Domestic dogs as reservoirs for African trypanosomiasis in Mambwe district, eastern Zambia

The control of African trypanosomiasis (AT) in Eastern and Southern Africa, including Zambia, faces huge challenges due to the involvement of wild and domestic animal reservoirs. Free-roaming dogs in wildlife-populated and tsetse-infested villages of Zambia’s Mambwe district are exposed to infectious tsetse bites. Consuming fresh raw game meat and bones further exacerbates their risk of contracting AT. We focus on the reservoir role of such dogs in maintaining and transmitting diverse species of trypanosomes that are infective to humans and livestock in Zambia’s Mambwe district. A cohort of 162 dogs was enrolled for follow-up at 3 different time points from June to December 2018 in selected villages of Malama, Mnkhanya, and Nsefu chiefdoms of Mambwe district, eastern Zambia. Blood and serum were screened for AT by microscopy, GM6 ELISA, PCR (ITS1 and SRA), and Sanger sequencing. Out of the 162 dogs in the cohort, 40 were lost to follow-up and only 122 remained traceable at the end of the study. GM6 ELISA detected Trypanosoma antibodies in 121 dogs (74.7%) and ITS1-PCR detected DNA involving single and mixed infections of T. congolense, T. brucei, and suspected T. simiae or T. godfreyi in 115 dogs (70.9%). The human-infective T. b. rhodesiense was detected by SRA PCR in 67 dogs (41.4%), and some sequence data that support the findings of this study have been deposited in the GenBank under accession numbers OL961811, OL961812, and OL961813. Our study demonstrates that the Trypanosoma reservoir community in Zambia is wider than was thought and includes domesticated dogs. As dogs are active carriers of human and livestock-infective trypanosomes, they pose a risk of transmitting AT in endemic villages of Mambwe district as they are neglected and left untreated. To fully bring AT under control, countries such as Zambia where the role of animal reservoirs is important, should not limit their prevention and treatment efforts to livestock (especially cattle) but also include dogs that play an integral part in most rural communities.


Detection by microscopy
Microscopy was conducted on all dogs at the first time point in June.The prevalence of trypanosomes in the blood was (8.0%; 95% CI 4.8-13.2).In most cases the parasitaemia was low, but when blood was injected into mice the parasitaemia increased several days post-inoculation (Fig. 3).

Detection by ELISA
The GM6 ELISA detected Trypanosoma antibodies in 121 dogs (74.7%; 95% CI 67.5-80.8)at a calculated cut-off value of 0.215.The cohort had a mean OD value of 0.372 that ranged from 0.028 to 2.414.Seroprevalence in the cohort varied during follow-up with more cases detected in the rainy season (Table 1).

Correlation among diagnostic tests
The 13 cases detected by microscopy were all accurately detected and confirmed that both tests were 100% sensitive: ELISA (kappa 0.134) and ITS1-PCR (kappa 0.146).However, there was only a slight agreement of 59.3% in case detection between ELISA and ITS1-PCR with a Kappa value of − 0.028.ELISA detected 36 positives that had no DNA and did not detect 30 samples with DNA (Table 3).Generally, the number of trypanosome-positive dogs in the cohort increased with time.Both ELISA and PCR detected more cases at the onset of the rainy season in December.Although trypanosomes were detected in dogs in all chiefdoms, cases were most prevalent in Malama.Of the 11 cases with corneal opacity, 7 (63.6%)were positive by ELISA and 9 (81.8%) by PCR.The two undetected cases were euthanized before subsequent follow-up.www.nature.com/scientificreports/amino acid sequence, asparagine was replaced by a serine thereby distinguishing it from known references (Fig. 5).

Discussion
Our study confirms that the reservoir community of trypanosomes in Zambia is wide and includes man's best friend, "dogs".Scanty trypanosomes or low parasitaemia suggest that most dogs had chronic infections.This is perhaps why some dogs recorded persistent infections throughout the 6-month follow-up period, long enough for infection to be passed on to other hosts.The propagation of scanty trypanosomes in mice indirectly implies that in the presence of competent arthropod vectors such as tsetse flies, tabanids and stomoxys flies which are abundant in Mambwe district, canine infections could be transmitted to wildlife, livestock, and humans.
The occurrence of single and mixed infections of T. congolense, T. brucei, and T. b. rhodesiense is concerning given that the former and latter cause devastating diseases in cattle and humans, respectively.However, the presence of these trypanosomes in dogs is not coincidental here, they have previously been reported in these chiefdoms 39 .More attention needs to be given to dogs in this area by relevant authorities.The detection of the SRA gene in trypanosomes isolated from many dogs poses a public health threat considering the close relationship between humans and dogs.Our research agrees with 41 who also amplified the SRA gene from dogs in Zambia.Other potential reservoirs in Zambia where the SRA gene has been amplified include cattle 36,37 , and wildlife (i.e., buffalo, vervet monkey, sable) 18,19 .The SRA gene has also been amplified from tsetse flies 35 , and humans 50,51 in Zambia.Whereas detecting "suspected" T. simiae and T. godfreyi positive cases is a new and interesting finding in dogs, our study did not verify these cases by sequencing but on the size of ITS bands on agarose gels.
Accurate diagnosis is therefore critical in detecting and preventing AT in endemic rural areas.However, most rural health facilities do not have modern sensitive tests and equipment to perform ELISA or PCR other than low-sensitive microscopy.Using GM6 ELISA (74.7%) and ITS1-PCR (70.4%), we were able to demonstrate in this study that microscopy (8%) was about 9 times less sensitive in detecting cases with kappa values of 0.134 and 0.146, respectively.The fact that the two tests picked up more positives than microscopy suggests that parasitaemia was below detectable levels on a microscope.Although we have compared GM6 ELISA and ITS1-PCR, we also want to point out that the two tests do not target the same thing.The former detects antibody titres that show exposure rather than potential active infections detected by the latter.Therefore, recent infections might mean no antibody detection or vice versa, so the two tests might show slight agreement in detecting cases as observed in our current findings.As this was the first attempt at using GM6 ELISA to detect AT seroconversion in naturally infected dogs, more research is needed to optimize the protocol.Otherwise, GM6 ELISA is a cheaper alternative diagnostic tool to ITS1-PCR that can be utilized by AT endemic facilities that rely on microscopy.Moreover, dogs in AT-endemic areas such as Mambwe district are neglected and do not receive treatment when infected or injured 40 .
In the absence of diagnostic tools, dog owners living in tsetse-infested areas such as Mambwe district can use corneal opacity as a sign to suspect canine AT.Corneal opacity is a common clinical feature of canine AT associated with T. brucei-subspecies infections 41,[52][53][54][55][56][57] .Amongst the pathognomonic clinical features of canine AT, corneal opacity was found to be one of the reliable predictors that can be used by dog owners in Mambwe district to suggest the involvement of trypanosomes 39 .By ELISA and PCR, we found 63.6% and 81.8% positivity amongst dogs with corneal opacity.The 11 cases of corneal opacity recorded in the current study mostly developed at the onset of the hot season from August onwards when cases were seen to increase.This corresponds with the influx of tsetse flies around human settlements "villages" in the hot season.Whereas infections could be due to infectious tsetse bites, the risk of AT in these dogs is further exacerbated by consuming fresh raw game meat and bones that can be acquired through oral lacerations 40 .
Our study demonstrates that the Trypanosoma reservoir community is wide and dogs that play an integral part in human establishments are active carriers.Humans and other mammalian hosts living close by are at risk as dogs are commonly left untreated.To fully bring AT under control, countries such as Zambia where the role of animal reservoirs is important, should not limit their prevention and treatment efforts to livestock (especially cattle) but include dogs.We are also cognisant of the fact that Mambwe district is not the only region in Zambia impacted by AT and tsetse flies, so it would be beneficial for other studies to investigate the current state of canine trypanosomes in other affected areas.Future research should compare the prevalence of trypanosomes in dogs owned by new settlers to those owned by long-term residents, as new settlers are more likely to live in www.nature.com/scientificreports/areas with higher tsetse fly infestations as is the case in Mambwe district.Moreover, as our study was limited in the duration of follow-up, we recommend that future studies should examine dogs for a much longer period to generate meaningful inferences on seasonal data.

Study area
Mambwe District, located in the Eastern Province of Zambia, spans approximately 5724.5 km 2 and has a human population of 119,313, with a density of 20.8 persons per km 240 .The canine population is estimated at 3155 (according to the Department of Veterinary Services, Mambwe District).Situated within the tsetse-infested Luangwa Valley basin, a historical hotspot for African trypanosomiasis (AT), the district's altitude can drop as low as 300 m above sea level.Mambwe experiences three seasons: a cold-dry season from May to early August, a hot-dry season from mid-August to mid-November with temperatures ranging from 32 to 43 °C, and a rainy-hot season from mid-November to April.The district is prone to localized flooding due to its location, with rainwater from higher neighbouring districts draining into Mambwe.Prolonged rainfall can overwhelm local drainage systems, rivers, and streams, causing flooding in low-lying areas and rendering most areas inaccessible by road except by boat or helicopter from November to May.Mambwe District is rich in wildlife.Its game management area (GMA) serves as an extension of the South Luangwa National Park (SLNP), hosting a variety of wildlife, including mammals, reptiles, and birds that migrate in and out of the district 58 .This study was conducted in wildlife-populated and tsetse-infested villages of Malama, Mnkhanya, and Nsefu chiefdoms in Mambwe district.The three chiefdoms were selected to encompass a variety of natural and human habitats and on previous cases of canine AT 39 .
i. Malama Chiefdom is situated in a densely wildlife-populated area on the outskirts of the SLNP.This remote location is heavily infested with tsetse flies and at the time of sampling had almost no livestock, except fowls and dogs.With the recent influx of new settlers, the livestock situation is more likely to change.ii.Mnkhanya Chiefdom is located on the edge of a wildlife zone, far from the SLNP.It has a very mild presence of wildlife and tsetse flies, with livestock and dogs being common.iii.Nsefu Chiefdom lies within a moderately wildlife-populated area.It has a moderate tsetse infestation, and livestock and dogs are prevalent.

Selection criteria
Building on the findings of Lisulo et al. 39 regarding the presence of AT in indigenous dogs in the Malama, Mnkhanya, and Nsefu chiefdoms of Mambwe district, we extended our current study to every dog-keeping village within these chiefdoms.Participation was voluntary, and only villages where dog keepers expressed a willingness to participate were included in the study.Individuals who did not consent or were below the minimum age of 12 were excluded.Consequently, we focused on 58 dog-keeping villages that expressed a willingness to participate: Malama (15 villages), Mnkhanya (18 villages), and Nsefu (25 villages).These included villages with previously identified cases of AT in dogs and additional villages that kept dogs.The selected villages were typically small, comprising one or more households from the same family.All three chiefdoms are predominantly inhabited by Kunda-speaking people, with other Zambian tribes as minority settlers.We enrolled male and female dogs aged over 3 months as part of the cohort.Although puppies as young as six weeks have been used in Trypanosoma in vivo studies elsewhere 55 , we excluded puppies less than 4 months old because they could still carry maternal antibodies against trypanosomes, which was one of the things we tested for in this study.
No additional households or dogs were included after the initial enrolment, and puppies born to females in the cohort were not considered 40 .

Study design
A cohort of 162 indigenous dogs (82 males and 80 females) from 93 dog-keeping households were enrolled for follow-up.On average, each household had 1.7 dogs (± 1.1), with the number of dogs per household ranging from one to five.The cohort's median age was 30 months (± 25), with ages spanning from four to 156 months.The dogs, from Malama (31 dogs), Mnkhanya (68 dogs), and Nsefu (63 dogs) were repeatedly examined for pathogens, injuries, illnesses, and mortalities.The cohort of dogs was monitored for 6 months, from June to December 2018, to track changes in their health and demographics.Throughout this period, the dogs were visited at three distinct time points, corresponding to different seasons: the initial visit and enrollment took place in June (cold-dry season), the second visit occurred in September (hot-dry season), and the final visit was in December (rainy-hot season) 40 .
As an incentive to the dog owners, each dog in the cohort was freely vaccinated against rabies at enrollment, dewormed at the second or third visit with intramuscular ivermectin (0.5 ml/25 kg body weight), and where necessary treatment was provided for injured dogs (i.e., wounds were cleaned and antiseptics applied) and those confirmed with motile trypanosomes were administered with 0.1 ml REDNIL per 2 kg live mass (Ready-to-use injectable solution containing diminazene aceturate 70 mg/ml, Phenazone 375 mg/ml).Our study did not involve the use of anaesthesia or euthanasia.www.nature.com/scientificreports/ the general body conditions, and measured anal temperatures.Therefore, any clinical change in the cohort such as fever, emaciation, pallor of visible mucous membranes, enlarged superficial lymph nodes, general oedema, corneal opacity, and other related clinical features was recorded.

On-site activities
Blood sampling About 2 ml of whole blood was collected from the cephalic vein of each dog using sterile vacutainer needles and tubes (Plain and EDTA tubes).

Buffy coat examination
The buffy coats and upper layers of red blood cells were placed on glass slides, covered with coverslips, and examined on-site.Trypanosomes were identified by their motility characteristics: T. brucei (varied sizes, rapid movement with undulating membrane creating light pockets), T. congolense (small, sluggish, adheres to red blood cells), T. vivax (large, very active, quickly crossing the field with occasional pauses), and T. theileri (large, tends to rotate or stay immobile with whip-like movements) 60 .If trypanosomes were present, thin Giemsa-stained smears were prepared from blood in EDTA for detailed analysis.Thin blood smears were made on-site, fixed in methanol for 3 min, air-dried, and stored in slide boxes.At the KTTRS laboratory, the slides were stained in 10% Giemsa for 30 min at room temperature and examined using an Olympus cx22 light microscope at × 1000 magnification.

Mice inoculation
Where possible, infected dogs were resampled, or their blood in EDTA tubes was used to inoculate mice intraperitoneally with 0.5-1.0 ml of infected blood to detect subpatent Trypanosoma infections.The inoculated mice were labelled and monitored at KTTRS, with daily checks for parasitaemia using tail blood and Giemsa-stained smears.

Sample transportation and storage
Blood in EDTA and plain tubes were chilled on ice in a vehicle-operated thermo-electric cooler box and transported on the day of collection to KTTRS for processing and temporary storage.Blood in EDTA was directly stored at − 20 °C while blood in plain tubes was allowed to separate and clot overnight at room temperature.The resultant serum was transferred into 1 ml sterile vials and stored at − 20 °C.Serum and blood in EDTA tubes were stored at − 20 °C for about three weeks at each time point before being transported to the main laboratory in Lusaka (Zambia's capital city).The frozen blood in EDTA tubes and serum samples were transported on ice in a vehicle-operated thermo-electric cooler box for molecular and serological diagnosis at the University of Zambia's School of Veterinary Medicine.

Sample analysis
Blood in EDTA was thawed and used to extract genomic DNA at the University of Zambia, School of Veterinary Medicine's trypanosomiasis laboratory.DNA was extracted using a mammalian blood DNA isolation kit (Roche Applied Science).About 200 μl of dog DNA per sample was eluted into sterile Eppendorf tubes and stored at − 80 °C until use.Serum samples were also stored at − 80 °C for future use.

Serodiagnosis
The ELISA assay followed a protocol described by Thuy et al. 61 with some modifications.The recombinant GM6 antigen (rTeGM6) derived from T. evansi and the anti-dog IgG antibody was used to detect Trypanosoma antibodies in the extracted dog serum.Briefly, each well of microplates (Maxisorp, Nalgene-Nunc) was coated overnight with 50 µl (1 μg/ml) of the rTeGM6 antigen.The antigen-coated wells were washed once with phosphate-buffered saline (PBS) containing 0.05% Tween 20 (PBS-T) and incubated with PBS containing 5% skim milk at room temperature for 1 h.The wells were washed once with PBS-T and serum samples (50 µl/well) diluted at 250 times with PBS containing 5% skim milk were added and incubated at room temperature for 2 h.The wells were washed thrice with PBS-T before adding 100 µl per well of 10,000 times diluted Goat anti-Dog IgG antibody conjugated with horseradish peroxidase (BETHYL Laboratories).After 1 h of incubation at room temperature, the wells were washed thrice with PBS-T.Tetramethylbenzidine (100 µl) was added to each well and incubated for 30 min at room temperature.Finally, 100 µl stop solution (1 M sulphuric acid) and the absorbance read at 450 nm.After subtracting the blank well OD value, the negative samples' standard deviation (SD) and average OD values were calculated.The cut-off ELISA values were calculated as mean ± 3SD of OD values using three negative control sera 48 .was used as a positive control and nuclease-free water as a negative control.

SRA PCR
Species-specific primers that amplify a 284 bp fragment of the serum resistance-associated (SRA) gene were used to confirm the human-infective T. b. rhodesiense 62 .The SRA PCR was run in a final reaction volume of 10 µl containing; 2 µl of eluted DNA, 5 µl 2 × Ampdirect® Plus buffer, containing MgCl₂ and dNTPs (Shimadzu, Kyoto, Japan), 0.05 µl BioTaq HS (Bioline, UK), 2.55 µl nuclease-free water, and 0.2 µl each of 10 µM SRA primers.Thermal cycling was performed at 95 °C for 10 min, followed by 40 cycles of 94 °C for 30 s, 60 °C for 1 min and 72 °C for 1 min, and final extension at 72 °C for 5 min.Amplicons were visualized in a UV transilluminator on 2% agarose gel stained with GelRed nucleic acid stain 19 .

Sequencing and phylogenetic analysis
Positive PCR products were purified using a Monofas Purification kit (GL Science) according to the manufacturer's instructions.Sequencing of purified PCR products was performed by BigDye™ Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems; Thermo Fisher Scientific, Foster City, CA).Purified sequence products were subjected to capillary electrophoresis using the ABI 3130 Genetic Analyzer (Applied Biosystems Japan Ltd., Tokyo, Japan).

Statistical analysis
Data were analysed using STATA 17.1 statistical software.Diagnostic results for microscopy were compared with those of GM6 ELISA and ITS1-PCR using the Kappa coefficient.PCR was used as the gold standard reference method to identify trypanosomes against GM6 ELISA as microscopy was not carried out on all dog samples.The sensitivity and specificity of GM6 ELISA were calculated as the number of ELISA positive and negative divided by the total number of PCR positives and negatives, respectively.Geneious ® 10.2.6 Bioinformatics software was used to edit and analyse raw DNA sequenced data (AB1 file extensions). https://doi.org/10.1038/s41598-024-69834-1www.nature.com/scientificreports/

Figure 1 .
Figure 1.Flow chart showing follow-up of dogs in the cohort from June to December 2018.

Figure 2 .
Figure 2. Corneal opacity in some dogs in the cohort.(a) Bluish eyes, and (b) greyish eyes.

Figure 5 .
Figure 5. Alignment of our dogs' SRA amino acid sequences with several GenBank references.Highlighted letters show the positions of alterations in the sequences.

Table 1 .
Summary of GM6 ELISA and ITS1-PCR prevalence based on seasons.

Table 2 .
Prevalence of single and mixed trypanosomes in the cohort.