Detection of Tick-Borne Pathogens in Ticks from Cattle in Western Highlands of Cameroon

This study aimed to detect and identify microorganisms in ticks collected in the Western Highlands of Cameroon. Quantitative real-time and standard PCR assays, coupled with sequencing, were used. A total of 944 ticks collected from cattle in five distinct sites in Cameroon were selected for the analyses. They belonged to five genera (Amblyomma, Hyalomma, Rhipicephalus, Haemaphysalis, and Ixodes) and twelve species. Real-time PCR revealed that 23% (n = 218) of the ticks were positive for Rickettsia spp., 15% (n = 141) for bacteria of the Anaplasmataceae family, 3% (n = 29) for Piroplasmida, 0.5% (n = 5) for Coxiella burnetii, 0.4% (n = 4) for Borrelia spp., and 0.2% (n = 2) for Bartonella spp. The co-infection rate (3.4%, n = 32) involved mainly Rickettsia spp. and Anaplasmataceae. Of the Rickettsia spp. positive ticks, the targeted PCR and sequencing yielded Rickettsia africae (78.9%), Rickettsia aeschlimannii (6.4%), Rickettsia massiliae (7.8%), Candidatus Rickettsia barbariae (0.9%), and Rickettsia sp. (0.9%). Anaplasmataceae included Anaplasma marginale (4.3%), Anaplasma platys (1.4%), Anaplasma centrale (0.7%), Ehrlichia ruminantium (0.7%), Wolbachia sp., Candidatus Ehrlichia rustica (13.5%), Candidatus Ehrlichia urmitei (7%), and an uncultured Ehrlichia sp. (4.2%). Borrelia theileri was identified in one Rhipicephalus microplus tick. Unfortunately, Piroplasmida could not be identified to the species level. This study demonstrates that in Cameroon, ticks harbour a wide variety of microorganisms and present a risk of zoonotic diseases.


Introduction
Ticks are obligate haematophagous arachnids, which are distributed globally and parasitise a wide range of vertebrates [1]. On a global scale, ticks are currently considered as the leading vectors of animal diseases and second biggest vectors of human diseases after mosquitoes [2,3]. Only about 20% of the world's cattle population remains unaffected by ticks and tick-borne diseases. However, the annual global cost losses are estimated to range from USD 14-19 billion [4,5]. With an estimated population of 6.5 million, cattle are regarded in Cameroon as the main source of animal protein in most households, according to the National Institute of the Statistic [6]. However, one of the most important constraints of small-and large-scale cattle production is the high prevalence of infectious diseases [7]. Of the about 900 currently known tick species, only about 10% are of significant veterinary and/or medical importance [8]. In sub-Saharan Africa, ticks of economic importance in livestock belong mainly to four genera, according to the previous studies, namely Amblyomma, Hyalomma, Haemaphysalis, and Rhipicephalus [9,10].
The direct effects of tick on their hosts are irritation, anaemia, inflammation, paralysis, abscesses, allergies, hypersensitivity, and skin deterioration at the biting site, which often leads to reduction in weight gain and milk yield [11]. Ticks are also responsible for indirect This study was done in the Western Highlands of Cameroon, which are one of the three major agro-ecological zones for livestock production ( Figure 1). It is in the mid-and high-latitude zones of the country. The annual average temperature is 20.6 • C and annual precipitation ranges from 1300 to 3000 mm. Two seasons can be distinguished as follows: the rainy season lasts eight months (March to October) and the dry season from November to February [20]. The choice of this zone is justified by the potential for the development of breeding livestock, which are the main host of ticks in Cameroon and Central Africa. In addition, there is a significant livestock trade network between this area and the Adamawa region, which is the primary livestock-producing region in Cameroon. It should also be noted that this area is one of the entry points for animals from West Africa, particularly neighbouring Nigeria [20,26].

Tick Collection and Morphological Identification
Cattle dwellings were visited in five sites in the Western Highlands (zone III), namely Dschang and Nkong-Ni (Bafou) in the Menoua division and Kouoptamo, Massangam, and Koutaba in the Noun division. Ticks were collected from cattle using blunt steel forceps and placed inside a collection tube containing 70% ethanol.
Ticks were first identified morphologically to the species by PhD-trained entomologists from the Vector Borne Disease Laboratory (VBID) at the University of Dschang in Dschang, Cameroon [10,27]. The samples were sent to France after obtaining import au-

Tick Collection and Morphological Identification
Cattle dwellings were visited in five sites in the Western Highlands (zone III), namely Dschang and Nkong-Ni (Bafou) in the Menoua division and Kouoptamo, Massangam, and Koutaba in the Noun division. Ticks were collected from cattle using blunt steel forceps and placed inside a collection tube containing 70% ethanol.
Ticks were first identified morphologically to the species by PhD-trained entomologists from the Vector Borne Disease Laboratory (VBID) at the University of Dschang in Dschang, Cameroon [10,27]. The samples were sent to France after obtaining import authorisation (number FR13-2020 from the French Ministry of Agriculture, Food and Forestry). At the IHU Méditerranée Infection in Marseille (France), the tick morphology was then rechecked using a Leica binocular lens (Leica Camera, Wetzlar, Germany) with an LED light source, by entomologists using previously established taxonomic identification keys [10]. Due to their engorged status and to morphological similarities between the ticks, some specimens in the Rhipicephalus genus were identified to the genus level only. The identification of tick specimens was then confirmed by molecular biology and further refined by MALDI-TOF mass spectrometry (Ngnindji et al., under review).

DNA Extraction and Molecular Detection of Microorganisms in Ticks Using Real-Time PCR
DNA was extracted from half of the ticks, as described previously [28]. To investigate the presence of pathogens using primers and probes targeting Anaplasmataceae bacteria, Bartonella spp., Borrelia spp., Piroplasmida, Rickettsia spp., and Coxiella burnetii, quantitative PCR (qPCR) was performed on the extracted DNA, using a CFX96 touch detection system (Bio-Rad, Marnes-la-Coquette, France). The reaction mix quantity and the programme for qPCR was the same as previously described [29]. For each qPCR run, DNA from Rickettsia montanensis, Bartonella elizabethae, Ehrlichia canis, C. burnetii, Borrelia crocidurae, and Theileria orientalis were used as a positive control. DNA from Rhipicephalus sanguineus s.l. raised in our laboratory, which were free of microorganisms, were used as negative controls. The qPCR tests were considered positive when the cycle threshold (Ct) was lower than 36 [28]. Positive samples for Rickettsia spp. were then submitted to the qPCR system, specifically for detecting Rickettsia africae [30]. Samples which were positive for C. burnetii using IS1111 gene were submitted to the second gene (IS30A) for confirmation. In our reference centre for Q fever and Rickettsial infections, only samples positive for both genes will be considered positive for C. burnetii [31]. The primers and probes used for quantitative real-time and conventional PCRs are summarised in Table 1. Table 1. Primers and probes used for quantitative real-time and standard PCR in this study.

Standard PCR, Sequencing, and Phylogenetic Analysis
Standard PCR was performed using a thermal cycler (Applied Biosystems, Paris, France) on the samples that were positive in qPCR, and then sequencing to identify the microorganism species. Samples that were negative for R. africae but positive for Rickettsia spp. were subjected to standard PCR to amplify a 632-bp fragment of the ompA gene [34,36]. The amplicons were sequenced, assembled, and compared to GenBank entries sequences by a BLAST search to identify the Rickettsia species. For sequences with a low percentage identity with the corresponding sequences in GenBank, PCR targeting a second gene (gltA) was used to amplify a 700 bp DNA fragment, followed by sequencing. Samples that were positive for the Anaplasmataceae family were subjected to amplifying and sequencing of 520 bp fragment of the 23S rRNA gene [32]. The samples positive for Borrelia spp. under qPCR were subjected to amplifying and sequencing of a 300 bp fragment of the flagellin (FlaB) gene [40]. Samples that were Piroplasmida-positive following qPCR were subjected to amplifying and sequencing of a 969 bp fragment of the 18S rRNA gene [29]. The obtained sequences were assembled and analysed using the CromasPro software (version 1.7.7) (Technelysium Pty. Ltd., Tewantin, Australia) and were then blasted against the reference sequences available in GenBank (http://blast.ncbi.nlm.nih.gov/ accessed on 20 July 2021).
Sequences from microorganisms were aligned using the BioEdit v 7.2.5.0 software (University of North Texas, Denton, TX, USA). The aligned sequences were imported into TOPALi v2.5 software (Biomathematics and Statistics Scotland, Edinburgh, UK) and phylogenetic trees were constructed using TOPALi v2.5 software [41]. The maximum likelihood (ML) phylogenetic tree model proposed by default by the software was used to construct the phylogenetic tree. Node numbers are percentages of bootstrap values obtained by repeating 100 interactions of the analysis to generate a majority consensus tree (only those with values equal to greater than 80 were retained).
DNA of Borrelia spp. and Bartonella spp. were found by qPCR for four and two ticks, respectively. For Borrelia spp., we succeeded in amplifying the flaB gene sequence only in one of four ticks. A BLAST search showed that sequence was 100% identical with Borrelia theileri (MK984606). This Borrelia specie was detected in a Ha. leachi specimen. However, all the standard PCR to identify the Bartonella species failed.
DNA from the Piroplamida was detected in 7.3% ( Two phylogenetic trees of Rickettsiae and Anaplasmataceae were constructed from the partial sequences of the ompA gene and the 23S rRNA gene sequences of our amplicons, respectively. These phylogenetic trees showed that sequences of the microorganisms detected in this study are close to their homologues available is GenBank, except for two sequences of Rickettsia sp. (Figures 2 and 3). Two phylogenetic trees of Rickettsiae and Anaplasmataceae were constructed from the partial sequences of the ompA gene and the 23S rRNA gene sequences of our amplicons, respectively. These phylogenetic trees showed that sequences of the microorganisms detected in this study are close to their homologues available is GenBank, except for two sequences of Rickettsia sp. (Figures 2 and 3).

Discussion
Several species of microorganisms were detected in ticks from Cameroon. DNA of Rickettsia spp. was found in 23% of the tested ticks, 78.8% of which was found in Am. variegatum and 92.2% was R. africae. Rickettsia africae is the aetiological agent of African tick-bite fever in humans [42,43]. The results show that the R. africae infection rate is high among Am. variegatum in the Western Highlands zone of Cameroon. This high prevalence of R. africae had already been identified in Cameroon [16,36]. This finding confirms those of previous studies conducted in other African countries [28,44]. Rickettsia africae was identified in 7 (6%) of 118 patients with acute fever of unknown aetiology in clinics along the coastal region of Cameroon [16]. Amblyomma variegatum is a vector of R. africae in sub-Saharan Africa. These ticks are not only vectors but also reservoirs of rickettsiae in sub-Saharan Africa with transstadial and transovarial transmission of R. africae infection in Am. variegatum ticks [42,45]. Rickettsia africae was also detected in 0.7% of Rh. microplus and 4.7% of Hy. truncatum in this study. These proportions are very low, compared to those obtained in Am. variegatum. However, the detection of a microorganism in an arthropod does not mean that this arthropod can act as a vector of the microorganism. This bacterium may have infected Rh. microplus and Hy. truncatum during co-feeding. Indeed, co-feeding transmission is the transmission, which can occur when ticks (infected and uninfected) feed in close spatial and temporal proximity on the same host. During this form of transmission, the host acts as a belt, bringing together infected and uninfected ticks to facilitate pathogen exchange [46]. Furthermore, it has been shown that in co-feeding transmission (or nonsystemic transmission), vector-to-vector transmission on the vertebrate host is essentially immediate [47]. This finding was shown in a previous study in Côte d'Ivoire [44]. In this study, other rickettsial DNA was also found in Hyalomma and Rhipicephalus tick species. Of the remaining 46 tick samples positives for Rickettsia spp., amplification and sequencing yielded sequences for 35 (78.2%) samples. The 11 positive Rickettsia qPCR samples for which amplification with the ompA gene was not possible may be due to a higher sensitivity of qPCR, compared to standard PCR [48]. This difference in sensitivity often results in qPCR positive samples with a high Ct (low bacterial load) not being amplified.
Rickettsia aeschlimannii was also identified in 75% of Hy. rufipes and 4% of Hy. truncatum. This microorganism is a recognised human pathogen, causing spotted fever and has been detected in many countries in sub-Saharan Africa, including Cameroon [28,36,42,44]. Rickettsia aeschlimannii was also found in 2.2% of Rh. sanguineus sl. This tick species is, however, not known to be a competent vector of R. aeschlimannii, although it is a rickettsia that has been frequently associated with Hyalomma spp. Co-feeding could, therefore, be the cause of infection of this tick species, as mentioned above.
In Rh. lunulatus (5.6%) and Rh. muhsamae (10%) samples, rickettsial sequences with homology to R. massiliae were identified. Rickettsia massiliae is a pathogenic rickettsia that is associated with Rhipicephalus spp. ticks. It has been described as a human pathogen in Europe and South America, but there has never been reports of human infections in Africa [42]. This is the second finding of R. massiliae in Cameroon, following a previous study conducted by Vanegas and collaborators in 2018 [36], and is the first in the Western Highlands area of the country. It was previously reported in Rh. lunulatus from Cameroon, in Rh. senegalensis (33%) from Côte d'Ivoire and in Rh. guilhoni (22%) from Senegal [36,37,44].
In this study, Candidatus Rickettsia barbariae, a SFG rickettsiae of unknown pathogenicity, was detected in 20% (2/10) of Rh. muhsamae. This Rickettsia has previously been reported in Sardinia, Italy, where it was detected in a Rh. turanicus ticks and named Candidatus Rickettsia barbariae [49]. Several other studies have reported the presence of Candidatus Rickettsia barbariae in ticks from livestock in some African countries, including Cameroon [36,50,51].
Two sequences of Rickettsia sp. from Rh. lunulatus were close to the corresponding sequence of R. slovaca (97.92 and 97.88%) using the ompA gene, and close to the corresponding sequences of R. parkeri at 98.09% similarity with citrate synthase (gltA) gene. Furthermore, these two species are not known in sub-Saharan Africa [42]. However, in order to be classified as known species, an isolate should exhibit more than one of the following degrees of nucleotide similarity with the most homologous validated species: 99.9% for the gltA gene and, when amplifiable, 98.8% for the ompA gene [52]. In this case, it might be a new species of Rickettsia and more work would be needed to characterise it properly.
Several bacteria from the Anaplasmataceae family were detected in our study. Some bacteria from this family are known to be pathogens of human and veterinary importance [53].
We found a Wolbachia sp., which was 99.78% identical to Wolbachia pipientis (KT827385) and Wolbachia endosymbiont at the same percentage identity (KT273261) deposited by Fromont and collaborators in 2015 (unpublished data). Wolbachia spp. are known to be obligate intracellular endosymbionts of arthropods [54]. Some studies have shown that environmental factors can influence the presence of Wolbachia in arthropods, such as mosquitoes [55]. Furthermore, Wolbachia may impact the reproductive biology of their hosts, through a wide range of interactions [56]. The presence of Wolbachia sp. has already been reported in several mosquito species in Cameroon [57,58]. This study reports, for the first time in the country, the presence of Wolbachia sp. in ticks. The presence of Wolbachia spp. in one Ix. ricinus has previously been reported in Algeria [59]. However, the mechanism of the transmission of Wolbachia in ticks and their consequences on tick biology remains elusive [60,61].
We also identified the bacteria named Candidatus Ehrlichia urmitei and Candidatus Ehrlichia rustica. The DNA of these two bacterial species were found mainly in Rh. microplus and Rh. lunulatus ticks in this study. These potential new species have previously been found in Am. variegatum and Rh. microplus collected from cattle in Côte d'Ivoire [44] and in Rh. microplus in Mali [28]. Candidatus Ehrlichia urmitei has been also identified in Corsica R. bursa ticks [62]. These bacteria have not been characterised to date. Further studies are needed to identify the pathogenicity of these bacteria.
Anaplasma marginale and A. centrale were detected in Rh. microplus. These two are obligate intracellular bacteria responsible for bovine anaplasmosis (gallsickness) worldwide, transmitted by tick species, mainly belonging to the Rhipicephalus genus. Anaplasma marginale would also be transmitted by Hy. Rufipes [10]. The presence of the bacteria A. marginale in Rh. Microplus is obvious, as this tick species is its principal vector. Rhipicephalus microplus is also known to be a competent vector of various tick-borne livestock pathogens, as previously described [20]. It has been shown that the introduction of this invasive tick is most often accompanied by the greatest economic losses in cattle breeding [63]. Several other studies have been conducted in Cameroon showing the presence of these Anaplasma spp. Both in ticks and their hosts [15,64,65]. The presence of A. marginale has previously been reported in cattle from south-western Ethiopia [66], in Rhipicephalus spp. In western Kenya [67], and in Malian Rh. Microplus ticks [28].
Anaplasma platys was also detected in one Rh. Microplus and one Rh. Sanguineus. This is a canine anaplasmosis agent, which exclusively infects platelets and causes cyclic thrombocytopenia in dogs. The main known vector is Rh. sanguineus s.l. Anaplasma platys has also been identified in other mammals, including cattle and humans, and ticks worldwide [68]. This Anaplasma has already been detected from blood samples from cattle in Cameroon [17], but our study marks the first time it has been discovered in Rh. sanguineus s.l. ticks from Cameroon. On the scale of Africa as a whole, A. platys is a species frequently found in the northern part of the continent. However, it had been previously detected in cattle from Nigeria [69]. The presence of A. platys in main vector in this area poses a risk to potential host dogs.
Ehrlichia ruminantium is the aetiological agent of heartwater or cowdriosis, which particularly affects domestic and wild ruminants. It is transmitted by the ticks of the genus Amblyomma, primarily Am. variegatum and Am. hebraeum [70]. The prevalence of 0.7% in main vector, Am. variegatum, was significantly lower in comparison to recently published data (8.3%) from Mali [28]. Ehrlichia ruminantium had already been reported in several studies conducted in Cameroon, both in ticks and on their hosts [16,17,71,72]. These bacteria were previously described in Am. variegatum from several countries in western Africa [28,44].
We found Borrelia theileri, a member of the tick-borne relapsing fever group [45]. This study reports, for the first time, the presence of B. theileri in ticks from Cameroon. This had already been detected in cattle blood in the country [17]. Rhipicephalus microplus, in which this bacterium was found in this study, is genetically related and overlaps in distribution with Rh. decoloratus and Rh. annulatus, the main known vectors of B. theileri [73]. The infection rate (0.3%) reported here is comparable to the study in Mali, which found 0.5% infected with B. theileri in Rh. geigyi collected from cattle. Nevertheless, the vector capacity of Rh. geigyi is unknown [74]. Other studies have revealed the presence of species of this genus in different tick species in Algeria [59], Mali [28], and Côte d'Ivoire [44]. Reported cases of tick-borne relapsing fever have proven to be responsible for economic losses in livestock [75].
For some of our tick samples, which were positive for Bartonella spp. by qPCR using the ITS2 genes and Piroplasmida using 18S Piro gene, no amplification was observed after standard PCR, despite the fact that our DNA quantification gave satisfactory results. This issue may be due to a higher sensitivity of qPCR, compared to conventional PCR [48]. It could also be due to DNA degradation or to PCR inhibition by cattle blood. However, several other microorganisms in the Piroplasmida order have already been found in cattle blood samples from Cameroon [17,64,76].
Coxiella burnetii, a strict intracellular Gram-negative bacterium, is responsible for Q fever affecting humans and a variety of animals [77]. Although Q fever is far more frequently airborne, at least seven hard and soft tick species, including Hyalomma spp., have formally been shown to be competent vectors of C. burnetii [78]. Our study is the first to detect C. burnetii in tick samples (0.5%) in Cameroon. The prevalence rate is comparable to the 0.6% rate reported in ticks from Côte d'Ivoire [44]. In addition, a study conducted in Mali and Nigeria reported a high prevalence of 37.6% and 14% in ticks, respectively [28,79]. Coxiella burnetii represents a real, albeit underappreciated threat to human and animal health throughout Africa [77] Finally, it has been reported that ticks are often co-infected after taking a blood meal from a host carrying several infectious agents [80]. Several studies have reported mixed infections in feeding ticks caused mainly by Rickettsia spp. and C. burnetii [44,79]. We also reported, for the first time, mixed infections in ticks from Cameroon involving mainly R. africae and Ehrlichia spp. The co-infected rate here (3.4%) is comparable to that (1.3%) observed in Nigeria [79].

Conclusions
This study demonstrates that ticks in Cameroon harbour a wide variety of tick-borne pathogens of veterinary and medical importance, in particular R. africae, the agent of ATBF. Humans are thus at risk of infection with R. africae, and Africa tick-bite fever should also be considered in patients presenting with febrile illnesses. This study also reports, for the first time in ticks from Cameroon, the presence of various agents, such as Borrelia theileri. Nevertheless, further studies are needed to ascertain the origin and zoonotic potential of the strains and their significance for animals and human health. These data can contribute towards future research, for example by providing an avenue for larger studies of ticks and the pathogens they harbour in Cameroon. Funding: This study was supported by the Institut Hospitalo-Universitaire (IHU) Méditerranée Infection, the French National Research Agency under the "Investissements d'avenir" programme, reference ANR-10-IAHU-03, the Région Provence Alpes Côte d'Azur and European ERDF PRIMI funding. Y.N.-Y. received a Ph.D. scholarship grant from IHU Méditerranée Infection. The funders had no role in the study design, data collection, and analysis; decision to publish; or preparation of the manuscript.
Institutional Review Board Statement: Permission to undertake the study was obtained from the Western Regional Delegation of the Ministry of Livestock, Animal Husbandry and Fisheries in Cameroon, reference number 21/20/L/RO/DREPIA-O/SRAG.

Informed Consent Statement:
The farmers provided informed consent for the use of their cattle and information for this study. For informed consent it was given verbally by the livestock owners.The collection of ticks from cattle did not involve national parks or other protected areas or endangered or protected species. Data Availability Statement: Not applicable.