Introduction

Bovine theileriosis is a tick-borne haemoprotozoan disease caused by the parasites of the genus Theileria, which infect bovine leukocytes and erythrocytes (Liu et al. 2009). The main symptoms of the disease are fever, anaemia, jaundice and superficial lymph node enlargement (Xu et al. 1997). The disease has a worldwide distribution. It is known that Theileria parva, Theileria annulata, Theileria mutans, Theileria taurotragi, Theileria velifera and Theileria orientalis can cause bovine infection (Onuma et al. 1998; Sugimoto et al. 1991). T. annulata, a malignant Theileria species causing tropical theileriosis, can be transmitted by ticks of the genus Hyalomma, such as Hyalomma detritum and Hyalomma anatolicum anatolicum (Dolan 1989), which are mainly distributed in North Africa, Southern Europe, India, the Middle East and Asia (Bilgic et al. 2010). This species mainly affects crossbred cattle, and infection is characterized by lymphadenopathy, splenomegaly, fever, anaemia, weakness, loss of body weight and sometimes death, which causes large economic losses (Omer et al. 2002; El-Deeb and Younis 2009). T. orientalis belongs to the group formerly known as the Theileria sergenti/Theileria buffeli/T. orientalis complex and is widely distributed throughout Japan, the Korean Peninsula, New Zealand and Australia (Wang et al. 2010a; Liu et al. 2010; Kamau et al. 2011a; Eamens et al. 2013; Pulford et al. 2016b). Animals infected by T. orientalis mainly exhibit apathy, fever, anaemia and lymphadenectasis (Jin et al. 2007). In China, T. annulata is distributed in the North Yangtze River region, where ticks of the Hyalomma genus are abundant. T. orientalis is the most prevalent bovine Theileria species, as its vector tick Haemaphysalis longicornis is the most widely distributed tick species in China. In addition, Bai et al. (2002) reported a new Theileria species infective to cattle and yaks, T. sinensis, that is transmitted by Haemaphysalis qinghaiensis and is thought to originate from Gansu Province. Usually, T. sinensis infection does not result in death, and due to limited geographical distribution, infections do not contribute to large economic losses. Thus, it is generally thought that the above-mentioned three Theileria species are responsible for bovine theileriosis in China.

Compared to traditional pathogenic identification through observations of blood smears, serological and molecular diagnostic methods are more sensitive and convenient for the detection of bovine Theileria infection. 18S rRNA, ITS, major piroplasma surface protein (MPSP) and T. annulata merozoite surface protein 1 (Tams1) genes were selected as target genes to develop a species-specific PCR assay for different Theileria species (Tanaka et al. 1993; d’Oliveira et al. 1995; Han et al. 2009; Kamau et al. 2011b). Recently, real-time PCR was utilized in the quantitative detection of Theileria infection (Ros-Garcia et al. 2012; Chaisi et al. 2013; Bogema et al. 2015; Pulford et al. 2016a). Additionally, reverse line blotting (RLB) was also used for the simultaneous detection of different Theileria infections (Salih et al. 2007; Iqbal et al. 2013). However, these detection techniques are time-consuming and require complex procedures and equipment, such as thermal cyclers, which resulted in the development of loop-mediated isothermal amplification (LAMP) for detecting Theileria infection (Wang et al. 2010a; Liu et al. 2013). These molecular diagnostic methods can detect and differentiate nearly all known Theileria species with a high sensitivity. These types of serodiagnostic techniques are essential tools for performing epidemiological studies of theileriosis and evaluating preventive measures. Compared to other serodiagnostic methods, such as complement fixation tests (CFT) and indirect fluorescent antibody tests (IFAT), enzyme-linked immunosorbent assays (ELISA) are more sensitive, non-subjective, easier to standardize, high throughput and suitable for large-scale epidemiological surveys. Thus, a number of ELISAs have been designed for the diagnosis of Theileria spp. infection (Schnittger et al. 2002; Salih et al. 2005; Wang et al. 2010b; Mohamed et al. 2012; Rajendran and Ray, 2014). Few studies, however, have focused on the development of simple, rapid, stable and routine serological techniques to simultaneously detect T. annulata, T. orientalis and T. sinensis to evaluate the prevalence of bovine theileriosis.

The MPSP homologues are a group of conserved glycoproteins on the surface of merozoites and piroplasms of Theileria species ranging from 30 to 34 kDa. They have excellent immunogenicity and can stimulate a strong immune response in infected animals, which suggests that these proteins are biologically significant in parasite development (Shiels et al. 1995; Onuma et al. 1997). These proteins are, thus, candidate antigens for vaccine and serodiagnostic development. Thus, in this study, rMPSPs of T. annulata, T. orientalis and T. sinensis were evaluated as candidate antigens for ELISA. A universal ELISA method that detects the three bovine Theileria infections in one reaction was developed based on the rMPSP protein of T. sinensis, and this assay was used in an epidemiological survey of bovine theileriosis in China.

Material and methods

Parasites

Bovine blood samples infected by T. annulata (Liaoyang strain from Liaoning Province), T. orientalis (Lintao strain from Gansu Province) and T. sinensis (Dingxi strain from Gansu Province) were cryopreserved in liquid nitrogen at the Lanzhou Veterinary Research Institute (LVRI), China.

Sera and genomic DNAs

Reference positive sera and genomic DNA of bovine piroplasms (T. annulata, T. orientalis, T. sinensis, Babesia bovis, Babesia bigemina and Babesia major), ovine piroplasms (Babesia motasi, Theileria luwenshuni and Theileria uilenbergi) and A. ovis were provided by the Vectors and Vector-borne Diseases (VVBD) Laboratory of LVRI.

Animal experimentation was approved (permit SYXK2010-0001) by the Science and Technology Department of Gansu Province, China. In this experiment, three calves were purchased from Gansu Province in China (Nos. 08840, 322 and 353) and were negative for Theileria species by microscopic examination of Giemsa-stained smears and PCR using primer 989/990 (Allsopp et al. 1993). The three animals were then infected with either 10 ml bovine blood containing T. annulata (Liaoyang strain), T. orientalis (Lintao strain) or T. sinensis (Dingxi strain). Sera, DNA extracts from blood and blood smears were made from the three infected calves during 19 weeks post-infection (wpi) for evaluation of the infections and antibody kinetics. The presence of the parasite was evaluated using microscopic examination and PCR, in which species-specific primer sets AnCb, SerITS and Tsin F/R were used to detect T. annulata, T. orientalis and T. sinensis, respectively (Liu et al. 2010, 2015).

Sera (n = 47) were from 47 cattle purchased from the Gansu Province of China and tested negative for T. annulata, T. orientalis and T. sinensis by microscopic examination of Giemsa-stained smears and PCR using primer 989/990 prior to the experiment. These samples were collected to evaluate the specificity of ELISA and mixed for use as a standard negative serum. Reference positive sera (n = 22) against bovine Theileria spp. provided by VVBD and sera (n = 45) collected from T. annulata and T. sinensis-infected calves during 4–19 wpi and that of T. orientalis during 7–19 wpi were used to evaluate the sensitivity of ELISA, and a mixture of these sera was used as standard positive serum.

Field-collected bovine sera were collected from 2473 cattle and yaks in 42 prefectures of 17 provinces from 2008 to 2013. The provinces sampled included Guangdong (Qingyuan, Maoming), Jinlin (Jilin, Changchun), Sichuan (Panzhihua, Luzhou), Liaoning (Anshan, Dalian and Benxi), Hunan (Huaihua, Yongzhou and Changde), Yunnan (Jinghong, Menghai, Ruili, Qujing, Wenshan and Honghe), Qinghai (Haibei, Xining), Shanxi (Yan’an), Gansu (Linxia, Zhangye and Maqu), Inner Mongolia (Hulunbeier, Baotou, Tongliao, Xing’anmeng and Xilinguolemeng), Tibet (Ihasa), Hebei (Tangshan), Guizhou (Kaili, Tongren, Anshun and Xingyi), Henan (Sanmenxia), Chongqing (Jiangjin, Wanzhou), Xinjiang (Zhaosu) and Guangxi (Chongzuo, Baise and Guilin).

Cloning and expressing MPSP gene

The specific primers were designed to amplify MPSP CDS based on the sequences of T. annulata, T. orientalis and T. sinensis (accession nos. JX683683.1, GQ180194.1 and GQ180193.1) in GenBank (Table 1). Subsequently, the PCR products were purified, digested with restriction enzymes EcoR I and Xho I and finally inserted into a pET-30a vector. The recombinant pET-MPSP vectors were confirmed by double restriction enzyme digestion and sequencing. The recombinant vectors were then transformed into chemically competent E. coli BL21(DE3) to express the rMPSPs as His-fusion proteins. Three hundred microliters of recombinant E. coli was inoculated into 30 ml fresh LB medium with 50 μg/ml of kanamycin and cultivated at 37 °C, 200 rpm. When the OD600 of the bacterium culture reached 0.6, IPTG was added at a final concentration of 1.0 mmol/L to induce protein expression at 37 °C for 8 h. Finally, the rMPSPs were identified and analysed by SDS-PAGE.

Table 1 The information of primers for amplifying MPSP genes
Full size table

The rMPSPs were purified according to the instructions of the Ni-NTA Purification System (Catalogue No. K950-01, Invitrogen, Carlsbad City, CA, USA). Briefly, the E. coli were pelleted and resuspended in native binding buffer (denaturing buffer for protein supernatant) and then lysed by freeze thawing and sonication. Following this, the supernatant was collected. After denaturation by guanidinium lysis buffer, the supernatant is soluble. The soluble proteins from supernatant were combined with the Ni-NTA resin for 1–2 h under gentle rotation. Following five washes with native or denaturing wash buffer, the eluate was collected into separate tubes for identification by SDS-PAGE.

Western blotting

Western blotting was performed using the method described by Guan et al. (2012). The three rMPSPs were separated by 12 % SDS-PAGE and electro-blotted onto nitrocellulose (NC) membranes (RPN303E, Amersham). The NC sheet was cut into strips and blocked with 10 % skimmed milk powder in 0.1 M Tris-buffered saline (pH 7.6) with 0.1 % Tween 20 (TBST), overnight at 4 °C. The NC strips were probed with each reference serum and mouse monoclonal anti-poly histidine antibody (H1029, Sigma). Following three washes with TBST, secondary antibodies, monoclonal alkaline phosphatase-conjugated goat anti-mouse IgG (A3562, Sigma), alkaline phosphatase-conjugated mouse anti-goat/sheep (A8062, Sigma) and alkaline phosphatase-conjugated rabbit anti-bovine IgG (A0705, Sigma) were applied. Following the four washes with TBST, positive signals were revealed using the 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium (BCIP/NBT) liquid substrate system (B1911, Sigma).

ELISA procedure

An ELISA procedure described previously by Chauvin et al. (1995) and Guan et al. (2010) was used in the present study. Briefly, 96-well microtiter plates (Nunc, Roskilde, Denmark) were coated with 50 μl recombinant antigen at a concentration of 2.8 μg/ml in a coating buffer (0.1 M carbonate–bicarbonate buffer, pH 9.6). The plates were washed three times with PBS containing 0.1 % Tween 20 (PBST) and incubated with 100 μl/well of a blocking solution (1 % BSA in PBST) for 1 h at 37 °C. After drying the plate, the samples, blanks (PBST) and standard positive and negative controls (dilution of 1:200) were distributed in duplicate and the plates were incubated at 37 °C for 1 h. After washing as described above, a peroxidase conjugate of rabbit anti-bovine IgG (A-5295, Sigma) diluted at 1:40,000 with PBST was added to each well and the plates were again incubated at 37 °C for 1 h. The plates were washed three times as described above, and subsequently, 50 μl of TMB (T0440-1L, Sigma, USA) was added to each well and incubated at room temperature for 20 min. The reaction was stopped by the addition of 50 μl of 0.2 M H2SO4 and the optical density (OD) was measured with an ELISA reader (microplate reader Model 680, Bio-Rad, USA) at a wavelength of 450 nm. The results are expressed as the percentage of the specific mean antibody rate (AbR%), determined using the formula, AbR% = (Sample mean OD − Negative control mean OD) / (Positive control mean OD − Negative control mean OD) × 100 %. Cutoff points, specificity and sensitivity of the ELISA were evaluated by receiver operating characteristic (ROC) analysis using MedCalc statistical software with 47 negative sera and 67 positive sera against T. annulata, T. orientalis and T. sinensis from cattle.

Results

Cloning MPSP gene, expression and purification of recombinant proteins

The complete MPSP CDSs of T. annulata, T. orientalis and T. sinensis were amplified successfully by PCR from genomic DNA, at the sizes of 846, 852 and 873 bp, encoding 281 (32 kDa), 283 (32 kDa) and 290 (33 kDa) amino acid residues predicted by an online software analysis, respectively. The SDS-PAGE results showed that the rMPSPs were 37 kDa (T. annulata), 37 kDa (T. orientalis) and 38 kDa (T. sinensis) in size including His-tag protein (approximately 5 kDa in size) (Fig. 1). T. orientalis and T. sinensis rMPSPs existed both in the supernatants (soluble protein) and precipitates (inclusion bodies) but T. annulata rMPSP only as inclusion body. The rMPSPs were purified according to the instruction of Ni-NTA Purification System. The concentrations of purified rMPSPs of T. orientalis and T. sinensis from supernatants were 0.2 and 0.28 mg/ml and that of T. annulata purified from inclusion bodies was 0.1 mg/ml.

Fig. 1
figure 1

SDS-PAGE analysis of purified rMPSPs. M protein molecular weight marker, 1 rMPSP of T. orientalis, 2 rMPSP of T. sinensis, 3 rMPSP of T. annulata

Full size image

Reactivity of recombinant proteins

The cross-reactivity between the three rMPSPs and positive reference sera against T. annulata, T. orientalis, T. sinensis, T. luwenshuni, T. uilenbergi, B. bovis, B. bigemina, B. major, B. motasi and A. ovis, were evaluated by western blotting. Each rMPSP had strong reactions with the positive sera against T. annulata, T. orientalis and T. sinensis but no cross-reactions with other piroplasm and A. ovis positive sera (Fig. 2). Meanwhile, all the three rMPSPs had a strong reaction with anti-polyhistidine antibody demonstrated by western blot, which indicates the exact expression of the three rMPSPs in the prokaryotic expression system. These results show that all the three rMPSPs can be used to develop a universal serodiagnostic method for evaluating infection with the three bovine Theileria species. However, this method does not differentiate between the three species.

Fig. 2
figure 2

Western blotting analysis of the three rMPSPs with sera against piroplasms and Anaplasma. Lane M the molecular weight markers; lane 1 anti-His-tag protein; 2 Negative sera; 3–5 Positive sera against T. annulata, T. orientalis, T. sinensis; 6-12 Positive sera against B. bovis, B. bigemina, B. major, ovine babesia, T. luwenshuni, T. uilenbergi, A. ovis

Full size image

Evaluating the cross-reactivity, specificity, sensitivity and threshold value of ELISA

Given the concentration of the purified rMPSPs, the rMPSP of T. sinensis was used as an antigen to develop a universal ELISA for detecting the infection of the three bovine Theileria species. The mean AbRs and standard diversity, calculated with Excel 2007 for ELISA results, showed the rMPSP of T. sinensis had strong reactions with the positive sera against T. annulata, T. orientalis and T. sinensis but no cross-reactions with other piroplasm and A. ovis positive sera (Fig. 3), which was consistent with the results from the western blot. Following the above procedure, 47 negative sera and 67 positive sera against T. annulata, T. orientalis and T. sinensis were used to evaluate the specificity, sensitivity and threshold value of the ELISA. The AbRs were catalogued in MedCalc statistical software. Results demonstrate the specificity of the assay is 95.7 % (0.855∼0.995 under 95 % CI) and the sensitivity is 95.5 % (0.875∼0.991 under 95 % CI) when the threshold value is 28.8 % (Fig. 4).

Fig. 3
figure 3

Cross-reactions of the three rMPSPs with sera against piroplasms and Anaplasma in ELISA

Full size image
Fig. 4
figure 4

Evaluation of the positive threshold value using mean antibody rates (AbR) of negative and positive sera. 0 = negative sera; 1 = positive sera

Full size image

Kinetics of anti-Theileria spp. antibodies in experimentally infected cattle

Three calves (nos. 08840, 322 and 353) were infected with bovine blood of either T. annulata (Liaoyang strain), T. orientalis (Lintao strain) or T. sinensis (Dingxi strain), and the sera during 1–19 wpi were collected and used to assess antibody kinetics. The antibodies of T. annulata and T. sinensis were detected at 3 wpi by ELISA, and the high level of antibodies lasted until 19th wpi. The antibodies against T. orientalis were positively detected between 6 and 19 wpi except in weeks 7, 9, 10 and 11 and the antibody levels were low and irregular (Fig. 5). Based on PCR tests, the three Theileria species were detected at 3 wpi, but the PCR test of T. annulata was negative from 14 to 19 wpi. In comparing blood smears to serological and PCR tests, blood smears showed T. annulata and T. sinensis were first observed at 4 wpi, T. annulata was not observed from 10 to 19 wpi and. T. orientalis was first observed at 7 wpi (Table 2).

Fig. 5
figure 5

Kinetics of anti-T. annulata, T. orientalis and T. sinensis antibodies in three experimentally infected calves

Full size image
Table 2 Dynamic changes of the three Theileria species from 1 to 19 weeks after infection based on microscopy of blood smears, PCR and ELISA test
Full size table

Serological epidemiology

In this study, a large-scale investigation of bovine Theileria infection in China was conducted using ELISA. Field-collected bovine sera (2473) from 42 prefectures of 17 provinces were evaluated. The results show 12.9 to 67.5 % of sera tested positive for Theileria infection with an average positive rate of 43.6 % in these 17 provinces (Table 3).

Table 3 Results of three bovine Theileria species antibodies detection using the ELISA in field-collected sera from 17 provinces in China
Full size table

Discussion

Bovine theileriosis is a widespread haemoprotozoan disease that causes great economic losses and can seriously hinder the development of cattle. Yearly, approximately 250 million animals are exposed to the risk of T. annulata infection (Radositis et al. 1997; Sudan et al. 2014, 2015), which results in roughly $800 million in losses worldwide. Recently, many diagnostic methods have been developed for detecting the infection of bovine Theileria. A multiplex PCR assay was used to detect and discriminate between T. annulata and T. orientalis (Liu et al. 2015). Since LAMP is a technique that has the advantage of high sensitivity and easy manipulation, Liu et al. (2013) developed a LAMP assay to identify the infection of T. orientalis and T. sinensis. Salih et al. (2007) developed an RLB assay to detect T. parva, T. mutans, T. velifera, T. taurotragi, T. buffeli and T. annulata infections in cattle, and Iqbal et al. (2013) applied an RLB assay to detect Theileria and Babesia sp. infections in goat and sheep. All of these methods can be used to detect two or more Theileria species simultaneously, but they are time consuming and inconvenient for large-scale epidemiological surveys due to the need for nucleic acid extraction compared to serodiagnostic methods. Examples of serodiagnostic methods include the assay developed by Gubbels et al. (2000), which used an indirect Tams1 ELISA for diagnosis of T. annulata infection in cattle, demonstrated that the antigen had no cross-reactivity with antisera positive for T. velifera, T. taurotragi, T. orientalis, B. bovis, and B. bigemina. Based on another membrane protein TaSP, Renneker et al. (2008) developed a competitive ELISA for specific detection of T. annulata infection. In 2010, Wang et al. (2010b) established an indirect ELISA to detect T. orientalis in water buffalo using recombinant P33, and cross-reactivity of the iELISA indicated there were no cross-reactions with the positive sera against B. orientalis, Schistosomiasis japonica, A naplasma marginale and Toxoplasma gondii in water buffalo. However, there is no serodiagnostic method to detect T. sinensis infection in cattle at present. Thus, to evaluate the infection status of bovine theileriosis in China, it is necessary to develop a serodiagnostic method for detecting T. annulata, T. orientalis and T. sinensis infection simultaneously.

In this study, we expressed the rMPSPs of T. annulata, T. orientalis and T. sinensis in a prokaryotic expression system. The three rMPSPs had strong reactivity with the positive sera against the three Theileria species but no cross-reactivity with other piroplasms and A. ovis as evaluated by western blotting. In addition, the rMPSP of T. sinensis had a high concentration of soluble protein in the supernatant. Thus, a universal ELISA method was developed using the rMPSP of T. sinensis for detecting the infections of the three bovine Theileria species simultaneously. The calculated specificity and sensitivity of the ELISA were 95.7 and 95.5 %, respectively, with a threshold value of 28.8 %.

Piroplasm-free calves, nos. 08840, 322 and 353, were infected experimentally with T. annulata, T. orientalis and T. sinensis to evaluate seroconversion using the ELISA. The results showed positive ELISA and PCR tests for T. annulata from the third wpi. Both the ELISA and PCR assays were more sensitive than microscopic examination of blood smears that first observed the parasite in the fourth wpi. However, the PCR tests were negative from the 14th wpi, but ELISA tests remained positive until 19 wpi.

It is well known that the piroplasm stage of T. annulata is the only stage of infection in recovered carrier animals that can be detected, and antibody titres of infected animals are typically higher in the later stages of infection (Renneker et al. 2009; Mohamed et al. 2012). Additionally, Seitzer et al. (2007) tested sera collected from experimentally infected cattle using a TaSP ELISA, which detected the parasite from 3 to 20 wpi. The PCR and ELISA results in this study are consistent with the results reported by Seitzer et al. (2007) which demonstrates that the developed ELISA method is reliable and useful for detecting animals with prior exposure to the parasite. It is also suitable for diagnosis in the later phase of T. annulata infection in contrast with PCR testing. Jenkins and Bogema (2016) utilized qPCR to test for T. orientalis, demonstrating detection from 3 wpi to the end of the experiment, which is consistent with our results. Meanwhile, Hammer et al. (2016) also obtained similar qPCR results in a high-dose group. In contrast, the antibody titre reached maximum in 6 wpi but then dropped sharply from 6 to 11 wpi. The serological response was stead from 12 wpi to the end of the experiment. However, the titres of four sera samples from 7, 9 10 and 11 wpi were under the cutoff value for positivity. In general, the kinetics of anti-T. orientalis antibodies in this study was similar to the results reported by Jenkins and Bogema (2016). However, the developed ELISA could result in underestimation of the prevalence of T. orientalis, because the antibody of some cases was negative but the animal still carried T. orientalis. According to the seasonal distribution of ticks in China, the ticks become active from March, so testing the field-collected bovine sera after June could reflect the prevalence of T. orientalis better. In a word, PCR and ELISA are more sensitive than microscopic examination, which first observed the parasite at 7 wpi. For T. sinensis, the data showed that ELISA results correlated well with PCR and microscopic examination, suggesting that ELISA may be useful for detecting active infection by T. sinensis. The results showed ELISA and PCR methods are more sensitive in detecting the three Theileria species infection in cattle or yak, compared to microscopic examination. However, during the later phase of the infection when parasitemia is low, it is difficult to detect the infection with microscopic examination and PCR. At this stage, the animals still have high antibody titres; therefore, the ELISA is a more suitable detection method for this condition.

In conclusion, the developed ELISA is a good choice to detect Theileria spp. infections, and it will be useful for large-scale sero-epidemiological investigations of bovine theileriosis.

To date, there have been no reports regarding large-scale epidemiological investigations of bovine theileriosis in China. Most of the surveys focus on one or several provinces. Jian et al. (2011) studied the prevalence of T. annulata in the Xinjiang regions using a recombinant Tams1 ELISA, demonstrating a positive rate of approximately 11 %. Hu et al. (2012) investigated T. orientalis infection using a PCR method in 206 cattle from Yanji, Jilin Province, and showed a positive rate of 62.6 %. Yang et al. (2014) detected the Theileria infection in 1235 cattle from 10 provinces using a genus-specific pan-Theileria FRET-qPCR. The results demonstrated a positive rate of 30.8 %. In the present study, we performed a large-scale sero-epidemiological survey of bovine theileriosis using a universal ELISA for three Theileria species. In total, 2473 field-collected bovine sera from 42 prefectures of 17 provinces were examined using the ELISA. Infection was found in all provinces investigated. The positive rates ranged from 12.9 to 67.5 % and the average positive rate was 43.6 % in these provinces, which indicates that bovine Theileria spp. are common in China. These data provide important information about bovine Theileria infection that is beneficial for management programmes of bovine theileriosis.