A Fluorescence In Situ Hybridization (FISH) Test for Diagnosing Babesiosis

Apicomplexan parasites of the genus Babesia cause babesiosis in humans and animals. The microscopic examination of stained blood smears, detection of serum antibodies by immunoassays, and PCR-based identification of parasite nucleic acid in blood are common laboratory methods for diagnosing babesiosis. The present study evaluated a commercially available Babesia genus-specific fluorescence in situ hybridization (FISH) test for detecting Babesia parasites in blood smears. The FISH test detected Babesia duncani and Babesia microti, two common species that cause human infections in the USA, and other Babesia species of human and veterinary importance in less than two hours. The Babesia genus-specific FISH test supplements other existing laboratory methods for diagnosing babesiosis and may be particularly useful in resource-limited laboratories.

Common laboratory tests for diagnosing babesiosis depend on detecting (i) parasites in stained blood smears by microscopy, (ii) serum antibodies by immunoassays, and (iii) Babesia nucleic acid in blood after PCR amplification [3,4]. Fluorescence in situ hybridization (FISH) tests are used for detecting malaria parasites in blood smears [9,10] and tuberculosis bacilli in sputum smears [11,12]. We now report the details of a commercially available FISH test designed to detect all known species of Babesia parasites in blood smears. The FISH test is based on DNA probes that hybridize to the multiple copies of 18S ribosomal RNA (rRNA) present in Babesia cells, and therefore the test does not entail PCR amplification of nucleic acid.

FISH Assay Reagents
The study utilized the Babesia genus-specific FISH assay kit (catalogue number BabGK04) available from ID-FISH Technology Inc. (Milpitas, CA, USA). The kit contained DNA probes targeting Babesia 18S rRNA, a hybridization buffer, a smear preparation reagent (SPR), necessary buffers, a counter stain, and a slide mounting medium. The DNA probes were covalently labeled with Alexa 488 green fluorescent dye.

FISH Test Method
The Babesia genus-specific test was performed according to the manufacturer's instructions provided with the kit similar to the FISH test procedures for Plasmodium falciparum and Plasmodium vivax [9] and Plasmodium knowlesi [10]. Three parts EDTA-treated whole blood or fresh cultures for testing were first mixed with 1 part SPR by volume. A thin smear on a glass microscope slide prepared from 4 µL of this mixture was air-dried and fixed with methanol. After adding 12 µL of hybridization buffer containing the probe mix to each fixed smear, it was covered with a plastic cover-slip and placed in a humid container at 37 • C for 30 min for hybridization. After 30 min, the smear was washed twice for 2 min each with wash buffer at an ambient temperature, followed by a rinse with rinse buffer. A drop of counterstain was added after drying the smear in complete darkness. The smear was then covered with a glass cover-slip and viewed at ×1000 magnification in an Olympus light microscope (Olympus, Tokyo, Japan) with an attached LED unit containing 492 nm excitation and 530 nm emission band pass filters (ID-FISH, Milpitas, CA, USA). Babesia parasites fluoresce green under these conditions.

Analytical Specificity of the Babesia Genus FISH Test
The FISH test was performed on smears prepared as described above from 12 pertinent species of bacteria and non-Babesia parasites in addition to Babesia duncani and Babesia microti grown in hamster blood as positive controls and a negative control of uninfected human blood (Table 1). An online BLASTn analysis at the National Centre for Biological Information (NCBI, Bethesda, MD, USA https://blast.ncbi.nlm.nih.gov) of the DNA probe sequences used in the Babesia genus FISH test was also performed against the human genome as well as Apicomplexa (taxid 5794), Theileria (taxid 5873), Cytauxzoon (taxid 27995), and Babesia (taxid 5864) genomes.

Detection of Different Babesia Species in the Babesia Genus-Specific FISH Test
The Babesia genus-specific FISH test was performed on thin blood smears prepared as described above from (i) hamster blood infected with B. microti and B. duncani (provided by Dr. Alan Ashbaugh, University of Cincinnati, OH, USA), (ii) an in vitro culture of B. divergens in human blood (Cat. No. CA-DIV-12, Fuller Laboratory, Fullerton, CA, USA) and (iii) bovine blood infected with B. bigemina and B. bovis (provided by Dr. Massaro Ueti, Washington State University, WA, USA). Blood smears from two patient blood samples received at IGeneX for routine testing and confirmed to be infected respectively with B. duncani and B. microti by DNA sequencing were also tested.

Analytical Sensitivity of the Babesia Genus FISH Test
The analytical sensitivity was measured as the limit of detection (LOD) as previously described for FISH tests on P. falciparum, P. vivax, and P. knowlesi [9,10]. The LOD was determined as the lowest concentration of parasites that could be detected by FISH in every one of 20 replicate smears prepared as described above at different serial dilutions of B. duncani (ATCC PRA-302) and B. microti (ATCC 30221D) grown in hamster blood provided by Dr. Alan Ashbaugh, University of Cincinnati, OH, USA. The starting parasitemia in hamster blood was determined from Giemsa-stained blood smears. The hamster blood was initially diluted to 1:1000 with uninfected human blood and then serially diluted two-fold in the same human blood to a final dilution of 1:32,000, before performing FISH tests on triplicate blood smears at each dilution. Once a preliminary LOD was determined, 17 additional smears made at the same dilution were tested for confirming the LOD.

Detection of B. duncani and B. microti in Clinical Blood Samples
From clinical blood samples originating in different US states that were received at IGeneX for routine testing and found to be positive in the Babesia genus FISH test, 44 were randomly selected for DNA sequencing after PCR amplification to further confirm that the test can detect both B. duncani and B. microti in clinical samples. For this purpose, DNA was first purified from 200 µL of EDTA-treated whole blood using the Generation Capture Column kit (Qiagen, Valencia, CA, USA). The purified DNA was then PCR-amplified in an Eppendorf Thermocycler 7000 using forward and reverse primers (Integrated DNA Technologies, San Diego, CA, USA) that targeted a species-variable region of Babesia 18S rDNA to yield an approximately 350 bp amplicon. For each PCR reaction, 9 µL of DNA was combined with 21 µL of the PCR reaction mixture (10 mM Tris-HCl pH 8.3, 50 mM KCL, 5 mM MgCl2, 0.001% gelatin, 1% glycerol, 0.5 µg of each primer, 0.1 mM dNTPs, and 0.25 U Amplitaq Gold from Applied Biosystems, Foster City, CA, USA). The PCR conditions employed were an initial denaturation at 95 • C for 10 min, followed by 50 cycles using a denaturation at 95 • C for 1 min, annealing at 57 • C for 1 min, and extension at 72 • C for 1 min. A final 10 min extension was performed at 72 • C. The PCR product was subsequently sequenced in both directions using the same forward and reverse primers to provide a minimum of 200 contiguous bases with a PHRED quality value of 20 or higher. The resulting DNA sequence was matched to the NCBI reference sequences for B. duncani and B. microti, with a cut-off BLASTn E-value of < 10 −30 .

Estimation of Clinical Diagnostic Parameters of the Babesia Genus FISH Test
A set of 154 clinical blood samples were received by IGeneX from the New York State Department of Health (NYSDOH) for proficiency testing for the Babesia genus FISH test. All 154 samples had been previously tested at NYSDOH by examining Giemsa-stained blood smears for Babesia parasites, and 51 were found to be positive. All 154 samples were tested by the Babesia genus FISH test at IGeneX in a blinded manner without prior knowledge of the Giemsa stain results.

Analytical Specificity
Among all the samples tested in Table 1, only Theileria equi in addition to B. duncani and B. microti reacted positively in the Babesia genus-specific FISH test. All the others were correctly identified as negative in the test. The online BLASTn analysis (NCBI, Bethesda, MD, USA) predicted that the Babesia genus-specific probes would detect 18S rRNA of T. equi but not 18S rRNA of other Theileria species or human DNA under the FISH test conditions. The BLASTn analysis also predicted that the Babesia genus probes would detect the 18S rRNA of all the animal and human Babesia parasite species in the NCBI database and may weakly detect Cytauxzoon felis 18S rRNA under the FISH test conditions.

Detection of Different Babesia Species
Photographs showing the results of the Babesia genus-specific FISH tests performed on blood smears containing B. microti, B. duncani, B. divergens, B. bovis, and B. bigemina are shown in Figure 1. They demonstrate that the Babesia genus-specific FISH test is able to detect the human parasites B. microti (Figure 1(A2,A3)) and B. duncani (Figure 1(B2,B3)) in hamster and patient blood as well as in vitro-cultured B. divergens (Figure 1(C2)). A1, B1, and C1 are respective photographs of Giemsa-stained smears of the B. microti, B. duncani, and B. divergens used in the FISH tests. In addition, the FISH assay is also able to detect B. bovis ( Figure 1D) and B. bigemina (Figure 1E), the two important parasites causing bovine babesiosis. The characteristic tetrad of merozoites, giving the appearance of a Maltese cross [3], was present in the Giemsa-stained (Figure 1(A1,B1,C1)) and FISH-reactive (Figure 1(A2,A3,B2,C2)) Babesia smears.

Analytical Sensitivity
The greatest dilutions of the Babesia-infected hamster blood at which B. duncani and B. microti could be detected in the FISH test in every one of the replicate smears were 1:16,000 and 1:8000, respectively. Based on the parasitemia of 904,176 per µL for B. duncani and 232,541 per µL for B. microti in the undiluted hamster blood, this corresponded to a LOD of 57 B. duncani and 58 B. microti per µL of blood or an approximate parasitemia of 0.001% for both species.

Detection of B. duncani and B. microti in Clinical Blood Samples
DNA sequencing showed that of the 44 randomly selected US clinical blood samples that were positive in the Babesia genus FISH test, 21 were infected with B. duncani and 23 with B. microti. Coinfections with B. duncani and B. microti were not observed in the 44 samples.

Estimated Clinical Diagnostic Parameters of the Babesis Genus FISH Test
Of the 154 proficiency test clinical blood samples from NYSDOH, 50 were found to be positive in the Babesia genus FISH test. Only the same 50 samples and one additional sample (n = 51) had been independently found by the NYSDOH to possess Babesia parasites on examination of the Giemsa-stained blood smears of all 154 samples. The remaining 103 samples were negative by both Giemsa staining and Babesia genus FISH. The clinical parameters of the Babesia genus FISH test relative to Giemsa staining estimated from these findings are presented in Table 2.

Discussion
Ixodes ticks that transmit Borrelia species causing Lyme disease and Relapsing Fever can also transmit Babesia to humans [4,13,14]. Babesiosis and borreliosis can present with similar clinical

Discussion
Ixodes ticks that transmit Borrelia species causing Lyme disease and Relapsing Fever can also transmit Babesia to humans [4,13,14]. Babesiosis and borreliosis can present with similar clinical manifestations and occur as coinfections in patients in the US [15][16][17][18][19]. The clinical symptoms of babesiosis also overlap with those of malaria [20]. The early intra-erythrocytic stages of human-infecting Babesia species in Giemsa or Wright's stained blood smears can be mistaken for the ring and trophozoite stages of P. falciparum, and this can lead to misdiagnosis in localities where babesiosis and malaria are co-endemic [4]. The ability of the Babesia genus-specific FISH test to differentiate Babesia infections in blood from those caused by Borrelia, Plasmodium and other pertinent blood-borne pathogens is therefore important for diagnosing babesiosis.
Our findings suggest that the Babesia genus-specific FISH test is useful for diagnosing human infections with different Babesia species well as B. bovis and B. bigemina infections in cattle and infections with other Babesia species in animals [21]. The Theileria species infecting livestock, with the exception of T. equi in horses, are not expected to be positive in the Babesia genus-specific FISH test, but additional investigations are needed to firmly establish this. Because Theileria equi was until recently classified as Babesia equi, it may be phylogenetically closer to Babesia than other Theileria species, leading to the cross-reactivity observed in the Babesia genus-specific FISH test. Cytauxzoon felis is a tick-borne apicomplexan parasite that infects felids worldwide and causes severe disease among domestic cats in the US [22]. The ability of the Babesia genus-specific FISH test to differentiate B. felis from C. felis requires further investigation, because in silico analysis predicts that there may be a weak cross-recognition of C. felis.
Serum antibody assays are widely used for diagnosing human and veterinary babesiosis, but the detection of antibodies does not readily distinguish between active and resolved Babesia infections [1][2][3]. Several types of PCR-based nucleic acid amplification tests for detecting Babesia parasites in blood are reported to have a greater sensitivity [23,24] than our finding of 57-58 Babesia parasites per µL in the Babesia genus FISH test. Real-time PCR (RT-PCR) tests are therefore recommended for screening donor blood for babesiosis in endemic areas of the US [25,26]. FISH tests targeting rRNA discriminate between viable and dead or dying parasites better than PCR tests targeting DNA because RNA is degraded more rapidly than DNA in dying cells [9,27], a property that is useful for monitoring the effects of drug treatment. Cost and infrastructure requirements also make FISH tests more practical than PCR-based tests for resource-limited laboratories. A LED/Filter unit ($3100 from IDFISH, Milpitas, CA, USA), a standard microscope (e.g., Olympus BX41, $3000, from Olympus, Tokyo, Japan), a 37 • C incubator ($410), and reagents/disposables ($2.60 per test) are required with minimal maintenance costs for FISH. FISH can be performed on a normal laboratory bench. On the other hand, a RT-PCR machine with analytical software ($22,500), two biosafety hoods ($6000), annual maintenance ($3800), and reagent/disposables ($3.50 per test) are estimated to be more costly. RT-PCR also requires a dedicated work area.
The Babesia genus FISH test allows the direct visualization of the important human parasites causing babesiosis in the US, viz. B. microti, B. duncani and B. divergens [1][2][3][4], as well as the bovine parasites B. bigemina and B. bovis [21]. In silico analysis suggests that the test will also detect other known species of Babesia that infect humans and animals worldwide. The FISH test reproducibly detects parasitemias as low as 57-58 Babesia parasites per µL or about 0.001%, and showed congruent results with a qPCR test performed independently in a different laboratory [24]. The analytical sensitivities in the Babesia genus FISH test for B. duncani and B. microti are comparable to those in FISH tests for P. falciparum (55-62 parasites per µL), P. vivax (56-59 parasites per µL), and P. knowlesi (61-84 parasites per µL), which were better than the examination of the corresponding Giemsa-stained blood smears [9,10].
Results with clinical samples estimate a clinical sensitivity, specificity, positive predictive value, and negative predictive value of 98%, 100%, 100%, and 99% respectively for the Babesia genus FISH test in comparison to independent detection by Giemsa staining. It is possible that the one sample positive by Giemsa staining at the NYSDOH but not by the FISH test at IGeneX may have been due to the presence of dead or dying parasites as a result of immune damage or drug treatment, but this could also have been a missed detection of low numbers of live parasites by FISH. Investigations into many different types of well-characterized clinical samples are needed to further establish the clinical diagnostic parameters of the Babesia genus FISH test.
The Babesia genus FISH test, like the similar FISH tests for malaria [9,10] and tuberculosis [11,12], provides a result in less than 2 h, detects only live parasites, requires minimal technical expertise and equipment, needs only an LED/filter unit attached to a standard light microscope for visualizing fluorescence, and utilizes stable reagents that do not require refrigeration. As nucleic acid amplification by PCR is not involved, the FISH test is not affected by interference from PCR inhibitors present in some clinical samples [28]. FISH using oligodeoxynucleotides to target rRNA was first shown in 1989 to distinguish closely related bacteria with appropriately-targeted probes labeled with different fluorescent dyes [29]. The development of species-specific FISH tests to distinguish different Babesia species in the manner shown possible for differentiating pathogen species in FISH tests for malaria [9,10] and tuberculosis [11,12] will be useful. Preliminary findings suggest that this may be possible for Babesia venatorum in China [30] as well as B. microti and B. duncani in the USA [31], but further investigations are required to confirm the species specificity of the respective probes.

Conclusions
The Babesia genus-specific FISH test has advantages that make it a useful supplement to the panel of tests currently available for the laboratory diagnosis of human babesiosis as well as babesiosis in pets and livestock.