DOI: 10.7589/2013-02-034 Journal of Wildlife Diseases, 49(4), 2013, pp. 1028–1032 # Wildlife Disease Association 2013

Neospora caninum Exposure in Overlapping Populations of (Canis latrans) and Feral Swine (Sus scrofa)

Sarah Bevins,1,4 Emily Blizzard,1 Luis Bazan,2 and Pat Whitley31Department of Biomedical Sciences, Colorado State University and USDA/APHIS National Wildlife Research Center, 4101 Laporte Ave., Fort Collins, Colorado 80521, USA; 2USDA/APHIS/Wildlife Services Texas, San Antonio, Texas 78269, USA; 3USDA/APHIS/ Wildlife Services Oklahoma, Durant, Oklahoma 74701, USA; 4Corresponding author (email: Sarah.Bevins@ colostate.edu)

ABSTRACT: Limited information exists on are exposed to N. caninum (Dubey and caninum transmission dynamics in

Schares, 2011), but the parasite has been Downloaded from http://meridian.allenpress.com/jwd/article-pdf/49/4/1028/2243275/2013-02-034.pdf by guest on 03 October 2021 wildlife. This coccidian parasite, whose pres- successfully isolated from only a few ence can lead to substantial economic losses in cattle operations, requires a canid definitive species. The primary interest in this host for reproduction. We examined exposure organism stems from it being a primary in a definitive host, coyotes (Canis latrans), and cause of abortion in dairy and beef cattle in overlapping populations of feral swine (Sus (Anderson et al., 1991). Transmission in scrofa) to determine if spatial proximity be- cattle is primarily vertical, with infected tween a definitive and incidental host influenc- es the likelihood of parasite exposure. Eighteen females passing the infection to calves. percent of coyotes (95% confidence interval Epidemic outbreaks have been document- [CI]514.2–21.8) and 15.8% of feral swine ed where .50% of dairy cows in a herd (95% CI512.5–19.2) had been exposed to N. abort within several weeks of each other, caninum, and this is the first report of exposure but recrudescence is unpredictable in in US feral swine populations. Analyses suggest infected adult cows, which generally that the parasite is present throughout the environment and that exposure is not tempo- exhibit no clinical signs. Economic im- rally or spatially linked to antibody-positive pacts associated with N. caninum are coyotes. Antibody-positive feral swine were substantial, with annual losses in individ- found in an area where the only definitive host ual US states approaching tens of millions is domestic (Canis familiaris), indicating of dollars (Trees et al., 1999; Larson et al., that wild canids are not required to maintain the parasite in the environment. 2004). In addition to cattle, N. caninum Key words: Canine, coccidian, feral swine, can cause clinical disease in dogs, sheep, Neospora, surveillance. and goats (Dubey and Schares, 2011). To understand N. caninum presence in the Neospora caninum is a coccidian para- environment better, blood samples from site that produces environmentally resis- coyotes, a definitive host, and feral swine, tant, infective oocysts that are passed an invasive species that is expanding its through the digestive system of definitive range and abundance, were screened for hosts as part of a complex reproductive N. caninum antibodies. cycle. The only definitive hosts for N. A majority of samples were opportunis- caninum identified to date are canids, tically collected in conjunction with work including coyotes (Canis latrans; Gondim conducted by United State Department of et al., 2004), domestic dogs (Canis famil- Agriculture (USDA)/Animal Plant Health iaris; Dubey et al., 1988b), and Inspection Agency (APHIS)/Wildlife Ser- (Canis lupus; Dubey et al., 2011). Neo- vices (WS). Samples were collected from spora caninum was only first recognized as 2009 through 2011, in cooperation with a separate parasite from Toxoplasma state and other federal agencies, from the gondii in 1988 (Dubey et al., 1988a) and southwestern and south central United the list of canid definitive hosts will likely States (Fig. 1). This region was preferen- increase. tially targeted because of overlapping Serologic surveys suggest that a large populations of both species. Sample number of domestic and wild mammals selection was refined by focusing on

1028 SHORT COMMUNICATIONS 1029 Downloaded from http://meridian.allenpress.com/jwd/article-pdf/49/4/1028/2243275/2013-02-034.pdf by guest on 03 October 2021

FIGURE 1. Sample locations for 71 Neospora caninum–positive coyotes and 74 N. caninum–positive feral swine in Oklahoma, New Mexico, and Texas, 2009–2011. counties where coyotes and feral swine with the use of a commercially available had been concurrently sampled. For competitive ELISA for N. caninum (VMRD, coyotes, blood was collected on Nobuto Inc., Pullman, Washington, USA) that has (Advantec MFS, Dublin, California, USA) been extensively used in multiple species, filter paper and stored at 220 C in the including S. scrofa, because of high spec- National Wildlife Disease Program No- ificity and sensitivity, as well as the low buto Sample Archive, at the USDA/ cross-reactivity to closely related parasites APHIS/WS/National Wildlife Research (Haddad et al., 2005; Almerı´a et al., 2007). Center in Fort Collins, Colorado, USA. All samples, positive controls, and negative For feral swine, blood was collected in controls were run in duplicate according to ethylenediaminetetraacetic acid or serum manufacturer’s instructions. For both as- separating tubes, processed according to says, the optical density sample-to-control protocol (Pedersen, 2012), and stored at ratio positive cutoff was $0.30. 280 C in the Feral Swine Serum Archive Data were mapped using ArcMap, v.10 at the NWDP. (ESRI, Redlands, California, USA). Mean Blood samples collected from coyotes antibody prevalence and 95% confidence on filter paper were eluted with phos- limits were determined with the use of a phate-buffered saline with the use of binomial distribution for prevalence in previously reported protocols, resulting both species. Separate logistic regression in a 1:10 sample dilution (Dusek et al., analyses were also run (SAS, v.9.2, Cary, 2011). Samples were screened with the use North Carolina, USA) for each species in of a commercially available N. caninum order to determine pathogen associated enzyme-linked immunosorbent assay risk in relation to animal sex, age (adult, (ELISA, BiovetH, Saint-Hyacinthe, Quebec, subadult), and their interaction. Both Canada), validated for use in multiple canine variables are known to impact parasite species, including coyotes (Wapenaar et al., exposure risk (Wouda et al., 1999). In 2007), at the manufacturer’s suggested 1:10 addition, we ran a bivariate K-function sample dilution. Feral swine sera were tested analysis in program R with the use of 1030 JOURNAL OF WILDLIFE DISEASES, VOL. 49, NO. 4, OCTOBER 2013

TABLE 1. Mean Neospora caninum antibody prevalence and 95% confidence intervals (CI) in coyotes and feral swine from the southwest and south central United States, 2009–2011.

Species Variable Category Prevalence (%)95% CI

Coyote (Canis latrans) Age Adult 17.4 13.30–21.43 Subadult 22 11.13–33.31 Sex Female 18.7 13.23–24.27 Male 17.4 11.98–22.76 Feral swine (Sus scrofa) Age Adult 19.3 14.81–23.72 Subadult 9.6 5.15–14.13 Sex Female 19.3 14.4–24.14 Male 11.7 7.40–16.06 Downloaded from http://meridian.allenpress.com/jwd/article-pdf/49/4/1028/2243275/2013-02-034.pdf by guest on 03 October 2021 spatstat (Baddeley and Turner, 2005) to CI51.001–2.865) and adults were more determine the degree of clustering be- likely to be infected than younger animals tween positive definitive hosts and positive (x256.25, P50.01; OR52.1, 95% CI5 swine. 1.17–3.85). Results from the bivariate-K Samples from 394 coyotes and 467 feral spatial analyses revealed that positive feral swinewerescreenedforN. caninum swine did not tend to cluster with positive antibodies (Table 1). In total, 334 of the coyotes. coyotes sampled were adults and 54 were We examined N. caninum exposure in a subadults. Feral swine samples consisted definitive host, coyotes, and in overlapping of 301 adults and 166 subadults. Sampling populations of feral swine. The absence of of males and females was roughly equal, differences in N. caninum antibody preva- with 192 female and 190 male coyotes lence between age classes or sexes sampled, and 254 female and 213 male suggests that exposure to the parasite is feral swine sampled. Both species showed ubiquitous. As a definitive host (Gondim exposure to N. caninum across Oklahoma, et al., 2004), coyotes are likely exposed New Mexico, and Texas (Fig. 1). Overall, through multiple routes of infection, in- 18% of coyotes (95% CI514.2–21.8) and cluding environmental contamination and 15.8% of feral swine (95% CI 5 12.5– ingestion of infected prey. This is supported 19.2) had been exposed to the parasite by our finding of exposure in both adult (Table 1). The mean inhibition for positive animals and younger animals, with the latter feral swine was 41.8% (SD514.0); the possibly being exposed in denning sites. mean optical density sample ratio for Conversely, feral swine showed signifi- positive coyotes 44.6 (SD516.9). Positive cant differences in antibody prevalence in control means were 56.4 (SD57.6) and relation to both age and sex. The higher 1.0 (SD50.0), respectively, for feral swine prevalence in adult animals may be and coyotes. indicative of an increased chance for Logistic regression revealed different exposure over time through rooting be- parasite exposure patterns for the two haviors. It is also possible that adult species. For coyotes, neither sex, age, nor animals are more likely to eat an interme- their interaction were significant predic- diate host (Barrett, 1978). A higher N. tors of N. caninum exposure; however, age caninum antibody prevalence in female (F56.26, P50.01) and sex (F53.8, feral swine may reflect different behaviors P50.04) were significant in feral swine. that impact exposure risk; female domestic The sex-by-age interaction was not signif- dogs have been found to have higher icant. Female feral swine were more likely prevalence than males, possibly because of to be infected than males (x253.86, higher recrudescence rates associated with P50.49; odds ratio [OR]51.69, 95% reproduction (Wouda et al., 1999). SHORT COMMUNICATIONS 1031

Positive feral swine did not cluster the USDA/APHIS/WS offices in Hawaii, spatially or temporally with positive coy- Oklahoma, Texas, and New Mexico that otes. This is not surprising, because N. contributed samples used in this study. caninum is a coccidian parasite, which Mention of trade names or commercial typically has infectious oocysts that persist products is solely for the purpose of in the environment for long periods of providing specific information and does not time once shed from the definitive host. imply recommendation or endorsement by There are multiple canid species in New the US Department of Agriculture. Mexico, Texas, and Oklahoma, and they move across large spatial scales, so oocysts LITERATURE CITED

are likely widespread in the environment, Almerı´a S, Vidal D, Ferrer D, Pabo´n M, Ferna´ndez- Downloaded from http://meridian.allenpress.com/jwd/article-pdf/49/4/1028/2243275/2013-02-034.pdf by guest on 03 October 2021 leading to exposure in a wide range of de-Mera MIG, Ruiz-Fons F, Alzaga V, Marco I, species. Coyotes, domestic dogs, and other Calvete C, Lavin S, et al. 2007. Seroprevalence of Neospora caninum in non-carnivorous wildlife canids could all contribute to the parasite from Spain. Vet Parasitol 143:21–28. load in the environment (Dubey and Anderson ML, Blanchard PC, Barr BC, Dubey JP, Schares, 2011); however, previous re- Hoffman RL, Conrad PA. 1991. Neospora-like search has not definitively linked exposure protozoan infection as a major cause of abortion in cattle to either domestic dogs or coyotes in California dairy cattle. J Am Vet Med Assoc 198:241–244. (Barling et al., 2000, 2001). As a corollary Baddeley A, Turner R. 2005. Spatstat: An R package to this study, feral swine samples collected for analyzing spatial point patterns. in Hawaii revealed N. caninum exposure Barling KS, Sherman M, Peterson MJ, Thompson JA, on all three islands tested—Hawaii, Oahu, McNeill JW, Craig TM, Adams LG. 2000. and Kauai—with antibody prevalences of Spatial associations among density of cattle, abundance of wild canids, and seroprevalence 24.1% (95% CI513.9–37.2), 33.3% (95% to Neospora caninum in a population of beef CI518–51.8), and 45.4% (95% CI528.1– calves. J Am Vet Med Assoc 217:1361–1365. 63.6), respectively (Bevins, unpublished). Barling KS, McNeill JW, Paschal JC, McCollum FT, There are no native canids on Hawaii, with III, Craig TM, Adams LG, Thompson JA. 2001. the only known source of infection being Ranch-management factors associated with antibody seropositivity for Neospora caninum domestic dogs. in consignments of beef calves in Texas, USA. We have presented evidence of wide- Prev Vet Med 52:53–61. spread N. caninum exposure in a definitive Barrett RH. 1978. The feral hog on the Rye Creek host species across a large spatial scale and Ranch, California. Hilgardia 46:283–355. provided the first reports of N. caninum Dubey JP, Schares G. 2011. Neosporosis in animals— The last five years. Vet Parasitol 180:90–108. exposure in US populations of feral swine, Dubey JP, Carpenter JL, Speer CA, Topper MJ, whose exposure may be indicative of cattle Uggla A. 1988a. Newly recognized fatal proto- infection risk. This rapidly expanding zoan disease of dogs. J Am Vet Med Assoc invasive species is a potential sentinel for 192:1269–1285. many infectious diseases, because of their Dubey JP, Hattel AL, Lindsay DS, Topper MJ. 1988b. Neonatal Neospora caninum infection in presence in a range of habitats and their dogs—Isolation of the causative agent and generalist diet. The analysis revealed experimental transmission J Am Vet Med Assoc, different levels of exposure in each 193:1259–1263. species, and spatial clustering analyses Dubey JP, Jenkins MC, Rajendran C, Miska K, Ferreira LR, Martins J, Kwok OCH, Choudhary suggest that the parasite is present across S. 2011. Gray (Canis lupus) is a natural the landscape. Much is still unknown definitive host for Neospora caninum. Vet about this disease system, and additional Parasitol 181:382–387. studies are required to understand expo- Dusek RJ, Hall JS, Nashold SW, TeSlaa JL, Ip HS. sure risk in better. 2011. Evaluation of nobuto filter paper strips for the detection of avian influenza virus antibody in We thank the many wildlife disease waterfowl. Avian Dis 55:674–676. biologists in the field who make this research Gondim LFP, McAllister MM, Pitt WC, Zemlicka possible, including Brian Mesenbrink and DE. 2004. Coyotes (Canis latrans) are definitive 1032 JOURNAL OF WILDLIFE DISEASES, VOL. 49, NO. 4, OCTOBER 2013

hosts of Neospora caninum. Int J Parasitol bovine neosporosis. Int J Parasitol 29:1195– 34:159–161. 1200. Haddad JPA, Dohoo IR, VanLeewen JA. 2005. A Wapenaar W, Barkema HW, Schares G, Rouvinen- review of Neospora caninum in dairy and beef Watt K, Zeijlemaker L, Poorter B, O’Handley cattle—A Canadian perspective. Can Vet J RM, Kwok OCH, Dubey JP. 2007. Evaluation of 46:230. four serological techniques to determine the Larson RL, Hardin DK, Pierce VL. 2004. Economic seroprevalence of Neospora caninum in foxes considerations for diagnostic and control options (Vulpes vulpes) and coyotes (Canis latrans)on for Neospora caninum–induced abortions in Prince Edward Island, Canada. Vet Parasitol endemically infected herds of beef cattle. JAm 145:51–58. Vet Med Assoc 224:1597–1604. Wouda W, Dijkstra T, Kramer AMH, van Maanen C, Pedersen K, Bevin SN, Schmit BS, Lutman MW, Brinkhof JMA. 1999. Seroepidemiological evi- Milleson MP, Turnage CT, Bigelow TT, Delib- dence for a relationship between Neospora Downloaded from http://meridian.allenpress.com/jwd/article-pdf/49/4/1028/2243275/2013-02-034.pdf by guest on 03 October 2021 erto TJ. 2012. Apparent prevalence of swine caninum infections in dogs and cattle. Int J brucellosis in feral swine in the United States. Parasitol 29:1677–1682. Hum Wildl Interact 61:38–47. Trees AJ, Davison HC, Innes EA, Wastling JM. 1999. Submitted for publication 12 February 2013. Towards evaluating the economic impact of Accepted 3 April 2013.