In Australia, bovine theileriosis is caused by the haemoprotozoan Theileria orientalis that infects erythrocytes, and can result in clinical signs of lethargy, fever, anaemia, jaundice, abortions and mortalities in naive cattle. Many outbreaks of clinical disease have occurred in NSW since 2007, and several have been reported in Victoria in 2011-2012. Clinical disease can be assessed by PCV and blood smear examinations to detect a regenerative anaemia and estimate the level of parasitaemia. However, theilerial infections are long-term and may be difficult to detect in normal cattle on smear examination as the level of parasites in the blood declines. T. orientalis has three common types (Ikeda, Chitose and Buffeli), based on differences in their major piroplasm surface protein (MPSP), and application of conventional (gel-based) PCR assays is currently required to differentiate these, using EDTA blood samples. The Ikeda type has been associated with severe outbreaks in cattle overseas, and may occur in association with other types. Some early disease outbreaks in NSW were shown to be associated with Ikeda and Chitose types,1 based on tests performed in Japan, while prior to 2007, theilerial infections in Australian cattle were not associated with disease and were attributed to the non-pathogenic Buffeli type found in Queensland. This type was considered to be transmitted by Haemaphysalis spp. ticks which have a distribution along the eastern seaboard of Australia2,3 but have also been reported further inland.4,5
Conventional PCR diagnostic tests for MPSP types require separate tests. Multiplexing (to detect multiple types in the one assay) is not possible because the end products are of similar size. Thus the different MPSP types would not be distinguished if PCR products were run concurrently on a gel. So while samples can be screened for the presence of a T. orientalis universal MPSP sequence (for example by using a p32 assay), they then require type-specific conventional PCR tests on positives. Application of quantitative PCR (also known as real time PCR) is a possible means of reducing testing time and cost by multiplexing, but will require extensive validation before it can be routinely used.
Serology for theileriosis due to T. orientalis has also received little attention. ELISA technology based on recombinant antigens would offer advantages over previously applied antibody-based tests that have relied on mechanically and chemically extracted protozoal fragments or fractions. Such antigens have been previously applied in tests including an indirect fluorescent antibody test (IFAT)6, a complement fixation test (CFT)7 and an ELISA.8,9 Apart from application of recombinant MPSP antigen in a latex agglutination test for cattle10 and an ELISA to detect antibodies in water buffaloes,11 no recombinant ELISA to detect T. orientalis in cattle, or MPSP type-specific assays have been described in the published literature.
Distribution of T. orientalis and its MPSP types
In a structured survey of herds in eastern Australia considered typical of their district but in regions where clinical disease due to theileriosis was infrequent or absent, testing of 516 cattle in 50 herds (NSW 20; Qld 20; south eastern Victoria 10 herds) was undertaken. Conventional PCR analyses of blood samples identified cattle that tested positive on the screening assay (p32) for theileriosis, and then confirmed which of the three major types (Ikeda, Chitose, Buffeli) were present in the positive samples. Chitose was the most prevalent type detected in all three states. Ikeda was more limited in distribution in this survey, appearing in three of five regions of Queensland (North, South and South East), and only one in NSW (North Coast), and was detected in only one animal in Victoria.
In investigations of diagnostic samples of suspect field cases in NSW, we also characterised, by MPSP type and location, 434 samples collected from 116 herds between May 2009-Oct 2011. The LHPAs of Mid-Coast and Cumberland each had 9-16 new confirmed herds detected annually. Trends indicated that confirmed herds were less frequently identified during the winter period (June - Aug) in each LHPA region, and that the numbers of known affected herds progressively increased in new areas, with 12 LHPAs having confirmed cases by October 2011. In that study, cumulative numbers of smear positive and Ikeda positive herds showed a divergence in the detection rates on samples after September 2010, coinciding with the introduction of a modified PCR method in which both neat and diluted DNA extracts were tested (Fig. 1). The total number of clinically infected herds reached 140 by 2011 and mirrored the number of Ikeda-infected herds detected. Later in the study period, PCR testing was less prevalent on diagnostic submissions, which explains the lesser number of Ikeda positive new herds found (Fig. 1).
In the Mid-Coast region, the number of herds detected as infected with Ikeda type continued to increase throughout the period, while in other regions the number of new herds with Ikeda infection increased steadily, with a period of a few to several months intervening before additional Ikeda-infected herds were recognised (Fig. 2). The main sources of Ikeda positive samples in the period were in five regions (Mid-Coast, Cumberland, Central North, Tablelands and New England), while the majority of the known Ikeda infected herds were found in four regions (Mid-Coast, Cumberland, New England and North Coast).
Significance of T. orientalis and its MPSP types in affected herds
The significance of the various theilerial types was examined by testing cattle within seven NSW herds with a history of clinical theileriosis and introductions of cattle. In these herds, Ikeda type was detected in all and Chitose in six. Pale and jaundiced mucosal surfaces were associated with clinically affected groups of cattle, and herds showing numbers of cattle with >=1% theilerias in erythrocytes on smear examination were associated with high prevalence of Ikeda type, with or without Chitose type. Further, for detection of theilerial infection, PCR testing of EDTA blood was more sensitive than traditional smear examinations, which can also be subjective, particularly when infection rates are low.
Infection with Ikeda and Chitose organisms was detected in calves as young as 1-2 weeks, and rapidly increased in prevalence within one month, with Ikeda prevalence increasing at the fastest rate. Direct contact with weaners introduced from coastal areas appeared a risk factor, with high prevalence rates detected in a single time point bleed in purchased adults on one property. Spread within herds appeared variable, and depended on the sampling time point. Cattle without contact with affected mobs may show no evidence of infection, while in others, the lack of direct contact with affected cattle had not prevented infection with the Ikeda type. Within groups showing infection, PCR testing showed repeatedly similar results at multiple samplings at one month intervals, so frequent sampling may be unrewarding once high prevalence is established.
Clinical pathology findings in diagnostic cases suspected to have T. orientalis infection
To investigate relationships between clinical pathology and findings in the conventional PCR assays, we examined diagnostic cases suspected to be infected with T. orientalis that were collected between Sept 2010-Nov 2011. 350 EDTA blood samples from cattle in 73 herds in NSW and Victoria were examined. These showed that beef cattle were predominantly affected, with Angus and Angus-crossbred cattle representing 48% of smear positive cases examined. Results of PCR tests for genes encoding the p32, Ikeda, Chitose and Buffeli MPSPs were compared with clinical pathology on stained blood smears (for parasitaemia) and packed cell volume (PCV). The PCR testing in these samples was much more sensitive than clinical pathology examinations in detecting T. orientalis infections, and the modified procedure of concurrent testing of neat and diluted extracts gave significantly more PCR positive results than testing of neat extract alone.
Significant associations and correlations were shown between PCR results of p32 and Ikeda assays with PCV levels indicative of anaemia, and with the level of parasitaemia estimated by smears. A high proportion of clinical cases had concurrent Ikeda and Chitose infection, and significantly more clinical cases of theileriosis were associated with the Ikeda MPSP type as the sole infection, compared with sole infection with types Chitose or Buffeli. These findings indicate Ikeda type organisms were significantly associated with clinical parameters of theileriosis in cattle herds in eastern Australia, and that this type is most likely to be responsible for outbreaks of theileriosis experienced in affected Australian herds. In New South Wales, 11 of 14 regulatory districts yielded Ikeda positive cases, with five (Mid-Coast, Cumberland, Central North, Hume and Lachlan) containing 234/307 (76%) of the Ikeda positive cases in this study period.
Development and application of quantitative (real time) PCR assays
In research studies of the T. orientalis MPSP gene, we have identified separate sequence targets suitable for qPCR that are theoretically specific for T. orientalis as a universal target, and specific for Ikeda, Chitose and Buffeli gene sequences. Our approach has been to develop a sensitive and specific fluorophore-based detection system (Taqman type assay). In our approach a DNA extract from a T.orientalis-positive sample results in an amplified product which is then exposed to specific probes with different fluorophore labels. These probes are able to attach at the specific DNA sites that encode a Universal and either an Ikeda, Chitose or Buffeli sequence in a multiplex assay. In samples that contain more than one MPSP type, multiple signals are generated. Our studies have intensively evaluated a set of approximately 50 samples known to be positive to different MPSP types by conventional PCR. In qPCR, assessment of DNA concentrations require DNA standards for each target. Since field material often contains multiple MPSP types, "purified" DNA was also manufactured by cloning Ikeda, Chitose and Buffeli genes to produce recombinant plasmids for each type. These plasmids are thus used as standards in each assay so the amount of specific DNA of each MPSP type in the test samples can be quantified.
In these studies, we have developed a reliable multiplex qPCR that detects Universal, Ikeda and Chitose elements, and a separate Buffeli qPCR. We have also demonstrated that test costs can be reduced using an in-house DNA extraction procedure, and are undertaking further work with robotic DNA extraction systems. The multiplexing will be further developed to see if it can also accommodate an internal DNA extraction control, to confirm that a negative result is not due to PCR inhibitors.
Development of ELISA based on recombinant MPSP antigen
We have also used recombinant DNA technology to produce Ikeda, Chitose and Buffeli recombinant MPSP antigens for ELISA. Evaluation was undertaken with sera from cattle where PCR testing had been carried out. A cocktail ELISA (cELISA) based on a mixture of recombinant Ikeda, Chitose and Buffeli type antigens was shown to react strongly with sera from cattle known to be infected with one or more types. Sensitivity analysis was based on 294 sera from cattle with known PCR, smear and PCV status from concurrently-collected EDTA blood. Among samples from 25 herds, 191 of 290 samples from 21 herds showed T. orientalis infection (based on p32 PCR). 50% of those 191 cattle found positive in the p32 PCR and 51% of 182 cases found positive in the Ikeda PCR were seropositive in the cELISA.
In infected herds, 52% of 123 smear positive cattle, 74% of 43 anaemic cattle and 66% of 65 cattle with a parasitaemia of at least 1% were also seropositive. However, in 119 animals that were infected with Chitose type and 46 infected with Buffeli type, only 38.7% and 19.6% respectively were seropositive. In a herd considered free of T. orientalis infection, the cELISA showed high specificity (100%) in 50 tested cattle at the positive cutpoint.
Two additional ELISAs, using only a single recombinant component from either Ikeda or Chitose, were not of higher diagnostic merit than the cELISA when applied to samples that were seroreactive (i.e. classified inconclusive or positive) in the cELISA. These single antigen ELISAs for Ikeda and Chitose respectively detected 65% and 73% of 37 anaemic cELISA reactor cattle, 58% and 64% of 80 smear positive cELISA reactors, and 64% and 73% of 52 cELISA reactors with a parasitaemia of 1% or higher. This may be partly due to the common finding of concurrent Ikeda and Chitose infection in cattle in herds affected with anaemia due to T.orientalis.
A temporal ELISA study on a dairy herd near Taree confirmed high positive cELISA responses in 10 naive 2 year old dairy heifers when bled at 34 days after introduction to a herd known to be affected with T. orientalis. These heifers remained seropositive until at least 76 days (Fig. 3).
A very high seropositive response at 34-40 days corresponded with the time of anaemia and peak parasitaemia and maximal prevalence of T. orientalis infection due to mixed infection with Ikeda, Chitose and Buffeli MPSP types. Infection with Ikeda type also showed a faster rate of infection at 20 and 34 days compared with the other types (Fig. 4). The cELISA may have merit in large scale studies to identify higher risk herds for further characterisation by molecular assays.
The modified conventional PCR tests were shown to be robust across very large numbers of samples, more sensitive than existing methods and clearly able to detect the major theilerial types. The p32 PCR was also confirmed as a valuable screening assay, as very few samples were found positive in the type specific PCRs that were not also p32 PCR positive. The importance of the Ikeda type is shown from the finding that this type was significantly associated with clinical disease, and sole infection with an Ikeda strain was found significantly more often compared with sole infections with either or both Chitose or Buffeli types. However, combined Ikeda and Chitose infection was by far the most common finding in affected cattle.
In clinically normal cattle in affected herds with Ikeda type infection, evidence of the value and sensitivity of PCR was demonstrated by the finding that over half of the smear negative animals were detected as infected with Ikeda type on PCR testing, while 90% of smear positive cases gave PCR positive results for Ikeda type. This shows there is a large reservoir of infected but clinically normal animals in affected herds.
This work included the first structured survey of cattle herds in the key states where traditional vectors of theilerial strains are known to occur (and where the avirulent Buffeli type may be present). The results indicate that introductions to cattle herds from areas where possible virulent strains of Ikeda occur can be focussed on particular regions in NSW (to include regions where most disease was recorded from diagnostic submissions and the North coast area detected in the survey), as well as multiple regions in Queensland, and certain coastal regions of Victoria. In addition it is clear that a very high proportion of young calves and adult cows in an affected herd, or in mobs of cattle that have been in contact with affected or high risk cattle, can be infected with Ikeda and Chitose type organisms, often but not always concurrently. The progress of infection can be rapid, despite the normal clinical appearance of the stock.
These studies have given a better understanding of epidemiological aspects from both a regional perspective and from the herd (infected or high risk) perspective, and will assist diagnostic laboratories to provide better tests to define infection status of different groups of cattle of variable risk status. The PCR assays are important in evaluating animal status when assessing alternate tests, such as serological assays, and the impact of treatment and preventative strategies such as therapeutic drugs or vaccines designed to reduce or eliminate infection.
In addition, the numerous blood samples and DNA extracts characterised in this work will provide a resource of future materials to investigate possible genetic differences between strains of T. orientalis that are currently categorised into the same type by PCR. These samples will also be valuable in improving diagnostic tests in the future, and developing more cost-effective diagnostic assays that can be used to rapidly assess and even quantify the level of infection with qPCR.
Funding from Meat and Livestock Australia and the McGarvie Smith Institute is gratefully acknowledged. Graham Bailey was integrally involved in the design and coordination of the MLA Theileria project, as well as contributing to sample collection. Jocelyn Gonsalves and Shayne Fell provided excellent technical support in undertaking the conventional PCR and qPCR analyses in the above work. Andre Poublon provided additional technical support for the MLA project. Colleagues interstate (QDAFF, Victorian DPI) and numerous DVs provided samples for the distribution study, while the significance study was particularly assisted by inputs from Drs Elizabeth Braddon, Helen McGregor, Ian Poe, Bruce Watt and Steve Whittaker. Dr Alan Glassop provided valuable samples for the temporal PCR and ELISA study in the Taree area.