A high-resolution melting approach for the simultaneous differentiation of five human babesiosis–causing Babesia species

Background Six species of apicomplexan parasites of the genus Babesia, namely B. microti, B. divergens, B. duncani, B. motasi, B. crassa–like and B. venatorum, are considered to be the primary causal agents of human babesiosis in endemic areas. These six species possess variable degrees of virulence for their primary hosts. Therefore, the accurate identification of these species is critical for the adoption of appropriate therapeutic strategies. Methods We developed a real-time PCR–high-resolution melting (qPCR-HRM) approach targeting 18S ribosomal RNA gene of five Babesia spp. based on melting temperature (Tm) and genotype confidence percentage values. This approach was then evaluated using 429 blood samples collected from patients with a history of tick bites, 120 DNA samples mixed with plasmids and 80 laboratory-infected animal samples. Results The sensitivity and specificity of the proposed qPCR-HRM method were 95% and 100%, respectively, and the detection limit was 1–100 copies of the plasmid with the cloned target gene. The detection level depended on the species of Babesia analyzed. The primers designed in this study ensured not only the high interspecific specificity of our proposed method but also a high versatility for different isolates from the same species worldwide. Additionally, the Tm obtained from the prepared plasmid standard is theoretically suitable for identifying isolates of all known sequences of the five Babesia species. Conclusions The developed detection method provides a useful tool for the epidemiological investigation of human babesiosis and pre-transfusion screening. Graphical Abstract Supplementary Information The online version contains supplementary material available at 10.1186/s13071-023-05839-5.


Background
Human babesiosis caused by Babesia species, which are blood protozoa belonging to the family Babesiidae (Order Piroplasmida; Phylum Apicomplexa) [1], is a re-emerging tick-borne disease.Cases of transfusion-transmitted and naturally acquired Babesia infection in humans represent a serious public health problem and have been frequently reported worldwide in recent years [2,3].The pathogens responsible for human babesiosis are predominantly transmitted via tick bites and blood transfusion and rarely via perinatal routes and organ transplantation [4][5][6][7][8].To date, more than 100 Babesia species have been identified worldwide [9].
Babesia spp.were previously believed to only infect wild and domestic animals.However, in 1957, the first case of human infection with Babesia was reported in Yugoslavia [10], and the disease has gradually become a new global parasitic disease in humans, with the potential to be a health threat to people with weaker immune systems.Six Babesia species, namely B. microti, B. venatorum, B. divergens, B. duncani, B. motasi, and B. crassalike, are well known as the major pathogenic agents responsible for human babesiosis [11,12].The majority of Babesia cases in the USA are caused by B. microti, in areas where the disease is endemic, but B. duncani and B. divergens are also distributed in the USA [13].In Europe, B. divergens is mainly found in France and Ireland [12,14], B. venatorum has been reported in Austria, Germany, Italy and Sweden [14][15][16] and B. crassa-like infections have been reported in Slovenia and France [17,18].In Asia, B. venatorum and B. crassa-like are mainly reported in China [19,20].However, B. microti cases have been reported in southwestern China along the Myanmar border, Taiwan and Japan [21][22][23][24][25]. Babesia motasi strain KO1 (DQ3466955) is distributed in Korea [12,26].The sequences of B. motasi KO1 and Babesia sp.KCDC-1 (MK930513) strains isolated from two human cases in South Korea showed great similarity with the sequence of the B. motasi hebeiensis strain in China [26].It has also been observed that B. duncani, which was first reported in Washington State, USA, is now prevalent in Canada [27][28][29].
People with a functioning immune system may remain asymptomatic post-Babesia infection, although intermittent and low-grade parasitemia can occasionally last for > 2 years [30,31].Thus, Babesia-infected patients who are asymptomatic may spread the pathogen to others via blood transfusion or organ transplantation [32].In the USA, parasitic infections related to blood transfusions are almost entirely caused by Babesia [33].It has also been observed that transfusion-transmitted babesiosis causes severe complications and death in approximately one of every five cases [30].However, in human babesiosis-epidemic areas, blood screening for babesiosis is still lacking or only performed at a limited number of blood collection centers [34].
The diagnostic techniques currently used for Babesia (natural hosts and animal models) detection include microscopic examination (ME), immunoassays and molecular techniques [2].Although the traditional ME has a high specificity, it is characterized by a low sensitivity and requires highly experienced microscopists to carry out the analysis.Common molecular techniques, such as nucleic acid detection, perform well but are relatively time-consuming, and some existing nucleic acid testing techniques for diagnosing babesiosis have high requirements for instruments [35][36][37].The immunological detection method is efficient, time-saving and sensitive; however, false positive results caused by crossreaction often occur during the detection process, and only one pathogen can be detected at a time.The simultaneous detection of several species is a labor-intensive process, and distinguishing current infections from past infections using serological tests is still challenging [38,39].While real-time PCR (qPCR)-based techniques, with high specificity and sensitivity, have been proposed for the detection of babesiosis, the number of Babesia species that can be differentially diagnosed at the same time using this technique is limited [40][41][42][43].Current diagnostic strategies are typically used to detect major species, notably B. microti and B. divergens.In contrast, methods with high sensitivity and specificity are lacking for lowprevalence species, such as B. crassa-like and B. motasi [18].The true incidence and distribution of human babesiosis may be much higher than the currently reported levels [17,44].
In areas where multiple Babesia spp.are co-endemic, species differentiation is critical for guiding appropriate treatment and monitoring.Tick vectors commonly transmit different Babesia species, leading to the simultaneous occurrence of several types of Babesia infections in some areas, which not only poses great difficulties in preventing and controlling the resulting diseases and establishing accurate detection methods, but leads to incorrect classification and low detection rates.Notably, the pathogenicity and reproductive ability of B. duncani are superior to those of B. microti [45], and the former is also more resistant to currently recommended anti-Babesia drugs [46].Therefore, in the future, different treatment methods should be selected according to the virulence and drug resistance of the Babesia species.
In this study, we designed a single-round qPCR-highresolution melting (qPCR-HRM) assay using a single pair of primers and investigated the performance of this assay in simultaneously identifying five zoonotic Babesia species, namely B. microti, B. divergens, B. duncani, B. motasi hebeiensis and B. crassa-like.The 18S ribosomal RNA (rRNA) gene, which contains the highly conserved and variable regions of Babesia, was selected as the target amplification region of the qPCR-HRM.The designed primers of HRM do not allow cross-reaction with other zoonotic pathogens, such as Toxoplasma gondii.Specifically, the HRM approach is based on the principle that during the dissociation of double-stranded DNA into single-stranded DNA, as the temperature increases, the melting curve and melting temperature (T m ) can be generated by monitoring the fluorescence of a dye in the double-stranded DNA.Different melting curves, different T m values, and different genotype confidence percentage (GCP) values can thus be obtained based on differences in the amplicon compositions of the different species [47].It has been reported that a higher GCP value is indicative of a higher degree of nucleic acid sequence similarity between the sample and positive control.Based on our results, this assay can be used for the differential diagnosis and epidemiological investigation of babesiosis in endemic areas.

Preparation of laboratory-infected animal specimens and DNA samples
Six-month-old sheep, negative for piroplasm infection as determined via microscopy and PCR, were purchased from Jingtai County, Gansu Province, China [48].Splenectomy was performed on the sheep 1 month before the experiment.The strains of B. motasi hebeiensis and B. duncani were obtained from the VVBD Laboratory.In brief, to prepare the infected animals, a suspension of B. motasi hebeiensis, preserved in liquid nitrogen, was rapidly thawed in water at 37 °C, and a 50-ml (approx. 1 × 10 9 infected erythrocytes) aliquot of this solution was injected into the sheep via the jugular vein.
Lakeview Golden (LVG) Syrian hamsters (age: 3 months) were purchased from the Beijing Vital River Laboratory Animal Technology Co., Ltd (Beijing, China).Similar to the infected sheep, after the rapid resuscitation of frozen B. duncani in a water bath at 37 °C, five hamsters were inoculated intraperitoneally with 200 μl of the B. duncani suspension (approx.2.0 × 10 6 infected red blood cells [iRBCs]).
At 4 and 7 days after the inoculation of the hamsters and sheep, respectively, blood samples were collected for the preparation of smears.After Giemsa staining of the blood smears, parasitemia was calculated by ME.Approximately 3000 RBCs were counted per blood smear to calculate the percentage of parasitized RBCs.When parasitemia varied in the range of 8% to 10%, whole blood samples, with different parasitemia levels, were collected in tubes containing ethylenediaminetetraacetic acid (EDTA) anticoagulant.Total DNA was extracted from 200 μl of these blood samples using a commercial DNA extraction kit (QIAamp DNA Blood Mini Kit; Qiagen, Hilden, Germany) according to the manufacturer's instructions.
The DNA of T. gondii, and Trypanosoma evansi were obtained from the blood of infected animals collected in the VVBD laboratory.In the specificity experiments, the DNA was directly amplified as a template without any treatment.The negative control DNA was isolated from whole blood obtained from healthy humans, sheep and piroplasm-free tick tissue and examined by ME of blood smears and by nested PCR [48].A NanoDrop 2000 spectrophotometer (NanoDrop Technologies) was used to assess DNA concentration.The extracted DNA was then stored at − 20 °C until use.
These experiments were approved by the LVRI Animal Ethics Committee of the Chinese Academy of Agricultural Sciences.All procedures were performed in accordance with the Animal Ethics Procedures and Guidelines of the People's Republic of China.

Preparation of DNA samples mixed with plasmids
The plasmids with a cloned target gene of B. microti or B. divergens were added to DNA samples from healthy individuals or ticks at different concentrations to prepare experimentally infected (positive) control samples.The number of plasmid copies added to obtain the positive samples ranged from 1 to 10 4 copy number(s)/μl.

Collection of clinical samples
Blood samples were collected from 492 patients with a history of tick bite who had visited the Second Hospital of Lanzhou University between May 2017 and July 2019 and resided in the Gannan Tibetan Autonomous Prefecture (Gansu, China).All clinical samples were collected using approved protocols and after obtaining written informed consent from each patient.Total DNA was extracted from 200 μl of each clinical blood sample using the QIAamp DNA Blood Mini Kit (Qiagen).

Target gene sequence selection and primer design
For this study, we designed primers with high specificity for distinguishing and identifying Babesia species.The 18S rRNA gene sequences of five Babesia species identified in different countries based on the NCBI database were analyzed using the DNAMAN 9.0 software package (Lynnon Biosoft, Vaudreuil-Dorion, QC, Canada), and the homology of the different isolates of each Babesia species was analyzed using MEGA version 7 software.The corresponding fragments of the same Babesia species from different regions that showed 90-100% sequence homology were screened out.There were several differences in base length between these common fragments of the five Babesia species (Additional file 1: Fig. S1; Additional file 2: Fig. S2).To distinguish the five Babesia spp., we designed universal primers using Primer v.5.0 software (PREMIER Biosoft, San Francisco, CA, USA) targeting common fragments for the qPCR-HRM analysis.

Evaluation of primer efficiency at various annealing temperatures
To ensure that the primers could simultaneously amplify the target regions of the 18S rRNA gene of all five Babesia spp.without producing nonspecific bands or primer dimers, both of which would interfere with the interpretation of the results in subsequent qPCR-HRM analysis, we performed a series of gradient PCR assays to determine the optimal annealing temperature for the primers.Briefly, the PCR mix (20 μl) containing the Premix Taq DNA polymerase (10 μl; Takara), DNA template (1 μl), each primer (10 μM/μl) and RNase-free water (7 μl) was prepared, and gradient PCR was then performed using a T100 ™ Thermal Cycler (Bio-Rad Laboratories, Hercules, CA, USA).Thereafter, the mixture was heated at 95 °C for 2 min, followed by 40 cycles of denaturation at 95 °C for 30 s, annealing at a gradient temperature of 54 °C to 61 °C for 30 s, extension at 72 °C for 30 s, with a final extension at 72 °C for 5 min.After gradient PCR amplification, the products were resolved by electrophoresis in 1.5% (w/v) agarose gels to evaluate primer specificity and amplification efficiency at the different annealing temperatures.

qPCR-HRM analysis
The qPCR amplification was performed in a 20-µl reaction solution containing the forward and reverse primers(10 nM), Forget-me-Not ™ qPCR Master Mix (10 μl; Biotium, Hayward, CA, USA), 40× template buffer (0.5 μl), RNase-free water (6.5 μl) and the DNA sample (1 µl).The qPCR-HRM assays were performed using a Rotor-Gene Q6000 Real-Time PCR system (Qiagen) with an initial denaturation step at 95 °C for 2 min, followed by 45 cycles of 95 °C for 5 s and 55 °C for 30 s. HRM was performed from 75 °C to 85 °C using a 0.2 °C step with a 2-s hold time at each acquisition step.Finally, the test results were analyzed using the High-Resolution Melt software v.2.3.1 program (Qiagen), which automatically generates the melting curve, T m and GCP values.

Evaluation of the analytical specificity and sensitivity of the qPCR-HRM approach
The specificity of the assay was assessed using 20 ng/μl plasmids with a cloned target gene of zoonotic Babesia, including from B. divergens, B. motasi hebeiensis, B. crassa-like, B. duncani and B. microti, with the samples diluted by DNA extracted from blood samples collected from healthy humans, to ensure that a pair of primers could be used to accurately distinguish the five Babesia spp. in a single reaction.Five replicates were prepared for each plasmid, and the standard T m and GCP values of each Babesia strain were obtained based on five independent qPCR-HRM experiments.The specificity of this assay was also evaluated using the DNA extracted from other organisms, such as T. gondii, T. evansi, ticks and hamsters.Genomic DNA extracted from whole blood obtained from healthy humans was used as a negative control.
Analytical sensitivity was assessed using tenfold serial dilutions of the five plasmid DNA standards, ranging from 1 to 10 7 copy number(s)/μl.This process was repeated 3 times to ensure the accuracy of the results.Standard curves were then plotted using the High-Resolution Melt software v.2.3.1 (Qiagen) to evaluate the efficiency of the assay.

Evaluation of qPCR-HRM assay using experimental and clinical samples
The performance of the qPCR-HRM assay was evaluated using standard positive samples (experimentally infected animal specimens, DNA samples mixed with plasmids and plasmids), as well as clinical samples (Table 1).The presence of B. duncani and B. motasi hebeiensis in the collected blood samples was detected by microscopy.The previously established nested qPCR method for specific detection of B. duncani was used to detect 80 DNA samples mixed with plasmids and 80 laboratory-infected animal samples, and to compare detection performance with the qPCR-HRM assay for detecting B. duncani.A total of 492 clinical samples, 120 DNA samples mixed with plasmids, and 80 laboratory-infected animal samples were also randomized, relabeled and blindly detected using the qPCR-HRM method.Simultaneously, the qPCR products of all the samples were purified using a gel recovery kit, and 18S rRNA gene sequencing was performed to verify the qPCR-HRM results.

Statistical analysis
One-way analysis of variance (ANOVA) was performed using SPSS software (IBM, Armonk, NY, USA) to analyze whether there were significant differences in the T m values corresponding to the five Babesia species.The T m values of B. crassa-like and B. motasi hebeiensis were further analyzed by performing a t-test.

Primer design and testing
A pair of universal primers (Babesia-3F: 5′-ACG AGA CCT TAA CCT GCT AA-3′, Babesia-3R: 5′-CAC AGA CCT GTT ATT GCC TTA-3′) was selected to amplify the target sequence located in the 18S rRNA gene of the Babesia spp.by screening multiple primer pairs (Fig. 1).According to the Basic Local Alignment Search Tool (BLAST) analysis of the primers, the PCR products varied in length from 118 to 127 bp.The upstream and downstream primers target the conserved regions on both sides of the V3 hypervariable region of the 18S rRNA gene of Babesia spp., allowing species differentiation via melting curve analysis.Electrophoresis results obtained at an annealing temperature of 55 °C showed the appearance of a strong band, without a nonspecific band in the negative control (Additional file 3: Fig. S3).

Verification of specificity
The qPCR-HRM assay successfully distinguished all five Babesia spp.considered in this study.Specifically, as shown in Fig. 2, the melting curve of different Babesia species could be clearly separated, and the T m of each species could be determined.No melting curves were observed for DNA samples belonging to non-Babesia pathogens.Further, the negative control containing the host, background DNA did not exhibit amplification.Thus, the T m and GCP values obtained could be used as reference standards for the identification of Babesia spp.(Table 2).The Babesia species were also identified based on similarities in standard curves and T m values between the analyzed samples and reference samples.

Verification of sensitivity
The detection limit of the qPCR-HRM assay was evaluated using tenfold serially diluted plasmid DNA samples from the five Babesia spp. in three independent reactions, with the concentrations ranging from 1 to 10 7 copies/µl.The detection limit of the assay for B. microti was one copy, whereas for B. divergens and B. duncani, it was 10 copies.Both B. motasi hebeiensis and B. crassa-like Fig. 1 Sequence alignment of common sequences of five Babesia spp.amplified by the universal primers used in this study.The sequence for the conserved region of the 18S rRNA common to the Babesia spp. is indicated in red.The sequence for the 18S rRNA variant region common to the Babesia spp. is indicated in blue.The white portion is the differential base of the 18S rRNA of the five Babesia species.Dots represent the base missing in this region of the strain compared to other species.rRNA, Ribosomal RNA could be at as low as 100 copies of the target sequence (Fig. 3).

Test on the clinical samples
The discriminatory power of this assay was assessed using a blinded test comprising 200 Babesia-positive samples (DNA samples mixed with plasmids and laboratory-infected animal samples) and 492 clinical samples.The observed sensitivity and specificity of the qPCR-HRM assay were 95% and 100%, respectively.Four B. divergens-specimens were detected in clinical samples from patients with a history of tick bites (Table 3).The results also showed that the qPCR-HRM approach can be used to detect Babesia in different host samples, such as ticks, humans and rodents, and no cross-reaction with host DNA was observed.In addition, samples containing different plasmid copy numbers of Babesia were successfully detected.Taken together, these observations indicated that the qPCR-HRM assay can specifically amplify the 18S rRNA gene of Babesia spp.The melting curves of the DNA samples extracted from different batches showed high repeatability, and even though there was a slight change in the T m values, no change in the melting peak shape or in the T m gap between the samples was observed.The results of one-way ANOVA showed significant differences in the T m among the five Babesia spp.(p < 0.0001), sufficiently demonstrating that the qPCR-HRM approach can be used to clearly distinguish Babesia species.In the melting curve analysis, the GCP value of the automatic analysis results was 90-100%, further indicating that the species differentiation results are quite convincing.
Even though the T m produced by the target sequences of B. crassa-like and B. motasi hebeiensis were relatively similar, these species could still be clearly distinguished given that the p-value obtained from the t-test between their T m values was < 0.0001.Additionally, B. crassa-like and B. motasi hebeiensis could be distinguished based on the shape of the melting curves of the mixed samples.The qPCR product of the plasmid standard of B. crassa-like was mixed with the qPCR products of DNA samples of B. crassa-like or B. motasi hebeiensis at a 1:1 volume ratio.The HRM analysis was repeated, and it was observed that the melting curve of B. motasi hebeiensis formed a double peak, whereas that of B. crassa-like formed a unimodal peak (Fig. 4).
The ability of our newly developed method to detect B. duncani was evaluated through comparison with the results from microscopy and nested qPCR using 120 DNA samples mixed with plasmids and 80 laboratoryinfected animal samples.The results obtained indicated that our proposed method, qPCR-HRM, shows good performance in B. duncani detection (Table 4).

Discussion
In recent decades, there has been a gradual increase in the number of human babesiosis cases, largely due to increased human activities in Babesia endemic areas, insufficient public awareness regarding human babesiosis, expansion in the range of major tick vectors and climate change.Existing technologies for Babesia spp.detection suffer from several limitations [44].At the present time, the gold standard for the confirmation of babesiosis infection is ME [49], but several Babesia spp. that infect humans are not only morphologically similar to Plasmodium but also similar to each other [2].The low parasitemia level obtained may result in these samples being misdiagnosed as false negatives during ME [50].Several PCR-based assays for the detection of Babesia are available, such as droplet digital PCR, cobas ® Babesia (Roche Molecular Systems Inc., Branchburg, NJ, USA) and a licensed nucleic acid test targeting the Babesia 18S rRNA gene [35][36][37].Although these methods have high sensitivity and specificity and can be used to detect multiple pathogens simultaneously, a major disadvantage that limits their applicability in babesiosis-endemic countries is that they require complex and expensive instruments.Fluorescence in situ hybridization can also be used to detect active infections; however, this method has shown limited performance in detecting hypoparasitemia during the early stages of infection [51].Immunofluorescence assays, enzyme-linked immunosorbent assays and multiplex bead format assays can also be used to detect antibodies produced within 1 week of infection.However, the antibodies produced by several Babesia spp. in the host appear to be different and, therefore, multiple analyses are required to rule out Babesia infection, which is a laborious process [52,53].Babesia antibodies are usually detectable in the blood for several months after infection remission, and distinguishing between current and previous infections is still challenging [50,54,55].Additionally, these above-mentioned diagnostic methods do not include the specific detection of B. crassa-like and  , which is more time-consuming than the method proposed herein.The time required to run a sample using the qPCR-HRM method we report here, including amplification curve detection and melting curve analysis, was 1 h and 10 min.The previously reported discrimination criteria for HRM are T m and melting curve.In addition to these two criteria, we used the GCP value for auxiliary identification in this study.
Since the difference in T m is directly related to differences in the sequence of the entire target region, once the T m value and the shape of the melting curve change, misidentification of the target species will result.In this study, HRM primers provided sufficient T m differences among the five Babesia species.For reliable discrimination, T m values > 0.25° C are considered to be indicative of different species [47].The amplicons of B. duncani and B. motasi hebeiensis have the lowest and highest T m , respectively, with a difference of about 1.7 °C.There was no overlap between the T m range and melting curve characteristics of the five Babesia species, which could be accurately distinguished by HRM analysis.Considering the practicability of the qPCR-HRM in the diagnosis of babesiosis, the designed primers ensure high interspecific specificity of the method.Additionally, owing to the high homology existing between the different isolates, the T m obtained from the prepared plasmid standard showed applicability for distinguishing all the known sequences of targeted species.These results lead to the conclusion that this method is highly versatile for different isolates of the same species worldwide.During the melting curve analysis, the GCP of the automatically invoked results was 90-100%, and the GCP provided strong support for the results of species discrimination.
Wilson et al. used a droplet digital PCR to discriminate between B. microti and B. duncani, a method which can detect 10 gene copies in a reaction volume [35].However, the detection limit of our newly developed qPCR-HRM method was one copy for B. microti and 10 copies for B. duncani.Stanley et al. used cobas Babesia (Roche Molecular Systems, Inc.) to detect B. duncani, B. microti, B. divergens and B. venatorum, and the detection limits for these four Babesia species were reported to range from 6.1 to 50.2 iRBCs/ml [36].Nevertheless, because the number of DNA copies is the target unit of qPCR-HRM detection, the results of the present study cannot be compared with the sensitivity of the determination developed by Stanley et al. using iRBCs/ml as the measurement unit.
The clinical detection performance of the qPCR-HRM assay was evaluated using clinical specimens and laboratory-infected animal specimens.The results strongly suggest that the assay has the same ability to accurately diagnose Babesia as currently available techniques.Babesia divergens infection was detected in clinical samples obtained from patients with a history of tick bites.This is consistent with the results previously reported by Wang et al. based on the use of nested PCR [48].The detection results of blood samples from hamsters infected with B. duncani showed that the sensitivity of the qPCR-HRM method to detect B. duncani was significantly higher than that of microscopy, and the detection performance was comparable to that of nested-qPCR.Using the qPCR-HRM method, not all of the targeted sequences of samples can be successfully amplified, which is affected by the sensitivity of the diagnostic method.Considering that DNA samples were only extracted from 200 μl of blood, which can lead to false negative results when parasitemia in these samples is extremely low or DNA extraction is improperly performed.Therefore, even if the qPCR-HRM diagnostic assay is relatively sensitive, DNA extraction is a key factor in determining the diagnostic accuracy.
The benefits of the method proposed here notwithstanding, it has some limitations.For example, it does not allow for quantitative detection.Another limitation of this study is that fewer specimens of Babesia human cases were collected, and there was a lack of performance data on the applicability of the designed qPCR-HRM assay in clinical practice.In the future, it will be necessary to optimize this technology, such as, for example, by adding species-specific molecular probes, to ensure a higher specificity and detection using more positive clinical specimens.

Conclusions
The qPCR-HRM assay developed in this study is a promising tool for the simultaneous identification of five human-infecting Babesia spp.using only one primer pair.This strategy constitutes an improvement on traditional methods, which are characterized by a low detection performance and can only detect individual species.This assay may be a practical and potential alternative for the rapid and accurate diagnosis of Babesia infection, as it also offers the possibility to detect species with low prevalence, such as B. crassa-like and B. motasi hebeiensis.Furthermore, the proposed qPCR-HRM assay can also be applied for blood screening, especially in areas where babesiosis is endemic.

Fig. 2
Fig. 2 Detection and discrimination of the five Babesia spp. a Raw data from melt curve analysis, b normalized HRM plots for 18S rRNA amplicon, c normalized difference curves, d derivative plot analyses.The melting curve and T m for each species can be very well discerned.HRM, High-resolution melting; T m , melting temperature

Table 1
Classification of test samplesFor mixed plasmid DNA samples and animal DNA samples, there were 40 specimens for each Babesia species

Table 2
Standard melting temperature and genotype confidence percentage values of five Babesia species in the realtime PCR-high-resolution melting analysis Values are presented as the mean ± standard deviation (SD) GCP Genotype confidence percentage, T m melting temperature

Table 3
Analysis results of clinical samples and laboratoryinfected samples Differential diagnosis of B. crassa-like and B. motasi hebeiensis.1C PCR product of plasmid of B. crassa-like, 2C PCR product of plasmid of B. motasi Hebei, 1 mixed samples of PCR products of plasmid and DNA of B. crassa-like, 2 mixed sample of PCR product of B. motasi hebeiensis DNA and PCR product of B. crassa-like plasmid.The qPCR product of the plasmid of B. crassa-like was mixed with the qPCR products of two unknown Babesia samples at a volume of 1:1, and then HRM analysis was performed

Table 4
Comparison of the performance of microscopy, nested real-time PCR and the newly developed real-time PCR-highresolution melting method for the detection of B. duncani B. motasi hebeiensis; thus, they are not suitable for the specific diagnosis and epidemiological investigation of these species.Consequently, it is necessary to develop a high-performance technique that offers the possibility to simultaneously detect multiple Babesia spp.Compared with other molecular diagnostic methods, qPCR-HRM has several advantages, namely low cost, high accuracy and rapidity; the same advantages as multiple qPCR based on the use of species-specific probes; and no sequencing or gel electrophoresis is required to analyze the PCR products, thus potential laboratory contamination of PCR products is avoided[56].In some studies, HRM technology has been applied for the detection of pathogens, and it has been observed that HRM targeting the 18S rRNA gene is a highly reliable strategy for detecting and differentiating species.Bakheitet al. distinguished Theileria equi and Babesia caballi by HRM [57], Chua et al. explored HRM as a technique for the simultaneous detection of five Plasmodium species infecting humans [58] and recently Keatley et al. performed species-level identification of Trypanosome infections in Australian wildlife [59].Rojas et al. also used HRM-qPCR to detect and quantify Spirocercalupi in fecal samples from dogs with spirocercosis [60].Additionally, Wang et al. used HRM to differentiate four Babesia species causing bovine babesiosis [47].The mPCR-HRM assay developed by Bielicka et al. also offers the possibility to detect B. microti, B. divergens, B. venatorum and B. canis but requires three pairs of primers for multiple reactions [61]