELISA results was performed using Cohen's kappa (κ) statistic and McNemar's test (Bennet and Underwood 1970; McHugh 2013). Seroprevalence comparisons between male and female, and between urban and wildland, for either IFAT or ELISA results were assessed by chi-square tests. All chi-square calculations were performed for a 95% confidence interval with 1 df. We compared the ELISA OD values for male versus female and urban versus wildland using Wilcoxon Rank Sum test and Welch's t-test (Welch 1938; Wilcoxon 1945).
Of the 170 samples tested by ELISA, 108 were designated true negatives and 60 were designated as true positives based on comparison to IFAT results. Thus the overall seroprevalence was 36% (62/170) for IFAT and 35% (60/170) for ELISA. The calculated sensitivity of the ELISA was 96.7% and the calculated specificity was 100%. There was no significant difference between the results of the IFAT and the ELISA tests (df=1, χ2=0.05, κ=0.970, 98.7% agreement) as measured by Cohen's κ statistic. This result was corroborated by the McNemar's test (odds ratio=1.000, P=0.617).
Fifty-eight samples were from male bears, 40 from females, and sex was not recorded for 72 animals. Based on ELISA results, seroprevalence was 31% (18/58) in males and 45% (18/40) in females. No statistically significant difference was found between the sex-based seroprevalence for either the IFAT (df=1, χ2=1.103, P=0.293) or the ELISA (df=1, χ2=1.987, P=0.1587) results. No statistically significant difference was found between the ELISA OD results for seropositive male versus female bears based on the Wilcoxon rank sum test (P=0.516) or Welch's t-test (P=0.810).
Of the 119 semiurban bear samples, 37 were seropositive (31%). The 45 wildland bear samples yielded 18 seropositives (40%). No statistically significant difference was found when comparing these two groups (df=1, χ2=1.155, P=0.280) or when comparing the OD results between wildland and urban bears using the Wilcoxon rank sum test (P=0.627) or Welch's t-test (P=0.771).
The results of this study strongly supported the viability of Protein A/G peroxidase modification of the commercial ELISA as a method for detecting anti-T. gondii IgG antibodies in ursine serum samples versus the more labor-intensive IFAT. We believe this ELISA test method might be adapted to monitor Toxoplasma exposure in other nondomestic species. Ideally, such tests would be validated by in-depth investigations such as experimental infections or parasite isolation and genotyping, but that is beyond the scope of this study.
As with most serosurveys, we found no significant difference in Toxoplasma exposures of male versus female bears. Our hypothesis that T. gondii exposure would be more common among semiurban bears was not supported by our results. Our assumption that Toxoplasma-infected felids are more prevalent in such settings might be incorrect, but another explanation could be that felids do not have a central role as a source of ursine Toxoplasma infection. The carnivorous aspect of transmission may be more important in ursine infection than the numbers of oocysts contaminating a given environment, with prey type and availability being the key feature. The genotype or strain of T. gondii may be a more critical determinant in successful parasite transmission than the density of infected felids. Dubey et al. (2010) found that a strain of Toxoplasma derived from a black bear appeared to be less productive in experimentally infected domestic cats. Studies indicate that different genotypes of Toxoplasma may dominate in sylvatic settings (Dubey et al. 2011; Jiang et al. 2018).
It is unfortunate that we lacked detailed information regarding the ages of animals sampled, as age is almost certainly a confounding factor for ursid exposure to Toxoplasma (Briscoe et al. 1993; Nutter et al. 1998). However, we are confident that very few of these animals were cubs or yearlings.
The overall seropositivity (36%; 62/170 by IFAT) compares favorably with other serosurveys of western US black bear populations: 45% in a survey of Oregon, California, and Washington (Mortenson 1998), 28% for northern California (Stephenson et al. 2015), 37% for New Mexico (Bard and Cain 2019), and 42% for Alaska (Zarnke et al. 2000). Bear populations in the eastern US often have a much higher seroprevalence, ranging from 62–84% (Dubey et al. 1995; Nutter et al. 1998; Bronson et al. 2014; Cox et al. 2017). The causes for this east-west discrepancy are unclear, although impaired survival of oocysts in the generally more-arid western US environs should be considered. A recent Oklahoma serosurvey (Scimeca et al. 2020) also found a 73% overall seroprevalence, perhaps arguing against that particular environmental influence. Further investigations into which genotypes or strains dominate in the western US may prove instructive (Dubey et al. 2013; Scimeca et al. 2020), as would determining dietary differences between eastern and western black bear populations.
The authors thank Chris Morris, Carl Lackey, Heather Reich, and the Western Region Law Enforcement officers of the Nevada Department of Wildlife for sample collection and both the Nevada Department of Wildlife and the Department of Biomedical Sciences, Carlson College of Veterinary Medicine for funding.