Molecular studies on Babesia, Theileria and Hepatozoon in southern Europe Part I. Epizootiological aspects

Transcription

1 Veterinary Parasitology 113 (2003) Molecular studies on Babesia, Theileria and Hepatozoon in southern Europe Part I. Epizootiological aspects A. Criado-Fornelio a,, A. Martinez-Marcos a, A. Buling-Saraña b, J.C. Barba-Carretero b a Laboratory of Parasitology, Microbiology and Parasitology Department, Faculty of Pharmacy, University of Alcalá de Henares, Alcalá de Henares, Spain b Dr. Barba Veterinary Diagnosis Laboratory, Segovia 45, Madrid, Spain Received 12 November 2002; accepted 25 January 2003 Abstract Molecular epizootiology of piroplasmids (Babesia spp., Theileria spp.) and Hepatozoon canis was studied in mammals from southern Europe (mainly from Spain, but also from Portugal and France). Partial amplification and sequencing of the 18s rrna gene was used for molecular diagnosis. In some particular cases (B. ovis and B. bovis) the complete 18s rrna gene was sequenced. Blood samples were taken from domestic animals showing clinical symptoms: 10 dogs, 10 horses, 10 cows, 9 sheep and 1 goat. In addition, DNA samples were isolated from blood of 12 healthy dogs and from spleen of 10 wild red foxes (Vulpes vulpes). The results of the survey were the following: Piroplasmid infections: Approximately from 50 to 70% of wild or domestic mammals (symptomatic) were infected. Piroplasmids detected in ruminants were: (1) Cow: B. bovis, T. annulata and Theileria sp. (type C). (2) Sheep and goat: B. ovis. Piroplasmids present in canids were: Babesia canis vogeli, Babesia canis canis, Theileria annae and B. equi. The only piroplasmid found in asymptomatic dogs was B. equi. Piroplasmids found in horse were: B. equi and B. canis canis. H. canis infections in canids: H. canis was absent of domestic dog samples, whereas all foxes studied were infected by this protozoa. Nucleotide sequence data reported in this paper are available in the GenBank TM database under the accession numbers AY150056, AY to AY150064, AY150067, AY and AY Corresponding author. Tel.: ; fax: address: angel.criado@uah.es (A. Criado-Fornelio) /03/$ see front matter 2003 Elsevier Science B.V. All rights reserved. doi: /s (03)

2 190 A. Criado-Fornelio et al. / Veterinary Parasitology 113 (2003) Genetic analysis showed that most of piroplasmid and Hepatozoon isolates from southern Europe matched unambigously with previously described species, as demonstrated by the high level sequence identity between them, usually between 99 and 100%. Minor differences, usually detected in hypervariable regions of 18s rrna gene are probably due to strain variations or rare genetic polymorphisms. A possible exception was B. bovis, which shows a relatively lower degree of homology (94%) with regard to other B. bovis isolates from several countries. The same is true for B. ovis, that showed a 94% identity with regard to Babesia sp. from South African cow and a 92% with rapport to B. bovis from Portugal Elsevier Science B.V. All rights reserved. Keywords: Babesia sp.; Theileria sp.; Hepatozoon sp.; Ribosomal small subunit; Molecular diagnosis; Epizootiology 1. Introduction The arrival of molecular diagnosis methods has led to the discovery of some new piroplasmids as well as Hepatoozon-related organisms in the lapse of just a few years (H. americanum, described by Vincent-Johnson et al., 1997; Theileria annae by Zahler et al., 2000a; B. leo by Penzhorn et al., 2001). Present procedures for molecular diagnosis of piroplasmids and Hepatozoon range from the classical single PCR with detection in ethidium bromide stained gels (Bashiruddin et al., 1999; Baneth et al., 2000) to the more sophisticated techniques based on use of DNA probes on membranes (Gubbels et al., 1999). The use of such methods has prompted a better knowledge of epizootiology in quantitative terms, but as sequencing of isolates is seldom undertaken, qualitative aspects remain less well understood. Another drawback in protozoa epizootiology is that molecular diagnosis is many times focused in those species that are clinically more relevant (e.g. B. equi in horses Nicolaiewsky et al., 2001, or B. canis in dogs Ano et al., 2001), which means that detection of recently discovered forms is usually disregarded as they are considered of lesser importance or present in mammals with low frequency. This may lead to a underdetection of parasites. The only approach that can partially overcome this difficulty is the use of the reverse line blotting technique (Gubbels et al., 1999) that allows to some extent simultaneous detection and discrimination of piroplasmids. These authors remarked, however, that this kind of assay can not identify new piroplasmids and in that case sequencing PCR products would be necessary in order to include appropriate diagnostic probes. We have used already a one-tube seminested PCR assay with universal primers (combined with sequencing) for piroplasmid detection in cats, finding unexpected infections by T. annae and Babesia canis canis (Criado-Fornelio et al., 2003). In the present work we intend to demonstrate that this kind of approach may reveal unexpected (but probably not infrequent) piroplasmid infections in unusual hosts from southern Europe. Besides sequence data obtained from piroplasmid/hepatozoon isolates in this molecular survey will be used to determine their phylogenetic relationships with other isolates already sequenced and available in GenBank TM database, this matter being the subject of the second part of our article.

3 2. Materials and methods A. Criado-Fornelio et al. / Veterinary Parasitology 113 (2003) Clinical/biological samples Domestic animals (cow, sheep, goat, dog and horse): veterinary practitioners from Spain, Portugal and France sent suspect samples (symptomatic) for diagnosis to our laboratory. In the case of dogs, samples from control animals (asymptomatic) were also obtained from Veterinarians. Blood samples were collected from client-owned animals, placed into EDTA and transported to the laboratory in a cold pack. Wild fox: we used frozen DNA samples obtained from spleen of 10 foxes captured in central Spain (Guadalajara) to carry out a parasitological survey during the period Details on location and capture methods were published in a former article (Criado-Fornelio et al., 2000) DNA analysis To design diagnostic primers, several 18s rrna gene sequences from piroplasmids and Hepatozoon were used for computer analysis. The following GenBank TM entries were chosen: B. canis canis (AY072926) B. felis (AF244912), B. bovis (L19077), B. leo (AF244911), B. equi (Z15015), B. caballi (Z151049, T. annae (AF188001), T. annulata (M64243), Hepatozoon canis (AF176385), Human (K034032) and horse (AJ311673). Sequences were aligned with the CLUSTALW computer program (Thompson et al., 1994) to search for suitable (specific) diagnostic regions. Location of restriction enzyme sites and design of a universal primer set for amplification of piroplasmid/hepatozoon DNA was performed with the aid of the PC/GENE software package (Intelligenetics, Mountain View, CA) DNA isolation, amplification and sequencing DNA was isolated from blood with the aid of the Blood Spin kit purchased from MO- BIO (Solana Beach, California, USA). DNA was obtained from fox spleen tissue by digestion with proteinase K and sodium dodecyl sulphate (SDS) as described previously (Criado-Fornelio et al., 2000). Primers used for DNA amplification/sequencing are described in the Appendix A. Diagnosis of Babesia/Theileria was performed with the seminested PCR assay (Criado-Fornelio et al., 2003). Diagnosis of H. canis was done with universal primers (BTH-1F and BTH-1R). The same primers and the set BT2-F and BT2-R were used for sequencing. Primers CAN-F/CAN-R were used to confirm the presence of dog DNA in samples where the existence of unexpected piroplasmid species would be considered a possible mistake in sample handling. Usual precautions to avoid DNA contamination (laminar flow hoods, separated work areas for reaction mixture preparation, DNA extraction, amplification and analysis of PCR products, etc.) were used in our laboratory to prevent carry over of amplified products (Lysby, 1999). The amplification mixture for single PCR contained 10 mm Tris, ph 8.3, 50 mm KCl, 1.5 mm MgCl 2, 200 M each datp, dctp, dgtp and dttp, 1 M each primer

4 192 A. Criado-Fornelio et al. / Veterinary Parasitology 113 (2003) and 2.5 units Amplitaq Gold DNA polymerase from Applied Biosystems (USA) in a final volume of 50 l. Amplification was carried out in an Eppendorf Personal thermal cycler. Optimum primer annealing temperatures were determined empirically. PCR diagnosis was performed with different thermal cycling profile depending on the primer combination: (1) Seminested PCR assay was performed as described by Criado-Fornelio et al. (2003). (2) Primers BTH-1F/BTH-1R: Initial enzyme activation and hot start 10 min at 94 C, 40 cycles of 30 s at 95 C and 1 min at 68 C, with a final extension of 10 min at 72 C. (3) Primers CAN-F/CAN-R: Initial enzyme activation and hot start 10 min at 94 C, 40 cycles of 30 s at 95 C, 30 s at 60 C and 30 s at 72 C, with a final extension of 10 min at 72 C. For sequencing purposes, amplified products were separated in agarose gels and visualized with ethidium bromide. Bands of interest were isolated with the Ultraclean 15 kit from MOBIO (Solana Beach, CA, USA). Purified fragments were cloned using the pgem T-Vector system (PROMEGA, Madison, WI, USA), following manufacturer s instructions. Sequencing was carried out in an ABI (Applied Biosystems Inc., Foster City, CA, USA) automated sequencer. At least three clones containing the amplified fragment were sequenced. Fragments obtained by seminested PCR were sequenced for identification of isolates (390 bp long for Babesia, 410 bp for Theileria). Additional sequence information was obtained only in those cases where amplification with alternative primers was possible using single round PCR. In most cases 1000 bp were sequenced in total at the 5 end of the 18s rrna gene. Only in B. bovis and B. ovis the complete 18s rrna gene was sequenced. 3. Results Seminested PCR assay showed the presence of piroplasmid infections in 60% of symptomatic mammals (mean of all host groups except foxes). Single-round PCR detected H. canis in 100% of foxes but no infections were found in domestic dogs. In general, isolates of the same parasite species showed identical sequences when studied in different animals. In a few cases there were single base variations between different clones of the same isolate, but they were not introduced in GenBank TM entries as we considered them as polymerase errors or rare polymorphisms in hypervariable regions. Details on parasitic protozoa detected in different hosts are presented in Table 1. Table 2 shows the results of a BLAST search of sequences found in southern European isolates against GenBank TM database. Results can be summarized as follows (presented by host groups) Ruminants Seven cows out of 10 were infected with piroplasmids. Species present were B. bovis (three cows from Portugal), T. annulata (one cow from France, one cow from Spain), Theileria sp. (two cows from Spain). The complete 18s rrna gene of Portuguese B. bovis isolate was sequenced.

5 A. Criado-Fornelio et al. / Veterinary Parasitology 113 (2003) Table 1 Parasitic protozoa species found by seminested PCR and sequencing in several mammals Horse Cow Sheep/goat Dog (symptomatic/asymptomatic) Fox Animals studied /12 10 Non-infected /9 B. equi 5 1 a /3 a B. bovis 3 T. annulata 2 Theileria sp. (type C) 2 T. annae 5 a B. ovis 6 B canis canis 1 a 3 B. canis vogeli 1 H. canis ND b ND ND 10 a New host for this piroplasmid species. b Not done. T. annulata isolates from France and Spain were identical between them and 99% identical to the sequence previously available in GenBank TM. Theileria sp. isolate was 99% identical to Theileria type C (Korea). Finally, we found that the Portuguese B. bovis isolate had only a 94% identity compared to a B. bovis isolate from Mexico. The only goat analysed was infected with B. ovis, as demonstrated by microscopic observation of blood smears and PCR. Five out of nine symptomatic sheep (negative by blood smear observation) were also infected with B. ovis as determined by seminested PCR and sequencing. The complete 18s rrna gene of B. ovis was sequenced in the infected goat. BLAST comparison of sequences demonstrated that B. ovis was related (94% identity) with a Table 2 BLAST comparison of sequences from southern European isolates in GenBank TM database Species/host/accession no. Closest GenBank TM entry/origin Percent identity B. canis canis (horse and dog) AY a AY (B. canis canis/croatia) 100 B. canis vogeli (dog) AY b AY (B. canis vogeli/france) 99 B. equi (dog) AY a Z15105 (B. equi/south Africa, horse) 99 B. equi Spain 1 (horse) AY b Z15105 (B. equi/south Africa) 99 B. equi Spain 2 (horse) AY b Z15105 (B. equi/south Africa) 99 T. annae (fox) AY a AF (T. annae/spain, dog) 100 Theileria sp. (cow) AY b U97051 (Theileria sp. type C, Korea) 99 T. annulata (cow) AY b M64243 (T. annulata) 99 B. bovis (cow) AY c L31922 (B. bovis, Mexico) 94 B. ovis (goat) AY c U09834 (Babesia sp./south Africa, cow) 94 H. canis (fox) AY b AF and AF (H. canis Israel/Japan) 98/99 First column are isolates reported in present paper. a In these isolates only the fragment obtained in seminested PCR was sequenced (length ranging from 390 to 410 bp). b In these isolates the length of fragment sequenced ranged from 640 to 1000 bp. c In these isolates the complete 18s rrna gene was sequenced.

6 194 A. Criado-Fornelio et al. / Veterinary Parasitology 113 (2003) Babesia sp. isolate from a South African cow. Identity with B. bovis (Portugal) reached only a 92% Canids All foxes were infected with parasitic protozoa: five were coinfected by H. canis and T. annae and five were infected only with H. canis. A BLAST search against GenBank TM revealed a 100% similarity of Spanish fox isolate sequence and T. annae found in dog. Spanish H. canis isolate from fox was 98% similar to a H. canis isolate from dog sequenced in Israel and a 99% with regard to H. canis from Japan, although in the last case the fragment that could be compared was only 409 bp long. Five dogs out of 10 (symptomatic) were infected with piroplasmids. Species present were: B. canis canis (three), B. canis vogeli (one) and B. equi (one). H. canis was absent from domestic dogs. Sequences obtained from piroplasmids infecting Spanish dogs were compared with GenBank TM entries by BLAST. B. canis canis from Spain was 100% identical to a Croatian isolate, whereas B. canis vogeli was 99% identical to a French isolate. An interesting finding in our dog survey was that one symptomatic animal was infected with B. equi. Although one may be tempted to think that we used a horse sample by mistake, the presence of dog DNA in this sample was confirmed with an additional PCR assay (Fig. 1). After finding this result, 12 asymptomatic dogs (showing both blood biochemistry and Fig. 1. PCR detection of dog DNA in mammalian samples with unusual piroplasmid infections (dog infected with B. equi and horse infected with B. canis canis). Amplification was performed with primers CAN-F and CAN-R, which specifically amplify a 290 bp product of canine mitosin gene. Fragments were separated in a 5% polyacrylamide gel and stained with Ethidium bromide. Lane 1: amplification performed with uninfected horse DNA. No product was obtained. Lane 2: amplification with B. equi-infected dog DNA. A band of 290 bp appears. Lane 3: amplification with uninfected dog DNA. The same 290 bp band is present. Lane 4: amplification done with DNA from a B. canis canis-infected horse. No product was obtained. Lane 5: amplification performed with B. equi-infected horse DNA. No band was obtained. Lane 6: digestion of product obtained in lane 2 with restrictase Xba I. The expected fragments of 155 and 135 bp are present. Lane 7: digestion of product obtained in lane 3 with restrictase Xba I. Same products as in lane 6 are obtained. Lane 8: negative amplification control.

7 A. Criado-Fornelio et al. / Veterinary Parasitology 113 (2003) Fig. 2. Map of Spain showing the distribution of B. equi infections in asymptomatic dogs. cytology normal values in all samples) were analysed by seminested PCR and sequencing, to ascertain if the infection was present in a non-biased sample. Interestingly, three healthy dogs were positive for B. equi. All dog isolates (from either asymptomatic or symptomatic animals) shared the same sequence in the 410 bp fragment obtained by seminested PCR. Fragment sequence was 99% identical to the only B. equi entry available in GenBank TM.A map of Spain showing geographic location of B. equi infections detected in asymptomatic dogs (Fig. 2) indicated that the parasite was not restricted to a single focus, since it was found in three different Spanish provinces (Madrid, Sevilla and Alicante) Horse Six horses out of 10 were infected with piroplasmids. Species present were B. equi (five animals) and B. canis canis (one). All of B. equi-infected horses (as well as B. equi-infected dogs) presented the same sequence in the 410 bp fragment obtained by seminested PCR. Additional sequence data was obtained from the central region of the 18s rrna gene in two B. equi isolates from horse, resulting in two different types (isolates B. equi Spain 1 and B. equi Spain 2). An interesting result of our study is that one horse was infected with B. canis canis (100% identical to dog isolate from Croatia). A PCR assay performed in this sample with

8 196 A. Criado-Fornelio et al. / Veterinary Parasitology 113 (2003) primers CAN-F and CAN-R (used to detect dog DNA, as explained before), produced no amplification (Fig. 1). This excluded the possibility of a wrong diagnosis due to a sample mishandling. 4. Discussion 4.1. Piroplasmid infections Seminested PCR demonstrated a high level of piroplasmid infection in symptomatic mammal populations. This is in agreement with recently published papers using nested PCR in horse (Nicolaiewsky et al., 2001) and dog (Ano et al., 2001). All piroplasmids found in Spanish mammals were in general the same described up to now in the Mediterranean area. There are however some important issues that deserve to be mentioned after considering the results of our study Ruminants In their study on Korean Theileria spp. isolates from cow, Chae et al. (1998) pointed out that the type C isolate was geographically restricted, since it was found only in one animal. We have found a cow isolate closely related to the Korea C in two Spanish cows, so that this geographical restriction is not so evident. Clearly the small sample size (14 cows) used by Chae et al. (1998) makes difficult to address conclusions from the relative frequencies of Theileria isolates. As we also used a small sample size, we do not dare to draw conclusions from the frequency of Theileria isolates in cows. We must point out however, that our data are in agreement with previous findings pointing out that T. annulata is one of the most common species of pathogenic Theileria in Spain (Almeria et al., 2001a). Likewise, B. bovis was another common pathogen of cattle in the Iberian Peninsula (Almeria et al., 2001b) detected in our PCR assays. The 18s rrna gene of B. bovis from Portugal shares, with other isolates of the same species, the main sequence features pointed out by Calder et al. (1996): presence of a hypervariable region (positions in L31922) and a conserved region that differentiates B. bovis from B. bigemina (positions ). This region, used for Calder et al. (1996) to design a hybridization probe is also a valid target for blot detection of the Portuguese B. bovis isolate. Our molecular survey on sheep piroplasmids is in agreement with previous works from other authors (Habela et al., 1990; Ferrer et al., 1998) pointing out that B. ovis is the most frequent piroplasmid present in sheep and goats in Spain. B. ovis is genetically related to other Babesia species that appear in the family Bovinae, like a Babesia sp. from South African cow and in a lesser extent, to B. bovis Canids In general, piroplasmids found by us in dogs are the same described already in Spain (Navarrete and Nieto, 1999). There was, however, a notable exception: B. equi. This piroplasmid has never been described before as a parasite of canids. It appeared in several dogs from different Spanish provinces, and mostly linked to a rural-type habitat. Interestingly it infected more frequently asymptomatic dogs, which might be indicative of a

9 A. Criado-Fornelio et al. / Veterinary Parasitology 113 (2003) low-pathogenicity in that host. If this trend is confirmed, it could be interpreted as evidence of a long-term established parasitic relationship (Ebert, 1998). Since microscopical observations were negative for piroplasmids in both symptomatic and asymptomatic dogs, one may wonder if this means a wrong diagnosis. We have demonstrated that sample mishandling is unlikely to have occurred, and contamination seems also difficult to assume (otherwise crossed infections should have appeared in other mammals). The fact that exchange of specific parasites was only detected between horse and dog seems to be an additional reason to believe that infection of horses with B. canis and infection of dogs with B. equi is not unlikely to happen. Probably the reason why nobody detected before the presence of B. equi in dogs is simply that B. gibsoni and B. equi are morphologically very similar (Purnell, 1981; Zahler et al., 2000b). For us this is an overwhelming evidence that only the use of molecular methods can lead to a correct identification of piroplasmids. B. canis canis is considered a very common parasite of dogs in Europe (Caccio et al., 2002), which is in agreement with findings in this report. B. canis vogeli of Spanish dog is genetically very similar to the French isolate described by Caccio et al. (2002). The red fox seems to be an important wild reservoir of T. annae in central Spain. Recent reports indicate that this newly discovered parasite is not uncommon in dogs from Galicia, in the northwest of Spain (Camacho et al., 2001). However, as judged by data from the present report, its frequency in dogs from central and southern Spain (where most of the animals came from) seems to be rather low. We were unable to detect its presence in domestic dogs, although the sample size used is too small as to ensure that the parasite is absent from these Spanish regions. Probably these prevalence differences might be explained by the absence of the arthropod vector, but further work in this sense is needed to confirm such hypothesis. Finally it is important to remind that wild red foxes have peridomestic habits, which makes them likely vectors of parasites both for animal and man (a fact repeatedly remarked by several authors like Conceicao-Silva et al., 1988, and Criado-Fornelio et al., 2000) Horse B. equi is the predominant piroplasmid infecting spanish horses. This fact is in agreement with previous reports by other authors studying horses from the Mediterranean region (Bashiruddin et al., 1999). There were two genotypes in Spanish isolates (Spain 1 and Spain 2), which shows that there is some degree of polymorphism in parasite populations, something pointed out also by Nicolaiewsky et al. (2001) in their study on the presence of B. equi in Brazilian horses. These authors, however, did not study the 18s rrna gene. B. caballi was present with low frequency in Portuguese horses according to Bashiruddin et al., In our small sample, we did not found this piroplasmid, although we found B. canis canis instead. The importance of horses as hosts of B. canis canis is unknown, because our sample was too small as to draw any valid conclusion about its abundance. For us, the main reason explaining why nobody ever found B. canis canis to be a parasite of horses is simply that it is morphologically very similar to B. caballi (Purnell, 1981). Only molecular procedures have been able to detect this new piroplasmid infection. It is important to point out that some authors like Bashiruddin et al. (1999) have used totally specific primers for

10 198 A. Criado-Fornelio et al. / Veterinary Parasitology 113 (2003) detection of B. caballi and B. equi in horse, but as shown in the present work, this kind of approach may leave undetected some piroplasmid infections Hepatozoon infections in canids H. canis infection was absent from dogs and present in all foxes examined. This discrepancy may be due to at least two main reasons: (1) Differences in food intake (live prey or carrion in foxes compared to domestic food or commercial feedstuff in domestic dogs). Since H. canis is transmitted by ingestion of ticks (Christophers, 1907), those animals feeding on live prey may have higher probabilities of ingesting ticks together with food. (2) Different kind of samples used for molecular analysis (blood in dogs versus splenic tissue in foxes). The parasite seems to be present in high numbers in reticuloendothelial cells and lymphocytes (Conceicao-Silva et al., 1988), so that spleen samples are much more likely to harbour-infected cells than are blood samples. To our knowledge, this report represents the first time that this parasite is described in Spanish foxes. In the neighbour country of Portugal, the presence of H. canis in fox was first described by Leitao (1963). More recently Conceicao-Silva et al. (1988), observing blood smears from Portuguese foxes, detected a relatively high level of Hepatozoon sp. infection (48%). It does not sound unreasonable that in the present work using molecular procedures (far more sensitive than microscopy), we found a rate of infection in fox of 100%. In agreement with our finding is the survey carried out on wild dogs (Lycaon pictus) in South Africa by Van Heerden et al. (1995). They found a 93% of animals infected by H. canis just by microscopical analysis of blood smears. Finally, the report of vertical transmission of the parasite in dogs (Murata et al., 1993) may help to explain the high prevalence of H. canis infection in fox. 5. Conclusion Navarrete et al. (1999) have pointed out that host specificity of piroplasmids is low and that the use of more sensitive methods of detection would probably unveil some new hosts not described hitherto. These affirmations are fully supported by our findings, demonstrating that at least in Spain, horse and dog share a number of piroplasmid infections. Moreover, we have shown also in a former study (Criado-Fornelio et al. (2003)) that cats can be infected also by dog parasites like B. canis canis and T. annae. Chemotherapeutic failure in the treatment of piroplasmosis has been reported occasionally (Collett, 2000; Camacho et al., 2002) and death may be the outcome of infection. This may be due in some cases to a misidentification of parasites. It is important to remember that B. equi is a small piroplasmid species related to Theilerids but morphologically similar to a small Babesid, B. gibsoni. Accordingly, detection of unusual species with universal primers, followed by sequencing or RFLP analysis must be a priority in clinical or epizootiological studies (Caccio et al., 2002). There are alternative procedures like the reverse line blot hybridization technique (Gubbels et al., 1999) or the real time PCR. The later method,

11 A. Criado-Fornelio et al. / Veterinary Parasitology 113 (2003) although not yet available, is very interesting because it shows some important technical advantages: (1) Low risk of contamination with amplicons (Polanco et al., 2002). (2) Allows both quantification of parasite loads and analysis of melting temperatures of amplified products, the later property leading to possible species discrimination (Collantes et al., 2002; Nicolas et al., 2002). We are currently evaluating this alternative method in our laboratory in an attempt to improve present molecular procedures for detection of piroplasmid and H. canis infections. Appendix A. Primers used for amplification/sequencing in this study All the primers (except dog DNA-specific primers) amplify variable length fragments of the 18s rrna genes in the organisms indicated. A.1. Universal Babesia Theileria (seminested PCR) BT1-F: ggttgatcctgccagtagt BT1-R: gcctgctgccttcctta Both primers were combined with BTH-1R (see below) in seminested PCR assays. A.2. Universal Babesia Theileria Hepatozoon (single PCR) BTH-1F: cctgmgaracggctaccacatct BTH-1R: ttgcgaccatactcccccca Used for sequencing (and also for diagnosis in H. canis) of the central region of 18s rrna gene in H. canis, Babesia spp. and Theileria spp. A.3. Primers used for sequencing of the 3 region of the 18s rrna gene in Babesia and Theileria BT2-F: ggagtatggtcgcaagtctg BT2-R: cttctgcaggttcacctacg Primers BT1-F and BT2-R were combined for amplification of complete 18s rrna gene. A.4. Detection of canine DNA CAN-F: cttgtcacggtaaggttc CAN-R: ctgatgtatttcctgcaccaag These primers specifically amplify a 290 bp fragment of the mitosine gene of Canis familiaris which contains a Xba I site (digestion with this enzyme yields two fragments of bp).

12 200 A. Criado-Fornelio et al. / Veterinary Parasitology 113 (2003) References Almeria, S., Castella, J., Ferrer, D., Ortuno, A., Estrada-Peña, A., Gutierrez, J.F., 2001a. Bovine piroplasms in Minorca (Balearic Islands, Spain): a comparison of PCR-based and light microscopy detection. Vet. Parasitol. 99, Almeria, S., Castella, J., Ferrer, D., Sparagano, O., Estrada-Peña, A., 2001b. First report of Babesia bovis in Spain. Vet. Rec. 149, Ano, H., Makimura, S., Harasawa, R., Detection of Babesia species from infected dog blood by polymerase chain reaction. J. Vet. Med. Sci. 63, Baneth, G., Barta, J.R., Shkap, V., Martin, D.S., Macintire, D.K., Vincent-Johnson, N., Genetic and antigenic evidence supports the separation of Hepatozoon canis and Hepatozoon americanum at the species level. J. Clin. Microbiol. 38, Bashiruddin, J.B., Camma, C., Rebelo, E., Molecular detection of Babesia equi and Babesia caballi in horse blood by PCR amplification of part of the 16S rrna gene. Vet. Parasitol. 84, Caccio, S.M., Antunovic, B., Moretti, A., Mangili, V., Marinculic, A., Baric, R., Slemenda, S., Pieniazek, N., Molecular characterisation of Babesia canis canis and Babesia canis vogeli from naturally infected European dogs. Vet. Parasitol. 106, Calder, J.A., Reddy, G.R., Chieves, L., Courtney, C.H., Littell, R., Livengood, J.R., Norval, R.A., Smith, C., Dame, J.B., Monitoring Babesia bovis infections in cattle by using PCR-based tests. J. Clin. Microbiol. 34, Camacho, A.T., Pallas, E., Gestal, J.J., Guitian, F.J., Olmeda, A.S., Goethert, H.K., Telford, S.R., Infection of dogs in north-west Spain with a Babesia microti-like agent. Vet. Rec. 149, Camacho, A.T., Pallas, E., Gestal, J.J., Guitian, F.J., Olmeda, A.S., Babesia canis infection in a splenectomized dog. Bull. Soc. Pathol. Exot. 95, Chae, J., Lee, J., Kwon, O., Holman, P.J., Waghela, S.D., Wagner, G.G., Nucleotide sequence heterogeneity in the small subunit ribosomal RNA gene variable (V4) region among and within geographic isolates of Theileria from cattle, elk and white-tailed deer. Vet. Parasitol. 75, Christophers, S.R., The Sexual Cycle of Leucocytozoon canis in the Tick, vol. 28. Scientific Memoirs by Officers of the Medical and Sanitary Department of the Government of India, pp Collantes, E., Zaballos, A., Alvarez-Garcia, G., Ortega-Mora, L.M., Quantitative detection of Neospora caninum in bovine aborted fetuses and experimentally infected mice by real-time PCR. J. Clin. Microbiol. 40, Collett, M.G., Survey of canine babesiosis in South Africa. J. S. Afr. Vet. Assoc. 71, Conceicao-Silva, F.M., Abranches, P., Silva-Pereira, M.C., Janz, J.G., Hepatozoonosis in foxes from Portugal. J. Wildlife Dis. 24, Criado-Fornelio, A., Gutierrez-Garcia, L., Rodriguez-Caabeiro, F., Reus-Garcia, E., Roldan-Soriano, M.A., Diaz-Sanchez, M.A., A parasitological survey of wild red foxes (Vulpes vulpes) from the province of Guadalajara, Spain. Vet. Parasitol. 92, Criado-Fornelio, A., Martinez-Marcos, A., Buling-Saraña, A., Barba-Carretero, J.C., Presence of Haemobartonella felis and piroplasmids in cats from Southern Europe: a molecular study. Vet. Microbiol., in press. Ebert, D., Experimental evolution of parasites. Science 282, Ferrer, D., Castella, J., Gutierrez, J.F., Seroprevalence of Babesia ovis in sheep in Catalonia, northeastern Spain. Vet. Parasitol. 79, Gubbels, J., de Vos, A., van der Weide, M., Viseras, J., Schouls, L.M., de Vries, E., Jongejan, F., Simultaneous detection of bovine Theileria and Babesia species by reverse line blot hybridization. J. Clin. Microbiol. 7, Habela, M., Reina, D., Nieto, C., Navarrete, I., Isolation and identification of Babesia ovis in Extremadura (Spain). Vet. Parasitol. 35, Leitao, J.L., Hepatozoon. In: Parasitos des Animais Domesticos en Portugal Metropolitano. Ed. Instituto Alta Cultura-Fundaçao Galouste-Gulbekian, Lisboa, p. 45. Lysby, G., Application of nucleic acid amplification in clinical microbiology. In: Meltzer, S. (Ed.), Clinical Methods in Molecular Biology: PCR in Bioanalysis, vol. 92. Human Press Inc., Totowa, NJ, pp

13 A. Criado-Fornelio et al. / Veterinary Parasitology 113 (2003) Murata, T., Inoue, M., Tateyama, S., Taura, Y., Nakama, S., Vertical transmission of Hepatozoon canis in dogs. J. Vet. Med. Sci. 55, Navarrete, I., Nieto, L., Babesiosis, Hepatozoonosis, Citauxzonoonosis felina. In: Cordero-del-Campillo, M., Rojo-Vazquez, F.A. (Eds.), Parasitología Veterinaria. McGraw Hill-Interamericana, Madrid, pp Navarrete, I., Serrano, F., Reina, D., Babesiosis. In: Cordero-del-Campillo, M., Rojo-Vazquez, F.A. (Eds.), Parasitología Veterinaria. McGraw Hill-Interamericana, Madrid, p Nicolaiewsky, T.B., Richter, M., Lunge, V., Cunha, C., Delagostin, O., Ikuta, N., Fonseca, A., da Silva, S.S., Ozaki, L., Detection of Babesia equi (Laveran, 1901) by nested polymerase chain reaction. Vet. Parasitol. 101, Nicolas, L., Milon, G., Prina, E., Rapid differentiation of Old World Leishmania species by LightCycler polymerase chain reaction and melting curve analysis. J. Microbiol. Methods 51, 295. Penzhorn, B., Kjemtrup, A., Lopez-Rebollar, L., Conrad, P.A., Babesia leo sp. from lions in the Kruger National Park, South Africa, and its relation to other small piroplasms. J. Parasitol. 87, Polanco, J.C., Rodriguez, J.A., Corredor, V., Patarroyo, M.A., Plasmodium vivax: parasitemia determination by real-time quantitative PCR in Aotus monkeys. Exp. Parasitol. 100, Purnell, R.E., Babesiosis in various hosts. In: Ristic, M., Kreier, J.P. (Eds.), Babesiosis. Academic press, New York, pp Thompson, J.D., Higgins, D.G., Gibson, T.J., CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, Van Heerden, J., Mills, M.G., Van Vuuren, M.J., Kelly, P.J., Dreyer, M.J., An investigation into the health status and diseases of wild dogs (Lycaon pictus) in the Kruger National Park. J. S. Afr. Vet. Assoc. 66, Vincent-Johnson, N.A., Macintire, D.K., Lindsay, D.S., Lenz, S.D., Baneth, G., Shkap, V., Blagburn, B.L., A new Hepatozoon species from dogs: description of the causative agent of canine hepatozoonosis in North America. J. Parasitol. 83, Zahler, M., Rinder, H., Schein, E., Gothe, R., 2000a. Detection of a new pathogenic Babesia microti-like species in dogs. Vet. Parasitol. 89, Zahler, M., Rinder, H., Zweygarth, E., Fukata, T., Maede, Y., Schein, E., Gothe, R., 2000b. Babesia gibsoni of dogs from North America and Asia belong to different species. Parasitology 120,