Detections of Yersinia pestis East of the Known Distribution of Active Plague in the United States

Abstract

We examined fleas collected from black-tailed prairie dog (Cynomys ludovicianus) burrows from 2009 through 2011 in five national park units east of the known distribution of active plague across the northern Great Plains for the presence of Yersinia pestis. Across all national park units, Oropsylla tuberculata and Oropsylla hirsuta were the most common fleas collected from prairie dog burrows, 42.4% and 56.9%, respectively, of the 3964 fleas collected from burrow swabbing. Using a nested PCR assay, we detected 200 Y. pestis–positive fleas from 3117 assays. In total, 6.4% of assayed fleas were Y. pestis positive and 13.9% of prairie dog burrows swabbed contained Y. pestis–positive fleas. Evidence of the presence of Y. pestis was observed at all national park units except Devils Tower National Monument in Wyoming. We detected the presence of Y. pestis without large die-offs, i.e., enzootic sylvatic plague, east of the known distribution of active plague and near the eastern edge of the present distribution of black-tailed prairie dogs. This study, in combination with previous work suggests that sylvatic plague likely occurs across the range of black-tailed prairie dogs and should now be treated as endemic across this range.

Introduction

Sylvatic plague is a wildlife disease caused by the bacterium Yersinia pestis, generally transmitted through the bite of infected fleas. Y. pestis is thought to have originated in Asia and was likely introduced to North America through San Francisco around 1900 (Gage and Kosoy 2005). Y. pestis spread eastward until expansion appeared to halt temporarily around the 103^rd meridian (Barnes 1982, Antolin et al. 2002, Adjemian et al. 2007), forming the “naturalized plague zone” or areas where wildlife populations support endemic plague.

Prairie dogs (Cynomys spp.) are an ecologically important species of short- and mixed-grass prairie ecosystems that are particularly susceptible to epizootic plague (Gage and Kosoy 2005, 2006). Their extirpation can have wide-ranging ecological effects (Eads and Biggins 2015). Prairie dog colonies east of the 103^rd meridian were thought to be safeguarded from plague (Antolin et al. 2002). In 2005, black-tailed prairie dog (Cynomys ludovicianus) colonies on the Pine Ridge Indian Reservation in South Dakota experienced the first documented epizootic east of the naturalized plague zone (Rocke et al. 2008). This was followed by epizootics in the Conata Basin, South Dakota, in 2008 (Griebel 2009, Abbott et al. 2012). Outbreaks east of the 103^rd meridian demonstrated formation of a new emergence zone, or area where sylvatic plague outbreaks had been historically rare and limited in distribution (Antolin et al. 2002, Adjemian et al. 2007). Spread of sylvatic plague out of its naturalized zone provided an opportunity to study how Y. pestis spreads on a landscape scale.

Until recently, the dominant paradigm for prairie dog–Y. pestis interactions focused on prairie dogs serving as highly susceptible epizootic hosts that experience nearly 100% mortality and exhibit large-scale outbreaks (Gage and Kosoy 2005, 2006). However, evidence for presence of enzootic plague (presence of Y. pestis without massive die-offs over large areas in short time spans) in black-tailed prairie dogs has recently accumulated from sites in Montana (Hanson et al 2007, Biggins et al. 2010, Jones and Britten 2010, Matchett et al. 2010, Jones et al. 2012). There is also evidence that individual prairie dogs survive exposure to Y. pestis (Cully et al. 1997, Pauli et al. 2006, Jones et al. 2012).

Limited research has been conducted in the recent sylvatic plague emergence zone (Rocke et al. 2008, Griebel 2009, and Lower Brule Sioux Tribe, unpublished data), but understanding the spread of Y. pestis beyond its previous distribution is very important to wildlife conservation. Studies documenting Y. pestis where plague is considered endemic may not provide an accurate picture of its dynamics in areas along the invasion front. Invasive species, such as Y. pestis, often exhibit fluctuating distributions across space and time (Lomolino et al. 2005, Lowell et al. 2015). This suggests plague transmission dynamics may differ between the naturalized plague zone and locations in the current emergence zone.

While prairie dogs have been well studied in this system, fleas are an understudied part of the sylvatic plague system despite being the primary mode of transmission of Y. pestis between individual hosts. By seeking blood meals from different hosts, fleas play a key role in the sylvatic plague cycle as transmitters of Y. pestis, yet they lack the capacity to disperse farther than a few meters on their own (Marshall 1982). In our study area, prairie dogs and their burrows are primarily infested with two species of “prairie dog” fleas: Oropsylla hirsuta and Oropsylla tuberculata (Brinkerhoff 2008, 2011, Jones 2010). Both Oropsylla species are capable of early-phase transmission of Y. pestis (Wilder et al. 2008a, b). Plague is commonly documented by prairie dog epizootics; however, detection of Y. pestis in fleas offers a more sensitive detection method (Hanson et al. 2007). An advantage of studying fleas directly is that there exists an effective way of reducing flea populations. Deltamethrin kills fleas, resulting in decreased transmission of plague and increased survival, even when plague-induced prairie dog die-offs are not apparent (Biggins et al. 2010, Matchett et al. 2010). Jones et al. (2012) demonstrated the efficacy of deltamethrin in preserving black-tailed prairie dog genetic diversity and increasing survivorship and densities of prairie dogs on dusted colonies during an epizootic in Montana.

We conducted surveys for Y. pestis in fleas on black-tailed prairie dog colonies across several national park units east of the naturalized plague zone. The objectives of this research were to detect enzootic plague and determine the prevalence of Y. pestis within fleas east of the naturalized plague zone, document the flea fauna of black-tailed prairie dog colonies, and suggest management strategies to prevent or mitigate plague epizootics.

Materials and Methods

Five national park units (National Park [NP], and National Monument [NM]) were examined for presence of sylvatic plague: Badlands NP (SD), Devils Tower NM (WY), Scotts Bluff NM (NE), Theodore Roosevelt NP (ND), and Wind Cave NP (SD), as well as colonies surrounding Badlands NP on Buffalo Gap National Grassland (SD) (Fig. 1). Localities were selected because they were east of the known distribution of plague and had prairie dog colonies of variable sizes that were of management interest to the National Park Service. Badlands NP, Buffalo Gap National Grassland (NG), and Wind Cave NP support reintroduced black-footed ferrets (Mustela nigripes). Within these parks, colonies were selected for survey based on presence of ferrets, colony size (1.6–305.3 ha, \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$\bar {x} = 79.5 \pm 104.5$$ \end{document} ), location (e.g., maximize park unit coverage), and suspicion of Y. pestis. Prairie dog colony boundaries were determined a priori by wildlife biologists at each national park unit (see Plumb et al. 2001).

FIG. 1.

Map of National Park Units where black-tailed prairie dog colonies were sampled for fleas: 1, Theodore Roosevelt National Park (ND); 2, Devils Tower National Monument (WY); 3, Wind Cave National Park (SD); 4, Badlands National Park (SD); and 5, Scotts Bluff National Monument (NE).

Each national park unit was surveyed over 3 years (2009–2011); in 2009, each park was surveyed once between June and August, in 2010 colonies were surveyed April–October, and in 2011 colonies were surveyed May–July. During the 2010 field season each park was surveyed twice with the exceptions of Devils Tower NM (once), Scotts Bluff NM (once), and Wind Cave NP (three). In 2011, each park was surveyed once between May and August with the exception of Scotts Bluff NM (twice). In 2010 and 2011, the sampling season began earlier in an attempt to capture as much flea diversity as possible.

Fleas were collected directly from prairie dog burrows along the longest axis of each colony using a swabbing technique developed by Ubico et al. (1988). A 30-cm²-square flannel cloth was inserted into a prairie dog burrow for 2 min. Each burrow was swabbed only once regardless of flea presence/absence. An attempt was made to collect fleas from a minimum of 20 burrows or when 100 prairie dog burrows were swabbed for fleas. GPS coordinates were taken for each burrow where fleas were collected. Fleas were labeled individually, stored in 70% ethanol, and identified using the keys of Lewis (2002), Furman and Catts (1982), and Hubbard (1947). Whole genomic DNA was extracted using PrepGem (Zygem, New Zealand) extraction kits. Fleas were tested for presence of Y. pestis using a nested PCR assay modified from Hinnebusch et al. (1993) by Hanson et al. (2007; see also Jones and Britten 2010, Jones et al. 2011). Due to small sample sizes, fleas from the Johnson's Plateau colony at Theodore Roosevelt NP in 2009, Sanctuary colony at Wind Cave NP in 2009, and EP08 and WR05 colonies at Buffalo Gap NG in 2010 were pooled by burrow and species when analyzed. Due to the large number of fleas collected in 2011, a subset of O. tuberculata and O. hirsuta were randomly selected from each colony for analysis. The total number of fleas collected, prevalence of fleas in burrows (number of burrows with fleas/total number of burrows), mean intensity of flea infestation in burrows (number of fleas/number of burrows with fleas), total number of Y. pestis–positive fleas, total number of burrows with Y. pestis–positive fleas, prevalence of Y. pestis–positive fleas (number of Y. pestis–positive fleas/total number of fleas tested), and prevalence of Y. pestis in burrows (number of burrows with Y. pestis–positive fleas/total number of burrows with fleas tested) were estimated for each colony and national park with 95% confidence intervals using Quantitative Parasitology version 3.0 (Margolis et al. 1982, Rózsa et al. 2000).

Aggregation was estimated for fleas in burrows and Y. pestis–positive fleas in burrows for each colony and national park. Coefficient of aggregation (k) and variance to mean ratio ( \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$\sigma^2: \bar {x}$$ \end{document} ) of fleas in burrows, as well as variance to mean ratio of Y. pestis–positive fleas in burrows were estimated using Quantitative Parasitology version 3.0 (Rózsa et al. 2000). The variance to mean ratio was used to assess aggregation, where \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$\sigma^2: \bar {x} > 1$$ \end{document} is considered to be an aggregated distribution (Wilson et al. 2002). The corrected moment estimate of the coefficient of aggregation (k) was also used for estimating aggregation because it takes into account sample size (Elliott 1977, Wilson et al. 2002), where values of k < 1 are considered to be aggregated (Rózsa et al. 2000, Wilson et al. 2002).

Results

We swabbed 5236 prairie dog burrows for fleas from 2009 through 2011. We collected 3964 fleas from 23.5% (1231/5236) of burrows swabbed (Table 1). The majority of fleas collected were O. tuberculata (42.2%; 1673/3964) and O. hirsuta (56.9%; 2257/3964, Table 2). O. tuberculata was proportionally the most abundant flea species collected from prairie dog burrows in March–May, whereas O. hirsuta was proportionally the most abundant from June–October (Fig. 2, Table 2). Less than 1% were other species (33/3964; 12 Oropsylla spp, one Foxella ignota, one Ctenophthalmus pseudayrtes, and 19 unidentifiable fleas). Across the colonies surveyed, the number of fleas collected from individual burrows ranged from 0 to 224 total fleas (0–177 O. tuberculata and 0–161 O. hirsuta; Fig. 2, Table 2; Mize 2015), mean intensity of infestation of burrows ranged from 0.0 to 14.6 total fleas (0.0–9.5 O. tuberculata and 0.0–14.6 O. hirsuta; Mize 2015), and prevalence of fleas in burrows ranged from 0.0 to 73.3% (0.0–70.0% O. tuberculata and 0.0–43.8% O. hirsuta; Table 2; Mize 2015).

FIG. 2.

Seasonal changes in abundance between O. tuberculata (black) and O. hirsuta (grey) in black-tailed prairie dog burrows at five national park units. Seasonal changes in prevalence of Y. pestis in O. tuberculata (circles) and O. hirsuta (squares). Numbers above bars are the total number of O. tuberculata and O. hirsuta observed during each month. Numbers below bars represent the total number of burrows swabbed for fleas during each month. Data were pooled from 2009 through 2011; there were no surveys in September. (*) One O. tuberculata flea was examined for Y. pestis and tested positive.

Table 1.

Summary of Sampling Effort for Fleas from Black-Tailed Prairie Dog Burrows from Five National Park Units from 2009 Through 2011

Year	National Park Unit	No. colonies sampled	Month sampled	No. burrows swabbed	No. burrows with fleas
2009	Badlands NP	5	June	463	68
	Buffalo Gap NG	3	June	361	43
	Devils Tower NM	1	July	65	6
	Scotts Bluff NM	2	August	138	39
	Theodore Roosevelt NP	3	July	187	25
	Wind Cave NP	5	August	127	23
2010	Badlands NP	7	July, October	605	111
	Buffalo Gap NG	4	May, October	159	36
	Devils Tower NM	1	June	90	9
	Scotts Bluff NM	2	June	92	29
	Theodore Roosevelt NP	10	May, July	595	217
	Wind Cave NP	8	March, April, July	853	179
2011	Badlands NP	5	May	329	131
	Devils Tower NM	1	May	96	34
	Scotts Bluff NM	2	May, June	201	54
	Theodore Roosevelt NP	5	July	270	11
	Wind Cave NP	7	May	605	216
	Total	71	March–October	5236	1231

NP, National Park; NM, National Monument; NG, National Grassland.

Table 2.

Summary of the Number and Prevalence of Fleas in Burrows for O. hirsuta and O. tuberculata from Black-Tailed Prairie Dog Burrows at Five National Park Units from 2009 Through 2011

Year	National Park Unit	No. colonies	No. burrows with fleas	No. O. hirsuta	Prevalence O. hirsuta	No. O. tuberculata	Prevalence O. tuberculata
2009	BADL	5	68	102	11.2% (8.5–14.5%)	29	5.4% (3.5–7.9%)
2009	BUGA	3	43	72	10.2% (7.3–13.9%)	7	1.9% (0.8–4.0%)
2009	DETO	1	6	6	9.2% (3.5–19.0%)	1	1.5% (0.0–8.3%)
2009	SCBL	2	39	127	28.3% (20.9–36.6%)	0	—
2009	THRO	3	25	35	12.8% (8.4–18.5%)	1	0.5% (0.0–3.0%)
2009	WICA	5	23	84	18.1% (11.8–25.9%)	0	—
	Total	19	204	426	13.5% (11.7–15.5%)	38	2.5% (1.8–3.5%)
2010	BADL	7	111	275	18% (15–21.3%)	2	0.3% (0.0–1.2%)
2010	BUGA	4	36	103	17% (11.5–23.7%)	33	10.7% (6.4–16.6%)
2010	DETO	1	9	4	4.4% (1.2–11.0%)	5	5.6% (1.8–12.5%)
2010	SCBL	2	29	134	31.5% (22.2–42.1%)	0	—
2010	THRO	10	217	255	20.5% (17.3–24%)	385	26.2% (22.7–30%)
2010	WICA	8	179	292	15.9% (13.5–18.6%)	466	9.7% (7.8–11.9%)
	Total	32	581	1063	17.8% (16.3–19.4%)	891	11% (9.8–12.3%)
2011	BADL	5	131	246	18.5% (14.5–23.2%)	456	33.7% (28.6–39.1%)
2011	DETO	1	34	55	30.2% (21.3–40.4%)	21	13.5% (7.4–22.1%)
2011	SCBL	2	54	171	25.9% (20–32.5%)	2	1% (0.0–3.6%)
2011	THRO	5	11	12	3% (1.3–5.8%)	3	1.1% (0.0–3.2%)
2011	WICA	7	216	284	20.3% (17.2–23.8%)	262	22.6% (19.4–26.2%)
	Total	20	446	768	18.2% (16.3–20.2%)	744	17.7% (15.8–19.8%)

BADL, Badlands National Park; BUGA, Buffalo Gap National Grassland; DETO, Devils Tower National Monument; SCBL, Scotts Bluff National Monument; THRO, Theodore Roosevelt National Park; WICA, Wind Cave National Park.

A total of 3117 fleas collected from 1053 burrows from 2009 through 2011 were assayed for plague. Of those fleas, 200 tested positive for Y. pestis with a minimum overall prevalence of Y. pestis–positive fleas of 6.4% (Mize 2015). Across colonies, the number of Y. pestis–positive fleas ranged from 0 to 21 total fleas (0–5 O. tuberculata and 0–21 O. hirsuta; Fig. 2, Table 3; Mize 2015) and prevalence of Y. pestis–positive fleas ranged from 0.0 to 100% (0.0–100% O. tuberculata and 0.0–53.8% O. hirsuta; Fig. 2, Table 3; Mize 2015). Overall prevalence of burrows with Y. pestis–positive fleas was 13.9%. The number of burrows on a colony with Y. pestis–positive fleas ranged from 0 to 12 burrows (0–6 O. tuberculata and 0–12 O. hirsuta; Mize 2015) and prevalence of burrows on colonies with Y. pestis–positive fleas ranged from 0.0 to 100% (0–100% O. tuberculata and 0–100% O. hirsuta; Table 3; Mize 2015).

Table 3.

Summary of the Number Burrows with Fleas Tested for the Presence of Y. pestis and the Burrow Prevalence of Y. pestis–Positive Fleas for O. hirsuta and O. tuberculata from Black-Tailed Prairie Dog Burrows at Five National Park Units from 2009 Through 2011

Year	National Park unit	No. O. hirsuta burrows tested	Y. pestis + O. hirsuta burrows	O. hirsuta burrow Y. pestis prevalence	No. O. tuberculata burrows tested	Y. pestis + O. tuberculata burrows	O. tuberculata burrow Y. pestis prevalence
2009	BADL	52	2	3.8% (0.5–13.2%)	25	1	4% (0.1–20.4%)
2009	BUGA	37	5	13.5% (4.5–28.8%)	7	1	14.3% (0.4–57.9%)
2009	DETO	5	0	0	1	0	0
2009	SCBL	39	13	33.3% (19.1–50.2%)	0	0	0
2009	THRO	24	4	16.7% (4.7–37.4%)	1	0	0
2009	WICA	23	8	34.8% (16.4–57.3%)	0	0	0
	Total	180	32	17.8% (12.5–24.2%)	34	2	5.9% (0.7–19.7%)
2010	BADL	109	2	1.8% (0.2–6.5%)	2	1	50% (1.3–98.8%)
2010	BUGA	27	5	18.5% (6.3–38.1%)	17	7	41.2% (18.4–67.1%)
2010	DETO	4	0	0	5	0	0
2010	SCBL	29	8	27.6% (12.7–47.2%)	0	0	0
2010	THRO	122	7	5.7% (2.3–11.5%)	156	11	7.1% (3.6–12.3%)
2010	WICA	136	10	7.4% (3.6–13.1%)	83	7	8.4% (3.5–16.6%)
	Total	427	32	7.5% (5.2–10.4%)	263	26	9.9% (6.6–14.2%)
2011	BADL	61	15	24.6% (14.5–37.3%)	38	4	10.5% (2.9–24.8%)
2011	DETO	18	0	0	7	0	0
2011	SCBL	52	18	34.6% (22–49.1%)	0	0	0
2011	THRO	8	0	0	3	0	0
2011	WICA	83	22	26.5% (17.4–37.3%)	25	3	12% (2.5–31.2%)
	Total	222	55	24.8% (19.2–31%)	73	7	9.6% (3.9–18.8%)

The coefficient of aggregation for fleas in burrows was variable across colonies, years, and species, with k ranging from 0.04 to 0.96 and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$\sigma^2 : \bar {x}$$ \end{document} ranging from 0.91 to 109.62 (Mize 2015). However, the majority of colonies demonstrated high levels of aggregation of fleas in burrows; 75% of colonies had values of k≤0.42 and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$\sigma^2 : \bar {x} > 1.85$$ \end{document} for all fleas, k ≤ 0.47 and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$\sigma^2 : \bar {x} > 1.0$$ \end{document} for O. tuberculata, and k ≤ 0.33 and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$\sigma^2 : \bar {x} > 1.62$$ \end{document} for O. hirsuta (Mize 2015). Measurements of aggregation of Y. pestis–positive fleas were variable, with k ranging from 0.16 to 3.19 and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$\sigma^2 : \bar {x}$$ \end{document} ranging from 0.5 to 11.95 (Mize 2015). The majority of colonies had unmeasurable values of k, and 75% of colonies had low to no measurable level of aggregation of Y. pestis–positive fleas in burrows ( \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$\sigma^2 : \bar {x} \leq 1.08$$ \end{document} for all Y. pestis–positive fleas, \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$\sigma^2 : \bar {x} \leq 1.0$$ \end{document} Y. pestis–positive O. tuberculata, and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$\sigma^2 : \bar {x} \leq 1.0$$ \end{document} Y. pestis–positive O. hirsuta; Mize 2015).

Discussion

Our observed Y. pestis–positive flea prevalence was similar to previous estimates for O. tuberculata (4.37%); however, O. hirsuta prevalence was two times higher than estimates (4.9%) for fleas collected from black-tailed prairie dog colonies in Colorado (Wilder et al. 2008a, b). Several other studies detected Y. pestis in fleas from burrows on black-tailed prairie dog colonies, with burrow prevalence ranging up to 44.4% in Montana (Hanson et al. 2007, Jones et al. 2011) and 19% in Colorado (St. Romain et al. 2013). These studies examined colonies within the same complex or nearby complexes. Our study is the first to examine fleas from colonies across a large geographic range. This may be in part why our range of prevalence of Y. pestis in burrows is larger than previous studies. In addition, our sampling protocol may not have been timed to catch the peak of O. tuberculata abundance in prairie dog burrows. Accordingly, our observed prevalence of Y. pestis–positive O. tuberculata should be considered a conservative estimate of Y. pestis prevalence at our study locations.

We provide evidence of enzootic plague across a large geographic area east of the naturalized plague zone. Considering that sylvatic plague has been documented as far north as Canada (Antonation et al. 2014) and as far south as Texas (Cully et al. 2010, Johnson et al. 2011, Rocke et al. 2012), the results from this study suggest Y. pestis is endemic across the range of black-tailed prairie dogs.

During this study, we observed either enzootic plague, population declines, or epizootic plague on prairie dogs colonies at all of the park units except for Devils Tower NM (Mize 2015), where we never detected Y. pestis. There are several possible reasons for a lack of detection of Y. pestis at Devils Tower NM: (1) Low sample size, (2) the sole prairie dog colony at Devils Tower NM is one of the most isolated in the study and this relative isolation may provide some protection against the arrival of Y. pestis, or (3) Y. pestis may have been suppressed by climatic conditions (Adjemian et al. 2007, Stenseth et al. 2006, Snäll et al. 2008).

Findings from the Wind Cave NP Bison Flats colony are of interest because fleas were sampled within a refugium for western tiger salamanders (Ambystoma mavortium) established prior to 2009. While the surrounding colony was treated with deltamethrin to control fleas, the burrows within the enclosure were not dusted, in part to assess the effects of the insecticide on the salamander. Y. pestis burrow prevalence declined steadily within the enclosure from 2009 through 2011 until dusting was curtailed outside the enclosure in 2012, after which, the Bison Flats colony boundary decreased in spatial extent and the remaining colony appeared less continuous. This is probably due in part to the movement of Y. pestis–positive fleas into the area after dusting.

The untreated enclosure is similar to “refugia” recommended in integrated pest management strategies to maintain gene flow between susceptible naïve populations and those exposed to insecticides, potentially leading to genetically acquired resistance (Coles and Dryden 2014, Halos et al. 2014). Further investigation is warranted to determine if such a dusting regime is an effective plague mitigation strategy.

We observed O. tuberculata and O. hirsuta, both competent vectors of Y. pestis (Wilder et al. 2008a, b), from burrow swabs at each sampled national park unit, as expected from previous studies (Lewis 2002, Brinkerhoff et al. 2006, Salkeld and Stapp 2008, Jones and Britten 2010, Jones et al. 2011). We observed a shift in the flea community on black-tailed prairie dog colonies from O. tuberculata to O. hirsuta in late May to early June. This is consistent with previous work describing the peak abundance of O. tuberculata in late winter to spring and of O. hirsuta in late spring through fall (Salkeld and Stapp 2008, Wilder et al. 2008b, Friggens et al. 2010). This transition in abundance between species may correspond to peaks in plague activity during spring and again in late summer continuing into fall, which we and Wilder et al. (2008a, b) observed. Given this seasonality, current deltamethrin application procedures mostly affect O. hirsuta and may not provide effective coverage of O. tuberculata at its peak abundance in May. We recommend shifting dusting activities, weather permitting, to May or even as early as April to control both flea species. Such action may aid in reducing early transmission of Y. pestis between prairie dogs as they are emerging from burrows in the spring and prevent epizootics from occurring before park personnel can observe population declines.

The preservation of the black-tailed prairie dog, and the prairie community it helps maintain, depends on our ability to detect this exotic pathogen early and to understand how it persists and spreads in the environment. We recommend routine surveillance of prairie dog colonies (e.g., densities, molecular assays, etc.) as a means of early detection. Vigilant monitoring of prairie dog colonies at regular intervals throughout the year is probably the most cost-effective way to identify colonies in the early stages of plague epizootics because it can be used by managers to target plague mitigation strategies more efficiently (Sidle et al. 2001, Cully et al. 2010, McDonald et al. 2011). Plague activity should trigger management activities in colonies near those colonies experiencing epizootics because these are more likely to die off (Stapp et al. 2004, Collinge et al. 2005, Cully et al. 2010). We note that Wind Cave NP and Badlands NP effectively used our yearly Y. pestis prevalence data to help inform their colony dusting efforts over the 3 years of our study. A fairly aggressive dusting program took place during our study, eventually including six of our nine study colonies at Wind Cave NP and four of 10 study colonies at Badlands NP. We attribute the continued existence of many of the treated colonies on these two national park units to early detection of Y. pestis and prompt management action.

With the spread of Y. pestis into the emergence zone in an enzootic state, a proactive approach is recommended (Griebel 2009, Cully et al. 2010, McDonald et al. 2011). History has shown that Y. pestis is almost certainly capable of dispersing east of the distribution of black-tailed prairie dogs. Monitoring should be continued or implemented to track expansion of plague eastward, both on prairie dog colonies where plague has not historically occurred and beyond the distribution of prairie dogs into new species.

Footnotes

Acknowledgments

We wish to thank the National Park Service Natural Resource Preservation Project for providing funding for this study. Additional resources were provided by The University of South Dakota. We would also like to thank National Park Service personnel from each of the study locations, in particular: J. Delger from Badlands National Park; M. Biel, R. Ohms, and A. Wetz from Devils Tower National Monument; R. Manasek from Scotts Bluff National Monument; M. Oehler and C. Sexton from Theodore Roosevelt National Park; B. Muenchau, D. Roddy, and G. Schroeder from Wind Cave National Park; D. Licht from Midwest Regional Office; and R. Griebel from Buffalo Gap National Grassland. We are especially grateful to the many field and laboratory technicians who helped collect and process thousands of fleas. This manuscript has been greatly improved thanks to comments from D. Biggins, M. Dixon, and two anonymous reviewers.

Author Disclosure Statement

No competing financial interests exist.

References

Abbott

, Osorio

, Bunck

, Rocke

. Sylvatic plague vaccine: A new tool for conservation of threatened and endangered species?. EcoHealth, 2012; 9:243–250.

Adjemian

JCZ

, Foley

, Gage

, Foley

. Initiation and spread of traveling waves of plague, Yersinia pestis, in the western United States. Am J Trop Med Hyg, 2007; 76:365–375.

Antolin

, Gober

, Luce

, Biggins

, et al. The influence of sylvatic plague on North American wildlife at the landscape level, with special emphasis on black-footed ferret and prairie dog conservation. Trans 67th North Am Wildlife Nat Res Conf, 2002; 67:104–127.

Antonation

, Shury

, Bollinger

, Olson

, et al. Sylvatic plague in a Canadian black-tailed prairie dog (Cynomys ludovicianus). J Wildlife Dis, 2014; 50:699–702.

Barnes

. Surveillance and control of bubonic plague in the United States. Symp Zoolog Soc Lond, 1982; 50:237–270.

Biggins

, Gobey

, Gage

, Carter

, et al. Vector control improves survival of three species of prairie dogs (Cynomys) in areas considered enzootic for plague. Vector Borne Zoonotic Dis, 2010; 10:17–26.

Brinkerhoff

, Markeson

, Knouft

, Gage

, Montenieri

. Abundance patterns of two Oropsylla (Ceratophyllidae: Siphonaptera) species on black-tailed prairie dog (Cynomys ludovicianus) hosts. J Vector Ecol, 2006; 31:355–363.

Brinkerhoff

. “Mammal and flea occurrence in association with black-tailed prairie dog (Cynomys ludovicianus) colonies: implications for interspecific plague transmission.”. 2008. Doctoral Dissertation, University of Colorado, Boulder Co.

Brinkerhoff

, Martin

, Jones

, Collinge

. Population genetic structure of the prairie dog flea and plague vector, Oropsylla hirsuta . Parasitology, 2011; 138:71–79.

10.

Coles

, Dryden

. Insecticide/acaricide resistance in fleas and ticks infesting dogs and cats. Parasites Vectors, 2014; 7:8.

11.

Collinge

, Johnson

, Ray

, Matchett

, et al. Landscape structure and plague occurrence in black-tailed prairie dogs on grasslands of the western USA. Landscape Ecol, 2005; 20:941–955.

12.

Cully

, Barnes

, Quan

, Maupin

. Dynamics of plague in a Gunnison's prairie dog colony complex from New Mexico. J Wildl Dis, 1997; 33:706–719.

13.

Cully

, Johnson

, Collinge

, Ray

. Disease limits populations: Plague and black-tailed prairie dogs. Vector Borne Zoonotic Dis, 2010; 10:7–15.

14.

Eads

, Biggins

. Plague bacterium as a transformer species in prairie dogs and the grasslands of western North America. Conserv Biol, 2015; 29:1086–1093.

15.

Elliott

. Some methods for the statistical analysis of samples of benthic invertebrates. Scientific publication, 25; 1977.

16.

Friggens

, Parmenter

, Boyden

, Ford

, Gage

, Keim

. Flea abundance, diversity, and plague in Gunnison's prairie dogs (Cynomys gunnisoni) and their burrows in montane grasslands in northern New Mexico. J Wildlife Dis, 2010; 46:356–367.

17.

Furman

, Catts

. Order Siphonaptera. In: Furman

, Catts

eds. Manual of Medical Entomology, 4th ed. Cambridge University Press, New York, 1982.

18.

Gage

, Kosoy

. Natural history of plague: Perspectives from more than a century of research. Annul Rev Entomol, 2005; 50:505–528.

19.

Gage

, Kosoy

. Recent trends in plague ecology. In: Roelle

, Miller

, Godbey

, Biggins

, eds. Recovery of the Black-Footed Ferret: Progress and Continuing Challenges. US United States Geological Survey Scientific Investigations Report 2005-5293, Reston, VA, 2006:213–231.

20.

Griebel

. Wall Ranger District, Conata Basin 2008 plague control report. U.S. Nebraska National Forest, Buffalo Gap National Grassland, Wall Ranger District, Wall, SD, unpublished report, 2009.

21.

Halos

, Beugnet

, Cardoso

, Farkas

, Franc

, Guillot

, Pfister

, Wall

. Flea control failure? Myths and realities. Trends Parasitol, 2014; 30:228–233.

22.

Hanson

, Britten

, Restani

, Washburn

. High prevalence of Yersinia pestis in black-tailed prairie dog colonies during an apparent enzootic phase of sylvatic plague. Conserv Genet, 2007; 8:789–795.

23.

Hinnebusch

, Schwan

. New method for plague surveillance using polymerase chain reaction to detect Yersinia pestis in fleas. J Clinical Microbiol, 1993; 31:1511–1514.

24.

Hubbard

. Fleas of western North America. Iowa State College Press, Ames, IA, 1947.

25.

Johnson

, Cully

, Collinge

, Ray

, eFrey

, Sandercock

. Spread of plague among black-tailed prairie dogs is associated with colony spatial characteristics. J Wildlife Manage, 2011; 75:357–368.

26.

Jones

. “A molecular genetic approach to studying sylvatic plgue on the landscape level: testing black-tailed prairie dogs as major dispersers of plague.”. 2010. Doctoral Dissertation, University of South Dakota, Vermillion, SD.

27.

Jones

, Britten

. The absence of concordant population genetic structure in the black-tailed prairie dog and the flea, Oropsylla hirsuta, with implications for the spread of Yersinia pestis . Molecular Ecol, 2010; 19:2038–2049.

28.

Jones

, Washburn

, Britten

. Gene flow in a Yersinia pestis vector, Oropsylla hirsuta, during a plague epizootic. J Vector Borne Dis, 2011; 48:125–132.

29.

Jones

, Biggins

, Eads

, Britten

. Deltamethrin flea-control preserves genetic variability of black-tailed prairie dogs during a plague outbreak. Conserv Genet, 2012; 13:183–195.

30.

Lewis

. A review of the North American species of Oropsylla Wagner and Ioff, 1926 (Siphonaptera: Ceratophyllidae: Ceratophyllinae). J Vector Ecol, 2002; 27:184–206.

31.

Lomolino

, Riddle

, Brown

. Biogeography, 3rd ed. Sinauer Associates, Sunderland, MA, 2005.

32.

Lowell

, Antolin

, Andersen

, Hu

, Stokowski

, Gage

. Single-nucleotide polymorphisms reveal spatial diversity among clones of Yersinia pestis during plague outbreaks in Colorado and the western United States. Vector Borne Zoonotic Dis, 2015; 15:291–302.

33.

Margolis

, Esch

, Holmes

, Kuris

, Schad

. The use of ecological terms in parasitology (Report of an Ad Hoc Committee of the American Society of Parasitologists). J Parasitol, 1982; 68:131–133.

34.

Marshall

. The Ecology of Ectoparasitic Insect. Academic Press, London, 1982.

35.

Matchett

, Biggins

, Carlson

, Powell

, Rocke

. Enzootic plague reduces black-footed ferret (Mustela nigripes) survival in Montana. Vector Borne Zoonotic Dis, 2010; 10:27–35.

36.

McDonald

, Stanley

, Otis

, Biggins

, Stevens

, Koprowski

, Ballard

. Recommended methods for range-wide monitoring of prairie dogs in the United States: Scientific Investigations Report 2011–5063. Fort Collins Science Center, US Geological Survey, US Department of the Interior, 2011:36 pp.

37.

Mize

. “Tracking the spread of emerging sylvatic plague in black-tailed prairie dog metapopulations using a landscape genetics approach.”. 2015 Doctoral Dissertation, University of South Dakota, Vermillion, SD.

38.

Pauli

, Buskirk

, Williams

, Edwards

. A plague epizootic in the black-tailed prairie dog (Cynomys ludovicianus). J Wildl Dis, 2006; 42:74–80.

39.

Plumb

, Willson

, Kalin

, Shinn

, Rizzo

. Black-tailed prairie dog monitoring protocol for seven prairie parks. Northern Prairie Wildlife Research Center Inventory and Monitoring Protocol. US Geological Survey, US Department of the Interior, 2001.

40.

Rocke

, Smith

, Marinari

, Kreeger

, Enama

, Powell

. Vaccination with F1-V fusion protein protects black-footed ferrets (Mustela nigripes) against plague upon oral challenge with Yersinia pestis . J Wildlife Dis, 2008; 44:1–7.

41.

Rocke

, Williamson

, Cobble

, Busch

, Antolin

, Wagner

. Resistance to plague among black-tailed prairie dog populations. Vector Borne Zoonotic Dis, 2012; 12:111–116.

42.

Rózsa

, Reiczigel

, Majoros

. Quantifying parasites in samples of hosts. J Parasitol, 2000; 86:228–232.

43.

Salkeld

, Stapp

. Prevalence and abundance of fleas in black-tailed prairie dog burrows: Implications for the transmission of plague (Yersinia pestis). J Parasitol, 2008; 94:616–621.

44.

Sidle

, Johnson

, Euliss

. Estimated areal extent of colonies of black-tailed prairie dogs in the northern Great Plains. J Mammal, 2001; 82:928–936.

45.

Snäll

, O'Hare

, Ray

, Collinge

. Climate driven spatial dynamics of plaque among prairie dog colonies. American Nat, 2008; 71:238–248.

46.

Stapp

, Antolin

, Ball

. Patterns of extinction in prairie dog metapopulations: Plague outbreaks follow El Niño events. Frontiers Ecol Environ, 2004; 2:235–240.

47.

St. Romain

, Tripp

, Salkeld

, Antolin

. Duration of plague (Yersinia pestis) outbreaks in black-tailed prairie dog (Cynomys ludovicianus) colonies of northern Colorado. EcoHealth, 2013; 10:241–245.

48.

Stenseth

, Samia

, Viliugrein

, Kausrud

, et al. Plaque dynamics are driven by climate variation. Proceed Natl Acad Sci, 2006; 103:13110–13115.

49.

Ubico

, Maupin

, Fagerstone

, McLean

. A plague epizootic in the white-tailed prairie dogs (Cynomys leucurus) of Meeteetse, Wyoming. J Wildlife Dis, 1988; 24:399.

50.

Wilder

, Eisen

, Bearden

, Montenieri

, Gage

, Antolin

. Oropsylla hirsuta (Siphonaptera: Ceratophyllidae) can support plague epizootics in black-tailed prairie dogs (Cynomys ludovicianus) by early-phase transmission of Yersinia pestis . Vector Borne Zoonotic Dis, 2008a; 8:359–367.

51.

Wilder

, Eisen

, Bearden

, Montenieri

, Tripp

, Brinkerhoff

, Gage

, Antolin

. Transmission efficiency of two flea species (Oropsylla tuberculata cynomuris and Oropsylla hirsuta) involved in plague epizootics among prairie dogs. EcoHealth, 2008b; 5:205–212.

52.

Wilson

, Bjørnstad

, Dobson

, Merler

, Poglayen

, Randolph

, Read

, Skorping

. Heterogeneities in macroparasite infections: Patterns and processes. In: Hudson

, Rizzoli

, Grenfell

, Heesterbeek

, Dobson

, eds. Ecology of Wildlife Diseases. Oxford University Press, New York, 2002; 6–44.