Abstract
Theileria orientalis Ikeda is a newly identified agent of bovine infectious anemia in the United States. Although T. orientalis Ikeda is transmitted by ticks other than the tick that transmits Anaplasma marginale—a bacterial etiology of bovine infectious anemia—the geographic distributions of these 2 infectious organisms overlap, with coinfection reported in some cattle. Only anaplasmosis has an approved effective treatment in the United States. To provide rapid diagnostic information for producers with anemic animals, we developed a duplex real-time PCR (rtPCR) for A. marginale and T. orientalis. With a cutoff of 38 cycles, the duplex assay has a sensitivity of 97.0% and a specificity of 100% for A. marginale; with a cutoff of 45 cycles, the duplex assay has a sensitivity and a specificity of 100% for T. orientalis, compared to existing tests. In addition to providing a tool for improved clinical decision-making for veterinarians and producers, our rtPCR facilitates the study of coinfection of cattle in Virginia. Of 1,359 blood samples analyzed, 174 were positive for T. orientalis, 125 were positive for A. marginale, and 12 samples were positive for both T. orientalis and A. marginale. Hence, coinfection by these 2 agents of bovine infectious anemia does occur within Virginia. It is likely that this pattern of infection will be seen in other regions where T. orientalis and A. marginale infections are endemic, despite the difference in tick vectors.
Keywords
The Theileria orientalis complex describes several genotypes of a species of non-transforming, theilerial hemoprotozoan. 9 Within the complex, the Ikeda and Chitose genotypes (T. orientalis Ikeda and T. orientalis Chitose, respectively) are capable of causing disease, and have been associated with outbreaks of anemia in Australia, New Zealand, and Japan.9,11,13,18 T. orientalis Ikeda in particular is increasingly implicated as a causative agent of bovine infectious anemia in the United States, a vector-borne disease characterized by hemolytic anemia, icterus, general malaise, ill thrift, and sporadic abortions.9,11,12 Although rarely fatal, affected animals are often poorly producing; consequently, this is a disease of economic importance in the countries in which it is found.9,11
T. orientalis is transmitted most effectively by the Ixodidae tick Haemaphysalis longicornis,10,18 a tick that has recently been discovered in several states along the U.S. Eastern Seaboard. 16 T. orientalis Ikeda has recently been identified in cattle in Virginia, affecting animals that were also parasitized with H. longicornis ticks 12 ; this tick has been confirmed as a competent vector in Virginia for T. orientalis Ikeda. 8 The animals that were clinically affected in this outbreak had the typical signs of anemia, icterus, and general malaise. In Virginia, this clinical presentation is identical to the blood infection caused by the bacterium Anaplasma marginale. H. longicornis is not capable of transmitting A. marginale, 6 and because A. marginale is transmitted by other species of tick, 17 in many other parts of the world, the geographic distribution of T. orientalis Ikeda and A. marginale do not overlap (Jenkins C., pers. comm., 2021 Jun 10). In Virginia, however, the initial outbreak of T. orientalis Ikeda occurred in areas where anaplasmosis has been diagnosed historically, suggesting an overlap in geographic distribution in this region. Further, the predicted range of H. longicornis based on modeling suggests its presence in areas of the country where anaplasmosis is common. 14 An animal included in the initial outbreak study in Virginia was positive for both T. orientalis Ikeda and A. marginale by conventional PCR and Sanger sequencing. 12 As a bacterial infection, anaplasmosis is treatable using oxytetracycline, a cost-effective drug approved for use in food animals. Theileriosis is markedly more difficult to treat, and no effective drugs are approved for use in food animals in the United States, although buparvaquone has been effective in Australia.5,18 Therefore, making an early distinction between anaplasmosis and theileriosis is critical for immediate and effective clinical decision-making that has animal welfare, productivity, and economic implications. To be of use to producers, it is important that the tools used to assist in this process are cost-effective.
In response, we developed a duplex real-time PCR (rtPCR) assay that detects both A. marginale and T. orientalis. Multiplex assays that can detect Theileria spp. and Babesia spp. protozoans, and Anaplasma spp. bacteria, have been developed for small ruminants1,7; we sought to fill the niche for cattle. We developed our assay in accordance with published standards. 15 Although analytical sensitivity was determined by examining the limit of detection (LOD), the reported specificity and sensitivity for A. marginale and T. orientalis are relative to the specificity and sensitivity of the tests against which they were compared. For T. orientalis in particular, there are few validated tests against which to compare, hence the interpretation of diagnostic sensitivity (DSe) and diagnostic specificity (DSp), although promising, must take that into consideration.
Materials and methods
Blood
Whole blood samples were submitted from privately owned cattle across the state of Virginia. Animals included clinical submissions by private veterinarians, as well as herds owned and maintained by the Virginia Department of Corrections (VADOC), and as part of an ongoing surveillance effort by the Virginia Department of Agriculture and Consumer Services (VDACS) of locally produced animals sent to auction. Whole blood was collected in K2-EDTA blood collection tubes (Vacutainer; Becton, Dickinson). Except for animals submitted by referring veterinarians as part of a routine clinical diagnostic workup, all animals were randomly sampled at auction houses, and the presence and degree of clinical signs were unknown. Convenience-sampled animals were submitted as part of an ongoing surveillance effort by VDACS. These samples were taken from cattle at the time of auction. A number of animals from each pen within the auction house were selected for a blood draw, regardless of the presence or absence of clinical signs and attempting to avoid collecting samples from animals from the same herd. Although the samples themselves were collected in limited geographic locations (the auction houses), they represented animals from herds across Virginia. In total, 1,359 blood samples were available for evaluation.
DNA extraction
DNA was extracted from K2-EDTA anticoagulated blood (DNeasy blood and tissue kit; Qiagen) following the manufacturer’s protocol with modifications. 11 The initial blood volume for DNA extraction was 100 µL, and control DNA (Applied Biosystems VetMAX Xeno internal positive control DNA; Thermo Fisher) was added to the lysis buffer of the Qiagen kit (Buffer AL) at 20,000 copies per sample. DNA was eluted twice in 50 µL of Buffer AE of the Qiagen kit, pre-heated to 56°C, for a total elution volume of 100 µL. All extractions were carried out within a biosafety cabinet.
Duplex rtPCR
Our duplex assay is a TaqMan-based assay that utilizes primers and probes designed to detect the major surface protein 1b (msp1b) gene of A. marginale, and the major piroplasm surface protein (MPSP) gene of T. orientalis. The primers and probes for MPSP are sensitive for T. orientalis but are not genotype-specific; to further characterize genotype, a second assay is run on those samples that are positive for T. orientalis.
The gene target for A. marginale, msp1b, is a highly expressed, conserved gene that persists between tick and cattle transmission cycles and is the target of an A. marginale–specific rtPCR. 3 The MPSP gene of T. orientalis was chosen because it is the major antigenic target and the gene used for genotyping the T. orientalis complex.
Amplification for the msp1b and MPSP genes were accomplished in the same reaction. The duplex rtPCR reaction consisted of 10 μL of master mix (Applied Biosystems TaqMan environmental master mix 2.0; Thermo Fisher), 0.6 μL of each T. orientalis forward and reverse primer, 1.2 μL of each A. marginale forward and reverse primer, 0.2 μL of T. orientalis universal probe, 0.4 μL of A. marginale probe, 0.8 µL of primer–probe mix (Xeno VIC; Thermo Fisher), and 2 µL of DNA template in a 20-µL reaction. The primers and probes for T. orientalis and A. marginale have been published previously.2,4 Fluorophores and quenchers were altered from the published probes to allow appropriate multiplexing (Table 1). The probes utilize 3 separate fluorescent tags: FAM for A. marginale, NED for T. orientalis, and VIC for the internal positive control. Amplification was completed (Applied Biosystems 7500 fast real-time PCR system; Thermo Fisher) using standard cycling and the following run method: 95°C for 10 min, followed by 45 cycles composed of 15 s of 95°C for annealing, and 1 min of 60°C for extension. Reactions were completed (Applied Biosystems MicroAmp Fast 8-tube strips, Applied Biosystems MicroAmp Optical 8-cap strips; Thermo Fisher). A sample lacking DNA template was included as a negative control with each run. A T. orientalis known-positive sample, an A. marginale known-positive sample, and 2,000 copies of internal positive control DNA (VetMAX Xeno; Thermo Fisher) were included in separate reactions with each run to serve as positive controls for the respective fluorophore channels and negative extraction controls for the other targets.
Primer and probe sequences used for Theileria orientalis and Anaplasma marginale duplex PCR assay.
Validation: limit of detection
To determine the LOD for analytic sensitivity for T. orientalis, 3 identical rtPCR runs were completed per the protocol described above. Eight, 10-fold serial dilutions of gBlock gene fragments (Integrated DNA Technologies) with the universal MPSP gene sequence capable of detecting each of the genotypes Ikeda, Chitose, and Buffeli were used. The LOD was determined as the lowest dilution at which samples on all 3 plates were positive, and a LOD was determined for each of the 3 T. orientalis genotypes of interest. Similarly, the LOD was determined for A. marginale using serial dilutions of an A. marginale–positive blood sample confirmed positive by external rtPCR and shown to have 1.22% infected RBCs.
Validation: repeatability
For our intra-assay repeatability test of precision, we examined the high (undiluted), medium (1:100 dilution), and low (1:10,000 dilution) concentrations of reference DNA in quintuplicate in one run on one day. Reference DNA was extracted from samples confirmed positive on in-house PCR. In this assay, the average cycle threshold (Ct) was compared to the SD of the 5 replicates of a given concentration.
For our inter-assay repeatability tests, we examined high, medium, and low concentrations of reference DNA in quintuplicate on each of 6 consecutive days, in accordance with OIE recommendations. Inter-assay repeatability was calculated by plotting the Ct value versus the day. We used one-way ANOVAs of both inter- and intra-assay repeatability tests to examine the difference in Ct means across days or within a single run, respectively. All repetitions were performed on the same thermocycler used above.
Relative diagnostic sensitivity and specificity
The relative DSe and DSp of the T. orientalis and A. marginale targets within the duplex assay were determined with receiver operating characteristic (ROC) statistics. Our genotype-specific T. orientalis rtPCR described below was used as the reference standard test against which the duplex rtPCR results were compared. For A. marginale, rtPCR was completed either at the U.S. Department of Agriculture’s Agricultural Research Service, Animal Disease Research Unit (USDA-ARS-ADRU; Pullman, WA, USA) or at the Washington Animal Disease Diagnostic Laboratory (WADDL; Washington State University, Pullman, WA, USA) as the reference standard test.
Genotyping rtPCR
A second multiplex assay that utilizes distinct probes capable of differentiating the MPSP gene of 3 of the major T. orientalis genotypes—Chitose, Ikeda, and Buffeli—further characterized the genotype. The forward and reverse primers are the same as those used for T. orientalis in the duplex rtPCR protocol above, but with the use of a distinct set of probes to distinguish among the genotypes 2 (Table 2).
Genotype-specific probe sequences used for the Theileria orientalis genotyping assay.
DNA extracted from whole blood samples that were positive for T. orientalis on the duplex rtPCR were used for the genotype multiplex rtPCR. The 20-µL reactions consisted of 10 µL of master mix (Applied Biosystems TaqMan environmental master mix 2.0; Thermo Fisher), 0.6 µL of each T. orientalis forward and reverse primer, 0.5 µL of Ikeda probe, 0.2 µL of Chitose A probe, 0.3 µL of Chitose B probe, 0.2 µL of Buffeli probe, and 2 µL of DNA (Table 2). The instrumentation, run method, and plastics were the same as for the duplex rtPCR assay. A sample lacking DNA template was included with each run as a negative control. Positive controls for each of the 3 genotypes were included in each run, and runs were accepted only if controls worked as expected.
Genotyping
DNA that tested positive for the Chitose genotype of T. orientalis in the duplex rtPCR were subjected to 2 rounds of conventional PCR (cPCR) targeting the MPSP gene of T. orientalis. The reaction for the first round of cPCR consisted of DNA polymerase (Invitrogen platinum PCR supermix high fidelity; Thermo Fisher), 1 µL of each sense and anti-sense primer, and 10.5 µL of DNA in a 25-µL reaction volume. Primers (Integrated DNA Technologies) for MPSP were 5′-CTTTGCCTAGGATACTTCCT-3′ (sense) and 5′-ACGGCAAGTGGTGAGAACT-3′ (anti-sense). 11 The PCR program was as follows: 95°C for 2 min, following by 29 cycles consisting of 95°C for 15 s, 57°C for 30 s, 72°C for 1 min; after these 29 cycles, the program progressed to 72°C for 7 min, and then held at 12°C. A second round of cPCR was carried out using the amplicons from the first round as template, using the same reaction components and PCR program. Second round amplicons were electrophoresed on 1% agarose in Tris-borate with added ethidium bromide gel and visualized under UV light. DNA bands were purified from the gel (QIAquick gel extraction kit; Qiagen) and submitted to the Fralin Life Sciences Institute, Genomic Sequencing Center (Blacksburg, VA, USA), for Sanger sequencing. Sequences were then aligned to GenBank accessioned sequences for each of the Ikeda, Chitose, and Buffeli genotypes.
Results
Duplex rtPCR
Of the 1,359 individual DNA samples available for rtPCR testing, 186 (13.7%) were positive for T. orientalis, and 137 (10.1%) were positive for A. marginale. Included in these numbers are 12 samples (0.88%) that were positive for both T. orientalis and A. marginale.
Limit of detection
The LODs for T. orientalis genotypes Ikeda and Buffeli were 1 × 108 pmol of DNA. For the Chitose genotype, the LOD was 1 × 109 pmol. A. marginale was detected according to dilution rate, given the nature of the known-positive sample. The LOD of A. marginale was a 10-5 dilution of reference DNA, extracted from a clinically positive sample according to our in-house PCR. Although a standard curve was not run alongside this PCR assay, extrapolating from a similar assay run in our laboratory (supplementary data 8 ) suggests that 1 × 109 pmol of Chitose gBlock is ~16 gene copies; 108 pmol of Ikeda gBlock is ~25 gene copies.
Repeatability
As calculated, there was significant variability between intra- and inter-assay repetitions of the A. marginale component of the assay. However, the variability did not impact the ultimate interpretation of the test results; negative results remained negative, and positive results remained positive (Table 3).
Comparison of one-way ANOVA results following the repeatability assay.
The inter-assay repeatability was calculated by plotting the cycle threshold (Ct) value versus the day; intra-assay repeatability was calculated by plotting the Ct value versus the SD of 5 replicates of a given concentration.
Relative diagnostic sensitivity and specificity
For A. marginale, the assay DSe was 97.0% and the DSp was 100% when a cutoff of 38 cycles was applied (Tables 4, 5). For T. orientalis, both the DSe and DSp were 100% when a cutoff of 45 cycles was applied (the final cycle of the rtPCR program; Tables 6, 7). The reference standard utilized for A. marginale was an external rtPCR, run either at the USDA-ARS-ADRU or WADDL.
A comparison of the diagnostic sensitivity (DSe) and diagnostic specificity (DSp) values for Anaplasma marginale compared to an external real-time PCR run at either USDA-ARS-ADRU or WADDL.
Ct = cycle threshold; USDA-ARS-ADRU = U.S. Department of Agriculture, Agricultural Research Service, Animal Disease Research Unit, Pullman, WA, USA; WADDL = Washington Animal Disease Diagnostic Laboratory, Washington State University, Pullman, WA, USA.
Overview of the positive and negative cases for Anaplasma marginale compared to an external real-time PCR run at either USDA-ARS-ADRU or WADDL.
USDA-ARS-ADRU = U.S. Department of Agriculture, Agricultural Research Service, Animal Disease Research Unit, Pullman, WA, USA; WADDL = Washington Animal Disease Diagnostic Laboratory, Washington State University, Pullman, WA, USA.
A comparison of the diagnostic sensitivity (DSe) and diagnostic specificity (DSp) values for Theileria orientalis compared to a validated, in-house real-time PCR reference standard.
Ct = cycle threshold.
Overview of the positive and negative cases for Theileria orientalis compared to a validated, in-house real-time PCR reference standard.
The reference standard utilized for T. orientalis was the T. orientalis genotype rtPCR developed by the Virginia Tech Animal Laboratory Services (Virginia-Maryland College of Veterinary Medicine, Blacksburg, VA, USA), modified from a previous validated assay, 2 and described below.
Genotyping
Because a universal probe was used for the MPSP gene of T. orientalis, an additional genotyping assay was required for those samples that tested positive for T. orientalis in the duplex assay. The genotype multiplex assay was run using extracted DNA, applying the same T. orientalis forward and reverse primers, but utilizing genotype-specific MPSP probes. Samples of known genotype were included as positive controls. Of 186 T. orientalis–positive samples from the duplex PCR, 159 (85.5%) were consistent with the Ikeda genotype, 21 were consistent with the Chitose genotype, and the remaining 6 (3.2%) were undetermined. None were consistent with Buffeli. There were no mixed infections with multiple genotypes.
Discussion
The high DSe for A. marginale and T. orientalis makes our duplex rtPCR assay a useful screening tool for producers in Virginia and in other localities with concurrent occurrence of anaplasmosis and theileriosis. There is reported evidence within the literature that the Chitose genotype is capable of causing disease on its own10,11; given that numerous cattle in our study sample were positive for Chitose alone, further investigation is warranted, not just into the geographic distribution and possible overlap of A. marginale and T. orientalis Ikeda, but also into the pathogenicity of the Chitose genotype. This is of particular importance for Virginia because A. marginale and pathogenic genotypes of T. orientalis—individually and in combination—have been found in counties that account for much of the state’s cattle production. T. orientalis is a disease of economic importance in other countries in which it occurs, and being able to rapidly identify A. marginale, T. orientalis, or both agents in combination as the likely cause for clinical disease in cattle is important for determining treatment plans for veterinarians and producers.
A limitation of our experiment is the small number of reference tests available for T. orientalis. Because DSe and DSp of this proposed duplex PCR are calculated in reference to existing tests, those results rely heavily on the reference tests performing with high specificity and sensitivity. The reference test used for A. marginale was validated analytically at an AAVLD-accredited laboratory and published. 4 The reference test for T. orientalis was published, 12 but in addition, we confirmed a subset of the samples with cPCR and Sanger sequencing, improving confidence in the results. Our tool was designed to identify and distinguish between etiologic agents capable of producing identical clinical signs in cattle in Virginia, and, in this niche, it is fit-for-purpose.
Footnotes
Acknowledgements
We thank the veterinarians and producers involved in ongoing surveillance efforts.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
Portions of our study were funded by the Virginia College of Osteopathic Medicine/Virginia-Maryland College of Veterinary Medicine Center of One Health Seed grant program.
