Introduction

Babesia gibsoni, an intra-erythrocytic protozoon, is the most commonly reported pathogenic species responsible for canine babesiosis (Wozniak et al. 1997). Clinical symptoms of B. gibsoni include fever, hemolytic anemia, jaundice, hemoglobinuria, and weakness, and severe cases may cause death (Aboge et al. 2007b; Birkenheuer et al. 2003). B. gibsoni is widely distributed throughout the world and has been reported in some areas of China, such as Shanghai, Jiangxi, and Wuhan (Cao et al. 2015; He et al. 2017; Zheng et al. 2017).

Microscopic examination of Giemsa-stained thin blood smear is a simple method widely used in clinical diagnosis (Aboge et al. 2007a). However, microscopy is low in sensitivity and difficult to differentiate species with low parasitemia and mix infection. The diagnosis is usually inaccurate due to the influence of the experience of the observers in analysis of blood smears (Ikadai et al. 2004). Serological surveys by testing the specific antibodies of B. gibsoni infection with Dot-ELISA, sandwich ELISA, indirect ELISA, and indirect immunofluorescent antibody test (IFAT) have shown higher sensitivity and specificity as compared to the standard microscopic examination. However, they may cross-react with other apicomplexan parasites (Bose et al. 1995; Mandal et al. 2014; Muhlnickel et al. 2002). The polymerase chain reaction (PCR) assay as a variable temperature amplification technology is of extraordinary sensitivity (Birkenheuer et al. 2003; Yao et al. 2014), and it is commonly used for laboratory examination. However, the long reaction time and high cost of equipment limited its application for clinical diagnosis. Collectively, these methods are not suitable for clinical and rapid detection, because of the tedious process and the rigid requirement of professionals and equipment.

Recombinase polymerase amplification (RPA), a novel isothermal nucleic acid amplification technology developed by TwistDX Inc., depends on the specific combination of enzyme and protein (Piepenburg et al. 2006a), which reacts at isothermal temperatures (37–42 °C) within 5–30 min. The products could be detected by 2% gel electrophoresis or real-time fluorescence (Wang et al. 2017). Due to the advantages of visual detection, high sensitivity, simplicity in detection and operation, the RPA method combined with the lateral flow dipstick (LF-RPA) has been widely used for detection of different pathogens, such as bacteria, viruses, and parasites (Ahmed et al. 2014; Boyle et al. 2014; Crannell et al. 2015; Ge et al. 2018; Jaroenram and Owens 2014; Nair et al. 2015).

In this study, we developed an RPA assay combined with LF dipsticks (LF-RPA) for detecting B. gibsoni infection in canine. The LF-RPA system showed a high specificity and no cross-reaction with closely related parasites. The sensitivity examination indicated that it could detect as low as 0.5 parasite/μl infected blood in 25 min at 35 °C. Clinical sample test verified that the LF-RPA assay is reliable and effective in diagnosing B. gibsoni affection.

Materials and methods

Sample collection and DNA extraction

Fifteen blood samples were collected from potential B. gibsoni-infected dogs in veterinary clinics with the permission of the dogs’ owners, Wuhan, China. Genomic DNA was extracted from 200 μl anticoagulant blood using TIANamp Genomic DNA Kit (Tiangen, Beijing, CN) according to the manufacturer’s instructions. DNA samples were stored at − 20 °C until use.

Positive and negative control DNAs were originally obtained from experimental B. gibsoni-infected dogs and healthy dogs, respectively. DNAs were stored in our lab for further use. Seven DNA samples of related apicomplexan parasites, Babesia orientalis, Babesia bigemina, Babesia duncani, Theileria sergenti, Toxoplasma gondii, Plasmodium falciparum, and Haemaphysalis longicornis, were obtained in the protozoan laboratory of Huazhong Agricultural University.

Design of primers and probe

The cytochrome c oxidase subunit I (cox I) of B. gibsoni was chosen as the target gene. According to the manuals of the TwistAmp® nfo kits, the probe and primers for the recombinase polymerase amplification assay were designed in a conserved region of cox I (GenBank no. AB499087). After screening the candidate primers, a pair of optimal primers was identified for RPA assay. To enable detection by lateral flow dipsticks, both the probe and primers were chemically modified according to the TwistDX guidelines. A normal forward primer (5′-ATAGTTTATTGCTTCAGCCAATAGCTTTCTGTTTGG-3′) and the reverse primer contained a labeled biotin at the 5′end (5′-biotin-TATCTACAGTTTGACCAATTGATTTTAAAGCGCC-3′). The probe was designed by adding a 5′-fluorescein FAM, an internal tetrahydrofuran residue (THF), and a C3 spacer (SpC3) on the 3′ end (5′-FAM-ATAATATTTGGTTTACTTGCCTCAGGTATA-THE-GCTAGTGCTATGAGTG-SPC3-3′). All the primers and probe used for RPA assay were synthesized by Sangon Biotech (Shanghai, China).

Development of LF-RPA assays

The RPA assay was performed as described in the operation manual of the TwistAmp® nfo kit (TwistDx, Cambridge, UK). Briefly, each RPA assay was performed in a PCR tube with the following 47.5-μl reaction mixture: 2.1 μl forward primer (10 μM), 2.1 μl 5′-biotin labeled reverse primer (10 μM), 0.6 μl lateral flow probe (10 μM), 2 μl DNA template, 11.2 μl ddH2O, and 29.5 μl rehydration buffer; then, the mixture was transferred into a freeze-dried pellet which contained 40 mg freeze-dried recombinant enzyme. Next, 2.5 μl of 280 mM magnesium acetate (MgAc) was added to the lid of the pellet, and the reaction was started by centrifugation of the pellet, followed by incubation of the tubes at 37 °C for 30 min.

After incubation, 2 μl of the RPA-amplified products was diluted with 98 ml of LF buffer (Milenia Biotec, Germany) in a 1.5-ml tube, followed by the addition of a Hybridetect-1 LF dipstick (Milenia Biotec, Germany) and incubation at room temperature for 2 min. A colored test line on the dipstick was determined as positive detection of B. gibsoni, while no band on the test line was defined as a negative result, and a separate control line indicated that the dipsticks worked effectively. The optimum amplification temperature was determined by performing the RPA assay in the temperature range of 20 to 50 °C in a certain amplification time (20 min). The optimum time for amplification was investigated by conducting the reactions at the optimal temperature for 5 min, 10 min, 15 min, 20 min, 25 min, and 30 min using reference standard DNA as templates and analyzing the products by lateral flow dipsticks.

Conventional PCR

Conventional PCR based on the V4 variable region of 18S rRNA was performed as previously reported (He et al. 2017; Inokuma et al. 2004; Masatani et al. 2017; Singh et al. 2016). In this study, conventional PCR was used for comparison with the newly developed LF-RPA. The reaction was performed in a total volume of 25 μl consisting of 12.5 μl of 2xTsingke master mix (10 μM) (Tsingke, Beijing, CN), 1 μl of forward and reverse primer (10 μM), 8.5 μl of ddH2O, and 2 μl of genomic DNA. The PCR was performed at 95 °C for 5 min (polymeras e activation and initial denaturation), followed by 35 cycles (at 95 °C for 30 s, 55 °C for 30 s,72 °C for 30 s, and finally at 72 °C for 10 min). The products were analyzed by electrophoresis in 2% agarose gels.

Evaluation of the developed LF-RPA

In order to evaluate the newly developed LF-RAP, specificity assay was performed by using the DNA samples of B. bigemina, B. orientalis, B. duncani, T. sergenti, T. gondii, P. falciparum, and H. longicornis. Meanwhile, uninfected dog DNA was used as a negative control.

For sensitivity assay, B. gibsoni standard genomic DNA was seven-fold diluted (103, 102, 10, 5, 2.5, 1.75, 1, 0.5, 0.25, and 0.125 parasite per 1 μl blood). The serially diluted DNA samples were used to evaluate the sensitivity of the LF-RPA assay as well as the sensitivity of the conventional PCR assay.

A total of 15 blood samples were randomly collected from sick dogs in clinic with the permission of their owners. These dogs’ symptoms were similar to those infected with B. gibsoni, such as fever, anorexia, and anemia, and were considered to be the potential infected dogs. Then, all samples were subjected to both the LF-RPA and the conventional PCR assays for B. gibsoni detection.

Results

Optimal RPA reaction

The RPA amplification based on the cox I of B. gibsoni was performed at the room temperature. The target gene was successfully amplified by the special primers (Table 1), and a fragment of 233 bp was generated. The optimal reaction temperature and time of RPA were determined by using B. gibsoni standard genomic DNA (106 parasites/μl blood). As shown in Fig. 1a, the RPA amplification worked in the wide temperature range of 25 to 50 °C, with the best reaction temperature at 35 °C. Meanwhile, after 20–30-min amplification reaction, the products could be visually detected by the lateral flow dipsticks within 2 min (Fig. 1b). Therefore, the following LF-RPA detection would be performed at 30–35 °C in 20–30 min.

Table 1 Recombinase polymerase amplification primers and probe used in this study
Full size table
Fig. 1
figure 1

Optimization of the reaction temperature and time for RPA ampilifcation. a The test line can be detected in the temperature range of 30 to 45 °C, with the optimal reaction temperature being 30 or 35 °C. b Screening for the optimal reaction time. The test line can be detected after 20 min of amplification reaction

Full size image

Analytical specificity and sensitivity of LF-RPA

The LF-RPA specificity was evaluated using the reaction system under the optimal conditions with the DNA from seven specific pathogens. As shown in Fig. 2, the LF-RPA method had high specificity, with only the target DNA exhibiting a positive signal in the test line, indicating that this method produced no cross-reaction with other parasites (B. orientalis, B. bigemina, B. duncani, T. sergenti, T. gondii, P. falciparum, H. longicornis) or dogs, so the primers of the LF-RPA were specific to the target of B. gibsoni. The sensitivity experiment validated that the LF-RPA detection limit was up to 0.5 parasite/μl blood (Fig. 3a), while the 18s rRNA PCR detection limit was 10 parasite/μl blood (Fig. 3b).

Fig. 2
figure 2

Specificity assay of LF-RPA. P, positive control, B. gibsoni; N, negative control, healthy dog DNA; B, blank control, water. Lines 1–7 were DNA from B. orientalis, B. bigemina, B. duncani, T. sergenti, T. gondii, P. falciparum, and H. longicornis, respectively

Full size image
Fig. 3
figure 3

Comparison of the sensitivity of LF-RPA and conventional PCR. P, positive control. N, negative control. M, maker. a Sensitivity of LF-RPA as evaluated using serially diluted B. gibsoni DNA (10, 5, 2.5, 1.75, 1, 0.5, 0.25, and 0.125 parasites/μl blood). b Sensitivity of conventional PCR as evaluated using the diluted samples

Full size image

Detection of B. gibsoni from clinic samples

A total of 15 field samples were detected separately by LF-RPA and PCR assays. Results showed that 13 of the 15 samples (86.64%) were identified as B. gibsoni positive and 2 samples as negative by LF-RPA (Fig. 4a), while 10 of the 15 samples (66.67%) were detected as positive by the conventional PCR (Fig. 4b).

Fig. 4
figure 4

Detection of field samples with LF-RPA and PCR assays. a Fifteen field samples were tested by LF-RPA. b The positive for PCR detection was 10. P, positive control. N, negative control

Full size image

Discussion

Babesiosis caused by B. gibsoni is a widespread protozoan disease of dogs in China (Cao et al. 2015). Diagnosis of babesiosis by detecting parasites through microscopic examination is the cheapest and fastest method, but only suitable for acute infection (Narantsatsral et al. 2011). Additionally, molecular diagnostics and serological technologies, such as polymerase chain reaction (PCR), indirect fluorescent antibody test (IFAT), and enzyme-linked immunosorbent assay (ELISA), were considered as the specific and sensitive detection methodology (Bicalho et al. 2004; Birkenheuer et al. 2003; Eichenberger et al. 2017), but they were not suitable for detection in pet hospitals due to lack of experimental apparatus, complex operation process, and long detection time. These infected dogs were hidden sources of B. gibsnoi for spreading diseases, because of untimely diagnosis and treatment. An accurate and rapid diagnosis is the precondition of treatment and prevention, suggesting the vital importance of a facile, feasible, sensitive, and specific method in clinical diagnosis (Jia et al. 2009).

Recombinase polymerase amplification is an isothermal amplification technology developed in 2006 (Piepenburg et al. 2006b), and it is widely used in various fields for the rapid detection of different pathogens due to its high sensitivity, high specificity, and simple operation (Ma et al. 2017). Loop-mediated isothermal amplification (LAMP) is also an isothermal amplification method, but it requires at least four pairs of specific primers on targeting gene and reacts at 60–65 °C. In contrast, the LF-RPA is more convenient and simple, with only one pair of primers needed and the reaction performed in a water bath (Poulton and Webster 2018). With the addition of the reverse transcriptase enzymes, the RPA can also be used for RNA virus amplification, such as Foot-and-Mouth Disease (Abd El Wahed et al. 2013). As a novel molecular detection technology, the RPA has been used by many scientists for diagnosis of parasitic pathogens such as Theileria annulata, Toxoplasma gondii, and Schistosoma haematobium (Wu et al. 2017; Yin et al. 2017). Recently, Castellanos-Gonzalez and White et al. have developed a RPA technique to detect different protozoan infections (Castellanos-Gonzalez et al. 2018).

In this study, the recombinase polymerase amplification (RPA)-lateral flow (LF) dipstick method was developed for specific diagnosis of B. gibsoni. This method could be carried out within 30 min at a temperature as low as 25 °C without thermal cycling. The specific validation test indicated that the method was only involved in the reaction with B. gibsoni and showed no cross-reaction with the other parasites (N = 7) used in the test. For the analytical sensitivity, the LF-RPA assay detection limit was 0.5 parasites/μl blood, which was more sensitive than the reported quantitative PCR detection limit (10 parasites/μl blood) (Matsuu et al. 2005). When compared with conventional PCR and quantitative PCR, the LF-RPA has the advantages of reaction at the room temperature without PCR inhibitors, visualization of results, and faster operation relative to the other diagnosis methods (Soliman et al. 2018). The LF-RPA assay was diagnostically validated with 15 clinical samples, demonstrating a good specificity and sensitivity of 86.64% versus the 66.67% of the conventional PCR, indicating that the LF-RPA method is more sensitive than the conventional PCR and very suitable for clinical diagnosis of B. gibsoni.

Here, we developed a detection method for the specific diagnosis of B. gibsoni by combining isothermal nucleic acid amplification with the lateral flow dipstick (LF-RPA) based on the cox I gene. The LF-RPA method could detect B. gibsoni at the room temperature in 30 min. All the integrated data demonstrate that it is a specific, sensitive, and convenient tool for rapid and accurate detection of B. gibsoni in pet hospitals or resource-limited conditions.