Tick-borne pathogens represent an important One Health issue, as many can cause disease in domestic and agricultural animals, wildlife, and humans. In the present study, we found evidence of exposure to and/or infection with numerous tick-borne pathogens in dogs from Chad. Several of these pathogens are, or have the potential to be, zoonotic, and many of the tick species found on Chadian dogs also infest humans [7]. Additional studies are needed in Chad to monitor the prevalence and transmission of these pathogens, specifically, to understand the risks they pose to the health of domestic animals and humans.
Similar to other studies of African domestic dogs, Hepatozoon spp., specifically H. canis, was the most common pathogen detected, with a 40–94% prevalence depending on the region and time point of sampling (e.g., [3, 19–21]). A multicountry study revealed a commensurately high prevalence of H. canis (average of 59%; Tanzania: 67–77%; Kenya: 54–85%; Uganda: 86–98%; Nigeria: 26–56%; Ghana: 46–68%; Namibia: 9–29%) [3]. Other studies also found a high prevalence in Sudan (42%), Ghana (40%), and Nigeria (41%) [19–21]. This characteristically high prevalence with wide distribution has been attributed to a large number of known vectors, including Rhipicephalus spp., and the potential for vertical transmission to puppies [22–24].
The detection of antibodies against Ehrlichia spp. and the molecular detection of E. canis were not surprising, as this pathogen has been reported in dogs from Chad and other African countries. Moreover, most dogs in this study were infested with R. sanguineus [11], the primary vector of E. canis [3, 11, 19–21, 25–31]. The 86% seroprevalence of Ehrlichia spp. in Chadian dogs in our study was considerably greater than that in a previous study in Chad (5/18 clinically normal military dogs, 28%) and against comparable studies in Ghana (21–35%), Sierra Leone (40%), and Nigeria (32–54%); however, our data were consistent with those of a study from Senegal (89%) [3, 20, 26–28]. Generally, the prevalence of E. canis antibodies in southern and eastern African countries was lower (e.g., Tanzania: 29–32%; Kenya: 15–22%; Uganda 4–10%; Namibia: 25–40%) [16]; however, variation does exist, and higher prevalence rates have been reported (e.g., 73% in Zimbabwe, 96% in Sudan, and 87% among sick dogs in Namibia) [28–30]. The PCR prevalence of E. canis in Chadian dogs ranged from 4–39%, depending on the region and time point of sampling, similar to the findings of studies in numerous other sub-Saharan African countries, including neighboring Nigeria [21, 31]. While 86.4% of dogs were seropositive for Ehrlichia spp. in May 2019, only 27.8% were confirmed to be actively infected with E. canis at that time. A similar trend was noted in dogs from Zimbabwe, as well as several other southern and sub-Saharan African countries [3, 30]. This can be explained by dogs having been infected previously and cleared the infection but still having antibodies present in the blood. Alternatively, these animals may have been infected by an Ehrlichia spp. other than E. canis, e.g., with E. ewingii, E. ruminatium, E. chaffeensis, and potentially new species of Ehrlichia also reported in West Africa [1]. Variation in prevalence similar to that documented in this study has been observed among other western and sub-Saharan African countries, including 20% of dogs from Ghana, 7.3% of dogs from the Ivory Coast, 12.7% of dogs from Nigeria, and 6.4% of dogs from Algeria [20, 21, 32, 33].
Canine cyclic thrombocytopenia, caused by A. platys, is a significant disease of dogs in many regions of the world, and similar to E. canis, R. sanguineus is a suspected vector [1, 6, 7]. This pathogen is also a rare zoonosis [34]. Our finding of 21% seroprevalence for Anaplasma spp. among dogs in Chad is comparable to that in other countries in southern and sub-Saharan Africa, e.g., Ghana: 0–30%; Sierra Leone: 19%; Kenya: 8–10%; Nigeria: 4–20%; Tanzania: 20–21%; Uganda: 4–24%; and Namibia: 8–23% [3, 20, 27]. In Zimbabwe, 10% of 225 samples were seropositive [30]. Interestingly, in May 2019, only 20.5% of the dogs were seropositive for Anaplasma spp., while 22% of the dogs were PCR positive for A. platys. This difference is likely due to recently infected animals not having yet mounted an antibody response, as the response is first detectable 16 days after infection [35]. In our study, the prevalence of A. platys via PCR varied from 6 to 24%, depending on the region and time point of sampling, which is similar to the findings in nearby countries in sub-Saharan Africa (Kenya (10–23%), Ghana: 10%; Ivory Coast: 1.5% and 0–30%; Gabon: 1.2%; and Nigeria: 6.6%), as well as northern Africa (Algeria: 5.4%) [7, 20, 21, 32, 33].
A small number of the dogs sampled in this study (n = 13) were positive for B. c. vogeli, one of three canine subspecies of B. canis (B. c. canis, B. c. rossi, and B. c. vogeli), which are distinguished by biological characteristics and molecular methods [10, 36]. For example, B. c. rossi is transmitted by Haemaphysalis spp., and infection is typically fatal, while B. c. vogeli is transmitted by R. sanguineus and is considered the least pathogenic [10]. The prevalence of Babesia spp. in dogs in sub-Saharan Africa varies considerably from 0–12% depending on country and rural vs urban area [3]. In countries neighboring Chad, 9% of dogs from Sudan were positive for Babesia spp. (five with B. c. rossi and two with B. c. vogeli) [19], and both B. c. rossi and B. c. vogeli have been detected in dogs in Nigeria, but B. c. rossi was more common [21, 37, 38]. The lack of B. c. rossi in our dogs may be due to the low number (n = 14) of dogs infested with Haemaphysalis leachii [11]. Additional studies to determine the distribution and factors related to the presence and intensity of H. leachii are needed to better understand the risk of severe babesiosis to the health of dogs in Chad.
The high number of dogs that were positive for antibodies against both Anaplasma spp. and Ehrlichia spp. is not surprising, given that both pathogens are transmitted by R. sanguineus [6, 27]. Among 53 dogs from Sierra Leone tested with the SNAP 4Dx test, 9.4% were positive for both Ehrlichia spp. and Anaplasma spp. antibodies, and 5.7% were positive for Ehrlichia spp. antibodies, Anaplasma spp. antibodies, and D. immitis [27]. Furthermore, antibodies against these pathogens have been shown to persist for months to years [39, 40]. Coinfections with two or three pathogens were detected by PCR in 24.2% of the samples in our study, with the most common combination being Hepatozoon spp. and Ehrlichia canis, and 4.0% of the samples had three pathogens detected. This finding is similar to that of a multinational study of African dogs, in which 30.9% of the dogs were coinfected with at least two pathogens, the most common combination (10.1%) being H. canis and E. canis, and 5.1% of the dogs had three or four pathogens in their blood[3]. Coinfections are not surprising given that these pathogens share at least one tick vector, Rhipicephalus sanguineus, and this tick was commonly detected on dogs in this study.
There were several significant spatiotemporal and demographic factors associated with the detection of exposure or infection with multiple pathogens included in this study. Dogs in the southern region were more likely to be seropositive for Ehrlichia spp. and Anaplasma spp. and to be infected with Hepatozoon spp. and E. canis than dogs in the northern region were. This may be explained by climate variation within Chad, with differences between regions that can impact tick populations: the northern study areas are more arid, and the southern region of Chad receives more rainfall [41]. For all three pathogens (Hepatozoon, A. platys and E. canis), the time point was a significant predictor of detection. In the northern region, Hepatozoon spp. were more likely to be detected later in the study (November 2019 and June 2020 > May 2019), whereas in the southern region, Hepatozoon spp. were more likely to be detected earlier in the study (May 2019 and June 2020 > November 2019). Overall, A. platys and E. canis were more likely to be detected earlier in the present study (May 2019 vs November 2019 and June 2020). Ehrlichia canis infection typically occurs during the dry-hot season when the tick Rhipicephalus sanguineus is active [8]. It is also possible that the removal of ticks from the study dogs at the three time points may have reduced the pathogen prevalence at later time points, as ticks were no longer present to transmit the pathogens of interest; however, that only represented a few days throughout the year.
Dog age was a significant predictor for the detection of A. platys by PCR. Our finding that younger dogs were more likely to be infected with A. platys agrees with the findings of previous work in Kenya and Ivory Coast that showed a 19.8% prevalence in dogs younger than one year, compared to 6.7% in adult dogs [7]. This finding supports our finding that dogs were more likely to be infected in May 2019 than at the two later time points of the study, as dogs were youngest at the first time point of the study. Furthermore, while infections with A. platys persist for months, dogs may clear infections after 100–150 days [35]. This is consistent with our finding that dogs were more often positive for A. platys at two consecutive time points (14 dogs) than at the first and third time points (six dogs) or at all three time points (two dogs).
The detection of antibodies against B. burgdorferi was unexpected based on the historical range of this pathogen in North America and Eurasia and its predominant association with Ixodes spp. ticks [42]. However, there are sporadic reports of this pathogen outside the expected range. In Africa, 1.4% of dogs in rural Kenya were seropositive for Borrelia spp. [3], and a single dog in Egypt and an associated Rhipicephalus sanguineus tick tested positive via PCR [43]. Another study in Egypt detected B. burgdorferi via PCR in 23% of dogs (n = 26), 16% of cattle (n = 25), 58% of dog-associated R. sanguineus (n = 12), and 21% of bovine-associated Hyalomma anatolicum excavatum (n = 14) [44]. Although no Ixodes were found on any of the dogs in this study, the three B. burgdorferi-positive dogs were infested with R. sanguineus. In addition to the true exposure of Chadian dogs to B. burgdorferi, there are other possible explanations for these findings. It is possible that our results represent cross-reaction with other Borrelia spp. or false positives. In Africa, relapsing fever group (RFG) Borrelia spp., such as B. recurrentis in countries east of Chad and B. crocidurae in countries north of Chad, have been reported, but rarely do RFG Borrelia cross-react with C6-based serologic tests [45, 46]. However, only a limited number of Borrelia species have been evaluated, so it is possible that some RFG Borrelia may cross-react. Additionally, a novel lineage of Borrelia, distinct from both the relapsing fever and Lyme disease groups, has been reported in Amblyomma spp. from Ethiopia and the Ivory Coast, including A. variegatum, a tick species found on dogs in Chad [11, 47, 48]. The cross-reactivity of this group with B. burgdorferi C6 assays is not known.
Aspects of this study limit the conclusions that can be drawn from the data. Importantly, ticks were opportunistically collected from dogs enrolled in an experimental therapeutic trial for the treatment and prevention of Guinea worms (Dracunculus medinensis) [13]. Therefore, sample size calculations and counts of total tick burden per dog were not performed, which limits the interpretability of the results. Additionally, there were only three time points of sample collection, and the SNAP 4DX tests were performed only at the first time point. More robust conclusions about prevalence trends could be drawn from data collected over many years, with multiple years of sampling during each season. Another limitation was that outwardly sick dogs were excluded based on the primary study criteria; therefore, analyzing data for associations between pathogen detection and clinical illness was not possible. Moreover, the number of dogs sampled decreased over time as dogs either died or moved with their owners away from the village where they were originally sampled. Finally, the ticks found on each dog were removed for subsequent testing at each time point, potentially reducing pathogen transmission and impacting prevalence estimates at later time points.