Ticks and Tick-Borne Pathogens Associated with Dromedary Camels (Camelus dromedarius) in Northern Kenya

Ticks and tick-borne pathogens (TBPs) are major constraints to camel health and production, yet epidemiological data on their diversity and impact on dromedary camels remain limited. We surveyed the diversity of ticks and TBPs associated with camels and co-grazing sheep at 12 sites in Marsabit County, northern Kenya. We screened blood and ticks (858 pools) from 296 camels and 77 sheep for bacterial and protozoan TBPs by high-resolution melting analysis and sequencing of PCR products. Hyalomma (75.7%), Amblyomma (17.6%) and Rhipicephalus (6.7%) spp. ticks were morphologically identified and confirmed by molecular analyses. We detected TBP DNA in 80.1% of blood samples from 296 healthy camels. “Candidatus Anaplasma camelii”, “Candidatus Ehrlichia regneryi” and Coxiella burnetii were detected in both camels and associated ticks, and Ehrlichia chaffeensis, Rickettsia africae, Rickettsia aeschlimannii and Coxiella endosymbionts were detected in camel ticks. We also detected Ehrlichia ruminantium, which is responsible for heartwater disease in ruminants, in Amblyomma ticks infesting camels and sheep and in sheep blood, indicating its endemicity in Marsabit. Our findings also suggest that camels and/or the ticks infesting them are disease reservoirs of zoonotic Q fever (C. burnetii), ehrlichiosis (E. chaffeensis) and rickettsiosis (R. africae), which pose public health threats to pastoralist communities.


Introduction
Kenya is home to over 3 million camels, representing about 6% of Africa's camel population [1,2]. In the northern parts of Kenya and the Horn of Africa, camel production is a major source of livelihood [1,3]. Since the 1960s, camel populations in this region have continued to increase despite numerous challenges brought about by climate change [2]. In response to increasingly frequent droughts, pastoralist communities that did not previously keep camels have started rearing them to supplement or replace income from cattle production [1]. In comparison with other ruminant livestock, camels are biologically and physiologically adapted to survive better in arid and semi-arid environments [4,5]. They provide a reliable source of meat and milk, even during dry seasons when production from other livestock species such as goats, sheep and cattle becomes insufficient [1]. Camels also play a role as beasts of burden [5].
Despite the economic importance and resilience of camels under harsh climatic conditions, camel production is constrained by pests, vector-borne diseases and parasites. Common haematophagous ectoparasites of camels, specifically in Marsabit County, northern Kenya, include biting flies (e.g., Tabanus, Stomoxys and Haematopota), the camel fly or camel ked Hippobosca camelina [6,7] and ticks, which are important disease vectors. While biting flies as mechanical vectors for trypanosomes in camels have been the subject of research for decades [7], very little is known about tick-borne pathogens circulating among camels in northern Kenya.
Tick-borne pathogens (TBPs) cause emerging and re-emerging diseases in Africa and beyond [8,9]. They are transmitted to animals and humans through tick bites and constitute major constraints to livestock production in Kenya [10]. Ticks are vectors of a wide range of pathogens, including viruses, bacteria and protozoa that can infect domestic and wild animals and humans [11][12][13]. These pathogens may cause bacterial diseases such as Q fever, rickettsiosis, ehrlichiosis and anaplasmosis, protozoal diseases such as babesiosis and theileriosis and viral diseases such as Crimean-Congo hemorrhagic fever [11]. Ticks of the genus Hyalomma are most commonly associated with camels and are known vectors of Theileria, Babesia, Anaplasma, Rickettsia and Ehrlichia spp. [14][15][16]. Other genera of ticks infesting camels in Kenya include Rhipicephalus and Amblyomma [3,17].
In Kenya, most of the studies undertaken on ticks and TBPs of livestock have been limited to cattle, sheep and goats; camels remain understudied. Climate change, as well as the extensive movement of camels and other ruminant livestock across semi-arid counties and the northern borders of Kenya with neighbouring countries such as Somalia, Ethiopia and South Sudan, potentially contribute to shifts in the distribution of ticks and TBPs in the region. An Ehrlichia sp. with a DNA sequence close to E. ruminantium was found in ticks infesting camels in herds affected by an outbreak of heartwater-like disease in dromedary camels in the Moyale Constituency of Marsabit County; the disease occurred in most of North Kenya's camel keeping region and caused significant losses in adult animals in 2016 [18]. The present study, in which blood samples and ticks were collected from dromedary camels and co-herded sheep in Marsabit County, northern Kenya, was carried out as part of a wider investigation into the possible involvement of E. ruminantium and heartwater in this novel camel disease. Co-herded sheep were included in the wider study as indicators for the presence of E. ruminantium infection in an area because they develop high and long-lasting levels of serum antibodies following exposure [19,20]; the results from the serological investigation in camels and sheep will be presented in a separate publication. Here we report the results from the morphological identification of tick species infesting healthy camels and co-herded sheep, molecular detection and the characterization of various TBPs in the ticks and host blood. To the best of our knowledge, this is the first detailed molecular study on tick species infesting camels in Kenya and on TBPs in blood and ticks from these camels and their co-grazing sheep.

Study Area
The study was conducted in February 2020 in Marsabit County in northern Kenya, with an area of~66,923 km 2 about 543 km north of Nairobi [21]. Marsabit County is located between longitudes 37 • 57 and 39 • [22]. The long wet season is from March to May, while the short wet season is from October to December [21]. Most of the County is located 300-900 m above sea level with an average annual rainfall ranging from below 150 mm to just over 1000 mm. Marsabit County is home to pastoralist camel keepers who rely on mobile livestock production for their livelihoods.
Blood samples and ticks were collected from healthy dromedary camels and from co-grazing sheep at 12 sites: Laisamis, Korr, Hula Hula, Kamboe, Shegel, Burgabo, Gola, Misa, Funanyatta, Dabel, Yabalo and Bori ( Figure 1). The wells located at these sites are important watering points for camels and other livestock.

Ethical Approval
This study was undertaken in strict adherence to the experimental guidelines and procedures approved by the University of Nairobi Biosafety, Animal Use and Ethics Committee (REF: FVM BAUEC/2019/200) and Kenya's National Commission for Science, Technology and Innovation (REF: NACOSTI/P/19/72855/27325). Animals were handled carefully to cause minimum discomfort. The camel pastoralists were informed about the study and, thereafter, sampling of camels was conducted after receiving verbal consent as most herders were unable to read or write in addition to the language barriers that required translation by our field assistants selected from the community.

Collection of Blood Samples and Ticks from Camels and Co-Herded Sheep
Sample collection from 296 healthy camels and from 77 healthy co-herded sheep was conducted during the dry season from February 2020 to March 2020. Co-grazing sheep were sampled as sentinel animals for a parallel serological study of E. ruminantium antibody levels in this combined livestock cohort, as part of the overarching study investigating the role of heartwater and other TBPs in camel diseases in Kenya. Four millilitres of blood were collected from individual animals via jugular venepuncture using vacutainer needles (18 gauge) and EDTA vacutainer tubes. Blood samples were kept under cold chain (4-10 • C) for up to 6 h before being preserved in liquid nitrogen for transportation to the Martin Lüscher Emerging Infectious Diseases (ML-EID) laboratory at icipe, Nairobi, for the molecular detection of TBPs.
Serrated forceps held firmly over the tick scuta and mouthparts were used to gently remove all visible ticks attached to the skin of sampled camels and sheep. Ticks were placed in cryovials and kept under cold chain (4-10 • C) for up to 2 h prior to preservation in liquid nitrogen for transportation to the ML-EID molecular biology laboratories for further analysis.

Morphological Identification of Ticks
Ticks were identified to species level using taxonomic keys [23]. The morphological features used for tick identification included the colour and ornamentation of the scutum, shape, size and distribution of punctations and grooves and colour of legs. The ticks were staged under a Stemi 2000-C microscope (Zeiss, Oberkochen, Germany) and photographed using a digital microscope connected to an Axio-cam ERc 5s camera (Zeiss). Fully-engorged ticks were removed during tick identification and excluded from subsequent analysis to minimise vertebrate host DNA in nucleic acid extractions. Ticks were pooled into groups of one to eight individuals based on species, host, sampling site and date of collection.

Extraction of DNA from Whole Ticks, Tick Leg Tissues and Blood
Two representative adult ticks from each of the eight tick species identified from camels were selected for molecular confirmation of identity. Legs were plucked from individual ticks for genomic DNA extraction. For TBP screening, whole ticks were first frozen in liquid nitrogen before homogenising them in 1.5-mL microfuge tubes containing 150 mg of 0.1-mm and 750 mg of 2.0-mm yttria-stabilised zirconium (YSZ) oxide beads (Glen Mills, Clifton, NJ, USA) and 100 µL of 1 × PBS using a Mini-Beadbeater-16 (BioSpec, Bartlesville, OK, USA) for 1 min. The ISOLATE II Genomic DNA extraction kit (Bioline, London, UK) was used to extract DNA from the leg tissues selected for tick identification and from homogenised whole tick and blood samples for pathogen screening following the manufacturer's instructions.

Molecular Identification of Ticks
In order to confirm findings of the morphological identification of tick species, we used the extracted tick genomic DNA in PCRs targeting fragments of the cytochrome oxidase subunit I (COI), 12S ribosomal (r)RNA and 16S rRNA genes ( Table 1). The PCRs were performed in 10-µL reaction volumes including 2 µL 5× HOT FIREPol ® Blend Master Mix (Solis BioDyne, Tartu, Estonia), 0.5 µL of 10 µM forward and reverse primers (Table 1) and 25 ng of DNA template in a ProFlex PCR systems thermocycler (Applied Biosystems, Foster City, CA, USA). The following conditions were used for amplification: Initial denaturation at 95 • C for 15 min followed by 35 cycles of denaturation at 95 • C for 30 s, annealing (at 55 • C for 16S rRNA and COI, at 48 • C for 12S rRNA) for 30 s, extension at 72 • C for 1 min and a final extension at 72 • C for 10 min. Successful PCR amplification of target amplicons was determined by resolving 5 µL of the PCR products by electrophoresis in 1.5% (w/v) agarose gels containing ethidium bromide and DNA fragments visualised under ultraviolet light using a Kodak Gel Logic 200 Imaging System (SPW Industrial, Laguna Hills, CA, USA). The remaining volumes of positive PCR amplicons were purified using ExoSAP-IT (Affymetrix, Santa Clara, CA, USA) according to the manufacturer's protocol and sequenced by Macrogen Inc. (Amsterdam, The Netherlands).

Molecular Detection of TBPs
In order to screen the DNA extracts of blood and ticks from camels and sheep for TBPs of the genera Anaplasma, Babesia, Coxiella, Ehrlichia, Rickettsia and Theileria, we conducted high-resolution melting (HRM) analysis of PCR products obtained using genus-specific primers ( Table 1). The analysis was carried out in Rotor-Gene Q (QIAGEN, Hannover, Germany), Mic qPCR (Bio Molecular Systems, Upper Coomera, Queensland, Australia) and Quant Studio 3 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) thermocyclers. The primer pairs Ehrlichia16S and AnaplasmaJV were used to amplify 200-bp and 300-bp fragments of Ehrlichia and Anaplasma 16S rRNA genes, respectively. Samples with unique Ehrlichia and Anaplasma 16S rRNA amplicon HRM profiles were re-amplified using longer primers targeting 16S rRNA (PER1, PER2) for Ehrlichia [31] and EHR16SD-1492R for Anaplasma [32,33]. Theileria and Babesia were amplified using primers targeting the 18S ribosomal gene (RLB_F and RLB_R) [35]. Rickettsial 16S rRNA genes were amplified using primers Rick-F and Rick-R [27]. Rickettsia-positive samples were re-tested using rickettsial outer membrane protein B (ompB) gene primers (28).
The PCRs were performed in 10-µL volumes including 2 µL HOT FIREPol ® EvaGreen ® HRM mix (Solis BioDyne, Tartu, Estonia), 0.5 µL of 10 µM forward and reverse primers and 25 ng of template DNA. For no-template controls, 1 µL nuclease-free water was used as a template. DNA samples of "Ca. Anaplasma camelii", "Ca. Ehrlichia regneryi", Theileria parva and Rickettsia africae from earlier studies [6,18,36,37] were used as positive controls. The PCR cycling conditions included an initial enzyme activation at 95 • C for 15 min; 10 cycles of denaturation at 94 • C for 20 s, step-down annealing from 63.5 • C to 53.5 • C (decreasing by 1 • C per cycle) for 25 s and extension at 72 • C for 30 s; 25 cycles of denaturation at 94 • C for 25 s, annealing at 50 for 20 s and extension at 72 • C for 30 s; and a final extension at 72 • C for 7 min. A 3 min hold at 72 • C was included after PCR cycling before HRM analysis by gradually increasing the temperature from 75 to 90 • C with fluorescence acquisitions after 2 s at 0.1 • C increments [34]. All samples with unique melt profiles were purified with an ExoSAP-IT PCR Product Cleanup kit (Affymetrix, Santa Clara, CA, USA) and submitted for Sanger sequencing by Macrogen (Amsterdam, The Netherlands). Chromatogram files were imported into the Geneious Prime software v. 2020.2.2 (created by Biomatters, Auckland, New Zealand) in which they were trimmed, edited and aligned to generate consensus sequences.

Phylogenetic Analysis
Nucleotide sequences obtained in this study were queried against known sequences in the GenBank nr database (http://www.ncbi.nlm.nih.gov/, accessed on 15 May 2021) using BLAST to confirm their identity and relation to existing deposited sequences [38]. Study sequences were then aligned with related tick or pathogen sequences available in the GenBank nr database using the MAFFT plugin in Geneious Prime software version 2020.2.2 [39]. Maximum-likelihood phylogenies were constructed using PhyML v. 3.0 with automatic model selection based on the Akaike information criterion. Tree topologies were estimated over 1000 bootstrap replicates with the nearest neighbour interchange improvements [40]. Phylogenetic trees were visualised using FigTree v. 1.4.4 [41].

Estimation of Tick Infection Rates
Estimated minimum infection rates (MIRs) of each of the TBPs obtained for each tick species were calculated as the number of positive pools per total number of ticks of that species tested × 100, with a conservative assumption that only one tick is positive per pathogen-positive pool.

Morphological and Molecular Identification of Ticks
A total of 2610 adult ticks removed from camels were morphologically identified as Hyalomma dromedarii, Hyalomma rufipes, Hyalomma impeltatum, Hyalomma truncatum, Amblyomma lepidum, Amblyomma gemma, Rhipicephalus pulchellus and Rhipicephalus camicasi. Hyalomma was the most prevalent genus and comprised three quarters (75.7%) of all adult ticks collected from camels, followed by Amblyomma (17.6%) and Rhipicephalus (6.7%). Tick infestation was low in sheep, from which we collected 86 adult ticks and identified them as Rh. camicasi, Am. gemma, Am. lepidum, Rh. pulchellus and Hy. rufipes. Rhipicephalus (53.5%) was the dominant genus sampled from sheep, followed by Amblyomma (45.4%) and Hyalomma (1.2%). Figure 1 shows the total numbers of ticks and the proportions of each species collected at each sampling site. Table 2 summarises the number of tick species collected from camels and co-herded sheep in northern Kenya. Photomicrographs of representative specimens of the eight tick species identified are shown in Figure 2.  BLASTn analysis of Am. gemma, Am. lepidum, Rh. camicasi, Rh. pulchellus, Hy. dromedarii, Hy. impeltatum, Hy. truncatum and Hy. rufipes sequences obtained in this study showed identities ranging from 99 to 100% with reference sequences from the Gen-Bank nr database (Table S1). Molecular identification based on partial 12S rRNA, 16S rRNA and COI gene sequences obtained from 15 representative samples was consistent with morphological identification and confirmed the wide diversity of tick species collected from camels ( Figure 3). The 12S rRNA and 16S rRNA molecular identification was more informative due to more consistent amplification as we were able to amplify COI sequences from only four of the tick samples. All tick sequences obtained in this study have been deposited in GenBank (COI gene accessions MT896151-MT896154; 12S rRNA gene accessions MT895851-MT895865; 16S rRNA gene accessions MT895169-MT895181).

Tick-Borne Pathogens and Endosymbionts Detected in Ticks
We screened 858 tick pools from camels for Ehrlichia, Anaplasma, Rickettsia, Coxiella, Babesia and Theileria species. In ticks sampled from camels, we detected 451-bp 16S rRNA gene sequences of E. ruminantium, "Ca. E. regneryi", E. chaffeensis and an Ehrlichia sp. Sequence similarities to reference sequences and to TBPs identified in camel herds of the study region in 2016 are indicated in Table 3 and the maximum likelihood phylogenetic relationships are shown in Figure 5A. We detected an E. ruminantium sequence, identical to that found in sheep blood, in Am. gemma and Am. lepidum; "Ca. E. regneryi" was detected in all three species of Hyalomma; E. chaffeensis (100% identity to E. chaffeensis strain Arkansas, GenBank accession NR_074500) was detected in Am. lepidum ticks; and an Ehrlichia sp. (100% identity to Ehrlichia sp. MN726921, detected in a Hyalomma anatolicum tick in Pakistan) was detected in Hy. rufipes, Am. gemma, Rh. camicasi and Rh. pulchellus ticks from different camels in different herds. Additionally, using the primer pair Ehrlichia16S F and Ehrlichia 16S R, we amplified short (163 bp) sequences identified as Paracoccus sp. (99% identity to Paracoccus sp. BRM2, GenBank accession KP003988, isolated from Tunisian phosphogypsum) in Amblyomma, Hyalomma and Rhipicephalus spp. collected from camels at five different sites (Table S3).
We detected identical C. burnetii sequences in camel blood and Hy. dromedarii, Hy. rufipes and Rh. pulchellus ticks. Additionally, we detected Coxiella endosymbionts in Am. gemma, Am. lepidum and Rh. pulchellus ticks using the Rick16S primers.
We detected "Ca. Anaplasma camelii" in all the species of Hyalomma, Amblyomma and Rhipicephalus spp. identified in this study; an Anaplasma sp. sequence in Hy. rufipes; Rickettsia aeschlimannii in Hyalomma and Rhipicephalus spp.; and Rickettsia africae in the two Amblyomma spp. identified in this study as shown in Table 3 and Figure 5B. In ticks sampled from sheep, we detected E. ruminantium (100% identity to E. ruminantium strain Welgevonden NR_074155) in Amblyomma spp.; R. africae in Amblyomma spp.; Theileria ovis in Rhipicephalus spp.; Anaplasma ovis in Amblyomma and Rhipicephalus spp.; and "Ca. A. camelii" in Rh. camicasi ( Table 4). The distributions of ticks and pathogens according to the sampling sites are shown in Table S2. We detected three pathogens, "Ca. Anaplasma camelii" (21.3%), "Ca. Ehrlichia regneryi" (3.4%) and C. burnetii (0.3%), in both blood and ticks from the same individual camels. Among sheep, we detected E. ruminantium, A. ovis and T. ovis in both ticks and blood from the same individual sheep (Table S3).

Discussion
This study provides critical insight on the diversity and abundance of tick species on camels and co-herded sheep in northern Kenya and the TBPs in ticks and blood from these animals. Tick species on camels identified in this study confirm earlier reports on camel ticks in North Kenya [17]. We also report for the first time that Hy. impeltatum ticks parasitise camels in Kenya. Notably, we found a diversity of ticks and tick-borne microorganisms associated with camel herds distinct from those recently identified on cattle in western Kenya [42]. We identified four TBPs, R. africae (in sheep ticks), E. ruminantium (in camel ticks, sheep ticks and sheep blood), E. chaffeensis (in camel ticks and sheep blood) and C. burnetii (in camel ticks and camel blood), that are of major economic, animal health and/or human health importance [11,13,43]. Information on tick species diversity, ecology and distribution will help improve the understanding of disease dynamics [44] and is a prerequisite for any future prophylaxis or control measures.

Species Diversity of Ticks Associated with Camels and Co-Herded Sheep in Northern Kenya
We identified eight epidemiologically-important tick species from three different genera, Hyalomma, Amblyomma and Rhipicephalus, parasitising camels and co-herded sheep. These tick genera have been reported to infest a broad range of vertebrate host species and transmit several important pathogens, including viruses, bacteria and protozoa of medical and veterinary importance [12,45].
The most prevalent tick species sampled from camels were Hy. dromedarii (35.21%) and Hy. rufipes (31.03%). Hyalomma dromedarii is considered to be the main tick species parasitising dromedary camels [16,17,46] and is the principal vector of Theileria spp. of camels in Egypt [47]. Hyalomma dromedarii may play a role in the epidemiology or transmission of emerging and re-emerging diseases such as rickettsioses [48][49][50], viruses such as Crimean-Congo haemorrhagic fever virus (CCHFV) and C. burnetii (responsible for zoonotic Q fever) [51,52]. A high prevalence of Hy. dromedarii has been reported in camels found in arid and hyper-arid regions of Kenya [17], Saudi Arabia, Sudan, Egypt, Iran and Tunisia, with infestation rates ranging between 49% and 89% [15,16,45,[53][54][55]. This tick can also infest other livestock, such as cattle, goats, sheep and horses [56,57], though we did not find this species on sheep co-herded with camels in the present study. Hyalomma rufipes, found on both camels and sheep, is known to be a vector of CCHFV as well as of R. aeschlimannii, Anaplasma marginale, Rickettsia conorii and Babesia occultans [58][59][60]. We also found, for the first time in Kenyan camels, Hy. impeltatum, which has previously been found on dromedary camels in Iran and northern Sudan [46,55,57]. Hyalomma impeltatum is known to have a wide range of animal hosts including buffalo, cattle and sheep [16,61] and has the potential to transmit CCHFV [62].
We found Am. lepidum and Am. gemma ticks on both camels and sheep. The economic importance of Amblyomma spp. ticks has long been recognised due to their ability to transmit multiple diseases to humans and animals [11]. They are the major vectors of E. ruminantium, which is the causative agent of heartwater disease in sub-Saharan Africa (SSA) and some Caribbean and Indian Ocean islands [11,[63][64][65]. Other tick species found on camels in our study include Hy. truncatum, Rh. camicasi and Rh. pulchellus; the latter two species were also found on sheep. Our report of Rh. camicasi infesting sheep and camels in northern Kenya extends knowledge about the geographic range and dynamics of this tick species in Kenya. Since most of these tick species have the potential to transmit diseases such as heartwater, anaplasmosis and Q fever, domestic animals and humans in the region may be exposed to a variety of tick-borne diseases.

Tick-Borne Bacteria Identified in Ticks, Camels and Co-Herded Sheep in Northern Kenya
Our findings show that E. ruminantium, E. chaffeensis, "Ca. E. regneryi", C. burnetii, "Ca. A. camelii", A. ovis and T. ovis are circulating among ticks from camels and sheep in the study area. The findings also show occurrence of distinct TBPs, with some overlap, in blood and ticks from camels and sheep in the study area.
Ehrlichia ruminantium was detected in Am. gemma and Am. lepidum ticks sampled from both camels and co-herded sheep in this study and in sheep blood, but not in camel blood. The bacterium is known to infect macrophages, neutrophils and vascular endothelial cells of ruminant hosts and is a major cause of livestock loss in SSA [63]. The absence of the pathogen in blood samples is not surprising considering the fact that E. ruminantium is mainly found in endothelial cells and can only be rarely detected in the bloodstream, except during clinical heartwater [66,67]. Our finding of E. ruminantium, which causes heartwater disease in ruminants, in Amblyomma ticks feeding on camels supports the recent reports on their potential impact on SSA camel populations [18,68], although it remains unknown if camels are susceptible to infection with E. ruminantium. Our findings in combination with the identification of Ehrlichia sp. with a DNA sequence close to E. ruminantium in Moyale Constituency, which is part of the study area in 2016 [18], and the isolation of the pathogen from Amblyomma spp. in eight districts across Kenya [69] suggest that there is continuous circulation of E. ruminantium across the country. Since 2016 and during the entire study period in 2020, no clinical heartwater cases were reported from camels, sheep and goats in Marsabit (Boku Bodha, unpublished observations). This is an indication that E. ruminantium may be endemic in Marsabit County. However, there is lack of information on the role of camels in the epidemiology of ehrlichiosis.
Ehrlichia chaffeensis DNA was detected in Am. lepidum ticks from camels and in blood from sheep. Ehrlichia chaffeensis, an emerging TBP, is known to cause human monocytic ehrlichiosis in humans [70]. Recent studies have found E. chaffeensis in Haemaphysalis leachi ticks collected from dogs in Uganda [71], Rhipicephalus sanguineus from dogs in Cameroon [72], Amblyomma hebraeum collected from both cattle and sheep in South Africa [73] and questing Amblyomma eburneum ticks in Kenya [34], which suggests that diverse tick species may vector this pathogen. To our knowledge, this is the first detection of E. chaffeensis in Am. lepidum ticks collected from dromedary camels. Our finding of E. chaffeensis in Am. lepidum ticks collected from camels and in blood from co-grazing sheep in northern Kenya suggests that this pathogen is actively circulating in the study area. Further investigation on the epidemiology of this pathogen is required.
We detected "Ca. E, regneryi" in camel blood and in Hy. rufipes, Hy. dromedarii and Hy. impeltatum ticks removed from camels, but not in other tick species feeding on camels. "Candidatus E. regneryi" is a novel Ehrlichia sp. first described in Saudi Arabia [74]. During an outbreak of heartwater-like disease in Moyale Constituency of Marsabit County in 2016, "Ca. E. regneryi" was found in blood from one camel that had reportedly recovered from the disease; however, the pathogen was not detected in ticks and blood from a severe clinical case of heartwater-like disease in a recumbent camel [18]. Our findings suggest that Hyalomma spp. are important vectors of the bacterium. "Candidatus E. regneryi" is phylogenetically closely related to Ehrlichia canis [74]. It is interesting to note that we did not detect the pathogen in blood and ticks from co-herded sheep, which suggests that it may be specific to camels. The pathogen was detected in apparently healthy camels, which suggests that the parasite in circulation is non-pathogenic and this is in line with the observations made in 2016 that the pathogen was not found in diseased camels [18]. Our findings suggest that camels are asymptomatic carriers of "Ca. E. regneryi" and further investigation into its pathogenicity, key vectors and zoonotic potential is required.
For Q fever, which is caused by C. burnetii, the association between camel exposure, seroprevalence in camels and human Q fever infections is well documented from Chad [52], and a recent study in Somalia found that C. burnetii infections were common in camel ticks [75]. Q fever is one of the most widespread neglected zoonosis worldwide with the highest seroprevalence rates recorded in female camels with a history of abortion [76]. Coxiella burnetii infection has been found in Hy. dromedarii and Hy. impeltatum ticks from camels in Tunisia [77]. While ticks facilitate a sylvan life cycle of C. burnetii in reservoir animals, domestic animals and humans are most commonly infected by contaminated aerosols [78]. We found C. burnetii in camel blood and in Hy. rufipes, Hy. dromedarii and Rh. pulchellus ticks from camels, which indicates that dromedary camels could be an additional reservoir species for this pathogen. In Laikipia, Kenya, just south of this study's geographic focus, 18.6% of camels have been found to have been exposed to C. burnetii by seropositivity [79]. The acute C. burnetii prevalence documented for healthy camels in this study (3.4%) is comparable to the prevalence in clinically asymptomatic cattle (4.3%) with a history of previous abortions and reproductive problems [80]. Coxiella burnetii is known to cause infections in a wide range of species, including domestic animals, birds and reptiles [29]. Ticks have been shown to transmit C. burnetii experimentally and could play a role as reservoirs in maintaining the bacterium in the environment between outbreaks due to their very long lives [81]. A study in Algeria suggested that Hyalomma spp. ticks could facilitate the transmission of C. burnetii among dromedary herds [51]. The detection of C. burnetii in Rhipicephalus and Hyalomma spp. corroborates previous reports on the same findings in Kenya [82][83][84] and Senegal [85]. Our results demonstrate that camels and their associated ticks in northern Kenya constitute an important epidemiological reservoir of C. burnetii, which increases human exposure and zoonotic risk of Q fever infection for camelkeeping communities, veterinarians and abattoir workers in the area. Antibodies against C. burnetii have been found in significant numbers of livestock handlers, indicating exposure to the pathogen [86,87]. Given the potential impact of C. burnetii on camel reproduction and the zoonotic risk for public health, further studies are required to better understand the role of camels in the epidemiology of Q fever.
Coxiella endosymbionts were detected in Am. lepidum, Am. gemma and Rh. pulchellus ticks using Rickettsia 16S rRNA gene primers. To the best of our knowledge, this is the first study to register Coxiella endosymbionts in ticks collected from northern Kenya. Previous studies in Kenya have revealed the presence of Coxiella endosymbionts in ticks collected from Busia [37], the Maasai Mara National Reserve [36] and the coastal region [34]. Coxiella endosymbionts assist in blood meal processing and egg production by providing the tick host with essential micronutrients and macronutrients [88,89]. The roles of these Coxiella endosymbionts in ticks are still not clear and require further investigation. Coxiella burnetii and Coxiella endosymbionts have different transmission routes and infectiousness, even though they are closely related [81]. Understanding the role of Coxiella endosymbionts in ticks may advance our understanding of Q fever.
We detected "Ca. A. camelii" in 78.7% of blood samples from 233 apparently healthy camels, indicating presence of an asymptomatic healthy carrier state. This high prevalence of "Ca. A. camelii" was found in camels carrying Amblyomma ticks with E. ruminantium infection rates between 5.2% and 12.4%. The fact that no heartwater cases were reported, observed or suspected in these camels throughout the study contradicts the notion that immunosuppression by "Ca. A. camelii" may be a contributing factor in development of clinical heartwater-like disease in camels, a hypothesis which could not be ruled out entirely during the 2016 outbreak [18]. The present study corroborates previous findings of "Ca. A. camelii" in blood from healthy camels in Kenya (6,18) and in other dromedary camel populations [53,[90][91][92]. Carrier status or persistence in the host is an important strategy for successful pathogen transmission to ticks and for developing resistance against the reinfection of hosts [93]. The high prevalence of "Ca. A. camelii" in healthy camels is an indication of endemic stability and/or that the bacterium is non-pathogenic. We detected "Ca. A. camelii" in all eight tick species removed from these 233 camels, with infection rates in tick pools ranging from 2.7% to 8.5%. We also detected "Ca. A. camelii" in one Rh. camicasi tick collected from co-grazing sheep, but not in sheep blood. Interestingly, "Ca. A. camelii" has also been found in hippoboscid flies (Hippobosca camelina) collected from camels in northern Kenya [6]. These flies can transmit "Ca. A. camelii" to small laboratory animals [94], indicating that hippoboscids might also play a role in the transmission of this organism. High infection rates of 88.3% found for A. ovis in clinically healthy sheep blood during this study suggest that sheep in northern Kenya may serve as reservoirs for this pathogen. Although A. ovis infection is generally a subclinical infection in small ruminants, more severe infections resulting in significant economic losses have been reported in Spain [95].
We found high infection rates for R. africae in Am. gemma (10.9%) and Am. lepidum (9.4%) tick pools from camels and in Am. gemma (14.3%) and Am. lepidum (16.7%) tick pools from co-herded sheep. The detection of R. africae in Amblyomma ticks collected from camels and sheep points towards the importance of camel-associated and sheepassociated Amblyomma ticks as significant reservoirs of zoonotic R. africae in North Kenya. For R. aeschlimannii, the infection rates in camel tick pools were 4.8% for Rh. pulchellus, 4.0% for Hy. truncatum, 2.7% for Hy. impeltatum, 1.1% for Hy. rufipes and 0.33% for Hy. dromedarii, respectively. These low infection rates suggest that camel-associated Hyalomma and Rhipicephalus spp. ticks represent minor reservoirs for R. aeschlimannii. Our findings correlate well with other studies that have predominantly detected R. africae in Amblyomma spp. and R. aeschlimannii in Hyalomma and Rhipicephalus spp. [59,60,96]. While R. africae and R. aeschlimannii were detected in both camels and their associated ticks in Algeria [97], we did not detect spotted fever group rickettsiae (SFGR) DNA in camel or sheep blood in the current study. The absence of SFGR may be due to the minute numbers of rickettsial organisms in the blood samples tested, the limited number of samples tested in this study or because the ticks are not actually transmitting the bacteria under normal circumstances. Camels in northern Kenya are kept close to homesteads and herds are in close association with other animals such as goats and sheep; thus, the presence of R. africae and R. aeschlimannii in ticks may present a health risk to humans. Healthcare providers in these areas should consider SFGR diseases in the differential diagnoses of patients presenting with fever of unknown origin and clinical signs compatible with rickettsioses.
We did not detect Theileria or Babesia spp. DNA in camel blood or in camel ticks. However, we did detect T. ovis in blood samples (80.5%) and Rh. camicasi ticks from healthy sheep. Similar high prevalence of T. ovis DNA in sheep blood has previously been reported in Ethiopia (91.9%) [98] and Sudan (88.6%) [99]; lower prevalences of 27-50% in sheep in Ghana were based on the morphological identification of piroplasms in blood smears [100].
We also detected Paracoccus sp. in Ambylomma, Hyalomma and Rhipicephalus spp. collected from camels; this raises the possibility of these bacteria being transmitted or harboured by ticks or by another invertebrate organism parasitising ticks. These bacteria were first associated with ticks feeding on horses at a single site in Brazil [101] and subsequently in Kenya with Amblyomma spp. ticks collected from livestock and tortoises at a single sample site [59], as well as with questing Haemaphysalis concinna ticks at two sites in Hungary [102] and Rhipicephalus microplus ticks removed from a collared peccary in Peru [103]. Further investigation is required to elucidate the relationship between Paracoccus bacteria and ticks and whether they pose any risk to animal or human health.

Conclusions
This is the first study to show tick and TBP point prevalence and infection rates among Kenyan camel herds. We found that Hy. dromedarii and Hy. rufipes are the most prevalent tick species on camels in northern Kenya and that camels are exposed to a range of TBPs. We also report for the first time that Hy. impeltatum ticks parasitise camels in Kenya. We report the presence of "Ca. E. regneryi", "Ca. Anaplasma camelii" and C. burnetii in camel blood and E. ruminantium, "Ca. E. regneryi", E. chaffeensis, "Ca. Anaplasma camelii", R. aeschlimannii, R. africae, C. burnetii and Coxiella endosymbionts in camel ticks in northern Kenya. Some of these pathogens, such as R. africae, E. chaffeensis and C. burnetii, are zoonotic and therefore have a potential to cause serious illness in humans. Presence of Coxiella endosymbionts in ticks raises exciting questions on the role they might play in tick physiology, population dynamics and the transmission of disease-causing pathogens. We found distinct TBPs, with some overlap, in blood and ticks from camels and sheep in the study area. These findings form a basis for strategic frameworks for research and development of novel control strategies, which are necessary to protect camels from threats that TBPs may pose. Further studies are required to identify the vectors of "Ca. E. regneryi" and "Ca. Anaplasma camelii" and for determining their effects on camel health and productivity. The epidemiology of E. ruminantium in camels needs to be investigated further to assess the potential involvement of this pathogen in heartwater-like disease of camels.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/microorganisms9071414/s1, Table S1: Morphological and molecular identification of tick samples collected from camels in Marsabit, northern Kenya, February-March 2020, Table S2: Minimum infection rates for TBPs identified in ticks and camel blood samples according to sampling sites in Marsabit, northern Kenya, February-March 2020, Table S3: Numbers of sampled camels and sheep with the same TBPs in ticks and blood or in ticks or blood only, Marsabit, northern Kenya, February-March 2020.