Br. vet../. (1996). 152, 139

REVIEW

EQUINE PIROPLASMOSIS AN UPDATE O N DIAGNOSIS, TREATMENT AND PREVENTION

A. BRONING* Impelqal CoUege of Science Medicine and Technolo~,, Department of Biology, West Wing,

P~qnce Consort Road, London SW7 2BB, UK

SUMMARY

Two haemoprotozoan parasites, Babesia cabaUi and Babesia equi, can cause equine piroplasmosis. Due to the presence of potential tick vectors in areas so far unaffected by equine babesias, import and export regulations often require the serum testing of animals for evidence of infection. Although the complement fixation test (CFT) has been recommended for detecting the presence of antibodies to Babesia spp., it has been demon- strated to have several disadvantages, including false-positive results and low sensitivity for detecting latent infections. An enzyme-linked immuno- sorbent assay (ELISA) may be an alternative for increased and sensitive detection of acute and latent babesial infections, but its development to date has been hindered by a limited antigen supply and poor specificity. h~ vitro cultivation of both parasite species and the identification of para- site proteins for diagnostic use has facilitated the development of a highly sensitive and specific ELISA. For the direct detection of the parasites, DNA probes are now available.

Several drugs are available for the treatment of equine piroplasmosis. For instance, diminazene diaceturate is effective in the chemosterilization of B. caballi and in the elimination of clinical signs in B. equi infections. Antitheilericidal drugs such as buparvaquone have been demonstrated to be effective in combatting disease due to B. equi and may--in combination with imidocarb--also eliminate the parasite. The control of equine piro- plasmosis must include effective tick control, seromonitoring of animals and the application of chemotherapy.

KzvwoRDs: Equine piroplasmosis; Babesia cabaUi; Babesia equi; diagnosis; treatment.

*Present address: Institute for Animal Health, Pirb,'ight Laboratory, Ash Road, Pirbright, Surrey, GU24 ONF, UK.

0007-1935/96/020139-13/$12.00/0 © 1996 Bailli~re Tindall

140 BRITISH VETERINARY JOURNAL, 152, 2

INTRODUCTION

Equine piroplasmosis is an often fatal, tick-borne disease of equines which is caused by two haemoprotozoan parasites, Babesia caballi and Babesia equi. Cases of equine piroplasmosis are reported throughout the world where the tick vectors are found. B. caballi and B. equi may occur together when a common vector is pre- sent but, in general, infections due to B. equi are more common than those due to B. caballi (Ershov, 1956; Levine, 1973; Friedhoff, 1982). Both species of parasite can be found in most tropical and subtropical areas of the world as well as in temperate climatic zones. Endemic areas include many parts of Europe, Africa, Arabia, and Asia (except Japan). Countries at risk of becoming exposed to the parasites due to increasing movement of horses include the USA, Canada, Australia, New Zealand and Japan. Ahhough epidemic foci in Florida of B. caballi (1961) and B. equi (1965) (Zaugg & Lane, 1992) and the introduction of B. equi (1976) into Australia after the importation of infected horses have been reported (Churchill & Best, 1976; Callow et al., 1979), the USA and Australia are still regarded as areas free of equine babesias and therefore at risk.

In March 1969, the complement fixation test (CFT) was introduced as the official serodiagnostic test for equine piroplasmosis (Friedhoff, 1982). It has been shown over the years, however, that in addition to a number of technical prob- lems, the CFT does not detect latent infections and may also give false-positive results (Friedhoff, 1982; Weiland et al., 1984; Weiland, 1986). As equine piro- plasmosis is becoming a disease of increasing international importance, and a major constraint to the movement of horses involved in events such as the Olympic games, this review will report on developments in its diagnosis, treatment and prevention.

THE PARASITES

The taxonomic status of the piroplasms remains unclear, ahhough they can be definitely assigned to the phylum Apicomplexa. Difficuhies arise at the ordinal, family and generic levels.

In the vertebrate host, the parasites undergo an asexual cycle of schizogony or bina~ T fission, while the remaining development occurs in the invertebrate hosts (ticks). Two genera, Babesia and Thdleria, are classified in the family Piroplasmi- dae, which is variously positioned within the phyltnn according to different authors. B. caballi is regarded as a true Babesia since in the vertebrate host it only undergoes an intraerythrocytic cycle with the parasites dividing into two parasites (merozoites). Characteristics of B. equi which distinguish it from many other species of Babesia are (1) its frequent division into four merozoites in e~ythrocytes--a feature that stimulated Franca (1909) to denote genus Nuttalia for it; (2) that transmission by ticks is transstadial (Friedhoff, 1982) and (3) that there is development in lymphocytes before the erythrocytic cycle, like species of Theileria (Schein et al., 1981; Moltmann et al., 1983).

Drug sensitivity studies have shown that the exoe~3,throcytic schizonts of B. equi are highly susceptible to schizoticidal drugs such as halofuginone, parvaquone, and oxytetracycline used in the treatment of boxfne theileriosis but not to the

EQUINE PIROPLASMOSIS 141

babesicidal drug diminazene diaceturate which is effective for the treatment of B. cabaUi (Zweygarth et aL, 1984). Recent phylogenetic studies using small subunit ribosomal RNA (sRNA) genes of species of dot:lbfful systematic position (including B. equi) suggest that B. equi belongs to a distinct paraphyletic group (Allsopp et al., 1994). There are thus good reasons for transferring B. equi to the genus Theileria or even for establishing a new genus. (Moltmann et al., 1983; Mehlhorn & Schein, 1993; Allsopp et al., 1994).

The establishment of short-term and long-term in vitro cultures of B. caballi (Canning & Winger, 1987; Br~ning, 1994) using the microaerophilous stage phase (MASP) technique for Babesia boris (Levy & Ristic, 1980) and the long-term in vitro cultivation of B. equi using an enriched chemically defined basal medium and serum replacement factors (Holman et al., 1994) have been recently reported. The ability to grow the parasites in cultures facilitates the in-depth study of the parasites, and particularly the identification of specific proteins for diagnosis, drug and vaccine development.

DIAGNOSIS

Diagnosis of clinical equine babesias or of subclinical infection with the parasites can be either direct or indirect. Direct detection includes the clinical examination of a suspected animal, particularly for evidence of tick infestation and consider- ation of any recent travel. The clinical status is recorded and thick and thin blood smears are examined for parasites. These are usually visible in peripheral blood during the acute phase of infection--particularly before haemoglobinuria appears--but they are generally not detectable during the latent phase. Since the severity of disease may vary depending on the strain of parasite and the suscepti- bility of the animal, clinical signs may be too general and not attributed to piro- plasmosis but to another disease. DNA probes for the detection of B. cabaUi and B. equi have been developed by Posnett and Ambrosio (1991) and Posnett et al. (1991). These workers were able to detect as little as 0.25 ng and 0.125 ng of B. caballi DNA and 0.15 ng B. equi DNA, corresponding to B. cabaUi parasitaemias of 0.12% and 0.06% respectively and of B. equi a parasitaemia equivalent to less than 0.0025%. The authors reported a heightened sensitivity of the probes in compari- son with microscopic examination of blood from naturally and experimentally infected animals as well as from carriers. DNA probes can also detect DNA in the blood of infected ticks (E. S. Posnett, personal communication), however, they have not yet been applied in routine diagnosis. Sensitivity of tests for the direct detection of the babesias could be even more increased by the development of the polymerase chain reaction (PCR) (Posnett & Ambrosio, 1991).

In cases of latent infections, indirect parasite identification or immunodiagnosis is essential. Immunodiagnosis is usually based on the detection of specific anti- bodies or antigen by a variety of immunoassays. None of the currendy available test systems for the detection of equine babesias meets all the requirements of sen- sitivity, specificity, reproducibility, simplicity, cost effectiveness and speed.

The CFT was developed by Hirato et al. in 1945. In March 1969 the test was accepted as the official test for equine piroplasrnosis by the American Department

142 BRITISH VETERINARY JOURNAL, 152. 2

of Agriculture, and is now used worldwide (Friedhoff, 1982). The underlying prin- ciple of the CFF is the fixation of complenaent during the reaction between specific antigen and antibody. If this reaction is carried ont in the presence of a known amount of complement and the remaining unfixed amount is detected with a separate antigen-antibody reaction, it is possible to obtain a measure of the complement fixing activity of the original serum (MAFF, 1986).

In the test system for equine piroplasmosis, B. caballi or B. equi antigen is added to heat inactivated test sera and allowed to react in the presence of a known amount of complelnent. Two units of complenaent are usually added enabling positive sera to fix up to 100% of the complement. After an incubation period, a haemolytic system is added. This consists of a suspension of sheep erythrocytes and a rabbit haemolytic serum diluted to a level which will enable the haenlolysis of 100% of the sheep erythrocytes in the presence of two units of complement. Complement fixation can be assessed by estimation of lysis as a percentage, i.e. 0%, 25%, 50%, 75%, and 100% which are accorded values of 4+ positive, 3+ posi- tive, 2+ positive, 1+ suspicious trace (negative) and COlnplete haemolysis (negative). Horses with sera which react positively at a dilution of 1:5 (i.e. less than 25% lysis) are considered positive. These regulations apply in particular to the USA, Canada, Australia, Japan, Mexico, and Brazil. Horses are retested by CFT on arrival in these countries and, if found seropositive, are either slaughtered or returned at the owner's expense.

The need for serodiagnosis of equine piroplasmosis has been well documented. Investigations in Hannover in 1981 and 1982, for example, showed that, of 765 sera from horses imported into Germany and the Netherlands, 42 (5.5%) were positive for B. equi and B. caballi, and two showed a double infection. The seroposi- tive animals originated from European and Middle Eastern countries (Friedhoff, 1982). Donnelly et aL (1982) reported a positive diagnosis of equine piro- plasmosis, due to one or other parasite or both, in four out of 14 foals born in England to seropositive dams fl'om Poland. When retested at the age of 4 months, CFT results for these animals were negative and it was concluded that the positive reactions were due to maternal antibodies.

A survey carried out at the National Veterinary Smwice Laboratories (USDA), Iowa, between 1986 and 1990 showed that of the 65 911 equine serum samples tested for B. caballi by the CFT, 2880 (4.4%) were positive for the parasites. Also tested were 67 517 sera for B. equi infection, of which 5246 (7.8%) were positive for B. equi (Zaugg & Lane, 1992). Similar results were obtained at the Central Vet- erinm'y Laboratory (CVL), Weybridge. Routine screening of equine sera for piro- plasmosis showed that between 1984 and 1993 of 17 765 samples, 499 (2.8%) were positive by CFT. In 1993 alone, for example, of the 1469 samples tested, 46 (3.1%) were positive for B. equi, three (0.2%) were positive for B. caballi and 12 (0.8%) were positive for both parasites. A further eight (0.5%) samples showed anti-comp- lementary reaction (P. Phipps, personal communication). The CFT detects anti- body titres from day 8 post-infection (p.i.) with the titres declining 2-3 months after infection. After spontaneous or therapeutic elimination of B. caballi or B. equi infections, reactions will become negative within 3-15 months in the case of B. cabaUi (Friedhoff, 1982) and 24 months in the case of B. equi (Weiland et al., 1984).

EQUINE PIROPLASMOSIS 143

It has been found that the CFT has technical drawbacks due to difficulties in the production of the reagents used in the test. For the production of the CFT antigen, for example, horses have to be splenectomized and then infected with the parasites. When a high parasitaemia is reached, whole blood is collected and the parasites are retrieved. For welfare and disease security reasons it would be preferable not to rely on the artificial infection and sacrifice of animals for antigen production. Moreover, there are many records of the occurrence of false- negative resuhs (Donnelly el aL, 1980a, b, 1982). Friedhoff (1982) refers to serum tests of animals, which reacted first negative and then positive. The problem of false-negative resuhs was further exemplified when horses that were negative by CFT still had B. equi parasites in their blood which could be demonstrated by blood smears (Friedhoff, 1982). Sigrist (1983) reported that nymphs of Rhipice- phahts turanicus and H~,alomma anatolicum could be infected with B. equi after feeding on a parasitologically and CFT negative horse. Furthermore, Rehbein and Heidrich-Joswig (1983) observed that the CFT failed to detect specific antibodies to B. equi schizonts. False-negative results also arise due to the anticomplementary action of the sera of some horses (Tenter & Friedhoff, 1986) which then have to be retested by indirect immunofluorescence test (IFAT). Another problem is cross-reactivity. Donnelly et al. (1982) reported that B. caballi antisera also reacted with B. equi antigen in the CFT. With all these drawbacks the CFT simply cannot be considered the "gold standard" for the diagnosis of equine piroplasmosis.

The IFAT is used in cases where the CFT has proved to be inconclusive. It was first used by Ristic and Sibinovic (1964) for the serodiagnosis of B. cabaUi infec- tion. In tiffs assay, parasite antigens are bound to glass slides and allowed to react with test sera. Bound antibodies are made visible under uhra-violet light after addition and binding of a fluorescein-labelled anti-species serum. Sera are con- sidered positive if they show strong fluorescence of the parasites at a dilution of 1:80 and above but estimation of the intensity of fluorescence is subjective. Several authors have examined sera of experimentally and naturally infected horses by comparing the CFT and IFAT. The IFAT was the more sensitive method (Madden & Holbrook, 1968; Friedhoff, 1982; Tenter, 1984; Tenter & Friedhoff, 1986). Weiland (1986) obsel-ved the earliest antibody responses of artificially infected horses to B. caballi or B. equi at day 3-20 post infection, with titres still detectable during the latent phases of infection. IFAT titres are on the whole detected more consistently .than CFT titres and sera remain positive by IFAT longer than by CFT (Friedhoff, 1982; Tenter, 1984). The IFAT can also be used for the detection of bovine, canine, and human piroplasmosis (Reif, 1980) and clearly shows advantages over the CFT for the diagnosis of equine piro- plasmosis. However, the test is time consuming and requires large amounts of antigen. Recent research, therefore, has been directed towards the development of an enzyme-linked immunosorbent assay (ELISA) for the detection of babesial infections.

Early work on other Babesia spp. showed the potential of ELISAs to detect infec- tions in experimentally and naturally infected animals. The diagnosis of infection by ELISA has been reported for bovine babesias including B. boris (Barry et al., 1982), Babesia divergens (Purnell el al., 1976; Young & Purnell, 1980), and Babesia major (Barry et al., 1982), and for canine babesias including Babesia canis (Kratzer,

144 BRITISH VETERINARY JOURNAL, 152, 2

1979; Weiland & Kratzer, 1979) and Babesia gibsoni (Weiland & Kra~er, 1979). This led to the development of ELISAs for the detection of equine Babesia infections.

The first ELISAs to be developed had been shown to be vm T sensitive in com- parison to the CFT with the earliest antibody responses reported on day 2 p.i. by various authors. However, these were not ve D, specific. G6t.z (1982), for example, demonstrated that sera positive for B. equi cross-reacted with B. caballi and Babesia rodhaini antigens. Merkle (1983) reported the cross-reactivity of B. caballi serum with B. equi antigen and Aicher (1984) showed cross-reactions between both species. Peymann et al. (1993) and B6se and Peymann (1994) reported the height- ened sensitivity of ELISA when detergent-treated B. caballi-infected erythrocytes were used as antigen. However, cross-reactions were observed with 20% of the B. equi immune sera tested. These authors recommended the combined use of ELISA and Western blotting for the detection of infection and identification of the causal agent. The problem of low specificity was attributed to the use of crude antigens in the tests, mainly erythrocytes infected with the equine babesias or crude whole parasites. To increase the specificity of the ELISA, recent develop- ments and optimization have been directed towards the identification of species specific antigens. Proteins, which are species specific, immunogenic, and con- served among geographically different strains of B. caballi and B. equi are ideal candidates. The ability to grow equine babesias in equine erythrocytes in vitro has facilitated the identification of such proteins since an unlimited amount of para- sites can be extracted from cultures. Bhushan el al. (1991) and B6se and Daemen (1992), for example, obtained B. cabaUi by the enrichment of B. caballi-infected erythrocytes using Percoll gradients. Furthermore, the combination of CO,_,-depri- ration and sequential centrifugation of MASP cultures of both B. cabaUi and B. equi led to the partial purification of the parasites from the cultures (Brfming, 1994). When this naethod was applied, up to 80% of the parasites in the cultures were recovered and could then be used for immunological studies or stored at -20°C until use.

Several authors have described the identification of babesial proteins that may be of diagnostic use. For B. caballi, B6se and Daemen (1992) analysed proteins of B. caballi (USDA strain)-infected mTthrocytes separated by electrophoresis under reducing conditions and blotted with sera from horses which had been exper- imentally and naturally-infected with B. caballi. Dominant babesial antigens in the range of 141-30 kDa were demonstrated, of which proteins of molecular weight (MW) 141, 112, 70, 50, and 48 kDa were recognized by all the immune sera tested. The proteins of MW 50 and 48 kDa were considered suitable for use in an immun- odiagnostic test for B. caballi infections since a wide range of B. caballi sera had recognized these antigens in Western blots (B6se, 1994).

Proteins of MW 70 and 49 kDa were recognized by inmmne sera from animals of different origin in Western blots of proteins of B. caballi merozoites (USDA strain) purified after CO,,-deprivation and sequential centrifugation. These pro- teins were highly immunogenic and specific in indirect ELISA tests (Br~ning, 1994).

Knowles el aL (1991a) used equine mTthrocytes infected with the USDA strain of B. equi and sera of naturally and artificially infected horses. Eleven major pro-

EQUINE PIROPLASMOSIS 145

teins ranging from 210-25 kDa were identified in immunoprecipitated prep- arations which had been separated by electrophoresis under reducing conditions. Ali et al. (1993) used B. equi (USDA strain) merozoites separated by Percoll density gradient centrifugation. After electrophoresis under reducing conditions, the pro- teins were immunoblotted with immune sera to B. equi. A range of proteins from 96-15 kDa was detected which were species specific. B6se and Hentrich (1994) identified nine major antigens of MW 41-19 kDa in B. equi (USDA) strain infected erythrocytes by Western blotting, of which only four were recognized by sera from field and experimentally infected European horses. A protein of MW 34 kDa of partially purified merozoites of B. equi (OP strain) was recognized by immune sera from naturally and artificially infected horses from different countries after West- ern blotting. This protein reacted specifically in indirect ELISA after purification from SDS-PAGE gels by excising and electroelution (Br/~ning, 1994).

A monoclonal antibody (mAb)-based competitive ELISA (cELISA) has been developed by Knowles et al. (1991a, b). A mAb (36/133.97) was selected from a panel of mAbs, which reacted with a 34 kDa surface protein of B. equi also detected by immune sera of horses infected with different strains of the parasite. In this test, antibodies in test sera compete with the mAb for the antigenic site in the merozoites. Positive titres are detected by a reduced colour reaction. The test proved to be more sensitive than the CFT and half of the results obtained for samples from artificially-infected horses, which were cELISA positive and CFT negative, were also positive when the samples were retested by immunoprecipit- ation of in vitro translation products from B. equi mRNA. The use of a mAb ensures high specificity making the retesting of samples by Western or IFAT unnecessary. The cELISA was optimized by the same authors with a recombinant protein instead of culture derived whole parasites (Knowles et al., 1992). The cEL- ISA detected latent infections of experimentally-infected horses which were not detected by CFT. The use of a recombinant antigen will facilitate standardization of the test and make unnecessary the long-term in vitro cultivation of the parasites or the artificial infection of horses for antigen production.

Feasibility studies for the development of an ELISA for the detection of B. caballi infections based on a mAb to the B. cabaUi protein of MW 70 kDa have been described elsewhere (Braining, 1994) and demonstrated that such an ELISA can be more sensitive than the CFT for the detection of B. caballi infections.

The "gold standard" for diagnosis of equine piroplasmosis should thus be a test that is (1) sensitive enough for both the early detection of acute infections and for the detection of latent infections; (2) specific for the differentiation between the two parasite species, and (3) economical with regards to materials and time. The CFT does not fulfil these criteria since it has been shown to be insensitive, non- specific, and short-lived with regards to materials. The IFAT is nearer to the ideal considering its sensitivity and specificity, but cannot be considered an economical test for routine testing. The same applies to Western blotting which may be a good means of confirmation of the causative agent. The ELISA, especially a mAb-based ELISA, offers the chance of a "gold standard" for the diagnosis of equine piro- plasmosis with regards to sensitivity and specificity and the availability of reagents. Monoclonal antibody based ELISAs could be easily commercialized and distrib-

146 BRITISH VETERINARY.JOURNAL, 152, 2

uted throughout the world making the standardized diagnosis of Babesia infec- tions possible.

TREATMENT AND PREVENTION

The treatment of equine babesiosis is aimed at the elimination of parasites from horses severely affected by the disease, or from horses travelling fi'om endemic regions into disease fi'ee areas.

For the treatment of piroplasmosis there are a number of drugs available. Bisazo dyes such as Trypan blue, which affect B. caballi but not B. equi tend to dis- colour many of the animal's tissues and are not favoured. Other drugs which are effective in eliminating B. cabaUi infections include diminazene diaceturate (Berenil, Hoechst), amicarbalide (Diampron, May & Baker) and imidocarb (Imizol, Mallinkrodt) (Friedhoff, 1982). However, diminazene diaceturate causes swelling and necrosis at the injection sites. Respiratory distress, depression and other signs of intoxication are often the result of toxic doses and must be carefillly monitored.

None of the babesicidal drugs are satisfactory for the elimination of B. eq.zti infections but may be effective in combatting disease. A horse which has been cleared of clinical signs may remain a life-long carrier of the parasites (Holbrook, 1969). Administration of hydronaphtacenes e.g. chlortetracycline hydrochloride (Aureomycin, Lederle Laboratories) and oxytetracycline hydrochloride (Terramycin, Pfizer) is only effective against B. equi with a daily intravenous injec- tion for 2 or more days at 5.5 mg kg -I (Janssen, 1953). Since the severity of disease may vary depending on the parasite strain and susceptibility of the affected animal, blood and /o r dextrose infusions and the intravenous administration of phospholipids, antibiotics and laxatives may he necessary (Janssen, 1953). Polyionic electrolyte infllsions should be given when animals are suffering from dehydration, anorexia or diarrhoea. Preliminary studies have shown that parva- quone (Clexon, Cooper), and buparvaquone (Butalex, Pitman-Moore Ltd), can be effective in the treatment of acute B. equi infections in horses (Kuttler et al., 1987; Zaugg & Lane, 1999). Since the greatest actixdty of these drugs has been reported to be against the schizont stages of the parasites, Zaugg and Lane (1992) suggested the combined use of bupara, aquone and imidocarb for the elimination of B. equi infections. Further trials are under way to determine the effect of other ttieilericidal drugs on B. equi and their ability to eliminate infection.

Some drugs may also be used in both B. caballi and B. equi infections. Acridine dyes e.g. euflavine (Gonacrine, May & Baker) and others in this group can be administered intravenously at a recommended dose of 4-8 ml 100 kg -j of a 5% solution with a maximum inoculum of 20 ml. However, there have been cases of resistance of B. equi to euflavine (Janssen, 1953). Aromatic diamidines and related compounds including diminazene diaceturate, amicarbalide diisethionate and imidocarb dipropionate are also effective against both parasite spp. (Frerichs & Holbrook, 1974). The mechanisms of drug action of man), of these drugs is not known, but in the case of imidocarb ultrastructural changes of B. equi have been recorded after drug administration (two doses of 5 mg kg -~) in ponies (Simpson &

EQUINE PIROPLASMOS1S 147

Neal, 1980). Dissolution of the parasite's nuclear envelope, vacuolation Of its cyto- plasm and concentration of chromatin were observe d by these authors. Recent studies have shown that steroid-immunosuppressed horses which had been exper- imentally infected with both B. equi and Ehrlichia equi were less affected by either pathogen than when infected with only one of the two species (Oladosu & Olufemi, 1992). This phenomenon was attributed to immunostimulation by cross- immunization and may provide opportunity for the development of a vaccine or prophylactic drugs.

Piroplasmosis should also be controlled by the prevention of tick vector-equine host interaction by, for example, applying acaricides. This may include the use of acaricides in phmge dips, spray races or by hand sprays. The elimination of tick vectors may be achieved by the regular and frequent use of acaricides. In free- ranging systems where animals are widely scattered and there is seasonality of tick infestation it is more feasible to allow an enzootically stable condition to develop by the strategic use of acaricides at times when animals are not being moved. So far no vaccines for equine piroplasmosis are available.

CONCLUSION

The growth of international equine sport and trade in horses has increased the risk of the introduction of pathogenic parasites into previously unaffected areas. For equine piroplasmosis this was demonstrated in 1961 when B. caballi was found in southern Florida following the importation of 30 ponies from Cuba. The occurrence of the tick vector Dermacenlor nitens in Florida allowed B. caballi to reach epidemic proportions by 1962. The epidemic was confined to Florida but the possibility remains that the disease could be introduced into other states. In Australia, B. equi was introduced in 1976 with a shipment of infected horses from Spain. The parasites did not spread and to date Australia remains free of equine babesias due to the absence of a suitable vector. However, transmission of B. equi from the infected imported horses to susceptible Australian horses with the use of contaminated needles led to disease.

So far, no vaccine has been developed, emphasizing the need for accurate diag- nosis and treatment of equine piroplasmosis. DNA probes may be useful in the routine direct diagnosis of babesial parasites in the horse. The currently used diag- nostic tests (CFT, IFAT) have disadvantages such as low sensitixdty and cross-reac- tivity or are not cost effective on a large scale. Thus there is a need to develop a sensitive and specific serodiagnostic test such as an ELISA. The development of long-term in vitro systems for both equine Babesia spp. ensures an unlimited supply of antigen and has facilitated the study of both parasites. Further investigations are needed to determine differences among the parasite species or differences within strains of each species. Protein profiles of geographically different strains of parasites may give a better understanding of their relatedness and the conser- vation of proteins, which may be of diagnostic value. International collaboration and the exchange of results in this field of research is especially important, since most veterina~ T institutions have access only to local strains of parasites and immune sera.

148 BRITISH VETERINARY .IOURN/~tL. 152, 2

Monoclonal antibody-based competi t ive ELISAs have been deve loped for both Babesia spp. and may replace the official serodiagnostic test, the CFT, as they mee t the r equ i rement s o f specificity, sensitivity, simplicity, and cost effectiveness. Th e early de tec t ion of acute and latent infections and the identif ication of the causa- tive agent of infect ion should also improve the therapy of infections.

The quest ionable taxonomic status o f B. equi has added a new dimension to the chem othe r a py o f equ ine Babesia infections. Theiler icidal drugs are promising can- didates for combat t ing disease due to B. equi, particularly impor t an t since animals may be life-long carriers of this parasite spp. T h e control of equine piroplasmosis should integrate prevent ion of infect ion by the applicat ion of acaricides and grazing policies with the use of specific and sensitive diagnostic tests and newly deve loped drugs.

ACKNOWLEDGEMENTS

I would like to thank Professor Elizabeth Canning for her guidance and support . Fu r the r more I would like to thank Paul Phipps for supplying me with the data of the serological survey carr ied out at the CVL and for his helpfld advice.

REFERENCES

AICHER, B. M. (1984). M6glichkeiten der serologischen Differenzierung der Babesia equi und Babesia caballi-Infektion des Pferdes mit Hilfe yon KBR, IFAT, nnd ELISA. Dissertation. Veterinal T Faculty, University of Munich.

At.I, S., St:(aMoro, C., MATSt'DA, M., St~(;w~, T., K~NEMARU TONUMA, M. & K~,xl..xt)..x, M. (1993). Protein characterization of Babesia equi piroplasma isolated from infected horse elythro- cytes. Parasitolo~' Research 79, 639-43.

At.LsoI, p, M. T. E. P., CAV.~UFR-S.xHTH, T., DEW,cal., D. T. & At.tSOl, P, B. A. (1994). Phylogeny and evolution of the piroplasms. PamsitoloKv 108, 147-52.

BARm; D. N., Rom~:u., B. J., Ti.xt,xls, P. & McGRE(;OR, W. (1982). A lnicroplate enzyme immunoassay for detecting and measuring antibodies to Babesia boris in cattle serum. Australian Vele~naT),Journa159, 136-40.

BI4USHAN, C., M(q~LER, I. & FRII:;DHOFF, K. T. (1991). Enrichment of Babesia cabaUi infected elTthrocytes from microaerophilous stationary-phase cultttres using Percoll gradients. Parasitology Research 77, 177-9.

BOSE, R. (1994). Polyclonal antibody characterization of Babesia caballi antigens. [nte~ national Journal for Parasitology, 24, 511-7.

BOSE, R. & DAEMEN, IL (1992). Demonstration of the humoral ilnmune response of horses to Babesia caballi by Western blotting. International Journal of Parasilolo~, 22, 627-30.

BOSE, R. & HENVmCH, B. (1994). Identification of antigens diagnostic for European isolates of Babesia equi by two-dimensional electrophoresis and Western blotting. Parasitology Research 80, 182-5.

BOSE, R. & PE~.~.-\Nx, B. (1994). Diagnosis of Babesia caballi infections in horses by enzyme- linked immunosorbent assay and Western blot. International Jour~ml for Parasitology 24, 341-6.

BRC'.X~NC, A. (1994). Serodiagnosis of equine piroplasmosis by ELISA. PhD Thesis, University of London.

C~LOW, L. L., McGRE(;OR, W., RODWELI., B.J., ROGERS, R.j., FI~-\SER, G. C., MAHONEY, D. F. & RO~ER'rSON, G. M. (1979). Evaluation of an indirect fluorescent antibody test to diagnose Babesia equi infection in horses. Australian Veterinmy Journa155, 555-9.

EQUINE PIROPLASMOSIS 149

CANNIN(;, E. U. & WIN(;ER, C. M. (1987). Babesiidae. In In Vitro Methods for Parasite Culti- vation, eds. A. E. R. Taylor &J. R. Baker, pp. 212. London: Academic Press.

Cm:~cml.l., R. C. & BLsr, D. R. (1976). Babesiosis of a horse in Australia. Australian Veteri~mu Journa152, 487.

DONNI'ZI.LV, J.,Jm~E~, L. P. & FRANK, C. (1980a). Quantitative epidemiological studies on the prevalence of babesiosis in horses in Kuwait. TrapicalAnimal Health Production 12, 253-8.

DONN~:Ia.V, J., .]m~l.:R, L. P., Gm-xltA,~l JONES, O. & ELHS, C. P. (1980b). A comparison of the complement fixation and immunofluorescent antibody tests in a survey of the preva- lence of Babesia equi and Babesia cabaUi in horses in the sultanate of Oman. Tropical A nimal Health ProductioTz 12, 50-60.

DONNIiI.I), ', J., Plllpps, L. P. & WATKINS, K. K. (1982). Evidence of maternal antibodies to Babesia equi and B. caballi in foals of seropositive mares. Equine Vetel~na U .]oul~al 14, 126-8.

E~slu)v, V. S. (1956). Parasitology and parasitic diseases of livestock. English translation from the Russian by Birrow A., Hechter H. G. & LengyJ. I. (1960). Israel Program for Scientific Translations.

Fp~.xxc..\, C. (1909). Sur la classification des piroplasmes et description de deux formes de ces parasites. Arquiv lnstitulo BacteHlo~co Camera Pestana 3, 11-8.

Fm-RICHS, W. M. & HOH*ROOK, A. A. (1974). Treatlnent of equine piroplasmosis (B. cabalh) with imidocarb dipropionate. Veterina O, RecoM 95, 188-9.

Fm~:lmovv, K. T. (1982). Die Piroplasmen der Equiden- Bedeutung flu" den internationalen Pferdeverkehr. Bediner und Manchener 7)'er:i~ztliche Wochenschrif195, 368-374.

Gorz, F. (1982). Untersuchulagen i'lber die Brauchbarkeit yon ELISA, IFAT, IHA und KBR zum Nachweis yon Babesia equi Infektionen, Dissertation. Veterinm y Facnl~, University of Munich.

Hm.-vro, K., N~xo.xm:~, M., Uw..\No, Y. & Kcqll, T. (1945). Studies on the complement fixation ,eaction for equine piroplasmosis, flqmneseJournal of Veterina U Sciences 77, 204-205.

HoH~Rootq A. A. (1969). BioloD, of equine piroplasmosis. Journal ofAme~4ean Veterina U and Medical Association 155, 453--4.

HoI.,M:xN, P.J., Cml-v~:s, L., F~I-RU:IJS, W. M., Olsox, D. & W.-\¢;XI.:R, G. G. (1994). Babesia equi eD.'throcytic stage continuously cultured in an enriched medium. Jou~al of Parasitolo~, 80, 232-6.

jaxss~:×, B. C. (1953). The parasiticidal effect of aureomycin (Lederle) on Babesia equi (Laveran 1899) in splenectomized donkeys. OnderstepoortJoumal of Vete14na U Research 26, 175-82.

I'MN()WI.ES, D. P., Lowl.:Ll., D. P., K-kPI'MI:~'I.~R, S., 8TII.I.ER, D., HENN.-\(;ER, S. G. 8c PERR'~MAN, L. E. (1992). Antibody to a recombinant merozoite protein epitope identifies horses infected with Babesia equi../om~al of Clinical Mio'obiolo~, 30, 3122-6.

tLxowH.s, D. P., P~:~ma~:\x, L. E., GOFF, W. L., Mna.eR, C. D., H.-\~mxcsox, R. D. & GoRHa,~,J. R. (1991a). A monoclonal antibody defines a geographically conserved surface epitope of Babesia equi merozoite. Infection and Dnmunit~, 59, 2314-7.

K.',:oWl.liS, D. P., PI-:I,',R~IAN, L. E., K.\ppMI-\I:R, L. S. 8,z HI~:NNA(;I-R, S. G. (1991b). Detection of equine antibody to Babesia equi merozoite proteins by a monoclonal antibody based com- petitive inhibition ELISA..]ournal of Clinical Microbiolo~, 29, 2056-8.

I~J~\rz~.:R, I. (1979). Die Brauchbarkeit der Babesia rodhaini Antigene zum Nachweis von Babesia Infektionen im ELISA. Dissertation. Veterinm T Faculty, University of Munich.

Ktcm.:R, K. L., Zac¢;¢;, J. L. & Gwsox, N. C. A. (1987). Imidocarb and parvaquone in the treatment of piroplasmosis ( Babesia equi) in equids. American Journal of 17ete~na O' Research 48, 1613-16.

Ll~:VlXl.:, N. D. (1973). The piroplasmasida. In Protozoan Parasites of Domestic Animals and of Man. 2nd Edn. pp. 317-46. Minneapolis, MN: Burgess Publishing.

L~.:\~', M. G. & RisNc, M. (1980). Babesia boris: continuous cultivation in a microaerophilous stationm T phase culture. Scie~ce 207, 1218-20.

M,-\DD~.:x, P. A. & HOLB~OOK, A. A. (1968). Equine piroplasmosis: indirect fluorescent anti- body test for Babesia caballi. America~t Journal of Veterina O, Research 29, 117-23.

MI:HI.HOI~,N, H. & SCHEIN, E. (1993). The piroplasms: "a long StOl T in short" or "Robert Koch has seen it". European]ou~ml of Pmlistolo~, 29, 279-93.

150 BRITISH VETERINARY .JOURNAL, 152, 2

MERKI.E, F.J. (1983). Umersuchungen fiber die Brauchbarkeit von KBR, IFAT ulld ELISA zum Nachweis der Babesia caballi-Infektion des Pferdes. Dissertation. Veterinal'y Faculty, University of Munich.

MINISTRV OF A(.RI(:UI:Ft'RE, FISIII-RII-S AND Foot> (MAFF) (1986). Manual :?f Veterina O, Parasito- logical Laboralo O, Techniques. 3rd Edn, pp. 138-45. London: H MSO.

MOI.TMA,NN, Y. G., Mr)It.llOR.,X, H., S(:HEIS, G., Vol(;'l', W. P. & ZWI.:WA~TII, E. (1983). Babesia equi (Laveran 1901). 1. Developlnent in horses and in lymphocyte culture. Zeitsch14fl der 7)op- enmedizin und Parasitologie 32, 223-7.

Ot..~DOSt,, L. A. & Ol.t'vt.:.~ll, B. E. (1992). Haematolo~, of experimental babesiosis and ehrlichiosis in steroid ilnlmmosuppressed horses. Jom'~ml of Vete)4na) 3, Medicine 39, 345-52.

PI:.~.~IAXN, B., BOSE, R. & FRIEI~IIOVV, K. T. (1993). Development of an ELISA and Westel'n blot for the diagnosis of Babesia caballi infections in horses and compalison with ilnlnUllotluo- rescence antibody test (IFAT) and complelnelat fixation test (CFT). In 14th l~tler~latimtal Co~!ference of the World Association for the Advancement ~?f Veteri~m O, Pamsitolo©,. Cambridge, UK.

POSNE'VI', E. S. & AMI~,R()SIO, R. E. (199l). DNA probes for the detection of Babesia caballi. Parasitoloff~' 103, 357-65.

PossE-i-r, E. S., FEIIRSES,J., DI. W.a..~l., D. T. & AxmROSlO, R. E. (1991). Detection of Babesia equi in infected horses and carrier animals using a DNA probe, l:eterina O' Parasitolo~' 39, 19-32.

Pt'RSI-:EI., R. E., HENDRV, D.J., BIDWI-:I.I., D. E. & Tt'RI', P. (1976). Microplate enzynae-lilaked imlnunosorbent assay for antibody to Babesia divergens in cattle. The Veterinm), Record 99, 102.

Rr:IIBEIS, G. & Hr:ll~Rl~:lI-Joswl¢'., S. (1983). Use of schizont and piloplasln antigens of Babesia equi in the indirect inmaulaofluorescelace antibody and complemelat fixation tests. Veterin- my Parasilolo~, 12, 135-44.

REly, I. (1980). Der serologische Nachweis der Babesia dive~gens Ilafektion des Rindes mit Hilfe der Immullofluoreszelaz, der indirekten Haelnagglutination und dem l lnmunoen- zym test. Dissertation. Veterinm 3, FaculD:, University of Munich.

RISTI¢:, M. & SIBINOVlC, S. (1964). Equine babesiosis. Diagnosis by a precipitation in gel and by a one-step fluorescent antibody-inhil~itiola test. Americml Journal of Vete~4nal), Research 25, 1519-26.

SC:IIEIN, E., R~.I-II~EIN, G., \'rOle;T, W. P. & ZWEV¢;..XR'I'II, E. (1981). Babesia equi (Laveran 1901). 1. Development in horses and in lyml)laocyte culture. Zeitschrift der Tropemnedizin und Parasi- tologie 32, 223-7.

SI¢;RIST, B. (1983). Ubertragung yon Babesia equi durch Hyalomma anatolicum anatolicum und Rhipicephalus tumnimLs. Dissertation. Veterinary Faculg,, University of Hannover.

SIMPSON, C. F. ~g2 NIL~I,, F. C. (1980). Ultrastructure of Babesia equi in ponies treated with imidocarb. A met4canJournal of l:eterina~), Research 41,267-71.

TI-STI-R, A. M. (1984). Serodiagnose experimenteller und natOrlicher Piroplaslneninfek- tionen der Pferde. Dissertation. Veterinal-y FaculD,, UniversiD, of Hannover.

TENTEI/, A. M. & FI/IEDHOFI:, K. T. (1986). Serodiagnosis of experimental and natural Babesia equi and Babesia caballi infections, l:eterinm 3, Parasilolo~, 20, 49-61.

WEIt_XNI), G. (1986). Species-specific serodiagnosis of equine piroplaslna infections by lnean of complement fixation test (CFT), i lnmunofluorescence (IF), and enzyme-lilaked imlnunosorbent assay (ELISA). Vete~fnm), Parasitolog 3, 20, 43-8.

WEIL~X'D, G., AJCHER, B. M. & BOON, J. (1984). Serodiagnostik und Therapiekontrolle der Pferdepil-oplasmose mit KBR und IFAT. Berliner u~d M~lm'hener Tier6~ztliche Wochensch~4fi 97, 341-9.

~¢V'I-II.-\ND, G. 8~2 Ka'L4.TZI:.R, I. (1979). Fluoreszenz und enzymserologische Ulaterst~chungela zum Nachweis der latenten Hundebabesiose. Berliner und Mfinchener Tierii~ztliche Wochensch~4fi 92, 398.

YouNg;, A. S. & PLTg.XrI.I., R. E. (1980). Evaluation of dried blood samples as a source of anti- body in the micro ELISA test for Babesia divergens. The Veterinmy Reco~ff 106, 60.

ZAU(;G, J. L. & L\sE, V. M. (1992). Efficacy of bupal'vaquolae as a therapeutic and clearing

EQUINE PIROPLASMOSIS 151

agent of Babesia equi of European origin in horses. American Journal of Veterinary Research 8, 1396-9.

ZW~:VC;ARVH, E., PuxM~:t~, j . s. & RV, HBEIN, G. (1984). The effect of halofuginone, Wellcome 993c, oxytetracycline and diminazene diaceturate on Babesia equi infected lymphoblas- toid cell cultures.Journal of Parasitology 70, 542-4.

( Accepted for publication 10 October 1995)