- Email: [email protected]
Molecular characterisation of Babesia canis canis and Babesia canis vogeli from naturally infected European dogs
Veterinary Parasitology 106 (2002) 285–292
Molecular characterisation of Babesia canis canis and Babesia canis vogeli from naturally infected European dogs夽 Simone M. Cacciò a,∗ , Boris Antunovic b , Annabella Moretti c , Vittorio Mangili d , Albert Marinculic e , Renata Rafaj Baric f , Susan B. Slemenda g , Norman J. Pieniazek g a b
Laboratory of Parasitology, Istituto Superiore di Sanità, Viale Regina Elena 299, I-00161 Rome, Italy Department of Animal Husbandry, Faculty of Agriculture, J.J. Strossmayer University, Osijek, Croatia c Department of Biopathological Veterinary Sciences, University of Perugia, Perugia, Italy d Department of Pathology, Diagnostic and Clinical Veterinary, University of Perugia, Perugia, Italy e Department for Parasitology and Invasive Diseases, Faculty of Veterinary Medicine, University of Zagreb, Zagreb, Croatia f Clinics for Internal Diseases, Faculty of Veterinary Medicine, University of Zagreb, Zagreb, Croatia g Division of Parasitic Diseases, Centers for Disease Control and Prevention, Atlanta, GA, USA Received 8 February 2002; accepted 5 April 2002
Abstract The morphologically small Babesia species isolated from naturally infected dogs in Europe, Japan, and US are described as Babesia gibsoni despite the fact that molecular techniques show that they should be assigned to two or three separate taxons. The morphologically large Babesia isolated from dogs in Europe, Africa, and US were generally classified as B. canis until it was proposed to distinguish three related, albeit genetically distinct subspecies of this genus, namely B. canis canis, B. canis rossi, and B. canis vogeli. The insight into the molecular taxonomy of canine piroplasms is, however, limited because only partial small subunit ribosomal RNA (ssrRNA) sequence data exist for two species from the B. canis group. In this work, we molecularly characterised natural Babesia infections in 11 dogs from Croatia, France, Italy, and Poland. These infections were diagnosed as caused by B. canis canis and B. canis vogeli based on the analysis of the complete sequence of the ssrRNA genes. Phylogenetic analysis confirmed that the large Babesia species of dogs belong the to the Babesia sensu stricto clade, which includes species characterised by transovarial transmission in the tick vectors and by exclusive development inside the mammalian host erythrocytes. The 夽 Nucleotide sequence data reported in this paper are available in the GenBankTM database under the accession nos. AY072925 and AY072926. ∗ Corresponding author. Tel.: +39-06-4990-2310; fax: +39-06-4938-7065. E-mail address: [email protected] (S.M. Cacci`o).
0304-4017/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 4 0 1 7 ( 0 2 ) 0 0 1 1 2 - 7
286
S.M. Cacci`o et al. / Veterinary Parasitology 106 (2002) 285–292
new data facilitate the reliable molecular diagnosis of the subspecies of B. canis. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Babesia canis; Dog; Small subunit ribosomal RNA gene; Phylogeny
1. Introduction Babesia species are tick-transmitted apicomplexan parasites that infect a wide range of vertebrate hosts and cause severe diseases in wild and domestic animals (Kuttler, 1988). The identification of individual species has traditionally been based on the host specificity and on the morphology of the intraerythrocytic forms (the piroplasms). Babesia parasites are usually classified as small (1.5–2.5 m) or large (3.0–5.0 m) piroplasms. Babesia canis and Babesia gibsoni are recognised as the two species that cause canine babesiosis worldwide. B. canis is a large piroplasm (4–5 m), which usually occurs as a single pear-shaped piroplasm or in pairs of merozoites divided by binary fission within the erythrocyte. Previous studies, on the basis of differences in the geographical distribution, vector specificity, and antigenic properties (Uilenberg et al., 1989; Hauschild et al., 1995), recognised that large canine piroplasms are subdivided into three subspecies, namely B. canis canis, transmitted by Dermacentor reticulatus in Europe, B. canis vogeli, transmitted by Rhipicephalus sanguineus in tropical and subtropical countries, and B. canis rossi, transmitted by Haemophysalis leachi in South Africa. B. gibsoni has been found to be associated with infection of dogs in Asia, North America, northern and eastern Africa, and Europe (Conrad et al., 1991; Casapulla et al., 1998; Birkenheuer et al., 1999). B. gibsoni is a small parasite that commonly appears as individual ring forms or piriform bodies of a size between 1 and 2.5 m (Conrad et al., 1991). However, it is difficult to morphologically distinguish this parasite from other small piroplasms such as B. equi (now reclassified as Theileria equi; Mehlhorn and Schein, 1998) or B. microti (Conrad et al., 1992). Traditionally, all small canine piroplasms were identified as B. gibsoni based on the assumption that no other small Babesia species infects dogs. This hypothesis was clearly wrong, as demonstrated by the recent genetic characterisation of small piroplasms of dogs from Spain and California (Zahler et al., 2000a,b; Kjemtrup et al., 2000a,b). Phylogenetic analysis of the small subunit ribosomal gene (ssrRNA) of these isolates showed that they are clearly distinct from B. gibsoni, being more closely related to B. microti (the B. gibsoni Spanish isolate) or to the Californian isolates from bighorn sheep and the mule deer (the B. gibsoni Californian isolate). These results show that molecular techniques, in particular the characterisation of the ssrRNA genes, represent an objective and precise method of species identification and phylogenetic classification. In spite of this fact, few data are available with respect to the large piroplasms of dogs, and a complete characterisation of the ssrRNA gene has been reported only for one isolate of B. canis rossi (Allsopp et al., 1994). Recently, two PCR–RFLP assays have been developed to discriminate the large canine piroplasms, based on polymorphisms of partial ssrRNA sequences (Carret et al., 1999), or of internally transcribed spacers (Zahler et al., 1998).
S.M. Cacci`o et al. / Veterinary Parasitology 106 (2002) 285–292
287
In this study, a molecular analysis of the complete ssrRNA gene was done to characterise Babesia parasites collected from 11 naturally infected dogs in Croatia, France, Italy, and Poland. 2. Materials and methods 2.1. Parasite isolates Blood from 11 dogs showing clinical signs of babesiosis, including apathy, fever, and anaemia, were collected close to Zagreb (Croatia), and in Perugia (Italy) during fall 2000 and spring 2001. No history of travel was recorded for any of the eight Croatian dogs, while two (referred to as Bc11 and Bc12) of the three Italian dogs acquired the infection during a stay in an area close to Nantes (France), and in an area close to Warsaw (Poland), respectively. Examination of Giemsa-stained blood smears revealed the presence of Babesia parasites, which have morphology (pear-shaped parasites, often occurring in pairs) and size (3–5 m) typical of the large Babesia, i.e. of B. canis. 2.2. DNA isolation, amplification and sequencing DNA was extracted from 200 l of whole blood using the QIAmp blood kit (Qiagen), according to the manufacturer’s instructions. Amplification of the ssrRNA gene was performed using universal apicomplexan primers: the forward primer 5 -AACCTGGTTGATCCTGCCAGTAGTCAT-3 , and the reverse primer 5 -GAATGATCCTTCCGCAGGTTCACCTAC-3 . The PCR mix consisted of 1× buffer containing 1.5 mM MgCl2 , 200 m of each dNTP, 20 pmol of each primer, 2.5 U of Taq Polymerase (Takara), and extracted template DNA. Reactions consisted of an initial 5 min denaturation step at 95 ◦ C, followed by 35–40 cycles of denaturation at 94 ◦ C for 30 s, annealing at 65 ◦ C for 30 s, and extension at 72 ◦ C for 2 min. Final extension was done at 72 ◦ C for 7 min, followed by a hold step at 4 ◦ C. PCR products were purified using the QiaQuick purification kit (Qiagen), and fully sequenced using the ABI prism dye sequencing kit (Applied Biosystems), and a set of internal primers. Sequencing reactions were analysed on the ABI 3100 automatic DNA sequencer (Applied Biosystems). Sequences were assembled by using the program SeqMan II (DNASTAR). The GenBank accession nos. for the complete ssrRNA sequences of B. canis vogeli and B. canis canis are AY072925 and AY072926, respectively. 2.3. Phylogenetic analysis The sequences of the ssrRNA gene from selected Babesia and Theileria species (see legend to Fig. 1) were retrieved from the GenBankTM database and aligned with the newly established ssrRNA sequences of B. canis canis and B. canis vogeli by using the program Clustal W (Thompson et al., 1994). Phylogenetic analysis was done with the DNAML, DNAPARS, DNADIST, and NEIGHBOR programs from the PHYLIP package (Felsenstein, 1989, version 3.573c) and with the TREE-PUZZLE program (Strimmer and von Haeseler,
288
S.M. Cacci`o et al. / Veterinary Parasitology 106 (2002) 285–292
Fig. 1. Phylogenetic tree based on full-length ssrRNA sequences of selected Babesia and Theileria species. Quartet puzzling maximum likelihood results are shown using Babesia microti as the outgroup. Numbers at the nodes indicate the quartet puzzling support for each internal branch. Scale bar indicates an evolutionary distance of 0.01 nucleotides per position in the sequence. Vertical distances are for clarity only. GenBank accession nos. of the sequences used for analysis: Babesia bigemina gene A, X59604; Babesia caballi, Z15104; Babesia canis canis, AY072926; Babesia canis vogeli, AY072925; Babesia canis rossi, L19079; Babesia microti, U09833; Babesia odocoilei, U16369; Babesia gibsoni (Oklahoma dog isolate), AF205636; Theileria parva, AF013418; Theileria sergenti, AB016074. New sequences (AY072925 and AY072926) are indicated by arrows.
1996, version 5.0). Phylogenetic trees inferred by these programs were drawn using the TreeView program (Page, 1996). 3. Results 3.1. Amplification and sequencing of the ssrRNA coding regions The amplification of the full-length region coding for the ssrRNA from 11 B. canis isolates yielded single products of approximately 1700 bp (data not shown). Sequence analysis
S.M. Cacci`o et al. / Veterinary Parasitology 106 (2002) 285–292
289
of the PCR products revealed the presence of two closely related, albeit different, ssrRNA sequences. The first sequence had a length of 1714 bp, and was found in all Croatian dogs (isolates Bc1–Bc8) and in two Italian dogs (isolate Bc10 and Bc12, of which the latter represents an infection acquired during a stay in Poland). No sequence differences were observed among the 10 isolates. BLAST search against GenBank revealed the highest similarity score (99.5%) with a partial, 364 bp long sequence from B. canis canis (accession no. AJ009795). There were two differences, namely a nucleotide deletion at position 434, and a nucleotide insertion at position 453 (numbering refers to the full-length ssrRNA sequence). Notably, the two differences were close to the primer used for PCR amplification and direct sequencing of AJ009795, and might therefore reflect sequencing errors; however, minor strain variations cannot be ruled out. The sequence from the isolate Bc10 was submitted to GenBank as the complete B. canis canis ssrRNA sequence (accession no. AY072926). The second sequence had a length of 1713 bp and was associated with the infection of another dog (isolate Bc11), which acquired the infection during a stay in France. A BLAST search against GenBank revealed the highest similarity score (98.6%) with a partial, 363 bp long sequence from B. canis vogeli (accession no. AJ009796). There were four nucleotide substitutions (at positions 488, 599, 632, and 789) and one nucleotide insertion (at position 776, numbering refers to the full-length ssrRNA sequence). Again, the observed differences may represent sequencing errors or minor strain variation. The sequence from isolate Bc11 was submitted to GenBank as the complete B. canis vogeli ssrRNA sequence (GenBank accession no. AY072925). A comparison between the complete ssrRNA sequences of B. canis canis and B. canis vogeli yielded an identity score of 98%. The remaining 2% represented 27 nucleotide substitutions and seven deletions/insertions. When compared to the B. canis rossi ssrRNA sequence, the identity scores were 95.7% for B. canis canis and 95.3% for B. canis vogeli. 3.2. Phylogenetic analysis The ssrRNA sequences of the Bc10 and Bc11 isolates were aligned with eight selected species from the Babesia and Theileria genera (see legend to Fig. 1 for the list of species and the GenBank accession no. for the sequences used). The alignment was flush-trimmed at the ends and all columns containing gaps and ambiguous characters were removed, leaving 1626 columns (the alignment is available from the authors upon request). B. microti, which is always placed outside the clade grouping the ‘true’ Babesia and Theileria species, was selected as the outgroup to orient the tree. All four phylogenetic methods used (parsimony, maximum likelihood, quartet puzzling, and distance analysis) generated a tree with the same topology (Fig. 1). All three B. canis subspecies form a well-resolved group, which is the sister clade to a group formed by B. odocoilei (a parasite of cervids) and B. gibsoni (Oklahoma dog isolate).
4. Discussion Canine babesiosis is a tick-borne infection caused by the intraerytrocytic apicomplexan parasites B. canis and B. gibsoni. The two species are easily distinguished by light
290
S.M. Cacci`o et al. / Veterinary Parasitology 106 (2002) 285–292
microscopy, as they differ considerably in size (Kuttler, 1988). B. canis is widespread in Europe, and affects more than 400,000 dogs per year in France alone (Moreau et al., 1988). Young dogs and pups are highly susceptible, with death occurring in the absence of treatment, and transplacental transmission has been suggested since high parasitemia has been observed in neonates (Farwell et al., 1982; Breitschwerdt et al., 1983). Previous studies recognised three subspecies within the B. canis group, on the basis of differences in the geographical distribution, vector specificity, and antigenic properties (Uilenberg et al., 1989; Hauschild et al., 1995). Moreover, the clinical manifestations of infections caused by the three subspecies were found to differ significantly. As a matter of fact, B. canis rossi, the prevalent subspecies in South Africa, is highly virulent and causes either a haemolytic disease (possibly immune-mediated) or an acute and overwhelming inflammatory response (Reyers et al., 1998). On the other hand, infection by B. canis canis results in transient parasitemia, usually below 1%, and clinical disease is associated with congestion of internal organs (Schetters et al., 1997). Infection by B. canis vogeli leads to a relatively mild disease, often without evident clinical signs. Moreover, no cross-protection was observed when dogs vaccinated with soluble parasite antigens (SPA) prepared from one strain were experimentally challenged with a different strain (Schetters et al., 1995). Antigenic variation has been considered as a possible explanation for the observed lack of cross-protection and, in agreement with this interpretation, the use of a mixture of SPA from both a European B. canis isolate and a South African B. canis rossi isolate induced a protective immunity against heterologous B. canis infection (Schetters et al., 2001). In spite of these important differences, the three subspecies cannot be distinguished at the morphological level and their differential diagnosis has to rely upon molecular methods. Thus far, the developed methods have been based on polymorphisms of partial ssrRNA sequences (Carret et al., 1999) and of internal transcribed spacers (Zahler et al., 1998). Such methods, however, present potential drawbacks for unambiguous species identification and phylogenetic analysis. On the other hand, sequencing of the full-length ssrRNA gene provides a reliable method for both species identification and phylogenetic analysis and allows new sequences to be compared with the ssrRNA sequences of other Babesia parasites. By using this approach, the number and the taxonomic status of Babesia species of dogs has been recently subjected to revision. While traditionally only two species have been recognised, i.e. the large B. canis and the small B. gibsoni, the use of DNA sequence data revealed a more complex situation, and at least three genetically distinct small piroplasms have been demonstrated (Kjemtrup et al., 2000b). Moreover, phylogenetic analysis revealed that small Babesia parasites of dogs fall into several taxonomic groups: B. gibsoni belongs to the ‘true’ Babesia group, while the Spanish isolate (for which the name Theileria anneae has been proposed; Zahler et al., 2000a) is closely related to B. microti, and the Californian isolate is related to piroplasms isolated from Californian bighorn sheep and mule deer (Kjemtrup et al., 2000a). The ssrRNA gene phylogeny presented in this work (see Fig. 1) indicates that the three B. canis subspecies form a monophyletic, well-supported group that is the sister group to the B. odocoilei/B. gibsoni group. In a previous analysis based on partial ssrRNA sequences (Carret et al., 1999), the position of B. canis vogeli was doubtful, and a closer affinity with the B. divergens/B. odocoilei group was suggested in most phylogenetic trees. In the present analysis, however, B. canis vogeli is clearly placed within the B. canis group,
S.M. Cacci`o et al. / Veterinary Parasitology 106 (2002) 285–292
291
and it is closely related to B. canis canis. This discrepancy illustrates the limited value of phylogenetic analyses based on partial sequences. All of these species belong to the so-called ‘true’ or sensu stricto Babesia group, which include species characterised by transovarial transmission in the tick vectors, and by the exclusive development inside the erythrocytes of the vertebrate host (Mehlhorn and Schein, 1984). However, a revision of the taxonomy of the three subspecies, and their elevation to the rank of species, as suggested by some authors (Zahler et al., 1998; Carret et al., 1999), would demand further studies. The presence of B. canis canis and B. canis vogeli, and absence of B. canis rossi, in infected dogs from Europe is in agreement with the known distribution of the vectors of the two subspecies. However, two factors suggest that this situation might change in the future, indeed, the expanding international mobility of pet dogs, and the presence of competent tick vectors, may cause the parasite to spread into previously non-endemic areas, while the increased contact of dogs with the wild environment potentially expose them to infection with new Babesia species (Shaw et al., 2001). In turn, this raises a concern about the possible role of dogs as reservoirs of zoonotic disease, and indicates that the future trends should be monitored by studying the prevalence of Babesia parasites in dogs and tick vectors. References Allsopp, M.T.E.P., Cavalier-Smith, T., de Waal, D.T., Allsopp, B.A., 1994. Phylogeny and evolution of the piroplasms. Parasitology 108, 147–152. Birkenheuer, A.J., Levy, M.G., Savary, K.C., Gager, R.B., Breitschwerdt, E.B., 1999. Babesia gibsoni infections in dogs from North Carolina. J. Am. Anim. Hosp. Assoc. 35, 125–128. Breitschwerdt, E.B., Malone, J.B., MacWilliams, P., Levy, M.G., Qualls, W., Prudich, M.J., 1983. Babesiosis in the greyhound. J. Am. Vet. Med. Assoc. 182, 978–982. Carret, C., Walas, F., Carcy, B., Grande, N., Precigout, E., Moubri, K., Schetters, T.P., Gorenflot, A., 1999. Babesia canis canis, Babesia canis vogeli, Babesia canis rossi: differentiation of the three subspecies by a restriction fragment length polymorphism analysis on amplified small subunit ribosomal RNA genes. J. Euk. Microbiol. 46, 298–303. Casapulla, R., Baldi, L., Avallone, V., Sannino, R., Pazzanese, L., Mizzoni, V., 1998. Canine piroplasmosis due to Babesia gibsoni: clinical and morphological aspects. Vet. Rec. 142, 168–169. Conrad, P.A., Thomford, J.W., Yamane, I., Whiting, J., Bosma, L., Uno, T., Holshuh, H.J., Shelly, S., 1991. Haemolytic anemia caused by Babesia gibsoni infection in dogs. J. Am. Vet. Med. Assoc. 199, 601–605. Conrad, P.A., Thomford, J.W., Marsh, A., Telford, S.R., Anderson, J.F., Spielman, A., Sabin, E.A., Yamane, I., Persing, D.H., 1992. Ribosomal DNA probe for differentiation of Babesia microti and B. gibsoni isolates. J. Clin. Microbiol. 30, 1210–1215. Farwell, G.E., LeGrand, E.K., Cobb, C.C., 1982. Clinical observations on Babesia gibsoni and Babesia canis infections in dogs. J. Am. Vet. Med. Assoc. 180, 507–511. Hauschild, S., Shayan, P., Schein, E., 1995. Characterization and comparison of merozoite antigens of different Babesia canis isolates by serological and immunological investigations. Parasitol. Res. 81, 638–642. Kjemtrup, A.M., Thomford, J., Robinson, T., Conrad, P.A., 2000a. Phylogenetic relationships of human and wildlife piroplasm isolates in the western United States inferred from the 18S nuclear small subunit RNA gene. Parasitology 120, 487–493. Kjemtrup, A.M., Kocan, A.A., Whitworth, L., Meinkoth, J., Birkenheuer, A.J., Cummings, J., Boudreaux, M.K., Stockham, S.L., Irizarry-Rovira, A., Conrad, P.A., 2000b. There are at least three genetically distinct small piroplasms from dogs. Int. J. Parasitol. 30, 1501–1505. Kuttler, K.L., 1988. World-wide impact of babesiosis. In: Ristic M (Ed.), Babesiosis of Domestic Animals and Man. CRC Press, Boca Raton, FL, pp. 1–22. Mehlhorn, H., Schein, E., 1984. The piroplasms: life cycle and sexual stages. Adv. Parasitol. 23, 37–103.
292
S.M. Cacci`o et al. / Veterinary Parasitology 106 (2002) 285–292
Mehlhorn, H., Schein, E., 1998. Redescritpion of Babesia equi Laveran, 1901 as Theileria equi Mehlhorn, Schein 1998. Parasitol. Res. 84, 467–475. Moreau, Y., Martinod, S., Fayet, G., 1988. Epidemiological and immunoprophylactic aspects of canine babesiosis in France. In: Ristic M (Ed.), Babesiosis of Domestic Animals and Man. CRC Press, Boca Raton, FL, pp. 191–196. Page, R.D., 1996. Treeview: an application to display phylogenetic trees on personal computers. Comput. Appl. Biosci. 12, 357–358. Reyers, F., Leisewitz, A.L., Lobetti, R.G., Milner, R.J., Jacobson, L.S., 1998. Canine babesiosis in South Africa: more than one disease. Does this serve as a model for falciparum malaria? Ann. Trop. Med. Parasitol. 92, 503–511. Schetters, T.P.M., Kleuskens, J., Scholtes, N.C., Bos, H.J., 1995. Strain variation limits protective activity of vaccines based on soluble Babesia canis antigens. Parasite Immunol. 17, 215–218. Schetters, T.P.M., Moubri, K., Precigout, E., Kleuskens, J., Scholtes, N.C., Gorenflot, A., 1997. Different Babesia canis isolates, different diseases. Parasitology 115, 485–493. Schetters, T.P.M., Kleuskens, J., Scholtes, N.C., Gorenflot, A., Moubri, K., Vermeulen, A.N., 2001. Vaccination of dogs against heterologous Babesia canis infection using antigens from culture supernatants. Vet. Parasitol. 100, 75–86. Shaw, S.E., Day, M.J., Birtles, R.J., Breitscwerdt, E.B., 2001. Tick-borne infections diseases of dogs. Trends Parasitol. 17, 74–80. Strimmer, K., von Haeseler, A., 1996. Quartet puzzling: a quartet maximum likelihood method for reconstructing tree topologies. Mol. Biol. E13, 964–969. Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acid Res. 22, 4673–4680. Uilenberg, G., Franssen, F.F.J., Perie, M., Spanjer, A.A.M., 1989. Three groups of Babesia canis distinguished and a proposal for nomenclature. Vet. Q. 11, 33–40. Zahler, M., Schein, E., Rinder, H., Gothe, R., 1998. Characteristic genotypes discriminate between Babesia canis isolates of differing vector specificity and pathogenicity to dogs. Parasitol. Res. 84, 544–548. Zahler, M., Rinder, H., Schein, E., Gothe, R., 2000a. Detection of a new pathogenic Babesia microti-like species in dogs. Vet. Parasitol. 89, 241–248. 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, 365–369.
Molecular characterisation of Babesia canis canis and Babesia canis vogeli from naturally infected European dogs夽 Simone M. Cacciò a,∗ , Boris Antunovic b , Annabella Moretti c , Vittorio Mangili d , Albert Marinculic e , Renata Rafaj Baric f , Susan B. Slemenda g , Norman J. Pieniazek g a b
Laboratory of Parasitology, Istituto Superiore di Sanità, Viale Regina Elena 299, I-00161 Rome, Italy Department of Animal Husbandry, Faculty of Agriculture, J.J. Strossmayer University, Osijek, Croatia c Department of Biopathological Veterinary Sciences, University of Perugia, Perugia, Italy d Department of Pathology, Diagnostic and Clinical Veterinary, University of Perugia, Perugia, Italy e Department for Parasitology and Invasive Diseases, Faculty of Veterinary Medicine, University of Zagreb, Zagreb, Croatia f Clinics for Internal Diseases, Faculty of Veterinary Medicine, University of Zagreb, Zagreb, Croatia g Division of Parasitic Diseases, Centers for Disease Control and Prevention, Atlanta, GA, USA Received 8 February 2002; accepted 5 April 2002
Abstract The morphologically small Babesia species isolated from naturally infected dogs in Europe, Japan, and US are described as Babesia gibsoni despite the fact that molecular techniques show that they should be assigned to two or three separate taxons. The morphologically large Babesia isolated from dogs in Europe, Africa, and US were generally classified as B. canis until it was proposed to distinguish three related, albeit genetically distinct subspecies of this genus, namely B. canis canis, B. canis rossi, and B. canis vogeli. The insight into the molecular taxonomy of canine piroplasms is, however, limited because only partial small subunit ribosomal RNA (ssrRNA) sequence data exist for two species from the B. canis group. In this work, we molecularly characterised natural Babesia infections in 11 dogs from Croatia, France, Italy, and Poland. These infections were diagnosed as caused by B. canis canis and B. canis vogeli based on the analysis of the complete sequence of the ssrRNA genes. Phylogenetic analysis confirmed that the large Babesia species of dogs belong the to the Babesia sensu stricto clade, which includes species characterised by transovarial transmission in the tick vectors and by exclusive development inside the mammalian host erythrocytes. The 夽 Nucleotide sequence data reported in this paper are available in the GenBankTM database under the accession nos. AY072925 and AY072926. ∗ Corresponding author. Tel.: +39-06-4990-2310; fax: +39-06-4938-7065. E-mail address: [email protected] (S.M. Cacci`o).
0304-4017/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 4 0 1 7 ( 0 2 ) 0 0 1 1 2 - 7
286
S.M. Cacci`o et al. / Veterinary Parasitology 106 (2002) 285–292
new data facilitate the reliable molecular diagnosis of the subspecies of B. canis. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Babesia canis; Dog; Small subunit ribosomal RNA gene; Phylogeny
1. Introduction Babesia species are tick-transmitted apicomplexan parasites that infect a wide range of vertebrate hosts and cause severe diseases in wild and domestic animals (Kuttler, 1988). The identification of individual species has traditionally been based on the host specificity and on the morphology of the intraerythrocytic forms (the piroplasms). Babesia parasites are usually classified as small (1.5–2.5 m) or large (3.0–5.0 m) piroplasms. Babesia canis and Babesia gibsoni are recognised as the two species that cause canine babesiosis worldwide. B. canis is a large piroplasm (4–5 m), which usually occurs as a single pear-shaped piroplasm or in pairs of merozoites divided by binary fission within the erythrocyte. Previous studies, on the basis of differences in the geographical distribution, vector specificity, and antigenic properties (Uilenberg et al., 1989; Hauschild et al., 1995), recognised that large canine piroplasms are subdivided into three subspecies, namely B. canis canis, transmitted by Dermacentor reticulatus in Europe, B. canis vogeli, transmitted by Rhipicephalus sanguineus in tropical and subtropical countries, and B. canis rossi, transmitted by Haemophysalis leachi in South Africa. B. gibsoni has been found to be associated with infection of dogs in Asia, North America, northern and eastern Africa, and Europe (Conrad et al., 1991; Casapulla et al., 1998; Birkenheuer et al., 1999). B. gibsoni is a small parasite that commonly appears as individual ring forms or piriform bodies of a size between 1 and 2.5 m (Conrad et al., 1991). However, it is difficult to morphologically distinguish this parasite from other small piroplasms such as B. equi (now reclassified as Theileria equi; Mehlhorn and Schein, 1998) or B. microti (Conrad et al., 1992). Traditionally, all small canine piroplasms were identified as B. gibsoni based on the assumption that no other small Babesia species infects dogs. This hypothesis was clearly wrong, as demonstrated by the recent genetic characterisation of small piroplasms of dogs from Spain and California (Zahler et al., 2000a,b; Kjemtrup et al., 2000a,b). Phylogenetic analysis of the small subunit ribosomal gene (ssrRNA) of these isolates showed that they are clearly distinct from B. gibsoni, being more closely related to B. microti (the B. gibsoni Spanish isolate) or to the Californian isolates from bighorn sheep and the mule deer (the B. gibsoni Californian isolate). These results show that molecular techniques, in particular the characterisation of the ssrRNA genes, represent an objective and precise method of species identification and phylogenetic classification. In spite of this fact, few data are available with respect to the large piroplasms of dogs, and a complete characterisation of the ssrRNA gene has been reported only for one isolate of B. canis rossi (Allsopp et al., 1994). Recently, two PCR–RFLP assays have been developed to discriminate the large canine piroplasms, based on polymorphisms of partial ssrRNA sequences (Carret et al., 1999), or of internally transcribed spacers (Zahler et al., 1998).
S.M. Cacci`o et al. / Veterinary Parasitology 106 (2002) 285–292
287
In this study, a molecular analysis of the complete ssrRNA gene was done to characterise Babesia parasites collected from 11 naturally infected dogs in Croatia, France, Italy, and Poland. 2. Materials and methods 2.1. Parasite isolates Blood from 11 dogs showing clinical signs of babesiosis, including apathy, fever, and anaemia, were collected close to Zagreb (Croatia), and in Perugia (Italy) during fall 2000 and spring 2001. No history of travel was recorded for any of the eight Croatian dogs, while two (referred to as Bc11 and Bc12) of the three Italian dogs acquired the infection during a stay in an area close to Nantes (France), and in an area close to Warsaw (Poland), respectively. Examination of Giemsa-stained blood smears revealed the presence of Babesia parasites, which have morphology (pear-shaped parasites, often occurring in pairs) and size (3–5 m) typical of the large Babesia, i.e. of B. canis. 2.2. DNA isolation, amplification and sequencing DNA was extracted from 200 l of whole blood using the QIAmp blood kit (Qiagen), according to the manufacturer’s instructions. Amplification of the ssrRNA gene was performed using universal apicomplexan primers: the forward primer 5 -AACCTGGTTGATCCTGCCAGTAGTCAT-3 , and the reverse primer 5 -GAATGATCCTTCCGCAGGTTCACCTAC-3 . The PCR mix consisted of 1× buffer containing 1.5 mM MgCl2 , 200 m of each dNTP, 20 pmol of each primer, 2.5 U of Taq Polymerase (Takara), and extracted template DNA. Reactions consisted of an initial 5 min denaturation step at 95 ◦ C, followed by 35–40 cycles of denaturation at 94 ◦ C for 30 s, annealing at 65 ◦ C for 30 s, and extension at 72 ◦ C for 2 min. Final extension was done at 72 ◦ C for 7 min, followed by a hold step at 4 ◦ C. PCR products were purified using the QiaQuick purification kit (Qiagen), and fully sequenced using the ABI prism dye sequencing kit (Applied Biosystems), and a set of internal primers. Sequencing reactions were analysed on the ABI 3100 automatic DNA sequencer (Applied Biosystems). Sequences were assembled by using the program SeqMan II (DNASTAR). The GenBank accession nos. for the complete ssrRNA sequences of B. canis vogeli and B. canis canis are AY072925 and AY072926, respectively. 2.3. Phylogenetic analysis The sequences of the ssrRNA gene from selected Babesia and Theileria species (see legend to Fig. 1) were retrieved from the GenBankTM database and aligned with the newly established ssrRNA sequences of B. canis canis and B. canis vogeli by using the program Clustal W (Thompson et al., 1994). Phylogenetic analysis was done with the DNAML, DNAPARS, DNADIST, and NEIGHBOR programs from the PHYLIP package (Felsenstein, 1989, version 3.573c) and with the TREE-PUZZLE program (Strimmer and von Haeseler,
288
S.M. Cacci`o et al. / Veterinary Parasitology 106 (2002) 285–292
Fig. 1. Phylogenetic tree based on full-length ssrRNA sequences of selected Babesia and Theileria species. Quartet puzzling maximum likelihood results are shown using Babesia microti as the outgroup. Numbers at the nodes indicate the quartet puzzling support for each internal branch. Scale bar indicates an evolutionary distance of 0.01 nucleotides per position in the sequence. Vertical distances are for clarity only. GenBank accession nos. of the sequences used for analysis: Babesia bigemina gene A, X59604; Babesia caballi, Z15104; Babesia canis canis, AY072926; Babesia canis vogeli, AY072925; Babesia canis rossi, L19079; Babesia microti, U09833; Babesia odocoilei, U16369; Babesia gibsoni (Oklahoma dog isolate), AF205636; Theileria parva, AF013418; Theileria sergenti, AB016074. New sequences (AY072925 and AY072926) are indicated by arrows.
1996, version 5.0). Phylogenetic trees inferred by these programs were drawn using the TreeView program (Page, 1996). 3. Results 3.1. Amplification and sequencing of the ssrRNA coding regions The amplification of the full-length region coding for the ssrRNA from 11 B. canis isolates yielded single products of approximately 1700 bp (data not shown). Sequence analysis
S.M. Cacci`o et al. / Veterinary Parasitology 106 (2002) 285–292
289
of the PCR products revealed the presence of two closely related, albeit different, ssrRNA sequences. The first sequence had a length of 1714 bp, and was found in all Croatian dogs (isolates Bc1–Bc8) and in two Italian dogs (isolate Bc10 and Bc12, of which the latter represents an infection acquired during a stay in Poland). No sequence differences were observed among the 10 isolates. BLAST search against GenBank revealed the highest similarity score (99.5%) with a partial, 364 bp long sequence from B. canis canis (accession no. AJ009795). There were two differences, namely a nucleotide deletion at position 434, and a nucleotide insertion at position 453 (numbering refers to the full-length ssrRNA sequence). Notably, the two differences were close to the primer used for PCR amplification and direct sequencing of AJ009795, and might therefore reflect sequencing errors; however, minor strain variations cannot be ruled out. The sequence from the isolate Bc10 was submitted to GenBank as the complete B. canis canis ssrRNA sequence (accession no. AY072926). The second sequence had a length of 1713 bp and was associated with the infection of another dog (isolate Bc11), which acquired the infection during a stay in France. A BLAST search against GenBank revealed the highest similarity score (98.6%) with a partial, 363 bp long sequence from B. canis vogeli (accession no. AJ009796). There were four nucleotide substitutions (at positions 488, 599, 632, and 789) and one nucleotide insertion (at position 776, numbering refers to the full-length ssrRNA sequence). Again, the observed differences may represent sequencing errors or minor strain variation. The sequence from isolate Bc11 was submitted to GenBank as the complete B. canis vogeli ssrRNA sequence (GenBank accession no. AY072925). A comparison between the complete ssrRNA sequences of B. canis canis and B. canis vogeli yielded an identity score of 98%. The remaining 2% represented 27 nucleotide substitutions and seven deletions/insertions. When compared to the B. canis rossi ssrRNA sequence, the identity scores were 95.7% for B. canis canis and 95.3% for B. canis vogeli. 3.2. Phylogenetic analysis The ssrRNA sequences of the Bc10 and Bc11 isolates were aligned with eight selected species from the Babesia and Theileria genera (see legend to Fig. 1 for the list of species and the GenBank accession no. for the sequences used). The alignment was flush-trimmed at the ends and all columns containing gaps and ambiguous characters were removed, leaving 1626 columns (the alignment is available from the authors upon request). B. microti, which is always placed outside the clade grouping the ‘true’ Babesia and Theileria species, was selected as the outgroup to orient the tree. All four phylogenetic methods used (parsimony, maximum likelihood, quartet puzzling, and distance analysis) generated a tree with the same topology (Fig. 1). All three B. canis subspecies form a well-resolved group, which is the sister clade to a group formed by B. odocoilei (a parasite of cervids) and B. gibsoni (Oklahoma dog isolate).
4. Discussion Canine babesiosis is a tick-borne infection caused by the intraerytrocytic apicomplexan parasites B. canis and B. gibsoni. The two species are easily distinguished by light
290
S.M. Cacci`o et al. / Veterinary Parasitology 106 (2002) 285–292
microscopy, as they differ considerably in size (Kuttler, 1988). B. canis is widespread in Europe, and affects more than 400,000 dogs per year in France alone (Moreau et al., 1988). Young dogs and pups are highly susceptible, with death occurring in the absence of treatment, and transplacental transmission has been suggested since high parasitemia has been observed in neonates (Farwell et al., 1982; Breitschwerdt et al., 1983). Previous studies recognised three subspecies within the B. canis group, on the basis of differences in the geographical distribution, vector specificity, and antigenic properties (Uilenberg et al., 1989; Hauschild et al., 1995). Moreover, the clinical manifestations of infections caused by the three subspecies were found to differ significantly. As a matter of fact, B. canis rossi, the prevalent subspecies in South Africa, is highly virulent and causes either a haemolytic disease (possibly immune-mediated) or an acute and overwhelming inflammatory response (Reyers et al., 1998). On the other hand, infection by B. canis canis results in transient parasitemia, usually below 1%, and clinical disease is associated with congestion of internal organs (Schetters et al., 1997). Infection by B. canis vogeli leads to a relatively mild disease, often without evident clinical signs. Moreover, no cross-protection was observed when dogs vaccinated with soluble parasite antigens (SPA) prepared from one strain were experimentally challenged with a different strain (Schetters et al., 1995). Antigenic variation has been considered as a possible explanation for the observed lack of cross-protection and, in agreement with this interpretation, the use of a mixture of SPA from both a European B. canis isolate and a South African B. canis rossi isolate induced a protective immunity against heterologous B. canis infection (Schetters et al., 2001). In spite of these important differences, the three subspecies cannot be distinguished at the morphological level and their differential diagnosis has to rely upon molecular methods. Thus far, the developed methods have been based on polymorphisms of partial ssrRNA sequences (Carret et al., 1999) and of internal transcribed spacers (Zahler et al., 1998). Such methods, however, present potential drawbacks for unambiguous species identification and phylogenetic analysis. On the other hand, sequencing of the full-length ssrRNA gene provides a reliable method for both species identification and phylogenetic analysis and allows new sequences to be compared with the ssrRNA sequences of other Babesia parasites. By using this approach, the number and the taxonomic status of Babesia species of dogs has been recently subjected to revision. While traditionally only two species have been recognised, i.e. the large B. canis and the small B. gibsoni, the use of DNA sequence data revealed a more complex situation, and at least three genetically distinct small piroplasms have been demonstrated (Kjemtrup et al., 2000b). Moreover, phylogenetic analysis revealed that small Babesia parasites of dogs fall into several taxonomic groups: B. gibsoni belongs to the ‘true’ Babesia group, while the Spanish isolate (for which the name Theileria anneae has been proposed; Zahler et al., 2000a) is closely related to B. microti, and the Californian isolate is related to piroplasms isolated from Californian bighorn sheep and mule deer (Kjemtrup et al., 2000a). The ssrRNA gene phylogeny presented in this work (see Fig. 1) indicates that the three B. canis subspecies form a monophyletic, well-supported group that is the sister group to the B. odocoilei/B. gibsoni group. In a previous analysis based on partial ssrRNA sequences (Carret et al., 1999), the position of B. canis vogeli was doubtful, and a closer affinity with the B. divergens/B. odocoilei group was suggested in most phylogenetic trees. In the present analysis, however, B. canis vogeli is clearly placed within the B. canis group,
S.M. Cacci`o et al. / Veterinary Parasitology 106 (2002) 285–292
291
and it is closely related to B. canis canis. This discrepancy illustrates the limited value of phylogenetic analyses based on partial sequences. All of these species belong to the so-called ‘true’ or sensu stricto Babesia group, which include species characterised by transovarial transmission in the tick vectors, and by the exclusive development inside the erythrocytes of the vertebrate host (Mehlhorn and Schein, 1984). However, a revision of the taxonomy of the three subspecies, and their elevation to the rank of species, as suggested by some authors (Zahler et al., 1998; Carret et al., 1999), would demand further studies. The presence of B. canis canis and B. canis vogeli, and absence of B. canis rossi, in infected dogs from Europe is in agreement with the known distribution of the vectors of the two subspecies. However, two factors suggest that this situation might change in the future, indeed, the expanding international mobility of pet dogs, and the presence of competent tick vectors, may cause the parasite to spread into previously non-endemic areas, while the increased contact of dogs with the wild environment potentially expose them to infection with new Babesia species (Shaw et al., 2001). In turn, this raises a concern about the possible role of dogs as reservoirs of zoonotic disease, and indicates that the future trends should be monitored by studying the prevalence of Babesia parasites in dogs and tick vectors. References Allsopp, M.T.E.P., Cavalier-Smith, T., de Waal, D.T., Allsopp, B.A., 1994. Phylogeny and evolution of the piroplasms. Parasitology 108, 147–152. Birkenheuer, A.J., Levy, M.G., Savary, K.C., Gager, R.B., Breitschwerdt, E.B., 1999. Babesia gibsoni infections in dogs from North Carolina. J. Am. Anim. Hosp. Assoc. 35, 125–128. Breitschwerdt, E.B., Malone, J.B., MacWilliams, P., Levy, M.G., Qualls, W., Prudich, M.J., 1983. Babesiosis in the greyhound. J. Am. Vet. Med. Assoc. 182, 978–982. Carret, C., Walas, F., Carcy, B., Grande, N., Precigout, E., Moubri, K., Schetters, T.P., Gorenflot, A., 1999. Babesia canis canis, Babesia canis vogeli, Babesia canis rossi: differentiation of the three subspecies by a restriction fragment length polymorphism analysis on amplified small subunit ribosomal RNA genes. J. Euk. Microbiol. 46, 298–303. Casapulla, R., Baldi, L., Avallone, V., Sannino, R., Pazzanese, L., Mizzoni, V., 1998. Canine piroplasmosis due to Babesia gibsoni: clinical and morphological aspects. Vet. Rec. 142, 168–169. Conrad, P.A., Thomford, J.W., Yamane, I., Whiting, J., Bosma, L., Uno, T., Holshuh, H.J., Shelly, S., 1991. Haemolytic anemia caused by Babesia gibsoni infection in dogs. J. Am. Vet. Med. Assoc. 199, 601–605. Conrad, P.A., Thomford, J.W., Marsh, A., Telford, S.R., Anderson, J.F., Spielman, A., Sabin, E.A., Yamane, I., Persing, D.H., 1992. Ribosomal DNA probe for differentiation of Babesia microti and B. gibsoni isolates. J. Clin. Microbiol. 30, 1210–1215. Farwell, G.E., LeGrand, E.K., Cobb, C.C., 1982. Clinical observations on Babesia gibsoni and Babesia canis infections in dogs. J. Am. Vet. Med. Assoc. 180, 507–511. Hauschild, S., Shayan, P., Schein, E., 1995. Characterization and comparison of merozoite antigens of different Babesia canis isolates by serological and immunological investigations. Parasitol. Res. 81, 638–642. Kjemtrup, A.M., Thomford, J., Robinson, T., Conrad, P.A., 2000a. Phylogenetic relationships of human and wildlife piroplasm isolates in the western United States inferred from the 18S nuclear small subunit RNA gene. Parasitology 120, 487–493. Kjemtrup, A.M., Kocan, A.A., Whitworth, L., Meinkoth, J., Birkenheuer, A.J., Cummings, J., Boudreaux, M.K., Stockham, S.L., Irizarry-Rovira, A., Conrad, P.A., 2000b. There are at least three genetically distinct small piroplasms from dogs. Int. J. Parasitol. 30, 1501–1505. Kuttler, K.L., 1988. World-wide impact of babesiosis. In: Ristic M (Ed.), Babesiosis of Domestic Animals and Man. CRC Press, Boca Raton, FL, pp. 1–22. Mehlhorn, H., Schein, E., 1984. The piroplasms: life cycle and sexual stages. Adv. Parasitol. 23, 37–103.
292
S.M. Cacci`o et al. / Veterinary Parasitology 106 (2002) 285–292
Mehlhorn, H., Schein, E., 1998. Redescritpion of Babesia equi Laveran, 1901 as Theileria equi Mehlhorn, Schein 1998. Parasitol. Res. 84, 467–475. Moreau, Y., Martinod, S., Fayet, G., 1988. Epidemiological and immunoprophylactic aspects of canine babesiosis in France. In: Ristic M (Ed.), Babesiosis of Domestic Animals and Man. CRC Press, Boca Raton, FL, pp. 191–196. Page, R.D., 1996. Treeview: an application to display phylogenetic trees on personal computers. Comput. Appl. Biosci. 12, 357–358. Reyers, F., Leisewitz, A.L., Lobetti, R.G., Milner, R.J., Jacobson, L.S., 1998. Canine babesiosis in South Africa: more than one disease. Does this serve as a model for falciparum malaria? Ann. Trop. Med. Parasitol. 92, 503–511. Schetters, T.P.M., Kleuskens, J., Scholtes, N.C., Bos, H.J., 1995. Strain variation limits protective activity of vaccines based on soluble Babesia canis antigens. Parasite Immunol. 17, 215–218. Schetters, T.P.M., Moubri, K., Precigout, E., Kleuskens, J., Scholtes, N.C., Gorenflot, A., 1997. Different Babesia canis isolates, different diseases. Parasitology 115, 485–493. Schetters, T.P.M., Kleuskens, J., Scholtes, N.C., Gorenflot, A., Moubri, K., Vermeulen, A.N., 2001. Vaccination of dogs against heterologous Babesia canis infection using antigens from culture supernatants. Vet. Parasitol. 100, 75–86. Shaw, S.E., Day, M.J., Birtles, R.J., Breitscwerdt, E.B., 2001. Tick-borne infections diseases of dogs. Trends Parasitol. 17, 74–80. Strimmer, K., von Haeseler, A., 1996. Quartet puzzling: a quartet maximum likelihood method for reconstructing tree topologies. Mol. Biol. E13, 964–969. Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acid Res. 22, 4673–4680. Uilenberg, G., Franssen, F.F.J., Perie, M., Spanjer, A.A.M., 1989. Three groups of Babesia canis distinguished and a proposal for nomenclature. Vet. Q. 11, 33–40. Zahler, M., Schein, E., Rinder, H., Gothe, R., 1998. Characteristic genotypes discriminate between Babesia canis isolates of differing vector specificity and pathogenicity to dogs. Parasitol. Res. 84, 544–548. Zahler, M., Rinder, H., Schein, E., Gothe, R., 2000a. Detection of a new pathogenic Babesia microti-like species in dogs. Vet. Parasitol. 89, 241–248. 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, 365–369.