Prevalence of Vector-borne Bacterial Pathogens in Riparian Brush Rabbits (Sylvilagus bachmani riparius) and their Ticks

The endangered riparian brush rabbit (Sylvilagus bachmani riparius) now occurs in only two populations in central California, following >95% loss of its historical habitat (Larsen 1993; Williams et al. 2002). In 2000, a recovery plan was implemented to re-establish habitat and perform captive breeding (Williams 2002). Six to 10 rabbits are removed from wild populations each winter into protected propagation pens. Predators are excluded through fencing, although birds, small mammals, and arthropods may enter the facility. Once juvenile rabbits reach targeted weights, they are medically evaluated and released into established or restored habitat.

Translocation, intensive management, and rerelease involve potential introduction and amplification of infectious diseases, including some that could become threats to humans and wildlife. Rabbits are handled during examinations and procedures (e.g., tagging, radio collar application), providing opportunities for zoonotic disease transmission directly and through arthropod vectors. There is potential for contact with other wild species in the captive propagation site and after release. Very limited data on infectious diseases have been collected from riparian brush rabbits.

In California, important zoonotic vector-borne pathogens for which reservoirs are rodents or lagomorphs include Rickettsia spp., Anaplasma phagocytophilum, Borrelia burgdorferi, and Bartonella spp. Rocky Mountain spotted fever, caused by Rickettsia rickettsii is a potentially fatal human or canine illness vectored by ticks; there were 29 human cases in California from 2002 to 2006 (Wikswo et al. 2008). An unreported subspecies of brush rabbit was positive for antibody to spotted fever group rickettsiae in Monterey County, California (Coultrip et al. 1973). Borrelia burgdorferi, the agent of Lyme disease, is present in many small mammals and Ixodes spp. ticks in California (Peavy et al. 1997). Anaplasma phagocytophilum occurs in cottontail rabbits (Sylvilagus floridanus) in Massachusetts (Goethert and Telford 2003) and rodents in Northern California (Foley et al. 2008). Of the 13 species or subspecies of Bartonella bacteria suspected to be pathogenic for humans (Chomel et al. 2009), only B. alsatica has been found in lagomorphs. We evaluated tick burdens on riparian brush rabbits and the prevalence of potentially zoonotic pathogens in rabbits and ticks.

Riparian brush rabbits were housed in three breeding pens ranging from 0.0049 to 0.0068 km² in San Joaquin County, California (38°10′3.914″N, 121°26′5.618″W). The adjacent pasture is irrigated heavily during spring and summer. Vegetation inside the pens consists of Himalayan blackberry (Rubus armeniacus). The pens exclude predators, but small mammals and birds have been observed by staff to enter the pens including black rats (Rattus rattus), California voles (Microtus californicus), house mice (Mus musculus), deer mice (Peromyscus maniculatus), California ground squirrels (Otospermophilus beecheyi), and a diversity of passerines and California quail (Callipepla californica).

From June to October 2010, 48 rabbits were trapped from pens, anesthetized with isoflurane, and examined. Whole blood was collected by jugular venipuncture into EDTA. Seven additional rabbits were examined only. Rabbits were systematically examined for ticks, and a maximum of 20 ticks were collected per rabbit and placed in 70% ethanol. Rabbits were either released to the wild or returned to breeding pens if they were injured, had low body weight, or were lactating.

Blood samples were stored at −80 C. We attempted culture for Bartonella spp. by centrifuging about 1 mL of whole blood and plating the pellet on heart infusion agar containing 5% domestic rabbit blood. Plates were incubated in 5% CO₂ at 35 C for 4 wk (Henn et al. 2007). Blood samples were analyzed for Rickettsia spp., B. burgdorferi, and A. phagocytophilum antibodies via indirect fluorescent assay with dilutions of 1∶25 in phosphate-buffered saline (PBS) applied to antigen slides (Veterinary Medical Research and Development, Pullman, Washington, USA). Slides were incubated at 37 C with moisture for 40 min, washed in PBS, and incubated for 40 min with fluorescein isothiocyanate–labeled goat anti-rabbit IgG, heavy + light chain (Kirkegaard Perry Laboratories, Gaithersburg, Maryland, USA) diluted in PBS at 1∶25. Slides were washed and stained with Eriochrome T-Black. Samples were considered positive if strong fluorescence was detected with a pattern characteristic for each antigen slide.

Ticks were identified to sex, stage, and species (Furman and Loomis 1984) and pooled if from the same rabbit in groups of up to five ticks of the same stage and sex. Two hundred microliters of TE buffer (pH 8.0) was added to each tube, and samples were ground with pestles, incubated at 95 C for 15 min, and diluted 1∶100 with water. DNA was extracted from rabbit blood samples using a kit (Qiagen, Valencia, California). DNA was stored at −20 C.

Real-time PCR was performed for the msp2 gene of A. phagocytophilum (Drazenovich et al. 2006), the gltA gene of Rickettsia spp. using primers 692F (5′-AATGCTTCTACTTCAACAGTCCGAAT-3′) and 773R (5′-GTGAGGCAATACCCGTGC TAA-3′) with probe 724T (5′-CTCATCCGGAGCTAACCCTTTTGCTTGT-3′), and B. burgdorferi with proprietary primers and probe courtesy of UC Davis TaqMan Services. All 12 µL reactions contained 5 µL of DNA, 1× Universal Master Mix (Life Technologies, Foster City, California), 2 nmol of each primer, and 400 pmol of probe with thermocycling conditions of 50 C for 2 min, 95 C for 10 min, and 40 cycles at 95 C for 15 sec, followed by 60 C for 1 min. Negative water controls were included during each run. A nested conventional assay targeted the ribosomal RNA 23S-5S (rrl–rrs) intergenic spacer region of A. phagocytophilum using external primers ITS2F, 5′-AGGATCTGACTCTAGTACGAG-3′ and ITS2R, 5′-CTCCCATGTCTTAAGACAAAG-3′, and internal primers ITS2iF, 5′-ATACCTCTGGTGTACCAGTTG-3′ and ITS2iR, 5′-TTAACTTCCGGGTTCGGAATG-3′ and thermocycler conditions of 94 C for 2 min, 35 cycles of 94 C for 30 sec, 58 C for 30 sec, and 72 C for 1 min, and then 5 min extension at 72 C. DNA was visualized on a 1% agarose gel stained with GelStar (Lonza, Rockland, Maine).

The rabbit tick Haemaphysalis leporispalustris was the only tick found, and it was detected on 47 of 48 rabbits (Table 1). There were 0 to >20 ticks per rabbit, with 48% (23 rabbits) having zero to five ticks, 15% (seven rabbits) from 6 to 10 ticks, and 25% (12 rabbits) from 11 to 16; 12% (6 rabbits) had >16 ticks. Of 375 ticks collected, 32% were adult females, 22% were adult males, 19% were nymphs, and 27% were larvae. All tick pools were negative by PCR for B. burgdorferi and R. rickettsii. Four tick pools were PCR positive for A. phagocytophilum including one pool comprising three adult males (C_T = 43.6), one larva (C_T = 42.3), and two pools each of three nymphs (C_T = 42.3 and 39.3). The minimum and maximum estimated prevalence for A. phagocytophilum was 1.1–2.7%.

Blood testing was performed on 41 rabbits with all rabbits antibody negative for Rickettsia spp. and B. burgdorferi and culture negative for Bartonella spp. Fourteen rabbits were PCR positive for A. phagocytophilum and six were antibody positive. One rabbit was PCR and antibody positive, and one PCR-positive rabbit was parasitized by a PCR-positive tick (Table 2). The weakly positive PCR products were insufficient for DNA sequencing.

In this study we found little evidence for vector-borne pathogens, although the high burden of ticks would serve as a risk for pathogen spread in the future. There was heavy infestation by H. leporispalustris, but not by other common ticks such as Ixodes spp. and Dermacentor spp., although Dermacentor spp., Ixodes spp., and H. leporispalustris have been described on brush rabbits in Monterey County, California (Coultrip et al. 1973) and on jack rabbits (Lepus californicus) in Mendocino County, California (Eisen et al. 2004). Ixodes pacificus could easily have entered the rabbit enclosures on birds or small mammals but may not be locally common or questing in this season, or they may not prefer to feed on riparian brush rabbits. The lack of Dermacentor spp. is unexpected because they are common in this region of California (Furman and Loomis 1984). Haemaphysalis leporispalustris is a three-host tick primarily of lagomorphs, although nymphs and larvae are sometimes found on birds (Sonenshine and Stout 1970). This allows for spread of disease because birds can bring infected ticks from long distances. The highest tick loads occurred in September, although sampling only occurred for 5 mo. In California, adult and nymphal H. leporispalustris ticks have been found on hosts year round, with increased activity in spring; larvae are reported in spring, summer, and fall (Furman 1984). Further surveillance for ticks during other seasons would be important.

There was no evidence of R. rickettsii, B. burgdorferi, or Bartonella spp. risk. The exact duration of detectable antibodies for these pathogens in brush rabbits is unknown, although antibodies would likely persist for at least a few months. A study in 203 rabbits concluded that H. leporispalustris is an incompetent vector for A. phagocytophilum (Goethert and Telford 2003). Anaplasma phagocytophilum has been identified in cottontail rabbits (Goethert and Telford 2003), but we could not confirm whether it was present in the riparian brush rabbits. The TaqMan results could have been positive because of possible cross-reaction with some non-Anaplasma target, although the primers and probe for this test have been verified by extensive BLAST searching to produce a very specific product (Drazenovich et al. 2006). More likely the samples contained only a small amount of DNA (as evidenced by the high TaqMan C_T values).

This captive population of brush rabbits carries a heavy tick burden, which may aid in the spread of tick-borne pathogens. Because H. leporispalustris appears to be an incompetent vector of A. phagocytophilum, there may be little risk of transmission of tick-borne diseases from riparian brush rabbits to people or other animals in this habitat. However, with the continued translocation of rabbits, further sampling should be done to more thoroughly investigate the presence of A. phagocytophilum and monitor for disease and ticks year-long to better define the risk of tick-borne disease in riparian brush rabbits.

We thank Daniel Rejmanek, Joy Worth, and Regina (Jazzy) Dingler for laboratory support; Kent Lloyd and the UC Davis STAR program for support of the research; Matthew Lloyd and Jeff Holt from the Endangered Species Recovery Program for extensive contributions in rabbit capture and handling; and Ray Wack and the Sacramento Zoo for contributing veterinary time, equipment, and facilities to this project.