Plague Epizootic Dynamics in Chipmunk Fleas, Sierra Nevada Mountains, California, USA, 2013–2015

We describe Yersinia pestis minimum infection prevalence in fleas collected from Tamias spp. chipmunks in the Sierra Nevadas (California, USA) during 2013–2015. Y. pestis–positive fleas were detected only in 2015 (year of plague epizootic), mostly in T. speciosus chipmunks at high-elevation sites. Plague surveillance should include testing vectors for Y. pestis.

T o better forecast vectorborne infection dynamics, characterizing disease cycles in both hosts and vectors is critical. The rate of infection of vector species can serve as a good indicator for risk during epizootic events, especially in areas with high human-wildlife overlap, but vectors are often poorly sampled. Yersinia pestis, the bacterium that causes plague, is carried by multiple flea species in western North America, where sciurids are often the primary reservoirs (1). Although human plague cases in this area are rare, in 2015, two cases were linked to exposures in Yosemite National Park, California, USA (2). In the investigation conducted to determine the source of these exposures, multiple Y. pestis-positive flea and rodent species were documented, and the lodgepole chipmunk (Tamias speciosus) was the host that was most frequently seropositive (2).
Plague surveillance in the western United States typically involves serologic testing of rodents and carnivores. Positive serologic results indicate prior plague activity. A Y. pestis-positive flea, however, indicates current plague transmission and is more likely to trigger control activities (3). Here, we sought to characterize Y. pestis infection in fleas of alpine (T. alpinus) and lodgepole (T. speciosus) chipmunks in Yosemite National Park and surrounding areas during 2013-2015. We focused on T. speciosus chipmunks because of their documented role in the 2015 epizootic (2) and on T. alpinus chipmunks because they co-occur with T. speciosus chipmunks (4) and little is known about their role in plague ecology. Our goals were to describe the proportion of T. speciosus and T. alpinus chipmunks harboring Y. pestis-positive fleas and the minimum infection prevalence of Y. pestis in fleas collected from these species across multiple sites and in years with and without known epizootic activity.
During June-October 2013-2015, we collected fleas from tagged chipmunks. Using a metal-pronged comb, we combed each animal 5 times down the dorsum, the tail, and each hind leg and placed collected fleas into vials containing 100% ethanol. These procedures were approved by the University of California, Berkeley, Animal Care and Use Committee (Berkeley, California, USA).
We found Y. pestis-positive fleas exclusively in 2015 at 5 of the 6 sites surveyed (Figure; Appendix Table 2). In 2015, 7.29% (14/192) of T. speciosus hosts carried >1 Y. pestis-positive flea. The minimum infection prevalence of Y. pestis in T. speciosus chipmunk-hosted fleas was 3.28% (assuming 1 positive flea per positive pool, 18 positive pools/548 total fleas in 280 pools tested). All 3 of the flea species (Ceratophyllus ciliatus mononis, Eumolpianus eumolpi, and E. eutamiadis) most commonly found on T. speciosus and T. alpinus chipmunks were found to be positive for Y. pestis (Appendix Table 1) (10). In 2015, a total of 5.13% (2/39) of T. alpinus hosts carried >1 Y. pestis-positive flea (Appendix Table 2). The infection prevalence (not minimum infection prevalence because each positive pool contained a single flea) of Y. pestis in T. alpinus chipmunk-hosted fleas was 2.47% (2 positive pools/81 total fleas in 50 pools tested). Unfortunately, these fleas were too damaged to identify morphologically, and molecular species identification was not possible.
Y. pestis-positive flea pools were detected at 5 of 6 high-elevation (2,650-3,200-m) study sites in 2015. Many of these sites are areas of high human activity, with popular hiking trails or established campgrounds. In 2015, plague risk assessments, including testing flea pools and rodent carcasses for Y. pestis DNA and rodent serology, also took place at lower elevation sites (1,778 ± 553 m) in and around the park; these surveys detected Y. pestis at 4 of 17 locations (2).
Altogether, our data indicate a dramatic shift in Y. pestis prevalence in fleas during a plague epizootic year in California. Our results support integrating flea testing, especially those at high-elevation sites, into regular surveillance.  B rucellosis can be transmitted between domestic animals and wildlife (1). Brucella melitensis has been isolated from wildlife, such as chamois (Rupicapra rupicapra) (2), Alpine ibex (Capra ibex) (3), and Iberian wild goat (Capra pyrenaica) (4). Badgers are major predators in forests and consume a broad spectrum of food items, including small terrestrial vertebrates and their cadavers (5), which might result in contact with pathogens from tissues of these vertebrates. We report an Asian badger (Meles leucurus) in China naturally infected with B. melitensis biotype 3.

About the Author
This study was approved by the Animal Ethics Committee of Shihezi University (approval no. AEC-SU2017-04). In 2017, a total of 7 illegally hunted and dying badgers in Nilka County, northwestern China, were confiscated by the local government.
We identified the animals as Asian badgers by using a PCR targeting the 16S rDNA gene (GenBank accession no. MH155253). We collected different organs or tissues, including heart, liver, spleen, lung, kidney, small intestine, large intestine, and blood, from all badgers. We separated serum from blood samples by centrifugation at 1,000 × g for 15 min and tested serum by using the rose bengal test (RBT) and serum agglutination test (SAT) (6). To detect Brucella antigens, we used immunohistochemical staining of liver and spleen tissue sections by pipetting mouse anti-Brucella melitensis IgG diluted 1:100 in 30% bovine serum albumin/phosphate-buffered saline onto each section. For comparison, we collected samples from 14 aborted sheep fetuses from Nilka County.
We extracted genomic DNA from all samples by using a commercial kit (Blood and Cell and Tissue Kit; BioTeke, http://www.bioteke.com). We used the partial omp22 gene (238 bp) encoding 22-kD outer membrane protein to identify the Brucella genus and the IS711 gene to identify Brucella species. We used PCRs that have been described (7). We used Brucella reference strains (B. melitensis 16M and B. abortus 2308) as positive controls and double-distilled water as a negative control.
We homogenized spleen samples of badgers and the 14 aborted sheep fetuses and inoculated these homogenates onto individual Brucella agar plates, which we then incubated at 37°C in an atmosphere of 5% CO 2 for 5 days. We tested putative Brucella colonies using by H 2 S production, dye inhibition, agglutination by monospecific serum, and sensitivity to bacteriophages (Appendix Table, https://wwwnc.cdc. gov/EID/article/26/4/19-0833-App1.pdf). We analyzed colonies by using a multilocus variable-number tandem-repeat analysis (MLVA) typing assay (8).
Only serum from badger no. 2 was positive for smooth Brucella antigen by RBT and SAT; the specific