Vector-Borne Infections in Tornado-Displaced and Owner-Relinquished Dogs in Oklahoma,USA

Introduction

Canine vector-borne disease (CVBD) agents are common in Oklahoma dogs, although prevalence estimates vary widely. A high incidence of tick infestation in dogs and coyotes results in antibodies reactive to Ehrlichia spp. in 10.8–64% and antibodies to spotted fever group (SFG) Rickettsia spp. in 38–60% (Murphy et al. 1998, Starkey et al. 2013). In a controlled exposure study, every dog (10/10) seroconverted to both Ehrlichia spp. and SFG Rickettsia spp. after 8 weekly walks in tick habitat (Barrett et al. 2014, Starkey et al. 2014). This high prevalence of antibodies is due, in part, to cross-reactivity (Barrett et al. 2014); recent specific peptide-based approaches suggest that only 1.0–7.2% of pet dogs in the state have antibodies to individual Ehrlichia sp. (Beall et al. 2012).

Mosquito-borne heartworm (Dirofiliaria immitis) infection is also common in dogs not on heartworm control products although a national serologic survey based on results of testing well-managed pet dogs using in clinic, membrane-bound ELISA assays reported that only 2.1% of Oklahoma dogs were infected (Bowman et al. 2009). Detection of D. immitis antigen can be masked by antigen–antibody complexing, resulting in false negative results, a phenomenon that may lead to underestimation of true prevalence (Velasquez et al. 2014). Here we describe the prevalence of select CVBD agents in a cross-section of dogs in central Oklahoma, including pet dogs displaced by a major tornado and dogs relinquished to an animal shelter.

Materials and Methods

Blood samples (serum and EDTA-anticoagulated blood) were available from 105 dogs in Oklahoma, including 44 dogs at a temporary shelter after the 2013 Moore, OK tornadoes and 61 dogs from a municipal animal shelter. All samples were collected as part of routine management practices at both facilities following approved shelter protocols. Patient-side assays using commercial ELISA for antigen of D. immitis and antibodies to Borrelia burgdorferi, Ehrlichia spp., and Anaplasma spp. (SNAP^® 4DX Plus^®; IDEXX laboratories, Inc., Westbrook, ME) were performed using whole blood according to manufacturer's instructions when samples were collected. Remaining samples were stored at −20°C until additional assays were performed.

Serum samples were tested for antigen of D. immitis before and after heat treatment using a commercial microtiter well assay (DiroCHEK^®; Synbiotics Co., Kansas City, MO) as previously described (Velasquez et al. 2014). Antibodies reactive to SFG Rickettsia spp. were identified through immunofluorescence antibody assays (IFA) for R. rickettsii, R. montanensis, and “R. amblyommii” as previously described (Barrett et al. 2014).

A commercial kit (Illustra blood genomicPrep Mini Spin Kit; GE Healthcare United Kingdom Limited, Buckinghamshire, United Kingdom) was used, according to the manufacturer's instructions, to extract DNA from whole blood samples. DNA of SFG Rickettsia spp. was identified by 17 kDa and rompA nested polymerase chain reaction (PCR) as previously described (Barrett et al. 2014). As previously described, all positive amplicons were purified and submitted for sequencing through institutional and commercial services; resultant sequences were analyzed and aligned (Barrett et al. 2014).

The Mann–Whitney U-test was implemented to determine statistical differences between pet and shelter dog prevalence of SFG Rickettsia spp. antibodies. Fischer's exact tests (two-tailed, α ≤ 0.05) were used to compare positive PCR and D. immitis test results between groups of dogs.

Results

Antibodies to Ehrlichia spp., but not to A. phagocytophilum or B. burgdorferi, were detected by commercial ELISA in 4/44 (9.1%) pet and 7/61 (11.5%) shelter dogs. Antibodies to SFG Rickettsia spp. were detected by IFA (>1:128) in 36/44 (81.8%) pet dogs and 56/61 (91.8%) shelter dogs. Shelter dogs were more likely to have antibodies to R. rickettsii than pet dogs (p = 0.03). Characteristic fragments of Rickettsia spp. were amplified from shelter dogs, with 13.1% (8/61) and 1.6% (1/61) identified through rompA and 17 kDa gene targets, respectively. Sequence of one amplicon was identical to “R. amblyommii” (GenBank CP003334.1); multiple attempts to sequence additional amplicons were unsuccessful.

Antigen of D. immitis was detected in 6/44 (13.6%) pet dogs and 5/61 (8.2%) shelter dogs using both the membrane-bound ELISA and the microtiter well ELISA before heat treatment and 10/44 (22.7%) pet dogs and 7/61 (11.5%) shelter dogs using the microtiter well ELISA after heat treatment of serum samples (Table 1).

Discussion

Antibodies to Ehrlichia spp. were identified in >10% of the dogs, and a majority (>80%) had evidence of past or current infection with SFG Rickettsia spp. The common nature of tick-borne infections in these dogs is not surprising. Multiple ixodid ticks, including Amblyomma spp. and Dermacentor variabilis, which are known to harbor or transmit a variety of Ehrlichia spp. and Rickettsia spp., are active in Oklahoma (Koch and Dunn 1980). Cross-reaction among antibodies to SFG Rickettsia spp. commonly occurs and likely elevated the seroprevalence seen in this study (Barrett et al. 2014). The finding that shelter dogs were more likely to have antibodies to R. rickettsii than pet dogs was somewhat unexpected because shelter dogs tend to be younger than pet dogs (New Jr. et al. 2000), but may reflect a more intense history of infestation in the former group.

This study also confirmed “R. amblyommii” by PCR in a naturally infected dog. Amblyomma americanum ticks have been shown to harbor “R. amblyommii” throughout their geographic range and are thought to be responsible for transmitting this infection (Barrett et al. 2014). A previous study documented research that dogs free of infection before walking in tick habitat seroconverted to and developed PCR evidence of “R. amblyommii” infection after a relatively brief tick exposure period (Barrett et al. 2014), and R. parkeri has been identified in shelter dogs by PCR (Grasperge et al. 2012), but the present article represents the first report of a naturally infected dog harboring “R. amblyommii.” Unfortunately, SFG Rickettsia spp. DNA is rarely detected in circulating whole blood even when infection is present (Barrett et al. 2014), and thus it is not surprising that the vast majority of dogs in this study were PCR negative to SFG Rickettsia spp.

Antigen of D. immitis was detected in more than 16% of dogs in this study, a prevalence almost eight-fold higher than previously reported from pet dogs in the region (Bowman et al. 2009). This increase is likely due to both the population of dogs considered and improved detection with the assays used. National prevalence values are derived from testing well-cared-for dogs in veterinary practices, an approach that likely creates negatively biased prevalence estimates (Bowman et al. 2009). Moreover, heat treatment of samples before testing disrupts immune complexes, freeing antigen for detection in some samples (Velasquez et al. 2014). In this study, we had the opportunity to test a cross-section of both pet dogs and shelter dogs, animals that may not be included in clinic-based surveys, providing a more accurate representation of the high prevalence of CVBD in the general dog population in the region.