Molecular detection of pathogens in ticks and fleas collected from companion dogs and cats in East and Southeast Asia

Background Ticks and fleas are considered amongst the most important arthropod vectors of medical and veterinary concern due to their ability to transmit pathogens to a range of animal species including dogs, cats and humans. By sharing a common environment with humans, companion animal-associated parasitic arthropods may potentially transmit zoonotic vector-borne pathogens (VBPs). This study aimed to molecularly detect pathogens from ticks and fleas from companion dogs and cats in East and Southeast Asia. Methods A total of 392 ticks and 248 fleas were collected from 401 infested animals (i.e. 271 dogs and 130 cats) from China, Taiwan, Indonesia, Malaysia, Singapore, Thailand, the Philippines and Vietnam, and molecularly screened for the presence of pathogens. Ticks were tested for Rickettsia spp., Anaplasma spp., Ehrlichia spp., Babesia spp. and Hepatozoon spp. while fleas were screened for the presence of Rickettsia spp. and Bartonella spp. Result Of the 392 ticks tested, 37 (9.4%) scored positive for at least one pathogen with Hepatozoon canis being the most prevalent (5.4%), followed by Ehrlichia canis (1.8%), Babesia vogeli (1%), Anaplasma platys (0.8%) and Rickettsia spp. (1%) [including Rickettsia sp. (0.5%), Rickettsia asembonensis (0.3%) and Rickettsia felis (0.3%)]. Out of 248 fleas tested, 106 (42.7%) were harboring at least one pathogen with R. felis being the most common (19.4%), followed by Bartonella spp. (16.5%), Rickettsia asembonensis (10.9%) and “Candidatus Rickettsia senegalensis” (0.4%). Furthermore, 35 Rhipicephalus sanguineus ticks were subjected to phylogenetic analysis, of which 34 ticks belonged to the tropical and only one belonged to the temperate lineage (Rh. sanguineus (sensu stricto)). Conclusion Our data reveals the circulation of different VBPs in ticks and fleas of dogs and cats from Asia, including zoonotic agents, which may represent a potential risk to animal and human health.

which may also infect humans [2][3][4]. The cat flea Ctenocephalides felis is the primary vector of Bartonella henselae, the main causative agent of cat-scratch disease [5,6], and is also considered as vector of Rickettsia felis [7].
East (EA) and Southeast Asia (SEA) are among the world's fastest-growing economic regions [8], which also resulted in a rise in the number of companion dogs and cats [9]. Companion dogs and cats live in close association with humans, potentially carrying ticks and fleas into human settlements. A large-scale survey conducted in EA and SEA reported that 22.3% of dogs and 3.7% of cats were infested by ticks, while 14.8% of dogs and 19.6% of cats were infested by fleas [10]. The most common flea species parasitizing dogs and cats in EA and SEA is C. felis, with Ctenocephalides orientis being increasingly observed in dogs [10,11]. Rhipicephalus sanguineus (s.l.), Rhipicephalus haemaphysaloides and Haemaphysalis longicornis represent the most common tick species reported in dogs and cats [10][11][12][13][14]. These tick species are responsible for the transmission of several species of apicomplexan protozoa of the genus Babesia. Babesia vogeli was reported in cats in Thailand and China [15,16] and widely reported in dogs in EA and SEA, including China, Cambodia, Thailand, the Philippines and Malaysia [17][18][19][20]. Additionally, dogs from Taiwan, Malaysia, China, and Singapore [10,21,22] were also diagnosed with Babesia gibsoni infection. Other apicomplexan parasite commonly found in dogs across this region is H. canis, which is transmitted by ingestion of Rh. sanguineus (s.l.). This protozoan is commonly found in dogs from Thailand, Taiwan, China, Cambodia, Malaysia, Vietnam and the Philippines [10,17,18,[23][24][25][26] and in cats from Thailand and the Philippines [10,24]. Of the tick-borne anaplasmataceae bacteria, Anaplasma platys was found in dogs from Malaysia [22] and cats from Thailand [27]. Apart from tick-borne pathogens, flea-borne pathogens are also increasingly recognized as important pathogenic agents to animals and humans. For instance, R. felis, the etiological agent of flea-borne spotted fever in humans, has been detected in dogs from Cambodia and China [18,28] and in C. felis from Taiwan, Laos, and Malaysia [29,30]. Other zoonotic flea-borne pathogens such as B. henselae and Bartonella clarridgeiae, agents of cat-scratch disease, were molecularly detected in cats and their fleas from the Philippines, Indonesia, Singapore, Thailand, Malaysia and China with the prevalence ranging from 10 to 60% [31][32][33][34][35][36].
Despite previous scientific investigations reported the circulation of vector-borne pathogens (VBPs) in dogs and cats in EA and SEA, there is a lack of similar studies conducted in their associated ticks and fleas. Therefore, the present study aimed to provide an overview of the pathogens circulating in ticks and fleas from companion dogs and cats in EA and SEA.

Samples collection and DNA isolation
Of the 2381 privately-owned animals examined (i.e. 1229 dogs and 1152 cats), ticks and fleas were collected from 401 infested animals (i.e. 271 dogs and 130 cats) from China, Taiwan, Indonesia, Malaysia, Singapore, Thailand, the Philippines and Vietnam under the context of a previous multicenter survey [10]. Ticks and fleas were collected and placed in labelled tubes individualized per host, containing 70% ethanol. Ticks and fleas (20%) were randomly selected from each tick/flea species and from each infested animal in all studied countries, giving a total number of 392 ticks (i.  Table 4 in Colella et al. [10]) Genomic DNA was isolated according to the procedures previously described [10,37].

Molecular detection and phylogenetic analysis of pathogens
Tick DNA samples were tested for the presence of apicomplexan protozoa (i.e. Babesia spp., Hepatozoon spp.), Anaplasmataceae (i.e. Anaplasma spp., Ehrlichia spp.) and Coxiella burnetii by conventional PCR (cPCR). Flea DNA samples were tested by using real-time PCR for Bartonella spp. The presence of Rickettsia spp. was also screened in both tick and flea samples. In particular, the first cPCR amplified a portion of citrate synthase (gltA) gene, which is presented in all members of the genus Rickettsia. Positive samples were then subjected to a second cPCR, which amplified a fragment of the outer membrane protein (ompA) of the spotted fever group (SFG) rickettsiae. All primers and PCR protocols used for the detection of VBPs are summarized in Table 1. For all reactions, DNA of pathogen-positive samples served as a positive control. Amplified cPCR products were examined on 2% agarose gels stained with GelRed (VWR International PBI, Milan, Italy) and visualized on a Gel-Logic 100 gel documentation system (Kodak, New York, USA). The cPCR amplicons were sequenced using the Big Dye Terminator v.3.1 chemistry in a 3130 Genetic analyzer (Applied Biosystems, California, USA). Nucleotide sequences were edited, aligned and analyzed using the BioEdit 7.0 software and compared with those available in the GenBank database using Basic Local Alignment Search Tool (http://blast .ncbi.nlm.nih.gov/Blast .cgi).
To assess the genetic variation of Rh. sanguineus (s.l.) and Rickettsia spp., the mitochondrial 16S rDNA sequences of Rh. sanguineus (s.l.) ticks generated previously [10] as well as the gltA and ompA gene sequences of Rickettsia spp. generated herein were subjected to phylogenetic analysis. Phylogenetic relationship was inferred by Maximum Likelihood (ML) method after selecting the best-fitting substitution model. Evolutionary analysis was conducted on 8000 bootstrap replications using the MEGA 7 software [44].

Statistical analysis
The percentage of detected pathogens was calculated and 95% confidence intervals (95% CI) (by the modified Wald method) were estimated by using Quantitative Parasitology 3.0 software [45]. Fisher's exact test was performed to analyze statistically significant differences in the detection of pathogens in fleas and ticks, and in the distribution of different Rickettsia spp. among different flea species using SPSS 16.0 software. Differences were considered significant at P < 0.05.
Representative nucleotide sequences for each detected pathogen displayed 99.4-100% identity with those available in GenBank database. In particular, A. platys nucleotide sequences (n = 3) revealed 99.6-100% identity with KU500914 (host: Canis lupus familiaris; origin: Malaysia), H. canis (n = 20) 99.7-100% identity with DQ519358 (host: C. lupus familiaris; origin: Thailand), E. canis (n = 7) and B. vogeli (n = 4) 100% identical to MN227484 (host: C. lupus familiaris; origin: Iraq) and KX082917 (host: C. lupus familiaris; origin: Angola), respectively. For Rickettsia spp. detection in ticks, the gltA sequence identified in one tick showed 99.7% identity with R. asembonensis (GenBank: KY445723; host: C. felis; origin: Brazil) and another was identical to R. felis (100% nucleotide identity with MG845522 (host: C. felis; origin: Chile) and 99.4% nucleotide identity with R. felis strain URRWXCal2; GenBank: CP000053 (source: cultivation; origin: USA)). The ompA gene amplification was successful for two samples which showed 100% nucleotide identity with unidentified Rickettsia sp. (GenBank: EF219467; host: R. haemaphysaloides; origin: Taiwan), which was Table 2 Pathogens detected in ticks according to their species, developmental stage, sex and host in East and Southeast Asia Abbreviations: C, cat; D, dog; L, larva; N, nymph; M, adult male; F, adult female a This female tick was reported as "Ixodes sp. " in [10]. Following reassessment of photomicrography images of this tick by one of the co-authors (F.D.-T.) the following morphological features were observed: auriculae and cornua present; porose area small, not contiguous, hypostome with a 2/2 dental formula on almost the entire hypostome; coxa I with slight internal spur, coxae III and IV each with external spur; syncoxae present on coxae I and II; trochanters lacking spurs. As such, this female shares several morphological features with Ixodes ovatus [46], but genetic data from a partial 16S rDNA sequence (percent identity: 90.7% with U95900) suggest that this may belong to a distinct species The phylogenetic tree based on the partial ompA gene sequences showed that all R. felis isolated from fleas were assembled together in one cluster, whereas Rickettsia sp. isolated from ticks clustered with R. rhipicephali and Rickettsia massiliae (Fig 1). In the gltA tree, phylogenetical analysis revealed that R. felis, R. asembonensis and "Ca. R. senegalensis" herein detected were formed together in a well-supported sister cluster include other R. felis-like organisms (RFLOs), close to the cluster of Rickettsia australis and Rickettsia akari (Fig. 2).  The ML tree of 35 representative mitochondrial 16S rDNA sequences of Rh. sanguineus (s.l.) gene showed that 34 sequences were identical to each other and identified as belong to the tropical lineage of Rh. sanguineus (s.l.) (100% identity with GU553075; origin: Brazil). One sequence from a tick collected from a dog in Beijing (northeast China) clustered with Rh. sanguineus (sensu stricto) (100% identity with GU553078; origin: Argentina) (Fig. 3).
Representative sequences of pathogens detected in this study were deposited in the GenBank database under the accession numbers MT499354-MT499356 (H.

Discussion
The results of this study reveal the presence of several pathogens in ticks (e.g. A. platys, B. vogeli, E. canis, H. canis and Rickettsia spp.) and fleas (e.g. Rickettsia spp. and Bartonella spp.) collected from dogs and cats in EA and SEA. The relatively low occurrence of pathogens herein detected in ticks is consistent with previous surveys conducted in ticks infesting owned dogs in Asia [13,26,[47][48][49]. Conversely, the occurrence of VBPs is higher in ticks collected from stray animals [24]. Although B. gibsoni was identified in dogs from China (2.3%; [10]), none of the tested ticks from these dogs was found positive for this parasite. The absence of B. gibsoni in tick populations is probably due to the low number of H. longicornis and H. hystricis, which are recognized as vectors of this pathogen [50,51]. The infection of E. canis (14.8% by serology) and H. canis (1.6% by cPCR) in host populations [10], along with the detection of these pathogens Fig. 1 Phylogenetic relationships of Rickettsia spp. isolated in this study (in bold) to other Rickettsia spp. based on partial sequences of the ompA gene. The analyses were performed using a Maximum Likelihood method with Tamura 3-parameter model. A discrete Gamma distribution was used to model evolutionary rate differences among sites. GenBank accession number, isolation source and country of origin are presented for each sequence in Rh. sanguineus (s.l.) in the sampling areas support the vector role of this tick species in the transmission of these VBPs in this region [11,14]. The finding of A. platys in dogs (7.1% by serology; [10]) and in Rh. sanguineus (s.l.) further suggests its vector competence for this pathogen. Additionally, the detection of R. felis and R. asembonensis in Rh. sanguineus (s.l.) is similar to previous results in Chile [52], Brazil [53] and Malaysia [54], consequently giving more concern about the role of Rh. sanguineus (s.l.) in the transmission of Rickettsia spp. other than R. conorii, R. massiliae and R. rickettsii [4,55]. Rickettsia sp. sequences herein obtained from Rh. haemaphysaloides are identical to one previously generated from the same tick species in Taiwan (named Rickettsia sp. TwKM01) [56]. This genotype and its closest related species R. rhipicephali remain of unknown pathogenicity to mammals [56,57]. Additionally, the vector role of Rh. haemaphysaloides needs further investigations since it was found harboring multiple pathogens such as R. rhipicephali, A. platys, E. canis, B. gibsoni [13,58].
Of the detected VBPs, R. felis stood out as the most important due to its wide distribution, association with Fig. 2 Phylogenetic relationships of Rickettsia spp. isolated in this study (in bold) to other Rickettsia spp. based on partial sequences of the gltA gene. The analyses were performed using a Maximum Likelihood method with Tamura 3-parameter model. A discrete Gamma distribution was used to model evolutionary rate differences among sites. GenBank accession number, isolation source and country of origin are presented for each sequence various arthropods, and importance as an emerging zoonotic pathogen [59]. In Asia, the first human case of flea-borne spotted fever attributed to R. felis was detected in the Thai-Myanmar border [60], since then several cases of human infection have been documented in Taiwan [61], Thailand [62], Laos [63], Vietnam [64] and Indonesia [65]. Although R. felis was detected in many arthropods, including non-hematophagous insect (i.e. the book louse Liposcelis bostrychophila) [66], the distribution of this rickettsia is highly affiliated with the distribution of C. felis [59]. Ctenocephalides felis is the most well-recognized vector of this rickettsia, which is transmitted transovarially and transstadially in the fleas [67], with dogs as proven mammalian reservoir hosts [68]. The high occurrence of R. felis in C. felis along with the high relative frequency of this flea species in host populations (65.1% in dogs and 98.7% in cats) [10] emphasizes the risk of R. felis infection in animals and humans.
The detection of R. asembonensis only from fleas collected on dogs (mainly C. orientis but in one case in C. felis) may suggest that dogs could act as amplifying hosts of this rickettsia, as they do for R. felis [68]. Rickettsia asembonensis is the most well-characterized genotype of RFLOs [69]. This rickettsia was initially described in fleas from dogs and cats in Kenya [70] and was then reported in various arthropods worldwide [69]. In Asia, R. asembonensis was also found in C. orientis from dogs [54] and in macaques from Malaysia [71]. Additionally, Rickettsia sp. RF2125, a genotype highly related to R. asembonensis, was reported with high incidence in C. orientis from India and Thailand [72,73], and was also found in a febrile patient from Malaysia [74]. Moreover, R. felis, R. asembonensis and "Ca. R. senegalensis" clustered in the SFG rickettsiae clade (Fig. 3), and while R. felis is a recognized Fig. 3 Phylogenetic relationships of the present Rhipicephalus sanguineus (s.l.) sequences (in bold) to other Rhipicephalus spp. based on a portion of the mitochondrial 16S rRNA gene. As most sequences (n = 34) were identified as belonging to the tropical lineage Rh. sanguineus (s.l.), representatives were selected for each country. The analyses were performed using a Maximum Likelihood method with Tamura 3-parameter model. A discrete Gamma distribution was used to model evolutionary rate differences among sites. GenBank accession number, isolation source and country of origin are presented for each sequence pathogen [67], the pathogenicity of other RFLOs is currently unknown.
Besides acting as vectors of Rickettsia spp., fleas have been well-recognized as vectors of Bartonella spp. [6,75]. The occurrence of Bartonella-positive fleas in our study was slightly lower than previous investigations in Laos [30], Malaysia [35] and Thailand [76]. Nevertheless, the occurrence of the two common Bartonella spp. (i.e. B. henselae and B. clarridgeiae) in dogs and cats from EA and SEA is relatively high; up to 60% [18,31,76,77]. Additionally, B. henselae infection in humans is usually associated to previous exposure to cats or cat fleas [78], emphasizing the role of cat fleas in Bartonella spp. transmission between animals and humans.
In the present study, all tested tick specimens were negative for C. burnetii although this pathogen was detected in Rh. sanguineus (s.l.) from dogs in Malaysia [79], in dogs in Taiwan [80] and in humans from Thailand [81]. Additionally, Coxiella-like endosymbionts were strongly associated with Rh. sanguineus (s.l.) tropical lineage [82], although the role of these endosymbionts in the biology and vectoral capacity of this tick lineage needs further investigation.
Finally, the genetic lineage of Rh. sanguineus (s.l.) from EA and SEA was investigated based on the 16S rDNA sequences. The finding of Rh. sanguineus (s.s.) in Beijing, a cold area, along with the existence of the tropical lineage in warmer localities, agreed with previous studies, which indicated that the tropical lineage is present in areas with annual mean temperature > 20°C, whereas the temperate lineage occurs in areas with annual mean temperature < 20°C [83]. This information is also relevant from a pathogen transmission perspective, considering that different Rh. sanguineus (s.l.) lineages may present variable vector competence and/or capacity for different pathogens; for instance, E. canis is primarily vectored by Rh. sanguineus (s.l.) tropical lineage [84].

Conclusions
Data herein reported updates the list of pathogens occurring in ticks and fleas from companion dogs and cats in EA and SEA. By sharing the common environment with humans, these parasitic arthropods could be responsible for the transmission of pathogens to humans (i.e. R. felis). Strategies to prevent tick and flea infestations in these animals are fundamental to decrease the risk of transmission of VBPs to animals and humans.