Zoonotic Bartonella Species in Cardiac Valves of Healthy Coyotes, California, USA

We investigated whether Bartonella spp. could cause endocarditis in coyotes or localize to cardiac valves before lesions develop. Bartonella DNA was amplified more often from coyote cardiac valves than spleen. Bartonella infection apparently leads to cardiac valve tropism, which could cause endocarditis, an often lethal complication in mammals, including humans.

coyotes has never been investigated. We hypothesized that B. v. berkhoffii or other Bartonella species could cause endocarditis in coyotes. We also hypothesized that bartonellae might preferentially localize to the aortic and/ or mitral valves before vegetative lesions develop. Hence, coyotes served as a naturally occurring epidemiologic and physiologic sentinel model for studying infection kinetics and pathology induced by this bacterium in a reservoir host (coyotes).

The Study
During the early 2000s, a total of 70 coyotes trapped in 9 northern California counties as part of a depredation control program were assessed for zoonotic heartworm (Dirofilaria immitis) disease (9). Coyote hearts and spleens were collected at that time and stored at -70°C in a manner to avoid DNA carryover during handling, storage, and processing. In 2012 and 2013, the hearts were dissected for macroscopic evidence of aortic and mitral valve vegetative endocarditis lesions. A board-certified veterinary pathologist examined possible valvular lesions or thickened regions; however, because the samples had been frozen, microscopic histopathologic examination was not conducted. We extracted DNA from aortic and mitral valves and spleens using DNeasy Blood and Tissue Kits (QIAGEN, Hilden, Germany). B. v. berkhoffii-spiked rabbit blood was the DNA extraction positive control. We tested extracted samples by PCR for Bartonella DNA targeting the citrate synthase gene (gltA) (10). PCR of spleen tissue was a substitute for blood culture detection of bacteremia. B. henselae DNA and distilled water were PCR-positive and -negative controls, respectively. Partial gene sequencing was performed on PCRpositive tissues. Nineteen aortic valve, mitral valve, and splenic DNA samples from 14 coyotes (B. v. berkhoffii PCR-positive animals by gltA PCR and sequencing) were genotyped by using primers targeting 16-23S intergenic transcribed spacer (ITS) region, as previously described (11) with minor modifications in annealing temperature (68°C for 15 s) and extension (72°C for 18 s). We conducted statistical analysis for differences in tissue tropism using Epi Info version 6 (Centers for Disease Control and Prevention, Atlanta, GA, USA).
We found no gross vegetative aortic or mitral valvular endocarditis lesions. Fifteen (21%) coyotes tested  (Table). Although a higher percentage of positive cardiac valves were aortic (53%) than mitral (33%), the difference was not significant. However, when we compared the number of Bartonella-infected cardiac valves (11 valves) with Bartonella-infected spleens (3 spleens), we found that Bartonella DNA was amplified 4.16 times (95% CI 1.02-24.12) more often from cardiac valves than from spleens.
Partial DNA sequencing showed that aortic valves from 8 (53%) of 15 coyotes were B. v. berkhoffii positive, compared with mitral valves from 4 (27%) and spleens from 3 (20%) coyotes. B. rochalimae was amplified from the spleen of coyote no. 99, and B. henselae DNA was amplified from the mitral valve of coyote no. 137 (Table). Of 14 coyotes tested for B. v. berkhoffii genotypes by 16-23S ITS PCR, 8 were positive, whereas Bartonella DNA was not amplified from the remaining 6 tissue DNA samples by using ITS primers. By sequence analyses, 4 coyotes were infected with B. v. berkhoffii genotype I, 3 with genotype II, and 1 with genotype III.
Nevertheless, the fact that ≈20% of the cardiac valve tissues were PCR positive for Bartonella perhaps indicates that the bacteria had localized to the valves of infected coyotes. B. v. berkhoffii can induce vasoproliferative lesions in animals (13); thus, cases of Bartonella endocarditis might represent only a small fraction of infected animals that have chronic cardiac valvular localization. All 3 B. v. berkhoffii genotypes identified in these coyotes have been previously involved in humans or dogs with endocarditis (14,15). To our knowledge, B. henselae and B. v. berkhoffii genotype III have not been previously identified in coyotes; thus, these mammals can be added to the list of susceptible species. Coyotes might be a natural reservoir for B. v. berkhoffii genotype III, which so far has been mainly described in California gray foxes (14).
In conclusion, Bartonella infection of a natural reservoir appears to lead to cardiac valve tropism. This tropism could result in development of endocarditis, a severe and often lethal complication of Bartonella infection.
Mr Kehoe is a fourth-year veterinary student at the University of California, Davis. His primary research interests include zoologic and wildlife medicine and health and emerging zoonotic infections.