Identification of 12 Piroplasms Infecting Ten Tick Species in China Using Reverse Line Blot Hybridization

Piroplasmosis is naturally caused by tick-transmitted, and generally host-specific, protozoan parasites in the genera Babesia and Theileria. It is a disease of domestic and wild animals as well as humans, and it is therefore responsible for economic losses. The impacts include lowered meat and milk production, abortions, fertility in bulls, control measure costs, trade or import restrictions, and a general impact on the global cattle industry (Bock et al., 2004).

Ticks are important vectors of a large variety of pathogens that infect animals and cause diseases of veterinary importance (Jonsson et al., 2008). Piroplasms are infective when both their definitive (ticks) or intermediate hosts (vertebrates) are encountered. Some tick species can carry more than 1 piroplasm species that affects the same host or different animal hosts. Furthermore, some piroplasm species are transmitted by different tick species, even those with close phylogenetic relationships (Yin et al., 2002).

Babesia cf. motasi and Babesia sp. Xinjiang were reported to cause ovine babesiosis in China (Liu et al., 2007). Haemaphysalis qinghaiensis and Haemaphysalis longicornis were described as the vectors of B. cf. motasi (Bai et al., 2002a; Guan et al., 2010). Hyalomma anatolicum was reported to transmit Babesia sp. Xinjiang (Guan et al., 2009). Four Babesia species, Babesia bovis, Babesia bigemina, Babesia major, and Babesia ovata, were described as responsible for cattle, buffalo, and yak babesiosis in China (Luo et al., 2005; Qin et al., 2015). Babesia bigemina and B. bovis, which are major causative agents of bovine babesiosis, share the same vectors Rhipicephalus (Boophilus) microplus, Rhipicephalus annulatus, and Rhipicephalus geigyi, and often present in co-infection (Ravindran et al., 2006; Tavassoli et al., 2013). Babesia major is transmitted by H. longicornis and Haemaphysalis punctata, while B. ovata is transmitted by H. longicornis, which have low virulence in the bovine species (Yin et al., 1996).

Ovine theileriosis is mainly caused by Theileria luwenshuni, Theileria uilenbergi, and Theileria ovis in China (Yin et al., 2007). Theileria uilenbergi and T. luwenshuni are considered highly pathogenic to small ruminants, such as sheep and goats, and are transmitted by H. qinghaiensis and H. longicornis (Li et al., 2009), while T. ovis is non-pathogenic or mildly pathogenic for small ruminants and can be transmitted by Hyalomma anatolicum (Li et al., 2010). Three Theileria species, Theileria annulata, Theileria sergenti, and Theileria sinensis, have been reported in China as agents of bovine theileriosis (Liu et al., 2010, 2015). Theileria annulata is transmitted by Hy. anatolicum, Hyalomma detritum, Hyalomma excavatum, Hyalomma dromedarii, and Hyalomma marginatum (Sayin et al., 2003). Theileria sergenti is the most prevalent, which is transmitted by H. longicornis (Liu et al., 2010; Zhao et al., 2017). Theileria sinensis infects cattle and yaks and is transmitted by H. qinghaiensis (Bai et al., 2002b).

Epidemiologic studies concerning tick-borne diseases provide information about endemic instability of these diseases (Ekici et al., 2012). Varied diagnostic methods have been used for the detection of Theileria and Babesia species in ticks, but the application of molecular methods provides improved sensitivity and specificity and allows direct detection of parasites (Altay et al., 2008; Aydin et al., 2013). Recently, PCR-RLB was used for rapid, simultaneous detection and differentiation of numerous species of Babesia and Theileria even in mixed infection (Schnittger et al., 2004; Niu et al., 2009; Iqbal et al., 2013). Although there have been several studies of the frequency of ixodid tick species and the prevalence of tick-borne diseases in most areas in China (Chen et al., 2014a; Li et al., 2014a; Liu et al., 2015; Yu et al., 2015), little information is available about the prevalence of pathogens in ticks. Thus, the investigations of known parasites and new tick-piroplasms associations are important and significant for the animal health and livestock industry.

In this study, our aim is to apply a previously developed PCR-RLB assay to detect ovine and bovine piroplasms in ixodid ticks collected from animals in different provinces of China, to better quantify the flow of piroplasm infection in field ticks, and to provide new molecular parasitological data.

In total, 652 adult ticks were collected from animals (cattle, yak, sheep, horses, and camels) from 13 different regions of 6 provinces: Xinjiang Uygur Autonomous Region, Gansu, Henan, Jilin, Guangdong, and Hunan provinces. Ticks were removed manually from the host body and placed in labeled bottles with water soaked cotton swabs. The ticks were identified to the species level as H. punctata (n = 29), H. longicornis (n = 107), Hy. anatolicum (n = 20), Hyalomma asiaticum (n = 14), Hy. detritum (n = 18), Dermacentor marginatus (n = 85), Dermacentor silvarum (n = 14), Rhipicephalus sanguineus (n = 168), R. (Boophilus) microplus (n = 168), and Ixodes ovatus (n = 29), according to taxonomic keys by Teng and Jiang (1991). Genomic DNA was extracted from ticks using a QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions and stored at −20 C.

To amplify the hyper-variable 4 (V4) region of the 18SrRNA gene of Theileria and Babesia species, primers RLB-F and RLB-R were used (Gubbels et al., 1999). The PCR and cycling conditions were as described in the previous study (Abdallah et al., 2016). The PCR products were purified using a MiniBEST DNA Fragment Purification Kit (TaKaRa®, Liaoning Province, China) and cloned into pGEM®-T Easy Vector Systems (Promega, Madison, Wisconsin), and then transformed into JM109 Escherichia coli cells according to the manufacturers' instructions. The plasmids with inserts were extracted by using MiniBEST Plasmid Purification Kit version 2.0 (TaKaRa®), and the inserts were sequenced. Positive colonies were selected and screened using vector primers (T7/SP6). Recombinant plasmids were extracted from overnight bacterial cultures and the V4 region of 18S rRNA was sequenced using vector primers.

Oligonucleotide probes (catch-all Theileria and Babesia, B. cf. motasi isolates, Babesia sp. Xinjiang, B. major, B. ovata, B. bigemina, B. bovis, T. uilenbergi, T. luwenshuni, T. ovis, T. annulata, T. sinensis, and T. sergenti) containing a N-(trifluoroacetamidohexyl-cyanoethyl, N, N-diisopropyl phosphoramidite [TFA])-C6 amino linker were synthetized (Gubbels et al., 1999; Schnittger et al., 2004; Niu et al., 2009, 2012; Abdallah et al., 2016). Positive plasmids from Babesia or Theileria species genomic DNA, 2 species of bacteria (Anaplasma marginale and Borrelia burgdorferi sensu stricto), and water were used as positive, negative, and blank control, respectively, to perform the specificity and sensitivity of the RLB assay (Abdallah et al., 2016). RLB procedure was performed according to Gubbels et al. (1999).

The 95% confidence intervals (CIs) for the overall prevalence values of each parasite were calculated using IBM SPSS Statistics version 19.0.

The most frequently detected parasite was T. annulata, mostly found in R. sanguineus from cattle in Huidong, reaching 33.9% (57/168). A high infected rate of T. annulata was also found in Hy. asiaticum (30%, 3/10), H. longicornis (26.3%, 21/80), and R. (B.) microplus (19.6%, 33/168) from the camels and cattle in Gansu, Xinjiang, and Hunan province, respectively. In addition, high T. luwenshuni (88.9%, 24/27) and T. sinensis (70.4%, 19/27) infections were observed in H. longicornis from sheep in Henan province. Species-specific positive signals were only obtained with 9 piroplasm species, and the average positive rates were 0.2% (95% CI = 0–0.7) for B. cf. motasi, B. major, and B. bovis; 0.6% (95% CI = 0–2.9) for Babesia sp. Xinjiang and B. bigemina; 3.9% (95% CI = 0.3–17.5) for T. luwenshuni; 18.7% (95% CI = 4.5–25.3) for T. annulata; 0.5% (95% CI = 0–1.9) for T. sergenti, and 2.9% (95% CI = 0–13.2) for T. sinensis (Table I).

Among total of 652 adult ixodid ticks, 157 (24.1%) ticks were infested with at least 1 piroplasm species, most of which corresponded to Theileria infection. No I. ovatus sample was found infected with any parasite. Single infections (20.6%, 134/652) were found in most of the positive ticks, most of which were infected by T. annulata (118/652, 18.1%). Single infections were also observed for other pathogens including Babesia sp. Xinjiang, B. major, B. bigemina, B. bovis, T. luwenshuni, and T. sergenti. Mixed infections (23/652, 3.5%) involved 7 pathogens, some of which could affect the same host (cattle or sheep), or 2 different animal hosts (cattle and sheep). The highest co-infection (18/652, 2.8%) was with T. luwenshuni + T. sinensis infecting H. longicornis. Other double infections were detected in Hy. anatolicum, R. sanguineus, and R. (B.) microplus with low infection rate. Triple infection was observed in 1 H. longicornis infected by B. cf. motasi + T. luwenshuni + T. sinensis (Table II).

The highest parasite infection occurred in H. longicornis (43%) and R. sanguineus (33.9%), while the positive rate of Babesia or Theileria in Hy. anatolicum, R. (B.) microplus, and Hy. asiaticum reached 20% (Table III).

Sixty infected tick samples were randomly screened and sequenced, based on the 18S rRNA of Babesia or Theileria species, or rhoptry-associated-protein-1 (rap-1) gene of sheep Babesia species, or cytochrome b (COB), as well as internal transcribed spacer (ITS) genes of Theileria species of cattle to further confirm our findings. Sequences were then blasted with the published sequences from GenBank. Based on the 18S rRNA and rap-1 genes to detect Babesia sp. Xinjiang, 2 Hy. anatolicum, and 1 H. longicornis samples were sequenced and showed 99% identity with these gene sequences (GenBank accession number: DQ159073 and KF811199). One R. sanguineus infected by B. bigemina was sequenced based on the 18S rRNA gene and showed 100% identity with this gene sequence (GenBank accession number: AY603402). In order to detect T. annulata infection, a total of 50 tick samples, including Hy. anatolicum, Hy. asiaticum, R. (B.) microplus, R. sanguineus, H. longicornis, H. punctata, D. marginatus, and D. silvarum samples, were sequenced based on the 18S rRNA and COB gene, and the similarity ranged from 87% to 100% with these gene sequences (GenBank accession number: KM288519; KP731977; KF732030). One R. (B.) microplus and 1 Hy. detritum DNA infected by T. sergenti were sequenced based on the ITS gene and showed 89% similarity (GenBank accession number: HM538261). One H. longicornis infected by T. sinensis was sequenced and showed 99% similarity with T. sinensis ITS gene from China (GenBank accession number: EF547931). The sequences of 18S rRNA gene from 2 D. marginatus DNA samples that are infected by T. luwenshuni showed 100% similarity with published sequence (GenBank accession number: JF719833). Few samples produced weak positive hybridization signals with the catch-all probe only. One Hy. detritum sample was sequenced based on the 18S rRNA gene and sequence blasted with 18S rRNA sequence of Babesia caballi.

All representative sequences obtained in this study have been deposited in GenBank with the following accession numbers: KY425610 (Theileria sp. ITS gene), KY464052–KY464056 (Theileria sp. COB gene), KY464057 (Theileria sp. 18S rRNA gene), KY464058–KY464062 (Theileria sp. COB gene), KY464045–KY464046 (Babesia sp. 18S rRNA gene), KY464047 (Babesia sp. rap-1b gene), and KY464048–KY464051 (Babesia sp. rap-1a gene).

In our study, the prevalence of 12 piroplasms was investigated in a broad geographic range in the field. The findings revealed that T. annulata was the most widespread species in the tested ticks. In China, T. annulata and T. sergenti are the most virulent bovine Theileria species and mainly distributed in northern China. At present, T. annulata infection was also found from Guangdong province of South China (Liu et al., 2015). In this study, the highest positive rate of T. annulata infection was observed in Guangdong, as well as Hunan provinces (south of China). Theileria annulata and T. sergenti are usually found as co-infections in the field samples (Gubbels et al., 1999). Here, only 1 R. (B.) microplus displayed mixed infections with these 2 species. The prevalence of T. sergenti was very low (0.3%), in contrast to a previous study in China (Liu et al., 2015). This variability might have been caused by the distribution of the pathogen in different samples (tick or host) and sampling sites. Ticks of the genera Hyalomma could transmit T. annulata (Sayin et al., 2003). In our case, except Hy. anatolicum and Hy. asiaticum, T. annulata infection was for the first time detected in H. longicornis, H. punctata, R. (B.) microplus, R. sanguineus, D. silvarum, and D. marginatus (Table IV). Except for a study in France that reported T. annulata infection in D. marginatus (Bonnet et al., 2013), no related studies have reported T. annulata infection in the other 6 tick species. This can be explained by the fact that most of these ticks parasitized cattle and they probably carried T. annulata from an animal infected with a T. annulata infection. We found H. punctata and D. silvarum from sheep; these ticks may have carried T. annulata from a previous blood meal or from the sheep they were collected from, as T. annulata can naturally infect sheep (Zaeemi et al., 2011). In our study, T. annulata infections were found in 3 Hy. asiaticum ticks collected from camel, consistent with studies that reported T. annulata infection in Hyalomma tick parasitizing camels and T. annulata being the most abundant piroplasm in camels (El Kammah et al., 2001; Youssef et al., 2015). Our finding further confirms the evidence of low host specificity of T. annulata. The T. sergenti infection was detected as single infection in vector H. longicornis, consistent with the literature (Liu et al., 2010). In addition, T. sergenti was also detected for the first time in Hy. detritum and R. (B.) microplus. This is not surprising since these ticks were all collected from cattle.

Infection by T. sinensis is widespread among cattle and yaks throughout China, and Lintao, Dingxi, and Weiyuan city from Gansu Province were described as endemic regions of T. sinensis (Yin et al., 2002; Liu et al., 2010). Here, T. sinensis infections were present in Linzhou of Henan province with prevalence (2.9%) comparatively lower than previous reports (Liu et al., 2010). Linzhou is likely a new endemic region for T. sinensis. In this study, T. sinensis was detected in 19 H. longicornis samples collected from sheep. This can be explained if H. longicornis could carry T. sinensis. Since H. longicornis can support diverse pathogenic microorganisms (Chen et al., 2012), this tick species may play an important role as the reservoir for T. sinensis in China.

Among ovine Theileria, only T. luwenshuni (3.9%, 26/652) was detected. A previous study reported a high infection rate of T. luwenshuni in ticks by RLB, or in small ruminants via PCR and FRET-qPCR (Niu et al., 2012; Li et al., 2014a; Yang et al., 2014). In our study, T. luwenshuni was mostly prevalent in Linzhou, Henan province, corresponding with a previous study, which reported T. luwenshuni infecting sheep in the same province (Chen et al., 2014b). We noticed that T. luwenshuni was mostly found parasitizing H. longicornis. Theileria luwenshuni was also detected in D. marginatus collected from cattle, but with a low infection rate. A study documented T. luwenshuni infection in Dermacentor niveus collected from Mongolian gazelle in northern China (Li et al., 2014b), but no finding concerning T. luwenshuni infecting D. marginatus. Ticks were collected from animals, and thus piroplasms infection could either come from previous meals or from the animal they were feeding on when collected. Thus, the potential transmitted role of this tick for T. luwenshuni has to be further studied, and blood from these animals should be tested for piroplasm infection.

The prevalence of B. bigemina and B. bovis reported was 0.6% and 0.2%, respectively, but high prevalence of B. bigemina and B. bovis in cattle was recently reported using PCR (Liu et al., 2014; Niu et al., 2015). Several studies showed that B. bigemina and B. bovis could be transmitted by the genus R. (B.) microplus, Rhipicephalus (Boophilus) decoloratus, Rhipicephalus evertsi evertsi, R. annulatus, R. geigyi, and Rhipicephalus bursa (Ravindran et al., 2006; Tavassoli et al., 2013). Our study showed that 3 and 1 R. (B.) microplus were positive in terms of B. bigemina and B. bovis infections, respectively (Table IV). An earlier study in China reported a higher prevalence of B. bigemina and B. bovis (7.3% and 5.8%) by multiplex PCR assay, and 2.3% and 1.5% by light microscopy, respectively (Liu et al., 2014). Moreover, 1 R. sanguineus displayed a co-infection of T. annulata and B. bigemina, which is a new finding and suggests that the role of R. sanguineus in B. bigemina transmission needs to be experimentally validated. In the past, B. major was isolated by infesting calves with adult H. punctata collected from pasture in Xinjiang Uygur Autonomous Region of China (Liu et al., 2014). In the present study, B. major with low prevalence (0.1%) was detected as a single infection in 1 H. punctata, which is also collected from Xinjiang. Although this H. punctata tick was removed from sheep, B. major was documented to be a bovine Babesia species (Yin et al., 1996).

The prevalence of the Babesia cf. motasi in our study was low (0.4%), consistent with previous data that detected Babesia cf. motasi in field blood samples by RLB assay (Niu et al., 2009). In the present study, only 1 H. longicornis collected from sheep was infected by Babesia cf. motasi (in a co-infection with T. luwenshuni and T. sinensis). It is possible that most of the H. longicornis tested in this study carried an undetectable amount of Babesia cf. motasi. Babesia sp. Xinjiang was originally isolated from a splenectomized sheep infested with R. sanguineus and Hy. anatolicum, but only Hy. anatolicum was experimentally shown to be a vector; R. sanguineus was incapable of serving as a vector for this parasite (Guan et al., 2009). In addition, a Babesia sp. Xinjiang infection was first detected in H. longicornis, removed from cattle. This tick might carry but not transmit Babesia sp. Xinjiang.

Additionally, 1 Hy. detritum DNA extract was found with B. caballi–like infection, for which species-specific probes were not included when developing the RLB method. This result reflects a great advantage of RLB. Moreover, Hy. detritum has never been reported to transmit B. caballi, especially when collected from cattle. This tick could have been collected in an area where cattle and horses shared the same habitat.

The present study shows that the prevalence of piroplasm species was in general lower than previously reported. Veterinary practitioners and stakeholders should be aware of the existence and prevalence of T. annulata in China, given the potential threat this parasite represents to the Chinese livestock industry. Many tick species examined during our investigation were associated with a broader pathogen range than previously reported, and the potential risk of these ticks to transmit those piroplasm species could exist. However, their vector competence needs to be validated by experimental transmission tests using tick and host blood sampled across the country.

This study was supported financially by the NSFC (Nos. 31502054, 31372432, 31201899, 31272556, 31402189, and 31471967); ASTIP; Creative Research Groups of Gansu Province (No. 1210RJIA006); NBCIS CARS-38; Special Fund for Agro-scientific Research in the Public Research (Nos. 201303035 and 201303037), MOA; the 973 Program (2015CB150300), Supporting Program (2013BAD12B03, 2013BAD12B05), MOST, China; and the Jiangsu Co-innovation Center Programme for Prevention and Control of Important Animal Infectious Diseases and Zoonoses, State Key Laboratory of Veterinary Etiological Biology Project. The research was also facilitated by CRP No. 16198/R0 IAEA.