Characteristics of tick and tick-associated microbes database

After removing 302 duplicates, the analysis of multi-source data revealed 13,906 records of ticks, of which 421 were from published papers, 1522 from GenBank, 11,708 from WRBU Vectormap, and 255 from GBIF, including 6708 in Tanzania, 6044 in Kenya, 636 in Uganda, 260 in Rwanda, 131 in Democratic Republic of the Congo (DRC), 89 in Burundi and 38 in South Sudan. These records represented a diverse array of 138 tick species from 5 genera of the Argasidae family, 8 genera of the Ixodidae family, and the monospecific Nuttalliellidae (Table S2). Among 1600 microbe records, 908 were associated with viruses, 610 with bacteria, and 82 with protists (Fig. 1).

Fig. 1

Study design and data sources. WRBU Vectormap the Walter Reed Biosystematics Unit Vectormap, GBIF Global Biodiversity Information Facility, CI Confidence interval. The tick genus within the thick outlines were identified to carry the microbes

Geographical distribution of ticks

As shown in Fig. 2, despite the variations in the geographical distribution of different tick genera, regardless of their genera, they were more abundant in areas with evergreen broadleaved forests and near water bodies. The most commonly encountered ticks belonged to the genus Rhipicephalus, which encompassed 45 species with 7299 records and distributed in the widest ranges of all seven EAC countries. Amblyomma was the second most abundant genus, with 3461 records of 23 species, and it showed a similar spatial distribution to Rhipicephalus ticks. The genera Hyalomma and Haemaphysalis had 1186 records of 8 species and 588 records of 14 species, respectively, both of which displayed various distributions. Notably, the Ixodes genus had fewer records (255) but more diverse species (31) than the above three tick genera. The geographical distribution maps of the remaining tick species are presented in Fig.S2. Moreover, tick distributions across the seven EAC countries revealed that ticks were more prevalent in regions with high altitudes and high population count (Fig.S3 and S4).

Fig. 2

Geographical distribution maps of ticks in East African Community. The base map is landcover. Shows recorded locations of each tick genus. Map approval No.: GS(2025)1426

Hosts of ticks in East African Community

Besides a few free-living ticks, the overwhelming majority of ticks in EAC countries were collected from a variety of animal families with microbe-associated ticks, including Bovidae (cows), Camelidae (camels), Canidae (dogs), Felidae (cats), Suidae (pigs), which were widely distributed in Kenya, Tanzania, and Uganda (Fig. S5). No records of animal hosts with microbe-associated ticks were reported in Rwanda, Burundi, DR Congo, and South Sudan (Fig. S5). A total of 125 species of animals could serve as hosts of 106 tick species, mainly in the genera Rhipicephalus, Amblyomma, Hyalomma, and Haemophysalis (Fig. S6).

Presence and prevalence of tick-associated microbe

A total of 40 tick species were identified as carriers of 64 microbial species, including14 from the genus Rhipicephalus, 11 Amblyomma, six Hyalomma, four Haemaphysalis, three Ixodes, and two Ornithodoros. The microbial diversity consisted of 22 viral species belonging to nine families, 26 bacterial species from five families, and 16 protist species from three families (Fig. 3). Except for 10 tick species, most species carried two or more microbes. R. appendiculatus was found to carry the highest number of microbes, with 19 species, including 10 viruses, five bacteria, and four protists. Furthermore, Amblyomma variegatum ranked second, harboring 17 microbes (5 viruses, 10 bacteria, and 2 protists). Notably, the more prevalent a tick species was, the more tick-associated microbes were identified (Fig. S7). On the other hand, several tick-associated microbes were observed to infect multiple tick species. For instance, Orthonairovirus dugbeense was discovered in 9 tick species, Coxiella burnetii in 8 tick species, and Orthonairovirus haemorrhagiae (the causative agent of CCHFV) in 6 tick species. Additionally, many unclassified Rickettsia, Ehrlichia, and Anaplasma species were each identified in over 11 tick species. Among the 64 recorded microbes’ species, 24 were classified as human pathogens (including those simultaneously pathogenic to both humans and animals), 19 were identified as animal pathogens (exclusively pathogenic to animals), whereas 21 were categorized as microbes with unknown pathogenicity (Fig. 3).

Fig. 3

Presence and prevalence of each tick-associated microbe in East African Community. In the Gantt chart, the violet grid indicates the presence of each tick-associated microbe, which is unable to calculate the positive rate. The blue grid indicates the positive rate for each microbe, with the color intensity and the number representing the value of the positive rate (percent scale). The merged positive rate and 95% confidence interval (CI) for each microbe in multiple tick species are provided at the end of each row

The meta-analysis was conducted on the tick-associated microbes, for which information on the number of tick samples tested and the number of positive samples was available. As a result, 11 viruses, 10 bacteria, and 3 protists were included to estimate their positive rates, which are displayed as blue grids in Fig. 3. The estimation of the positive rate for each microbe in a tick species is provided in Table S3 and Fig. S8. Four viruses across the Orthonairovirus genus were each identified in two or more tick species. Orthonairovirus amblyommae was identified in four tick species, with an overall positive rate of 8.47% (I2 = 77%, 95% CI 5.35–11.60). Orthonairovirus haemorrhagiae, the pathogen of CCHFV, was detected in Hy. marginatum, Hy. rufipes, Hy. truncatum, R. appendiculatus, R. decoloratus, and R. sp., with an overall estimated positive rate of 6.35% (I2 = 82%, 95% CI 2.50–10.20). The other three viral pathogens, i.e. Orthoflavivirus nilense, Orthobunyavirus bunyamweraense and Bandavirus bhanjanagarense, infected various tick species, each of which had an estimated positive rate of 14.49%, 4.71% (I2 = 67%, 95% CI 0.00–15.60), and 2.91% (I2 = 97%, 95% CI 0.10–87.90), respectively. Among the tick-associated bacteria, rickettsiae showed high infection rates. Candidatus Rickettsia africaustralis exhibited a highest pooled positive rate of 96.25%, followed by Candidatus Rickettsia africaseptentrionalis (I2 = 100%, 66.55%, 95% CI 4.15–100.00), and Rickettsia africae (I2 = 100%, 65.10%, 95% CI 26.16–100.00), which has been known to cause African tick-bite fever [15]. Rickettsia aeschlimannii, another known human pathogen [16], had a pooled positive rate of 21.31% (I2 = 92%, 95% CI 0.00–42.85). C. burnetii, the causative agent of Q fever [17], had a pooled positive rate of 19.27% (I2 = 89%, 95% CI 9.22–29.33). Anaplasma bovis had a pooled positive rate of 7.63% (I2 = 0%, 95% CI 0.00–16.49), which had been reported as the causative agent of human and animal infections [18]. Candidatus Ehrlichia regneryi, which is pathogenic to animals, had a prevalence rate of 13.33% (I2 = 81%, 95% CI 6.83–19.84). Besides, in the Theridiidae family, Theileria equi had a high pooled positive rate of 24.67% (I2 = 86%, 95% CI 0.00–57.42), while T. mutans and T. parva had pooled positive rates of 11.11% and 9.90% (I2 = 97%, 95% CI 6.01–13.78), respectively. The forest figure of combined positive rates for each microbe in all positive tick species can be found in Fig. S9.

Geographical distribution of tick-associated microbes

We mapped the geographical distribution of tick-associated microbes carried by various tick species (Fig. 4). Although ticks were distributed across all seven EAC countries, tick-associated microbes were only reported in ticks from Kenya, Uganda, Tanzania, and DR Congo, with no report in the remaining three countries. Most (89.63%) tick-associated microbes were reported in Kenya, where 21 species of viruses, 5 species of spotted fever group rickettsiae, 11 species within the family Anaplasmataceae, 8 Theileria species, and 2 Babesia species were detected (Fig. 4). Fourteen tick-associated microbes were identified in ticks from Uganda, including two viruses, 10 bacteria and two protists (Fig. 4). One virus, 5 bacteria, and 3 protists were detected in ticks from two regions of Tanzania (Fig. 4). Only two bacteria (Anaplasma platys and Ehrlichia sp.) were detected in DR Congo. Patient cases of CCHF caused by Orthonairovirus haemorrhagiae were reported in Kenya, Tanzania, Uganda, and the DR Congo [19,20,21,22,23]. Orthobunyavirus bunyamweraense and Rickettsia conorii were identified in febrile patients from Kenya [24, 25]. Patients infected with R. africae, Rickettsia conorii, and Coxiella burnetii infections were reported in Tanzania [26, 27]. Livestock disease ECF caused by T. parva has been reported in six EAC countries, excluding the DR Congo [28,29,30,31,32,33]. Additionally, animal infections with African swine fever virus, Rickettsia sibirica, Ehrlichia ruminantium, Candidatus Ehrlichia regneryi, Anaplasma marginale, Theileria bicornis, T. mutans, Theileria taurotragi, and Babesia bigemina were recorded in Kenya, Tanzania, and Uganda [34,35,36,37,38,39,40,41].

Fig. 4

Geographical distribution of tick-associated microbes in East African Community. The base map is density of human population. The solid diamond represents a microbe with known location. A hollow diamond indicates the presence of the microbe in the country but without a specific location; thus, it is marked at the center of the country and enclosed with a dashed box. A pie with the dark red circle outline and “*” indicates pathogens that infect humans, while a pie with a dark blue circle outline and “*” indicates pathogens that infect animals. Each color within the diamonds and pies corresponds to a specific microbe, as shown in the legend. Map approval No.: GS(2025)1426

Predicted suitable habitats of four dominant tick species

We predicted areas suitable for four dominant ticks species by ecological niche modelling. The model for Am. variegatum included nine independent variables and produced an average AUC value of 0.854 ± 0.0155 (Table S4, Fig. S10), indicating a respectable level of predictive ability. The modelling revealed that elevation and mean temperature of the coldest quarter were the primary variables influencing the geographical distribution of Am. variegatum ticks, contributing over 71.80% to the model (Fig. 5, Fig. S11). Furthermore, when including all the variables in the model, the predicted response curves showed that low and high mountainous areas were suitable habitats. Additionally, the most suitable areas for Am. variegatum was the northwest of Uganda, the north of Tanzania, and some eastern parts of DR. Congo (Fig. 5, Fig. S12).

Fig. 5

Potential distribution map of four dominant ticks’ species in East African Community. The predicted probabilities of suitability for each tick species are presented as continuous values ranging from 0 to 1, as generated by Maxent. The predicted areas suitable for Amblyomma variegatum, Haemaphysalis leachi, Hyalomma truncatum and Rhipicephalus appendiculatus. Map approval No.: GS(2025)1426

On the other hand, the prediction model for Hae. leachi included thirteen independent variables with an optimization AUC score of 0.907 ± 0.028 (Fig. S13), showing excellent prediction accuracy. The distribution of Hae. leachi ticks were mainly influenced by elevation, the mean temperature of the wettest quarter, the mean temperature of the coldest quarter, and the precipitation of the driest quarter (Table S5), accounting for 82.50% of the model’s distribution (Fig. 5, Fig. S14, S15). Kenya was identified as the most suitable site for Hae. leachi, particularly in its southwestern region. Some areas in the southern part of Tanzania were also found suitable sites for Hae. leachi. Twelve environmental variables were included in the model to predict Hy. truncatum distribution, achieving an AUC value of 0.888 ± 0.045 (Fig. S16). The four major variables, including annual precipitation, annual mean temperature, percentage of tree cover, and elevation, explained 75.30% of the model (Fig. 5, Table S6, and Fig. S17, S18). Furthermore, the southwestern part of Kenya and the northern part of Tanzania were identified as the most suitable habitat for Hy. truncatum. Ten variables were included in the prediction model for R. appendiculatus ticks, obtaining an AUC value of 0.867 ± 0.014 (Fig. S19). The distribution of R. appendiculatus was mainly influenced by elevation, mean temperature of the coldest quarter, tree cover percentage, and the precipitation of the driest month, explaining 83.40% of the model (Fig. 5, Table S7, Fig. S20, S21). Suitable habitats for R. appendiculatus were identified in Kenya as well as in the northern and southern parts of Tanzania. Most areas of Burundi,Rwanda and Uganda also exhibited suitable conditions for the distribution of R. appendiculatus.