Detection and analysis of tick-borne infections in communal dogs of northwest Zimbabwe

Domestic dogs (Canis familiaris) may serve as a reservoir or a sentinel for infectious disease pathogens that can affect human and wildlife health. To understand the role of tick-borne diseases in rural and lesser developed regions, we investigated the prevalence of several tick-borne pathogens in communal dogs of Zimbabwe. Blood samples from 225 dogs in northwest Zimbabwe were assessed by serology for Ehrlichia canis, Anaplasma phagocytophilum and Borrelia burgdorferi, and 241 samples were assessed by polymerase chain reaction (PCR) for Ehrlichia. There was a high seroprevalence (73%) of E. canis-specific antibodies in domestic dogs in northwest Zimbabwe, but follow up analyses via PCR and genetic sequencing indicated only 7.5% of the canines were actively infected with the organism. Whilst indicating that an organism serologically related to E. canis is likely present in the region, this data also shows that the organism is currently present in a relative minority of the domestic dogs in the region. Its presence as evidenced by both serologic and PCR analysis is significant because of the ‘one health’ paradigm, where humans and wildlife may be affected by the exposure to this pathogen in domestic dogs.


Introduction
Tick-borne diseases pose a significant risk to humans and domestic animals as well as wildlife in sub-Saharan Africa (Dantas-Torres, Chomel & Otranto 2012; Lightfoot & Norval 1981;Perry & Young 1995). Dogs may be the reservoirs of several infectious agents of significance, including tick-borne pathogens, and can serve as sentinels indicating when specific pathogens may be present in an area (Cleaveland et al. 2006;Day 2011;McRee et al. 2014). These agents directly impact the health and welfare of domestic dogs and may also have significant effects on wildlife and humans as well as other domestic species with which they interact (Cleaveland et al. 2006). Specifically, the 'One Health' paradigm, which recognises the interconnectivity of humans, animals and the environment, has been used to demonstrate that domestic dogs and cats are involved in transmission cycles of some tick-borne diseases (Skotarczak 2018). Likewise, endangered African wild dogs (Lycaon pictus) show increased pathogen exposure when in contact with domestic dogs (Woodroffe et al. 2012).
Ticks bites can transmit the bacterial pathogens Ehrlichia spp., Anaplasma spp. and Borrelia burdorferi (Groves et al. 1975;Johnson et al. 1984;Stuen 2007). Canine ehrlichiosis is known to occur in South Africa with seroprevalence levels of up to 75% with a higher prevalence in economically depressed areas, yet the causative organism has neither been isolated nor characterised (Inokuma et al. 2005;Pretorius & Kelly 1998). There is even less definitive data from other parts of Africa. A serological survey for antibodies reactive with E. canis done on 228 dogs in rural Zimbabwe in 2004 showed evidence of exposure in over a third of the dogs tested (Kelly, Eoghain & Raoult 2004). Only one of these pathogens, B. burdorferi (Lyme disease), can be prevented by vaccination, and Lyme disease has not yet been isolated in Zimbabwe. Furthermore, many people in rural Zimbabwe lack the financial resources to purchase vaccines for their dogs. Additionally, tick control is important for prevention of these tick-borne bacterial pathogens, although medications available for tick control in dogs are also expensive.
Good quality data on the importance and means of control of infectious diseases in sub-Saharan Africa are deficient partly because of lack of access to veterinary attention as well as economic Domestic dogs (Canis familiaris) may serve as a reservoir or a sentinel for infectious disease pathogens that can affect human and wildlife health. To understand the role of tick-borne diseases in rural and lesser developed regions, we investigated the prevalence of several tickborne pathogens in communal dogs of Zimbabwe. Blood samples from 225 dogs in northwest Zimbabwe were assessed by serology for Ehrlichia canis, Anaplasma phagocytophilum and Borrelia burgdorferi, and 241 samples were assessed by polymerase chain reaction (PCR) for Ehrlichia. There was a high seroprevalence (73%) of E. canis-specific antibodies in domestic dogs in northwest Zimbabwe, but follow up analyses via PCR and genetic sequencing indicated only 7.5% of the canines were actively infected with the organism. Whilst indicating that an organism serologically related to E. canis is likely present in the region, this data also shows that the organism is currently present in a relative minority of the domestic dogs in the region. Its presence as evidenced by both serologic and PCR analysis is significant because of the 'one health' paradigm, where humans and wildlife may be affected by the exposure to this pathogen in domestic dogs.

Detection and analysis of tick-borne infections in communal dogs of northwest Zimbabwe
Read online: Scan this QR code with your smart phone or mobile device to read online.

barriers (McDermott & Arimi 2002
). Since many infectious pathogens affect multiple hosts, it is often more feasible, less expensive and less risky to determine the prevalence of various infectious agents in a domestic species -one that shares an environment with human and wildlife populations.
Our objective was to investigate the seroprevalence of three bacterial agents carried by ticks (Ehrlichia canis, Anaplasma phagocytophilum and Borrelia burgdorferi). In addition, detection of the most seroprevalent infectious agent related to E. canis was undertaken using PCR on whole blood.

Research methods and design Sites
Dogs inhabiting the Hwange District Rural communal lands bordering both the Victoria Falls and the Zambezi National Park were used for this investigation. These lands are in the vicinity of wildlife preserves and encompass a major township in Zimbabwe. The annual rainfall in the Hwange District is less than 450 mm annually, and the temperatures are as high as 25 °C and above (Nhemachena et al. 2014

Sampling
Dogs were aged based on appearance, dentition and history when known. Dogs were classified as adults (> 3 years old), young adults (1-3 years old) and juvenile (< 1 year). Dog owners were encouraged to participate in the study by providing donated multivalent canine distemper-canine adenovirus-2-canine parvovirus-2 and rabies vaccines (Merial, Atlanta, GA, United States [US]) as incentives.

Enzyme-linked immunosorbent assay
Serum samples (n = 225) collected in 2012 were tested for antibodies to E. canis, A. phagocytophilum, B. burgdorferi and Dirofilaria immitis using the IDEXX 4DX SNAP enzymelinked immunosorbent assay (ELISA) (IDEXX, Inc., Portland, ME, US) following manufacturer instructions. Borrelia burgdorferi was primarily investigated because of its inclusion in the ELISA kit; however, evaluating potential spread of Lyme disease was a secondary benefit. Serologic testing on samples collected in 2014 was not done, as kits were unavailable.

Polymerase chain reaction (PCR) and genetic analysis
To detect Ehrlichia organisms, we first extracted whole blood using the QIAamp deoxyribonucleic acid (

Data analysis
Serological and PCR results were classified and recorded as positive or negative for each pathogen tested. Crude and factor-specific seroprevalence of each pathogen was then computed. Factors considered were animal age, sex and geographic location. Associations between seroprevalence and each of the factors were assessed using Fisher's exact test.
A p-value of < 0.05 was used to indicate statistical significance.
In 2014, 90% of the samples were collected from young adult dogs, and all samples were collected in mid-July.

Discussion
This investigation shows a significant level of exposure to an E. canis-related organism amongst communal dogs in Zimbabwe. For those whom serologic assays alone were done, it is unknown when exposure occurred. Genetic detection of the organism revealed a much lower prevalence level for active current infection. Testing was done opportunistically during June and July of 2012 and 2014. This corresponds to winter, or the dry season in Zimbabwe, when tick populations are not expected to be at their highest. Nevertheless, the organism was detected and found to be closely related to E. canis. The majority of dogs tested were young adults; even so, there was not an age association with exposure. That is, exposure likely occurs very early in a dog's life, as soon as the first tick season after birth. There was not an even distribution of exposed dogs in the region tested, but exposure was found at a considerable level in all areas.
Because there is extensive serologic cross reactivity amongst ehrlichial species, we cannot say for certain that other species of ehrlichial organisms such as E. chaffeensis and E. ruminantium or even uncharacterised organisms are not present (Kelly et al. 2004 (Mtshali et al. 2017). Related organisms that infect humans are also carried by these vectors and occur in Africa, including E. chaffeensis and E. ewingii (Brouqui et al. 1994;Ndip et al. 2005). In 2000, an organism related to A. phagocytophilum, the agent of human and canine granulocytic anaplasmosis was identified in dogs in South Africa (Inokuma et al. 2005). The significance of all these infectious agents amongst animals in Zimbabwe remains unknown. There is a critical need to determine the prevalence of both Ehrlichia and Anaplasma in the domesticated canine population to identify the potential vectors associated with those canines and to characterise the risk factors associated with each pathogen. Without this knowledge, it will not be possible to manage these ticks and minimise pathogen transmission of these emerging and preventable diseases.
In summary, testing of communal dogs in Zimbabwe revealed a high seroprevalence for organisms related to E. canis. Polymerase chain reaction detection of the organism indicated that there is significant identity with E. canis, but as only 18 samples produced positive results, it is possible other serologically related organisms may also be involved.

Ethical considerations
Research was approved by the University of Tennessee Animal Care and Use Committee. The Wild Horizons Wildlife Trust invited the authors to assist with disease surveillance and diagnosis at the Wildlife Laboratory and Clinic in Zimbabwe. The facility is the only one in the Kavanago Zambezi Trans-frontier Conservation Area (KAZA TFCA).
All samples were collected under the direct supervision of staff including a veterinarian from Victoria Falls Wildlife Trust in accordance with their mission, which states: 'To actively advance and promote environmental conservation and the sustainable use of indigenous resources in Southern Africa'.

Funding information
A.M. and R.E.T. were funded for travel by Morris Animal Foundation Veterinary Fellow Scholarship.

Data availability
The authors confirm that the data supporting the findings of this study are available within the article.