Molecular Evidence of Bartonella Species in Ixodid Ticks and Domestic Animals in Palestine

Ticks play an important role in disease transmission as vectors for human and animal pathogens, including the Gram-negative pathogen Bartonella. Here, we evaluated the presence of Bartonella in ixodid ticks and domestic animals from Palestine. We tested 633 partly engorged ticks and 139 blood samples from domestic animals (dogs, sheep and camels) for Bartonella using ITS-PCR. Bartonella DNA was detected in 3.9% of the tested ticks. None of the ticks collected from sheep and goats were positive for Bartonella. Seventeen R. sanguineus ticks (17/391; 4.3%) collected from dogs were infected with B. rochalimae (n = 10), B. chomelii (n = 6), and B. koehlerae (n = 1). Four H. dromedarri ticks (4/63; 6.3%) obtained from camels were infected with B. bovis (n = 2) and B. rochalimae (n = 2). Among canine blood samples (n = 110), we found one asymptomatic female dog to be infected with B. rochalimae (0.9%). The detection of zoonotic Bartonella species in this study should raise awareness of these vector-borne diseases among physicians, veterinarians and public health workers and highlight the importance of surveillance and preventive measures in the region.


INTRODUCTION
Tick-borne diseases comprise a group of globally distributed and rapidly spreading illnesses caused by a range of pathogens. Molecular approaches make it possible to screen ticks for pathogens of veterinary and public health importance and perform detailed epidemiological studies (Sparagano et al., 1999). Bartonellosis is an infectious disease caused by bacteria from the genus Bartonella, which infect erythrocytes and endothelial cells in humans (Harms and Dehio, 2012). These pathogens are transmitted by biting arthropod vectors and infect a wide range of wild and domestic mammals, including rodents, cats, dogs, and cattle. In humans, Bartonella is responsible for emerging and reemerging diseases worldwide with presentations that range from subclinical or self-limiting infection to severe, life-threatening disease.
Different Bartonella species appear to be adapted to specific mammalian hosts (Breitschwerdt and Kordick, 2000;Vayssier-Taussat et al., 2010). For example, cats are the main reservoir for B. henselae, which causes cat-scratch disease. Cats are infected through the bites of cat fleas, while humans are directly infected through scratches or bites from an infected cat (Billeter et al., 2008).
Although transmission of Bartonella to humans typically occurs through traumatic contact with infected animals or by blood-sucking insect vectors such as fleas, lice and sand flies, this group of species has also been widely reported from ticks. The potential for involvement of ticks in transmission of Bartonella spp. has been suggested by several molecular and serological epidemiological studies (Billeter et al., 2008). Bartonella was first detected in questing ticks from the USA, including Ixodes pacificus, Dermacentor, and R. sanguineus (Chang et al., 2000). Additional surveys conducted in the Netherlands, France, Poland, and Austria have demonstrated the presence of Bartonella DNA in Ixodes ricinus ticks obtained from vegetation (Billeter et al., 2008). In France, the species B. henselae has been identified in Ixodes ricinus isolated from vegetation (Vayssier-Taussat et al., 2013), as well as from humans exposed to tick bites (Vayssier-Taussat et al., 2016).
Clinical studies have supported transmission of Bartonella by ticks to humans, as infections have occurred after tick bites without any known contact with other arthropods (Morozova et al., 2005;Billeter et al., 2008). Bartonella infection was reported in three patients with scalp eschar and neck lymphadenopathy following tick bites (Angelakis et al., 2010). A recent study by Vayssier-Taussat et al. (2016) identified potentially zoonotic Bartonella strains in symptomatic patients who reported tick bites. Natural co-infections with Bartonella species and other tick borne pathogens such as Babesia, Anaplasma, and Borrelia have also been demonstrated (Hofmeister et al., 1998;Angelakis et al., 2010). In particular, co-infections of Bartonella and Borrelia have been reported in humans from the USA and Europe (Billeter et al., 2008). These observations provide indirect evidence for tick-borne transmission, even in the absence of direct proof of tick vector competence for Bartonella (Cotte et al., 2008;Reis et al., 2011). Furthermore, experimental transmission studies using infected ticks and live susceptible animals support the role of ticks in the natural lifecycles of some Bartonella species (Cotte et al., 2008;Reis et al., 2011).
In our previous studies, when we screened ticks from domestic animals, we identified several tick-borne pathogens, including Rickettsia from the spotted fever group, Babesia and Hepatozoon Azmi et al., 2016). Given the potential role of ticks as a source of zoonotic Bartonella infection in humans (Vayssier-Taussat et al., 2016), we wished to determine whether ticks carry Bartonella in Palestine. More specifically, we set out to extend our previous surveys and to assess the presence of Bartonella in a set of previously studied ticks and in the blood samples collected from their animal hosts throughout Palestine.

Study Sites, Tick, and Animal Samples
A total of 633 hard ticks were collected during January to October 2014. The ticks were collected from dogs, sheep, goats, and camels in nine districts of Palestine (Hebron, Jenin, Jericho, Nablus, Qalqilia, Ramallah, Salfit, Tubas, and Tulkarem), located in three zones in the central, northern and southern regions of the country (Figure 1). Ticks were identified based on morphological characteristics (Feldman-Musham, 1954).
A total of 110 blood samples were collected from outdoor domestic dogs, with simultaneous tick collection from the locations mentioned above (17 from Hebron, 25 from Jenin, 18 from Jericho, three from Nablus, 21 from Ramallah, 13 from Salfit, and 13 from Tulkarem). All samples were collected in FIGURE 1 | Distribution of Ixodid ticks collected from nine districts in Palestine from which Bartonella DNA was detected.
EDTA-anticoagulant tubes and stored at −20 • C until further use. Study animals were selected irrespective of sex and age. In addition, 29 blood samples were obtained from camels (n = 19), sheep (n = 7), and goats (n = 3) from Jenin, Nablus, and Jericho, in January 2016. None of the animals showed clinical signs; all were apparently healthy at the time of sampling. The animal owners were verbally informed about the goals of this research and the sampling protocol. All owners gave their written informed consent to collect blood from their animals. The ethics committee at Al-Quds University approved the study.

DNA Extraction
DNA was extracted from each tick using a DNA extraction kit (QIAGEN GmbH, 40724 Hilden, Germany) following the manufacturer's instructions. The eluted DNA (100 µl) was stored at −20 • C until used as templates for PCR amplifications. DNA was extracted from whole blood (200 µl) following the QIAamp animal blood and Tissue Kit procedure (QIAGEN GmbH, Hilden, Germany), adjusted in 200 µl of Tris-EDTA (TE) buffer and stored at −20 • C until further use.

Molecular Detection and Identification of Bartonella Species
For screening, conventional PCR was performed on all tick and blood samples (n = 772) targeting the Intergenic Transcribed Spacer (ITS) locus, using the following forward and reverse primers: (321s: 5 AGATGATGATCCCAAGCCTTCTGG and H493as: 5 -TGAACCTCCGACCTCACGCTTATC) as previously described (Maggi and Breitschwerdt, 2005;Gutierrez et al., 2014). PCR reactions were performed in 25-µl Syntezza PCR ready mix (Syntezza, Jerusalem), containing 1 µM of each set of primers and 5 µl of the extracted DNA. The thermal cycling procedure was as described previously (Norman et al., 1995;Renesto et al., 2001;Maggi and Breitschwerdt, 2005). Samples of PCR grade water were included as a negative (no-DNA) controls. To confirm amplicon identity, all PCR products from the positive samples underwent DNA sequencing; the nucleotide sequences were compared to those present in GenBank database using the Basic Alignment Search Tool (BLAST). 1 Statistical analysis was done using the SPSS program v20.

Bartonella DNA in Animal Blood
Among the canine blood samples (n = 110; 22 female: 88 males), one female dog sample was positive for Bartonella, yielding a PCR-amplified ITS sequence that showed 99% sequence identity to the reference sequence of B. rochalimae (FN645466.1). None of the blood samples from camels (n = 19), sheep (n = 7), and goats (n = 3) were positive for Bartonella.
Overall, 25 ticks (3.9%; 9 females, 11 males, and 5 nymphs) were positive for Bartonella DNA by ITS-PCR. Of these 21 ticks were collected from dogs and four from camels. None of the ticks from sheep or goats were positive for Bartonella ( Table 1). Identification of Bartonella was successful for 21 positive samples (84%). The initial attempt to sequence four amplicons failed and no additional DNA samples were available. Seventeen R. sanguineus ticks were infected with B. rochalimae (n = 10), B. chomelii (n = 6), and B. koehlerae (n = 1). All these ticks were obtained from dogs.
Four H. dromedarii ticks obtained from a single camel in Hebron and from two camels in Jericho were found to be infected with B. bovis (n = 2) and B. rochalimae (n = 2), respectively ( Table 2)

DISCUSSION
Here, we report Bartonella DNA in ixodid ticks and blood samples from domestic animals from Palestine. The overall prevalence of Bartonella DNA in ticks (3.9%) was in agreement with previous screening undertaken worldwide including Czech Republic, United States, Italy and Thailand (Sanogo et al., 2003;Hercik et al., 2007;Billeter et al., 2008Billeter et al., , 2012. Several Bartonella species have been identified in humans, animals, and their flea vectors in neighboring countries. However, Bartonella DNA was not detected in any of the Ixodid ticks examined in Israel and Egypt (Loftis et al., 2006;Harrus et al., 2011).
In our study, four Bartonella species were identified: B. rochalimae, B. chomelii, B. Bovis, and B. koehlerae. B. rochalimae was the predominant species among Bartonellapositive ticks (12 out of 25 samples; 48%). We examined seven hard tick species for Bartonella and found evidence of the pathogen in two of them: the brown dog tick, Rhipicephalus sanguineus and the camel tick, Hyalomma dromedarii. These findings represent the first detection of Bartonella in Ixodid ticks from the Middle East. Cases of human parasitism by the brown dog tick-the most widespread tick in the world-are  -Soto et al., 2006). Together, the exposure of animals to arthropod vectors and the proximity of infected animals to humans make some Bartonella species potential zoonotic agents. People dealing with dogs (e.g., veterinarians, shepherds, dog owners and pet shop workers) appear to be at particular risk of exposure to R. sanguineus and its pathogens. We found Bartonella DNA in a single sample of dog blood, with high sequence identity to a reference sequence from B. rochalimae (99% DNA sequence identity; GenBank accession number FN645466.1). The dog that provided the sample was apparently healthy, showing no signs of bartonellosis. Detection of Bartonella in asymptomatic dogs has also been reported from Peru (Diniz et al., 2009). We identified B. rochalimae in two R. sanguineus ticks obtained from the same dog; both tick samples yielded sequences 100% identical to each other and to the sequence obtained from dog. Although the blood and tick samples were collected from dogs at the same time, Bartonella DNA was more prevalent in the dog ticks (5.4%) than in the dog blood (0.9%) suggesting that Bartonella be carried in partially engorged adult ticks. These findings provide highly suggestive molecular evidence that R. sanguineus ticks can act as vectors of animal-associated Bartonella infection in Palestine. Furthermore, although the role of dogs as source of human Bartonella infection remains unclear, we speculate that they may present a risk for zoonotic transmission similar to that seen with B. henselae in cat scratch disease (Chomel et al., 2006).
Recently, a study conducted in Israel confirmed the presence of a novel species of Bartonella in camelids, which has been named Bartonella dromedarii sp. nov . In our study, we identified two species of Bartonella (B. rochalimae and B. bovis) in ticks obtained from camels, although we did not detect Bartonella DNA in blood samples from these camels. However, we cannot rule out bloodstream infections in the camels as the source of Bartonella in the ticks, as the blood samples were not taken at the same time as the ticks. Furthermore, even if camels are not reservoir hosts for these Bartonella species, they may have an important role as mechanical dispersers of infected ticks.
Our discovery of Bartonella in camel ticks from the species Hyalomma dromedarii is worrying because other species of hard ticks from the genus Hyalomma clearly bite humans (Psaroulaki et al., 2005;Bursali et al., 2011) and there are suggestions that H. dromedarii can do so too 2 . Other lines of evidence support the role of ticks in the natural cycles of some Bartonella species including those pathogenic for humans (Cotte et al., 2008;Reis et al., 2011;Liu and Bonnet, 2014). In particular, ticks can be infected in the larval or nymph stages by ingesting blood from an intermediate host carrying Bartonella. The pathogen can then survive in the midgut of ticks during molting and can be transmitted through feeding to an uninfected host.
In the present study, brown dog ticks were found to be infected with B. chomelii-a pathogen first isolated from French domestic cattle (Maillard et al., 2004) and found to be the most frequent Bartonella species infecting cattle grazing in Spain pastures (Antequera-Gomez et al., 2015). We also describe the first detection of B. koehlerae in a R. sanguineus tick obtained from dog. This species was first isolated from the blood of two pet cats in California (Droz et al., 1999). Since then, it has been detected in cats and their fleas in France, Israel and Palestine (Rolain et al., 2003;Gutierrez et al., 2013;Nasereddin et al., 2014), has been reported as causing endocarditis in humans and Boxer dogs in Israel (Avidor et al., 2004;Ohad et al., 2010). However, the presence of a microbial agent within a tick does not imply that the tick is a biological vector and might transmit it during the course of blood feeding (Telford and Wormser, 2010).
One of Bartonella -positive ticks we obtained from a dog harbored Rickettsia, a pathogen that is known to be ticktransmitted . This fits in with evidence from across the world that Bartonella is often found in ticks alongside well-known tick-transmitted organisms such as Anaplasma, Borrelia, and Rickettsia (Angelakis et al., 2010). Other studies supporting the hypothesis that Bartonella can be transmitted by ticks include a US report that dogs infected with Bartonella were also seropositive for Anaplasma phagocytophilum (MacDonald et al., 2004) and a human case study showing that patients infected with Borrelia burgdorferi after tick bites also carried Bartonella DNA in their blood (Podsiadly et al., 2003).

CONCLUSION
The detection of zoonotic Bartonella species in this study should increase the awareness of these vector-borne diseases among physicians, veterinarians and public health workers and highlight the importance of surveillance and preventive measures in the region. Additional epidemiologic surveys are required to enhance our understanding of the transmission dynamics of Bartonella in Palestine and in other parts of the Middle East.