Introduction

Piroplasma species are tick-borne parasitic protozoa which are differentiated into the genera Theileria and Babesia. A number of these parasites are highly pathogenic for cattle, sheep, and goats; and the diseases emerging from these infections are referred to as theileriosis and babesiosis, respectively. The economic losses due to theileriosis and babesiosis are enormous in tropical and subtropical areas (Mehlhorn and Schein 1984; Mehlhorn et al. 1994). Theileria ovis, Theileria lestoquardi (formerly T. hirci), and Theileria separata are recognized as the species that cause ovine theileriosis (Preston 2001), whereas ovine babesiosis is caused by Babesia ovis, Babesia motasi, and Babesia crassa (Uilenberg 2001).

In China, it was first recorded that the ovine babesiosis spreads in the Sichuan and Heilongjiang province, and its pathogen was suspected to be B. ovis due to its relatively high pathogenicity (Chen 1982; Zhao et al. 1986). Later, several ovine Babesia isolates were isolated from different geographical distributions (Yin et al. 1997; Guan et al. 2001). At present, different characters between Babesia sp. Kashi isolate and others were observed, and molecular taxonomy based on 18S rRNA gene had been introduced to make more accurate taxonomic relationship between ovine Babesia isolates (Liu et al. 2007), demonstrating that B. motasi and an unidentified Babesia sp. (Kashi) existed in China. Ovine Theileria was found in the Sichuan province firstly, and was found latterly in the Qinghai, Gansu, Liaoning, Inner Mongolia, Ningxia, and Shaanxi province in China. Based on 18S rRNA gene sequencing of the Chinese isolates and other species of Theileria available in Genbank, it was found that ovine Theileria spp. infective to small ruminants was composed of two species of Theileria, one was closely related to benign Theileria, such as Theileria buffeli, Theileria sergenti, while the other one was more close to Theileria mutans. These two species isolated in China were named Theileria luwenshuni and Theileria uilenbergi respectively. (Yin et al. 2007).

Generally, the detection and diagnosis of sheep piroplasmosis consist of three methods: (1) Traditional methods to detect and identify these parasites rely on light-microscopy examination of thin blood smears. However, this can be difficult in the case of carrier animals where presence of parasites is scant and even in acute cases at the onset of the disease. In addition, species identification based only on morphology is not easy, being particularly difficult if mixed infections occur. (2) Some of serological tests like the complement fixation test (CFT) and the indirect fluorescent antibody test (IFAT) can be useful to detect past infections, but cross-reactions between species have been reported (Bruning 1996; Papadopoulos et al. 1996). Furthermore, false positive and negative results are commonly observed in these tests. (3) Polymerase chain reaction (PCR) is the most commonly used molecular technique to detect piroplasms, and this technique is more sensitive and specific than conventional methods. (Nagore et al. 2004a; Schnittger et al. 2004; Alhassan et al. 2005; Altay et al. 2005), but PCR assays do not generally detect mixed infections, although there are some amplification protocols that can detect mixed piroplasm infections to some extent (Birkenheuer et al. 2003; Criado-Fornelio et al. 2003b).

In order to overcome these limitations, a reverse line blot (RLB) assay was developed to detect virtually all the species of piroplasms infecting our sheep population. RLB was initially developed as a reverse dot blot assay for the diagnosis of sickle cell anemia (Saiki RK et al. 1988), but the essence of both techniques is the hybridization of PCR products to specific probes immobilized on a membrane in order to identify differences in the amplified sequences. In the “line” approach, multiple samples can be analyzed against multiple probes to enable simultaneous detection. This approach was initially developed for the identification of Streptococcus serotypes (Kaufhold et al. 1994). The first application of RLB for the detection and differentiation of pathogens in ticks was developed for Borrelia spirochetes by Rijpkema et al (1995) and followed by an RLB for Mycobacterium tuberculosis strain differentiation (Kamerbeek et al. 1997) and was subsequently combined with Ehrlichia spp (Schouls et al. 1999). Successful application of RLB to detect and differentiate all known Theileria and Babesia species was carried out by Gubbels et al. (1999) and Schnittger et al. (2004).

In this study, we used RLB to detect and differentiate Theileria and Babesia species of importance in small ruminants on the basis of their differences in 18S rRNA gene sequences in China.

Materials and methods

Primers and probes

Species-specific RLB oligonucleotide probes were deduced from the hypervariable V4 region of the 18S rRNA gene sequences (Table 1). These probes were shown to bind only to their respective target sequences. The two Theileria species included in the assay are T. luwenshuni and T. uilenbergi. The two Babesia species included are B. motasi and Babesia sp. (Kashi). A catch-all Theileria and Babesia species control oligonucleotide is also included. All the specific oligonucleotide probes containing a N-(trifluoroacetamidohexyl-cyanoethyl, N,N-diisopropyl phosphoramidite [TFA])-C6 amino linker, were designed and diluted to give 50–1,200 pmol/150 ul in 500 mM NaHCO3 (pH 8.4). The sequence and optimized concentrations of the oligonucleotide probes used are summarized in Table 1. These sequences were aligned using MUTALIN online interface (http://www.ncbi.nlm.nih.gov/). The variable regions of these sequences were flanked by the sequences of the two PCR primers that were used for amplification.

Table 1 Sequence and concentration of specific oligonucleotide probes Theileria sp. and Babesia sp are as following
Full size table

Parasites isolates

All parasite stocks used in this study were described in detail previously (Schnittger et al. 2004; Yin et al 2004; Liu et al 2007). Five Babesia isolates originated from Kashi of Xinjiang region, Chengde of Hebei Province, Tianzhu, Ningxian, and Lintan of Gansu Provine, were used in this research. Among these isolates, Chengde, Tianzhu, and Ningxian isolates were obtained by inoculating field-collected blood of the asymptomatic sheep to the splenectomized sheep. Lintan isolate was isolated from a sheep infested with adult Haemaphysalis qinghaiensis ticks from Lintan, Gansu Province. The Kashi isolate was obtained from a sheep infested with a batch of mixed Rhipicephalus sanguineus and Hyalomma anatolicum anatolicum from Kashi, Xinjiang Region. These isolates were designated as B. motasi (Chengde), B. motasi (Lintan), B. motasi (Ningxian), B. motasi (Tianzhu), and Babesia sp. (Kashi) respectively; T. luwenshuni (Lintan, Madang, Ningxian), T. uilenbergi (Longde, Zhangjiachuan) as defined previously. All parasite isolates were stored as EDTA–blood stabilates in liquid nitrogen.

Field samples collection

Anticoagulated blood samples and sera, all of 117 samples, were collected from sheep and goats in Gansu (Tianzhu, Xiahe), Sichuan (Qianning), Liaoyang (Liaoning), Xinlong (Hebei), and Inner Mongolia (Tongliao) regions. All blood samples were collected in EDTA and stored at −20°C. Sera were detected by enzyme-linked immunosorbent assay (ELISA) and are described in detail (Gao et al. 2002). Anticoagulated blood samples were handled using Gentra Kit following the protocol of the manufacturer and amplified PCR.

Samples, 1,117, were collected from Biandu and Yangyong districts of Lintan County, Kache and Azitan districts of Zhuoni County and Ganjia districts of Xiahe County from July 2004 to June 2005 in Gannan Tibet Autonomous Region, Gansu Province by Guo et al. (2007).

DNA isolation and PCR amplification

DNA was isolated using a genomic DNA Purfication Kit (Gentra, USA) according to the manufacturer’s instructions. The amount of DNA isolated was assessed to 100 ng by photometry. Negative control DNA was isolated from the venous blood of uninfected sheep/goat. Genomic DNA (100 ng) was added to a reaction mixture (final volume of 67 ul). For each sample, the PCR mixer was prepared as follows: H2O 56.0 μl, 10× reaction buffer (200 mM Tris–HCl (pH 8.55), 160 mM (NH4)2SO4 and 20 mM MgCl2) 7.2 μl, 10 mM dNTP 1.6 μl, 10 pM sense primer (RLB-F) 3.6 μl, 10 pM antisense primer (RLB-R [labelled with biotin]) 3.6 μl, Taq polymerase 1.8 U (0.36 μl if the concentration is 5 U/μl). Prepare the PCR mixer according to number of samples and aliquot 66 μl to each tube. Add 1 μl of genomic DNA from filed samples. For blank control, nothing was added. PCR amplification was performed in an automatic DNA thermocycler (Eppendorf). The reaction was incubated at 94°C for 3 min to denature genomic DNA and the thermal cycle reaction program was 1 min at 94°C, 90 s at 55°C, and 90 s at 72°C for 40 cycles with a final extension step at 72°C for 5 min. Samples were held at 4°C until analysis. The PCR products were verified using agarose gel electrophoresis before it was analyzed by RLB hybridization.

Reverse line blot hybridization

Preparation of the membrane

Preparation of RLB membrane and hybridization were carried out as previously described (Gubbels et al. 1999) with the following adaptations: Cut Biodyne C membrane into 14.5 cm × 14.5 cm size without removing protective paper. Mark the membrane with a ball pen for identifying the direction late. Activate the Biodyne C membrane for 10 min by incubation in 10 ml freshly prepared 16% (w/v) 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDAC) in demineralized water, in a rolling bottle at room temperature. Place the membrane in a plastic container and shake with demineralized water for 2 min and place it on a supporting cushion in a clean miniblotter system.

Hybridization

The diluted probes were aliquoted into the miniblotter slots, and an incubation of 1 min for linking of oligonucleotides to the membrane occurred. After aspiration of solutions, the membrane was incubated in 100 mM NaOH for 10 min, washed at 60°C for 5 min and then at 42°C for 5 min in 2× SSPE, 0.1% (SDS). Subsequently, the membrane was placed perpendicular to its previous orientation into the miniblotter. Twenty microliters of PCR products was diluted with 2× SSPE, 0.1% SDS to a final volume of 150 ul, heated to 99°C for 5 min and then cooled on ice. The denatured samples were aliquoted into the slots of the miniblotter for 60 min at 42°C, then aspirated and the membrane washed at 42°C for 10 min in 2× SSPE, 0.1% SDS. Subsequently, the membrane was treated at 42°C for 30 min with peroxidase-labeled streptavidin diluted 1:4,000 in 2× SSPE/0.1% SDS, washed twice at 42°C for 10 min and twice at room temperature for 5 min in 2 × SSPE, 0.1% SDS. Finally chemiluminescence detection was performed according to standard procedures (Amersham). After examination and documentation, membranes were stripped as described by Gubbels et al. (1999) and could be reused about 15 times.

Results

Specificity of RLB of different isolates

Amplification of the V4 region of the 18S rRNA gene was performed by PCR on all isolates. Generated DNA fragments were hybridized to the oligonucleotide-linked membrane followed by chemiluminescence detection resulting in signals of equal intensity for each oligonucleotide. All probes bound to their respective target sequence only; therefore, no cross reaction was observed, resulting in the clear recognition of either individual strains, species, or groups. No signal was observed when ovine genomic DNA and water were used as the control, demonstrating that observed signals are due to the presence of parasite DNA in investigated samples. Each Theileria species is identified by three oligonucleotide probes: the catch-all probe (ca841–859), a probe recognizing Theileria parasites (T-all811–832), and the species-specific probes for either T. luwenshuni (T-l628-647) or T. uilenbergi (T-u678-697). Babesia piroplasms are recognized by the following and there oligonucleotides as well: a catch-all probe (ca841–859), a probe recognizing Babesia parasites (B-all745-766), and the species-specific probes for either B. motasi species (Bm466–487) or Babesia sp. (Kashi).

Sensitivity of RLB for ovine piroplasms

Sensitivity of the hybridization assay was assessed by RLB processing the serially diluted genomic DNA of Babesia and Theileria, accordingly; the presented RLB system is capable of identifying a parasitemia of about 10−6% (B. motasi), 10−3% (Babesia sp. Kashi), 10−8% (T. luwenshuni), 10−8% (T. uilenbergi), while the detection level of a traditional PCR, carried out as a control, was restricted to about 10−3%, 10−2%, 10−5%, 10−3%, respectively. The detection rate of RLB was higher than the PCR, obviously. For all species, just the sensitivity of Babesia sp. (Kashi) is relatively lower when compared with others. The sensitivity of RLB could be considerably enhanced to detect a parasitemia level of at least by amplification of PCR, respectively. The established RLB system was tested for its sensitivity, and observed detection limits were equal to those of a corresponding RLB assay designed for the detection of cattle-infecting piroplasms (Gubbels et al. 1999).

Detection of parasites in field samples collected from sheep in China

The field samples were collected from six regions, Liaoning, Sichuan, Hebei, Inner Mongolia, and Gansu (Tianzhu and Xiahe). The prevalence of each hemoparasite species identified by RLB, PCR, and ELISA is summarized in Table 2. RLB was more sensitive compared to the results obtained after PCR and ELISA examination. The number of positive samples with RLB was at least two times or even higher than those detected by other methods. Most of the samples analyzed were negative by PCR and ELISA. The number of positive samples to T. luwenshuni by RLB of these six regions were 17 (94.1%), 20 (100%), 24 (91.6%), 20 (60%), 20 (40%), 16 (62.5%) (Table 2), respectively. It has been revealed that most of the positive samples were infected by T. luwenshuni and/or T. uilenbergi genotypes, and prevalence of Babesia spp. was low with only three, three, and one sample positive to B. motasi out of the samples from Madang, Tongliao, and Xinglong, respectively, and no samples positive to Babesia sp. (Kashi). Most of the samples were negative to hemoparasites by PCR and/or ELISA (mainly samples from Sichuan). Furthermore, 1,117 samples from five areas in Gannan Tibet Autonomous Region were examined with RLB assay and compared with ELISA assay for corresponding samples. The average detection rate with ELISA was 71.3% for the whole samples, while mean detection rate with RLB was 80.3% (data not shown). It was found that the positive rate of the samples from Biandu area by RLB was much higher than that by ELISA test (Table 2). Only two samples, one is from Kache and the other one is from Biandu, were found to be positive to B. motasi. No sample was positive to Babesia sp. (Kashi).

Table 2 Comparison of the results of RLB, PCR, and ELISA
Full size table

Discussion

Babesiosis was first reported in small ruminants in China in 1982, and the pathogen was identified as B. ovis (Chen 1982). In 1985, an outbreak of babesiosis in sheep occurred in Heilongjiang, and the causative agent was again identified as B. ovis (Zhao et al. 1986). Although more than two decades passed, the data on the distribution and epidemiology of ovine babesiosis is very limited. Theileriosis of sheep and goats was first reported in Sichuan province, 50 years ago (Yang 1958), and an ELISA and PCR were developed for detection of the diseases (Gao et al. 2002; Yin et al. 2008). However, it has been demonstrated that T. luwenshuni, T. uilenbergi, and B. motasi are sharing the same vector as H. qinghaiensis, and in most endemic areas (Yin et al. 2007), the sheep and goats were infected by two or three of the species of piroplasms mentioned above. It would be very convenient and useful to develop a method being able to detect simultaneously all these three pathogens and other piroplasms infective to small ruminants for diagnosis, detection, quarantine, and epidemiological survey.

The RLB hybridization assay is a versatile diagnostic tool to sensitively and simultaneously detect and differentiate hemoparasites in blood, tissue, or ticks. RLB is based on simultaneous PCR amplification of related species. Each species can be identified by a species-specific oligonucleotide probe using a line-blotter apparatus. It is quickly becoming a standard molecular tool for diagnostic and epidemiological studies in an increasing number of laboratories all over the world. Centeno-Lima et al. (2003) used RLB for the characterization of Babesia divergens in a human case, whereas Nijhof et al. (2003) and Altay et al. (2007) discovered novel Theileria and Babesia species through the application of RLB. Furthermore, detection and differentiation by using RLB for many Babesia spp. and Theileria spp. occurring in small ruminants have been reported by Schnittger et al. (2004).

In this study, an RLB assay was developed based on the V4 region of the 18S rRNA gene of Theileria and Babesia species infecting sheep in China. All species included were detected by single species-specific oligonucleotides, and no cross-reactions were observed. The assay allows a remarkable higher specificity than other methods through comparison. The generic oligonucleotide probe (catch-all) and group-specific probes designed specifically to detect the 18S rRNA genes of the genera Theileria and Babesia were found to hybridize in all the cases where one or more species/genotypes were present, and the catch-all probe would indicate that a new species or genotype is present when the PCR products were hybridized with probes and no species-specific signal. Similarly, the Theileria genus-specific probe designed specifically detected any Theileria species present (T-all), and the Babesia genus-specific probe designed specifically detected any Babesia species (B-all) in this study. Meanwhile, four species-specific probe also designed used to detect distinct species (B. motasi; Babesia sp. Kashi; T. luwenshuni; T. uilenbergi). This assay can be used to identify the different ovine-infecting piroplasms species at present.

As there was very limited data on the distribution and epidemiological situation of Babesia spp. and Theileria spp. infection in China, sera and bloods samples were collected from different areas in China to evaluate the sensitivity and specificity of RLB and to assess the epidemiological situation of these parasites’ infection. The RLB technique revealed that 88 out of 117 samples were positive for the presence of ovine Theileria parasites, and the sensitivity was higher than PCR and ELISA obviously, and further confirmed the sensitivity of the RLB. On the other hand, as the difference of results of the samples from Liaoning region by RLB was obviously with other two assays, in order to certify whether it was caused by infection of parasites or the false positive, the sequence analysis was conducted for amplicons (PCR product) from these samples by random sampling. It was found that the fragment was about 500 bp, and the homology was 99.6% with the hypervariable region 4 (V4 region) of the 18S ribosomal DNA gene of T. luwenshuni by DNAstar analysis, suggesting that the sheep were infected by T. luwenshuni, and the RLB was a reliable diagnostic tool. The RLB has been used as a standard molecular tool for discrimination of piroplasm from nonpathogenic piroplasm species and the simultaneous detection and identification of Theileria and Babesia species when they occurred in the same animal (Nagore et al. 2004b; Georges et al. 2001; Brígido et al. 2004; García-Sanmartín et al. 2006).

From the results, it could be seen that Theileria sp. is widely distributed in China, and T. luwenshuni was the dominant species, which agreed with the previous data by PCR (Yin et al. 2008). Only nine out of 1,234 (1,117 + 117) samples were positive for B. motasi, suggesting that the prevalence of B. motasi was much less than T. luwenshuni or T. uilenbergi. However, the distribution of B. motasi is relatively large as the positive samples were from three provinces. In this study, no positive samples were detected for Babesia sp. Kashi. The possible reason is that all samples are from areas where the vector of Babesia sp. Kashi, Hyalomma anatolicum anatolicum, is free. As H. anatolicum anatolicum is only distributed in Xinjiang, it would be interesting to collect some samples from these areas and evaluate the situation of Babesia sp. Kashi infection in sheep and goats in the further study.

In conclusion, RLB should be considered as a piroplasm infection diagnostic tool not only for animals showing clinical signs and suffering acute infections, but for those apparently healthy animals with potential subclinical infections and seems the optimal approach for the detection and discrimination of these important parasites. We recommend its use for integrated epidemiological monitoring of tick-borne disease, since RLB can also be used for screening ticks and can easily be expanded to include additional haemoparasite species.