International Journal for Parasitology 36 (2006) 779–789 www.elsevier.com/locate/ijpara Description of Babesia duncani n.sp. (Apicomplexa: Babesiidae) from humans and its differentiation from other piroplasms* Patricia A. Conrad a,*, Anne M. Kjemtrup b, Ramon A. Carreno c, John Thomford d, Katlyn Wainwright a, Mark Eberhard e, Rob Quick e, Sam R. Telford III f, Barbara L. Herwaldt e a Department of Pathology, Microbiology and Immunology, School of Veterinary Medicine, University of California, 1126 Haring Hall One Shields Avenue, Davis, CA 95616-8745, USA b Vector-Borne Disease Section, California Department of Health Service, 1616 Capitol Avenue, MS 7307, P.O. Box 942732, Sacramento, CA 95899-7413, USA c Department of Zoology, Ohio Wesleyan University, Delaware, OH 43015, USA d Department of Biological Sciences, Mira Costa College, Oceanside, CA 92056, USA e Division of Parasitic Diseases, Centers for Disease Control and Prevention, 4770 Buford Highway, Atlanta, GA 30341-3724, USA f Division of Infectious Diseases, Cummings School of Veterinary Medicine, Tufts University, 200 Westboro Road, North Grafton, MA 001536, USA Received 12 December 2005; received in revised form 15 March 2006; accepted 29 March 2006 Abstract The morphologic, ultrastructural and genotypic characteristics of Babesia duncani n.sp. are described based on the characterization of two isolates (WA1, CA5) obtained from infected human patients in Washington and California. The intraerythrocytic stages of the parasite are morphologically indistinguishable from Babesia microti, which is the most commonly identified cause of human babesiosis in the USA. Intraerythrocytic trophozoites of B. duncani n.sp. are round to oval, with some piriform, ring and ameboid forms. Division occurs by intraerythrocytic schizogony, which results in the formation of merozoites in tetrads (syn. Maltese cross or quadruplet forms). The ultrastructural features of trophozoites and merozoites are similar to those described for B. microti and Theileria spp. However, intralymphocytic schizont stages characteristic of Theileria spp. have not been observed in infected humans. In phylogenetic analyses based on sequence data for the complete18S ribosomal RNA gene, B. duncani n.sp. lies in a distinct clade that includes isolates from humans, dogs and wildlife in the western United States but separate from Babesia sensu stricto, Theileria spp. and B. microti. ITS2 sequence analysis of the B. duncani n.sp. isolates (WA1, CA5) show that they are phylogenetically indistinguishable from each other and from two other human B. duncani-type parasites (CA6, WA2 clone1) but distinct from other Babesia and Theileria species sequenced. This analysis provides robust molecular support that the B. duncani n.sp. isolates are monophyletic and the same species. The morphologic characteristics together with the phylogenetic analysis of two genetic loci support the assertion that B. duncani n.sp. is a distinct species from other known Babesia spp. for which morphologic and sequence information are available. q 2006 Published by Elsevier Ltd on behalf of Australian Society for Parasitology Inc. Keywords: Babesia; Theileria; Babesiosis; Ultrastructure; 18S rRNA; ITS2 1. Introduction The order Piroplasmorida Wenyon, 1926, in the Phylum Apicomplexa, contains two families of tick-borne intraerythrocytic protozoa, the Babesiidae and Theileriidae (Levine, 1988). These families are comprised of the genera Babesia, Theileria * Note: Nucleotide sequence data newly reported in this paper are available in the GenBank database under the accession numbers AY998760–AY998772. * Corresponding author. Tel.: C1 530 752 7210; fax: C1 530 752 3349. E-mail address: paconrad@ucdavis.edu (P.A. Conrad). and Cytauxzoon, which, as a group, contain the piroplasm parasites, named as such because the intraerythrocytic merozoite stages are pear-shaped (Levine, 1971). The intraerythocytic stages of some species within these genera may have similar morphology; however, Babesia spp. do not have an exoerythrocytic stage in mammalian hosts which distinguishes the genus Babesia from the genera Theileria and Cytauxzoon. Most of the human cases of Babesia acquired in the United States have been due to small (1–3 mm) intraerythrocytic babesial parasites (reviewed by Kjemtrup and Conrad, 2000, 2001). Babesia microti, the most commonly identified etiologic 0020-7519/$30.00 q 2006 Published by Elsevier Ltd on behalf of Australian Society for Parasitology Inc. doi:10.1016/j.ijpara.2006.03.008 780 P.A. Conrad et al. / International Journal for Parasitology 36 (2006) 779–789 Table 1 WA1-type babesial parasites isolated from humans in the western United States since 1991 Name of parasitea Isolate of parasite obtainedb Year case/parasite identified Mode of transmissionc Characteristics of infected person Citations WA1 WA2 WA3 CA1 CA2 CA3 Viable (syntype isolate) Viable Viable DNA only DNA only DNA only 1991 1994 1994 1991 1992 1993 Index case-patient Blood recipient Blood donor Asplenic Asplenic Asplenic; died Quick et al., 1993 Herwaldt et al., 1997 Herwaldt et al., 1997 Persing et al., 1995 Persing et al., 1995 Persing et al., 1995 CA4 CA5 DNA only Viable (syntype isolate) 1994 2000 Tick-borne Blood-borne Tick-borne Tick-borne Tick-borne Tick-borne (tick bite 19 days before symptom onset) Tick-borne Blood-borne Persing et al., 1995 Kjemtrup et al., 2002 CA6 Viable 2000 Tick- borne Asplenic Blood recipient; premature infant Blood donor Kjemtrup et al., 2002 a The parasite names reflect the state in which the infected person lived (i.e., ‘WA’ for ‘Washington’ or ‘CA’ for ‘California’) and the chronologic order, within a state, in which cases/parasites were identified and characterized. The state of residence also is the state in which infection was acquired, with the possible exception of the person infected with CA6, a blood donor who lived in northern California but had travelled to Oregon (in 1999-2000), several months before donating blood (in 2000). Therefore, the origin of parasites CA5 and CA6 was either northern California or Oregon. b By intraperitoneal inoculation of hamsters (Mesocricetus auratus) with blood from the infected person. c All cases indicated as tick-borne were presumed to have been acquired by tick exposure (tick vector not yet identified). agent, is endemic in the northeastern and midwestern states (reviewed by Telford et al., 1993). In 1991, a babesial piroplasm isolated from a patient in Washington was shown to be morphologically similar to, but molecularly and physiologically distinct from, B. microti, and has since been referred to as WA1 (Quick et al., 1993; Thomford et al., 1994). Only two human cases of babesiosis acquired in the western United States had previously been reported (Scholtens et al., 1968; Bredt et al., 1981). Both cases occurred in California before molecular tools for the characterization of Babesia spp. were available, and the infecting species were not determined in either case. Since the characterization of the WA1-type parasite in 1991, eight more cases of human babesiosis in the western USA have been attributed to WA1-type or closely related babesial parasites (Table 1). In some cases, viable parasite isolates were obtained from these patients by blood inoculation into rodents, whereas in others only DNA was isolated. Samples of parasites taken from a single host are referred to in this paper as ‘isolates’, whether viable or not. Medical history review and patient follow-up determined that infections were acquired either through natural exposure (presumably tick-bite) or through blood transfusion. Viable isolates were obtained from both the asymptomatic blood donor (WA3) and recipient (WA2) in the first recognized case of human infection by blood transfusion of WA1-type babesial parasites (Herwaldt et al., 1997). More recently, a WA1-type Babesia was identified in a premature infant in California (CA5) who was the recipient of blood from a healthy donor with a subclinical babesial infection (Kjemtrup et al., 2002). Parasites isolated by hamster inoculation from the donor (CA6) 2 months after donating the infected blood proved to be virtually indistinguishable by analysis of the complete 18S rRNA gene from the parasites isolated from the infant (2 bp differences). The 18S rRNA gene from the infant’s isolate (CA5) was phylogenetically indistinguishable from WA1 (8 bp differences) (Kjemtrup et al., 2002). The only other piroplasm reported to cause human infection in the western United States was a Babesia divergenslike parasite in Washington State (Herwaldt et al., 2004). The acquisition and characterization of multiple isolates of WA1-type parasites from humans over the past 12 years now provides sufficient data to show that this is a distinct parasite species, which requires proper taxonomic identification. The description of the proposed new species is based on characterization data from two syntype isolates rather than one prototype isolate. The peripheral blood smears from the index case-patient in Washington (Table 1), although sufficient to identify the intraerythrocytic stages as piroplasms, were inadequate for definitive morphological description of the parasites. Therefore, the morphological descriptions and measurements of the intraerythrocytic trophozoites and merozoites provided in this paper are based on light microscopic examination of blood smears from the clinically affected infant from which CA5 was isolated. Previous analysis (Kjemtrup et al., 2002) has shown that the 18S rRNA genes of the WA1 and CA5 isolates are virtually indistinguishable with 8 bp differences within 1715 bp sequenced, occurring at positions 18, 463, 616 (insertion), 623 (insertion), 818–819 (insertion), 872 and 905 of the WA1 18S sequence (GenBank accession number AF158700). Both WA1 and CA5 isolates caused fatal infections when inoculated into Syrian hamsters (Mesocricetus auratus) as described (Quick et al., 1993; Thomford et al., 1994; Wozniak et al., 1996; Kjemtrup et al., 2002). Pathogenicity of WA1 parasites in experimentally infected hamsters and mice was markedly greater than seen in infections with equivalent doses of B. microti (Wozniak et al., 1996; Hemmer et al., 1999, 2000a, b). New data are also presented herein to show identity between these isolates in a second locus, the internal transcribed spacer region 2 (ITS2), as well as the phylogenetic relationships of the WA1-type parasites and other piroplasm parasites based on this sequence. Biological data relevant to this species has been reviewed P.A. Conrad et al. / International Journal for Parasitology 36 (2006) 779–789 (Kjemtrup and Conrad, 2000, 2001; Conrad et al., 2003) and are briefly described in this paper. The data described here indicate that the WA1-type parasite should be classified in the genus Babesia and is distinct from other characterized species. The proposed name for this new species is Babesia duncani n.sp. (henceforth referred to as B. duncani). 2. Materials and methods 2.1. Light microscopy Four thin blood smears from the California infant infected with B. duncani (CA5), were fixed in methanol, stained in 10% Giemsa solution for 1 h and examined with light microscope under oil immersion. On each slide, at least 500 red blood cells were counted to calculate percent parasitemia (number of infected red blood cells/total red blood cells counted!100). Additionally, 71 well-stained and clearly visible individual parasites were identified and measured. Images of parasites were classified either as a piriform, ring form or ameboid trophozoite or as a merozoite which may be in a dividing tetrad form consisting of four merozoites (Thomford et al., 1994). The length and width of each parasite was measured using an ocular micrometer. 2.2. Transmission electron microscopy Fresh infected blood samples from human cases (WA1, CA5) were not available for transmission electron microscopic studies. Therefore, ultrastructural studies were conducted using erythrocytes obtained directly from a parasitemic hamster that was experimentally infected with B. duncani originally from the WA1 human case. All animal experiments were humanely conducted with approval of the Institutional Animal Care and Use Committee (IACUC) at the University of California, Davis, which is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care, International (AAALAC). Blood was collected under anesthesia by cardiac puncture, fixed overnight in 2% glutaraldehyde in 0.085 M cacodylate buffer (pH 7.4), postfixed for 1 h in Karnovsky fixative and pelleted by centrifugation at 5000 rpm for 10 min in a 1.5 ml microcentrifuge tube. After centrifugation, the supernatant was discarded and the pellet was re-suspended in a drop of 1% agar to facilitate processing and handling of the specimen. After post-fixation for 1 h in 1% osmium tetroxide, the pellet was dehydrated through increasing concentrations of ethanol followed by two changes in pure propylene oxide and embedded in araldite resin. Serial thin (70 nm) sections were collected sequentially onto 1.0!0.25 mm slot grids coated with a thin film of formvar and were examined and photographed on a Zeiss EM10-C TEM. 2.3. Analysis of 18S rRNA gene The analysis of 18S rRNA gene of WA1 and CA5 was performed previously (Kjemtrup et al., 2002) and results 781 reprinted as Fig. 3 with GenBank accession numbers corrected. In brief, identification of parasite DNA was performed using PCR with primers targeting the complete babesial 18S rRNA gene and phylogenetic comparison was performed as previously described (Kjemtrup et al., 2000b, 2002). The source of sequences were either PCR products from whole blood from humans or animals from the western United States infected with piroplasms (WA1, CA5, CA1, CA3, CA4, WA2, CA6, MD1, BH3) obtained as described earlier (Herwaldt et al., 1997; Kjemtrup et al., 2000b, 2002) or from GenBank (nine additional Babesia spp., five recognized Theileria spp., ‘Theileria annae’, Cytauxzoon felis, and three protozoal outgroup sequences). These sequences were included because they had been previously documented in the United States or to provide adequate representation of the major piroplasm species. Alignment of sequences was performed as previously described (Kjemtrup et al., 2000b). Briefly, CLUSTAL W (Thompson et al., 1994) was used for initial alignment. Subsequent hand alignment using MacClade 3.0 (Maddison, W.P., Maddison, D.R. MacClade: analysis of phylogeny and character evolution. 3. 1992. Sunderland, MA, Sinauer Associates) refined ambiguous sections by following the secondary structure of the 18S small subunit rRNA (ssrRNA) gene molecules for Theileria spp. and Babesia spp. obtained from the ssrRNA database (Van de Peer et al., 1998). The neighbor-joining method was employed to estimate the phylogeny of the entire alignment using PAUP* 4.0 available through SeqLab (GCG) (Swofford, D.L. PAUP*: phylogenetic analysis using parsimony (and other methods). 4.0b10. 1998. Sunderland, MA, Sinauer Assoc.). The neighbor-joining method was employed because previous work on the 18S rRNA gene, using similar isolates, demonstrated that the neighbor-joining method provided a biologically robust tree compared with maximum parsimony and maximum likelihood-derived trees, though these trees did not differ statistically in tree length (Kjemtrup, 2001 Molecular Epidemiology of Human Babesiosis in California. PhD thesis. Epidemiology, University of California Davis, Shields Library Microcopy Collection). The Kimura two-parameter model for generation of the distance matrix was used. The analysis was performed with 500 bootstrap replications. 2.4. Analysis of ITS2 Sequences used for the ITS2 analysis are shown in Fig. 4. Available piroplasm isolates of good quality from humans and animals from the western United States as well as B. microti isolates were sequenced and included in the analysis. DNA was extracted from infected blood samples using the QIAmp DNA mini kit (Qiagen Inc., Valencia, CA, USA.). The PCR primers 5.8SFOR (5 0 -GGATGTCTTGGCTCACACAACG) and BABREV (5 0 -GCTTAAATTCAGCGGATAG) were used to amplify the second internal transcribed spacer from each isolate. PCR reactions of 25 ml consisted of 2 ml DNA, 0.5 mM of each primer, 200 mM deoxynucleoside triphosphates and a MgCl2 concentration of 2 mM. Proofreading polymerase (Finnzymes DyNAzyme EXT, MJ Research, Watertown, 782 P.A. Conrad et al. / International Journal for Parasitology 36 (2006) 779–789 Massachusetts) was used for amplification. Reaction conditions included denaturation at 94 8C for 3 min, followed by 36 cycles of 94 8C for 30 s, 52 8C for 30 s, and 72 8C for 1 min, followed by a post-amplification extension of 72 8C for 7 min. Amplified PCR products were prepared for direct sequencing using enzymatic treatment with exonuclease I and shrimp alkaline phosphatase (PCR Product pre-sequencing kit, USB Corporation, Cleveland, Ohio). Sequencing of PCR products was performed using ABI BigDye v3.0 (Applied Biosystems, Foster City, California) terminator sequencing chemistry and reaction products were separated and detected using an ABI 3100 capillary DNA sequencer. Some PCR products did not yield readable sequences in both directions on direct sequencing so were cloned using the pCRw-Blunt II-TOPOw plasmid from the Zero Bluntw TOPOw PCR Cloning Kit (Invitrogen Life Technologies). For all sequencing reactions for clones, either the original PCR primers or the universal M13 forward (K20) or M13 reverse primers were used. Sequence fragments were assembled using SeqMane II (DNASTAR, Inc., Madison, WI). Ambiguities in sequence data were noted following international universal base codes. The sequences were aligned (including gap insertion) using Clustal X version 1.53b (Thompson et al., 1997) and subsequently edited manually. Phylogenetic analysis of DNA sequences was performed using PAUP* 4.0b10 (Swofford, D.L. PAUP*: phylogenetic analysis using parsimony (and other methods) 4.0b10. 1998. Sunderland, MA, Sinauer Assoc.) and sequence alignments analyzed by maximum parsimony using branch-and-bound methods. The maximum parsimony method is not a model-based method, which makes assumptions about nucleotide substitution rates (Swofford et al., 1996). Because the evolutionary model is not known for the ITS2 and because maximum parsimony produced results consistent with other methods for the 18S rRNA gene analysis, the maximum parsimony method was chosen as a reasonable approach for ITS2 analysis. The outgroup sequences used were those of B. microti (two sequences), Theileria youngi and a Babesia sp. from a cat (Kjemtrup et al., 2000a). All of the characters in the alignment were used in a preliminary analysis. However, the alignment contained regions that were judged ambiguous with respect to positional homology inference and these regions were excluded from a subsequent analysis. These alignment-ambiguous regions included positions 1–35, 273– 289 and 391–456. Relative reliability of clades was assessed using 2000 replicates of bootstrap resampling. Phylogenetic trees were viewed and printed using Treeview (Page, 1996). 3. Results 3.1. Description of Babesia duncani nov. sp. (Syn: WA1-type Babesia) 3.1.1. Host Humans (Homo sapiens) are the type host for B. duncani Isolates of the parasite in human erythrocytes were infective and generally cause mortality when experimentally inoculated into gerbils, hamsters and mice as described (Quick et al., 1993; Thomford et al., 1994; Wozniak et al., 1996; Hemmer et al., 1999, 2000a, b). A splenectomized dog inoculated with blood from hamsters infected with the first B. duncani isolate from Washington State (WA1) seroconverted but did not become infected (Thomford et al., 1994; Krause et al., 1994). Although B. microti has never been serially propagated in vitro, B. duncani was continuously cultivated in hamster erythrocytes (Thomford et al., 1994). The natural reservoir for B. duncani has not been identified. B. duncani could not be assigned to any previously described Babesia sp. Our analysis demonstrates its distinction from similar parasites from canines and ungulates in California, and rodents for which sequence or other data is available. Small piroplasms have been noted from rabbits, raccoons, and sigmodontine rodents (Wood, 1952; Van Peenen and Duncan, 1968) but these were not assigned to species and to our knowledge, blood smears or other identifying materials are not extant and thus unavailable for analysis. Parasites infecting Alaskan rodents, tentatively assigned to B. microti (Fay and Rausch, 1969) have recently been analysed and comprise a distinct clade related to that taxon and clearly distinct from B. duncani (Goethert and Telford, 2003). 3.1.2. Locality To date, parasitologically confirmed cases of B. duncani infection have been identified only in Washington and California. However, one case (CA6; Table 1) was in a blood donor who lived in northern California but who had travelled to Oregon several months before the case was diagnosed and CA5 was isolated from the blood recipient. In addition, the specific locations in the western states in which this case and the cases caused by WA1 and WA3, which are presumed to be tickborne (Table 1), were acquired is not known with certainty. 3.1.3. Type specimens Type material, consisting of stained peripheral blood smears from the WA1 and CA5 human cases and hamsters inoculated with these isolates, are on deposit at the US National Parasite Collection (USNPC), collection numbers 095898.00 and 095899.00. Accordingly, the WA1 and CA5 strains may be designated as syntypes and are maintained as cryostabilites in research laboratories at UC Davis (Conrad) and the CDC (Eberhard). 3.1.4. Etymology This species is named in honor of Professor (Emeritus) ‘Duncan’ C.G.D. Brown in recognition of his extensive contributions to the biological understanding of piroplasm parasites, mentorship of graduate students in this area of research and passionate commitment to the control of hemoprotozoal diseases worldwide. 3.2. Morphologic and ultrastructural features The shapes and sizes of intraerythrocytic forms of the species are based on light microscopic examination of stained P.A. Conrad et al. / International Journal for Parasitology 36 (2006) 779–789 783 Fig. 1. Photomicrograph of merozoites of Babesia duncani from a Giemsa-stained blood smear of infected human case, CA5: a. piroplasm or ‘tear-drop’ form; b. ring form; c. ameboid form d. tetrad or ‘Maltese Cross’ form e. a multiply infected cell with a ring form and ameboid form f. multiple piroplasms in one red blood cell. Bar represents 5 mm. peripheral blood smears from the CA5 patient (Kjemtrup et al., 2002) prepared when the parasitemia of the blood smears was approximately 6.0%. Examples of typical intraerythrocytic stages are shown in Fig. 1. The ultrastructural features of trophozoites and merozoites were based on examination of blood from WA1-infected hamsters with parasitemias of 10– 20%. Despite careful examination by light and transmission electron microscopy of the blood samples available, parasite stages were not visualized in peripheral white blood cells of human patients or hamsters infected with WA1 or CA5. 3.2.1. Trophozoites Intraerythrocytic trophozoites were indistinguishable from those of B. microti. Trophozoites were generally round to oval, with the nucleus situated at the periphery (Fig. 1(a)–(c) and (e)). The piriform forms (nZ10) (Fig. 1(a)) measured 1.64G 0.15 mm by 2.24G0.33 mm. The diameter of ring forms (nZ17) (Fig. 1(b) and (e)) measured 2.4G0.18 mm. The ameboid forms (nZ12) (Fig. 1(c)) measured 2.4G0.21 by 3.2G0.35 mm. The ultrastructural features of trophozoites consisted of a single limiting plasmalemmal membrane, prominent nucleus, intracytoplasmic ribosomes, and acristate mitochondria (Fig. 2(a) and (b)). 3.2.2. Merogony Tetrads (syn. Maltese Crosses or quadruplet forms) of merozoites were formed by division within erythrocytes (Fig. 1(d)). The individual parasites that made up the tetrad (nZ32) (Fig. 1(d)) measured 0.72G0.04 by 1.41G0.10 mm and the total length and width of the tetrad formation (nZ6) measured 3.87G0.30 by 2.67G0.42 mm. Fig. 1(f) shows multiple merozoites in an erythrocytes. Fig. 2(c) and (d) shows the ultrastructural features of merozoites in the process of division. Depending on the plane of section, 2–4 merozoites were seen budding from the residual body by schizogony, in which nuclear division of the forming merozoites appeared to be completed prior to cytoplasmic division. Each merozoite had a single limiting plasmalemmal membrane, a prominent nucleus, intracytoplasmic ribosomes, acristate mitochondria and an apical complex consisting of a subplasmalemmal inner membrane and multiple electron dense rhoptries (Fig. 2(c) and (d)). 3.3. Phylogenetic relationship to other Apicomplexa The 18S rRNA and ITS2 sequences analyzed in this study have been deposited in GenBank with accession numbers shown in Figs. 3 and 4. Fig. 3 shows the phylogram of the 18S rRNA gene analysis. In this analysis, B. duncani n.sp. (WA1 and CA5), WA2 and CA6 were phylogenetically indistinguishable. Together, these isolates constituted a clade separate from other babesial parasites isolated from humans and large ungulates in California with 100% support. There is strong bootstrap support (100%) for placement of all human, large 784 P.A. Conrad et al. / International Journal for Parasitology 36 (2006) 779–789 Fig. 2. Transmission electron micrographs of Babesia duncani within hamster erythrocytes showing trophozoite (a. and b.) and merozoite (a. c. and d.) forms. All bars are 5 mm. a. An erythrocyte with both a trophozoite form and three merozoites (ME) in the process of division by schizogony. b. Two trophozoites adjacent to each other. Trophozoites had a single limiting plasmalemmal membrane (PM), prominent nucleus (N) with a double nuclear membrane, intracytoplasmic ribosomes (R) and acristate mitochondria (M). c. Three MEs in the process of division by intraerythrocytic schizogony in which nuclear division precedes cytoplasmic division and the merozoites bud-off, leaving a residual body (RB). d. One merozoite still attached to a RB. Merozoites had a single limiting PM, prominent N with a double nuclear membrane, intracytoplasmic R, acristate M, and an apical complex consisting of inner subplasmalemmal membranes (IM) and multiple electron dense rhoptries (Rh). ungulate, and the recently named B. conradae (Kjemtrup et al., 2006) from canines in California into their own clade, distinct from Theileria spp., B. microti and related species and the Babesia species sensu stricto (Babesia bigemina, Babesia divergens, B. canis, Babesia gibsoni). ITS2 sequence lengths ranged from 309 to 390 bp. Phylogenetic analysis of the entire ITS2 alignment yielded 39 equally parsimonious trees having a consistency index of 80.3% and a homoplasy index of 19.7%. When ambiguous alignment characters were excluded, three equally parsimonious trees were found, each having a consistency index of 80% and a homoplasy index of 20%. Strict consensus trees from both analyses had an identical topology (Fig. 4). In the analyses from which alignment-ambiguous regions were excluded, three distinct clades had moderate to strong bootstrap support. In the first, there was 100% support for a lineage P.A. Conrad et al. / International Journal for Parasitology 36 (2006) 779–789 785 B.bigemina (X59604) 59 B.divergens (U07885) 69 95 B. canis (L19079) 91 B. gibsoni (AF205636) B. bovis (L19077) 100 100 T. parva (L02366) T. annulata (M64243) T. buffeli (Z15106) 90 Cytauxzoon felis (L19080) 54 91 T. youngi (AF245279) T. equi (Z15105) CA 1 Human (AF158703) 71 100 52 CA 4 Human (AF158705) MD 1 Mule Deer (AF158706) CA 3 Human (AF158704) BH 3 Bighorn Sheep (AF158709) 100 WA 1 B. duncanin. sp.Human (AF158700) 100 WA 2 Human (AF158701) CA 5 B. duncanin. sp. Human (AY027815) 92 CA 6 isolate Human (AY027816) 93 B. conradae (AF158702) B. felis (AF244912) 100 B. rodhaini (M87565) 67 B. microti (U09833) “T. annae” (AF188001) Toxoplasma gondii (M97703) Crypthecodinium cohnii (M64245) Plasmodium falciparum (M19172) 50 changes Fig. 3. Results of neighbor-joining analysis of the 18S rRNA gene marker demonstrating phylogenetic relationship of the Babesia isolates. Numbers above the lines represent the percentage of replicates out of 500 bootstrap samplings in which the given branching pattern was obtained. Branch lengths are proportional to the number of base changes. The scale bar represents 50 nucleotide substitutions per 100 bases. GenBank reference numbers are shown in parentheses. Reprinted from Kjemtrup, A.M., Lee, B., Fritz, L., Evans, C., Chervenak, M., Conrad, P.A. Investigation of transfusion of a WA1-type babesial parasite to a premature infant in California: 1482–1487. Transfusion, 42 (11), Copyright (2002), with permission from Blackwell Publishers, Ltd. containing the B. duncani isolates WA1 and CA5, as well as the B. duncani-type parasites, CA6 and clone 1 of theWA2 isolate. There was also support (83%) for a subclade that excluded a sequence from a second clone of the B. duncani-type parasite, WA2, reflecting likely intraspecific variation in the ITS2 of the WA2 isolate. The second WA2 clone differed from the others in a total of four loci in the alignment of which two were excluded in the second analysis. 4. Discussion Over a decade has passed since the first identified case of babesiosis caused by a WA1-type parasite occurred in Washington State (Quick et al., 1993) and the parasite was shown to be antigenically, biologically and molecularly distinct from B. microti, the other then known cause of human babesiosis in the United States (Thomford et al., 1994). The WA1 parasite was not given a species distinction at that time because samples from the human case were inadequate to provide sufficient morphologic, molecular and biological comparisons with other characterized Babesia species. Since 1993, additional WA1-type Babesia isolates were obtained from blood donors with asymptomatic infections and recipients who developed clinical babesiosis in Washington State (Herwaldt et al., 1997) and California (Kjemtrup et al., 2002). All of these parasites were morphologically indistinguishable from each other and from other small Babesia spp., such as B. microti and B. conradae, (formerly called B. gibsoni) (see Kjemtrup and Conrad, 2006; Kjemtrup et al., in press). Initial phylogenetic analysis of partial 18S rRNA gene sequences for the WA1-type isolates showed that they were molecularly distinct from other known Babesia species for which sequence 786 P.A. Conrad et al. / International Journal for Parasitology 36 (2006) 779–789 T. youngi (AY998760) WA 2 clone 2 Human (AY998763) B. duncaniWA1 Human (AY998762) 100 B. duncani CA5 Human (AY998761) 83 CA 6 Human (AY965740) WA 2 clone 1 Human (AY965741) BH 1 Bighorn Sheep (AY965735) FD 1 clone 1 Fallow Deer (AY965737) 99 100 CA 2 Human (AY998766) CA 1 Human(AY998765) 60 MD 1 Mule Deer (AY965736) 60 93 RD 1 Reindeer (AY998769) 62 62 FD 1, clone 2 Fallow Deer (AY998764) BH 3, clone 1 Bighorn Sheep (AY998767) 70 BH 3, clone 2 Bighorn Sheep (AY998769) 100 B. conradae (AY965738) 100 B. conradae (AY965739) B. felis (AY965742) B. microti,clone 1 (AY998770) 99 100 B. microti, clone 2 (AY998771) Fig. 4. Strict consensus tree of three equally parsimonious trees inferred from the alignment of ITS2 sequences. Bootstrap percentage values from 1000 replicates are included for relevant clades. GenBank reference numbers are shown in parentheses. information was available (Thomford et al., 1994; Herwaldt et al., 1997). Analysis of parasites and DNA in fresh blood samples acquired in 2000 from a premature infant with a transfusion-transmitted infection (Kjemtrup et al., 2002) provided the necessary data to show that the CA5 and WA1 isolates were syntypes and confirm that they constituted a newly described species, herewith designated as B. duncani n.sp. (Apicomplexa: Babesiidae). This organism is assigned to the genus Babesia on the basis of the phenotypic characteristics of the intraerythrocytic stages, lack of identifiable exoerythrocytic stages and molecular phylogenetic analysis of the 18S rRNA gene and ITS2 sequences. The morphologic features of the intraerythrocytic trophozoites and merozoites of B. duncani in blood smears from infected humans were consistent with those described for other small (1–3 mm) Babesia parasites that are characterized by tetrad dividing forms (Levine, 1971, 1985, 1988; Mehlhorn and Shein, 1984). The small Babesia spp. differ from larger species of Babesia sensu stricto in that the intraerythocytic stages of these species are generally 3–5 mm and they divide primarily by binary fission to form two merozoites. Some differences were observed in the frequency of different intraerythrocytic forms when comparing B. duncani in the blood of patients with those in natural B. microti infections. In the latter, ameboid trophozoite forms with white or vacuolated cytoplasm are more common and tetrad dividing forms are rare and often asymmetrical (S. Telford, unpublished data). These differences may be due to host factors, including immune responses, which can affect the rate of parasite multiplication. The ultrastructural features were similar to those described for B. microti, Theileria parva and Theileria annulata, which all have trophozoite stages that differentiate and multiply to form tetrads of four merozoites within erythrocytes (Rudzinska, 1981; Conrad et al., 1985, 1986). The 18S rRNA gene was used in this analysis because it has shown to be sufficient to resolve phylogenies of piroplasms to the generic level (Allsopp et al., 1994; Mackenstedt et al., 1994; Chae et al., 1998). Based on 18S rRNA analysis, B. duncani lies in a distinct clade with other piroplasms from humans, dogs and wildlife in the western United States, which is separate from Babesia sensu stricto (e.g. B. bigemina, B. divergens, B. canis), Theileria spp. and B. microti. Thus far, intralymphocytic schizont stages, characteristic of Theileria spp., have not been observed in people infected with B. duncani or in rodents that were experimentally inoculated with infected erythrocytes. Exoerythrocytic stages, if they exist, probably would be present only in persons infected by sporozoites from infected vector ticks and then might be difficult to identify, particularly if they were present primarily or exclusively before the development of symptoms and the formation of intraerythrocytic piroplasm stages. Nonetheless, based on accepted criteria for the differentiation of parasites in the genera Babesia and Theileria, the absence of finding intralymphocytic schizont stages in any of the hosts for the parasites in the ‘western clade’, including B. duncani, necessitates that these parasites be classified as Babesia. Although most piroplasm phylogenetic relationships have been established based on 18S rRNA analysis, data from more gene sequences is deemed valuable in establishing species distinctions (Tenter et al., 2002). In this case, P.A. Conrad et al. / International Journal for Parasitology 36 (2006) 779–789 sequence analysis of the ITS2 region, a different locus of the small subunit ribosomal rRNA gene, was undertaken to better assess the phylogenetic relationship of isolates with 18S rRNA homology, most notably the B. duncani syntype isolates WA1 and CA5, as well as the B. duncani-type parasites, WA2 and CA6. The ITS2 is useful for evaluating relationships in lower levels, such as among genera (Hwang and Kim, 1998). The results showed that the ITS2 sequences of the B. duncani isolates (WA1, CA5), and the B. duncanitype parasites CA6 and one of the WA2 clones were homologous to each other. Their ITS2 sequence was distinct from the sequences of other piroplasms included in this study and also differed from other ITS2 sequences that have been deposited in GenBank (e.g. B. canis ‘subspecies’, other Babesia spp. and T. parva). The ITS2 analysis provides robust molecular support that the B. duncani isolates are monophyletic, the same species, and constitute a distinct species from other known Babesia spp. for which sequence information is available. The ITS2 data also complement additional molecular data from the much more highly conserved ssrRNA genes that have been analyzed from the isolates included in this study as well as from a broader group of piroplasms (Persing and Conrad, 1995; Kjemtrup and Conrad, 2000, 2001; Penzhorn et al., 2001). These new data included in this study allowed for the identification of several polymorphic ITS2 sequences from cloned isolates, including the B. duncani-type parasite WA2 and the fallow deer isolate sequenced previously (Kjemtrup et al., 2000b). Variability in internal transcribed spacer regions was previously reported within species from other piroplasms, including T. parva and Theileria spp. from ruminants (Chae et al., 1998; Collins and Allsopp, 1999). There was 100% bootstrap support for the four isolates in the B. duncani clade, separating them from other clades. However, the extent of polymorphism in different clones of WA2, with clone 2 being in a separate subclade from clone 1, indicates that heterogeneous populations exist within some isolates of B. duncanitype parasites that may warrant further investigation. With the exception of the cloned isolates from the B. duncani-type parasite WA2 and the fallow deer isolate none of the other clones of human Babesia isolates or additional Babesia spp. isolates analyzed, including B. microti, the small Babesia sp. from bighorn sheep and canine B. conradae, showed evidence of polymorphisms. Both the 18S and ITS2 analyses separate B. duncani isolates, including the syntype isolates (WA1 and CA5) and the B. duncani-type parasites WA2 and CA6, from other babesial parasites whose DNA was recovered from humans (CA1, CA3, CA4 in the 18S analysis and CA1, CA2 in ITS2 analysis) and wildlife (bighorn sheep and deer) in California. Though all these isolates share morphologic features and are serologically cross-reactive (Kjemtrup and Conrad, 2000, 2001), it is premature to consider the CA1–CA4 human isolates and deer and bighorn sheep isolates as B. duncani. These isolates from the western United States may represent a ‘species complex’. Clearly, characterization of more piroplasm isolates would increase our understanding of the 787 distribution and relationship of B. duncani and the parasites in this complex to each other, as well as to other species of piroplasms. Although it would be desirable to have a thorough understanding of the natural history of B. duncani, extensive efforts have failed to identify the reservoir host(s) and tick vector of the parasite (reviewed by Kjemtrup and Conrad, 2000, 2001; Conrad et al., 2003). Investigations of potential rodent hosts for the parasite in areas of northern California where the CA3 infection most likely occurred (Persing et al., 1995) resulted in the discovery of only one morphologically similar piroplasm, Theileria youngi (Kjemtrup et al., 2001), which was prevalent in woodrats (Neotoma fuscipes). The 18S rRNA gene and ITS2 analyses both confirmed that T. youngi is in a distinct clade from the B. duncani. The ITS2 analysis also showed a distinct phylogenetic separation between the isolates of wild ungulates and the clade of B. duncani, which argues against our earlier contention that mule deer and/or bighorn sheep were potential reservoir hosts for this parasite (Kjemtrup et al., 2000b). The ITS2 sequences of the parasites from two of the human patients with babesiosis in California (CA1, CA2) were most closely related to the bighorn sheep and deer isolates in this study but polymorphisms were still detected. The other important result of the ITS2 analysis was to confirm the phylogenetic separation between B. conradae isolated from dogs in California and the other human isolates, including the B. duncani isolates. Ironically, the Conrad laboratory initially became involved in collaborative efforts to characterize CA1 and WA1-type parasites so as to determine whether they were related to the piroplasm isolated from dogs in southern California, then thought to be B. gibsoni (Conrad et al., 1991, 2003) but since renamed B. conradae (Kjemtrup et al., 2006). This reclassification was prompted by the discovery that there are at least four morphologically similar small piroplasm parasites of dogs, two of which have been identified in the United States (Kjemtrup et al., 2000a; Kjemtrup and Conrad, 2006). The infectivity of B. duncani and potential to cause clinical disease in humans is well documented (Quick et al., 1993; Herwaldt et al., 1997; Kjemtrup et al., 2002). However, the extent of the parasite’s geographic distribution both in the western United States and elsewhere, as well as the public health impact of infection with B. duncani, remains unknown. Serologic reactivity to B. duncani has been detected in human populations in California (Persing et al., 1995; Fritz et al., 1997). However, thus far there have been too few confirmed clinical cases to validate the serologic test employed, as has been done for B. microti using the indirect fluorescent antibody test (Krause et al., 1994). Analysis of additional B. duncani infections would provide a better understanding of the ecology of this parasite. For patients presenting with compatible clinical signs (reviewed by Kjemtrup and Conrad, 2000) and probable exposure history (possibility of tick bite or recent blood transfusion), infection with B. duncani should be considered in the differential diagnosis, particularly by physicians in the United States. 788 P.A. Conrad et al. / International Journal for Parasitology 36 (2006) 779–789 References Allsopp, M.T., Cavalier-Smith, T., De Waal, D.T., Allsopp, B.A., 1994. Phylogeny and evolution of the piroplasms. Parasitology 108 (Pt 2), 147–152. Bredt, A.B., Weinstein, W.M., Cohen, S., 1981. Treatment of babesiosis in asplenic patients. J Am Med Assoc 245, 1938–1939. Chae, J., Lee, J., Kwon, O., Holman, P.J., Waghela, S.D., Wagner, G.G., 1998. 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, 41–52. Collins, N.E., Allsopp, B.A., 1999. Theileria parva ribosomal internal transcribed spacer sequences exhibit extensive polymorphism and mosaic evolution: application to the characterization of parasites from cattle and buffalo. Parasitology 118 (Pt 6), 541–551. Conrad, P.A., Kelly, B.G., Brown, C.G., 1985. Intraerythrocytic schizogony of Theileria annulata. Parasitology 91 (Pt 1), 67–82. Conrad, P.A., Denham, D., Brown, C.G., 1986. Intraerythrocytic multiplication of Theileria parva in vitro: an ultrastructural study. Int. J. Parasitol. 16, 223–229. Conrad, P., Thomford, J., Yamane, I., Whiting, J., Bosma, L., Uno, T., Holshuh, H.J., Shelly, S., 1991. Hemolytic anemia caused by Babesia gibsoni infection in dogs. J. Am. Vet. Med. Assoc. 199, 601–605. Conrad, P.A., Miller, M., Kjemtrup, A.M., Smith, W., Garner, I.A., 2003. The human–wildlife–domestic animal interface provides exciting research opportunities for parasitologists. J. Parasitol. 89 (Suppl.), S27–S36. Fay, F.H., Rausch, R.L., 1969. Parasitic organisms in the blood of arvicoline rodents in Alaska. J. Parasitol. 55, 1258–1265. Fritz, C.L., Kjemtrup, A.M., Conrad, P.A., Flores, G.R., Campbell, G.L., Schriefer, M.E., Gallo, D., Vugia, D.J., 1997. Seroepidemiology of emerging tickborne infectious diseases in a Northern California community. J. Infect. Dis. 175, 1432–1439. Goethert, H.K., Telford III, S.R., 2003. What is Babesia microti? Parasitology 127, 301–309. Hemmer, R.M., Wozniak, E.J., Lowenstine, L.J., Plopper, C.G., Wong, V., Conrad, P.A., 1999. Endothelial cell changes are associated with pulmonary edema and respiratory distress in mice infected with the WA1 human Babesia parasite. J. Parasitol. 85, 479–489. Hemmer, R.M., Ferrick, D.A., Conrad, P.A., 2000a. Role of T cells and cytokines in fatal and resolving experimental babesiosis: protection in TNFRp55K/K mice infected with the human Babesia WA1 parasite. J. Parasitol. 86, 736–742. Hemmer, R.M., Ferrick, D.A., Conrad, P.A., 2000b. Up-regulation of tumor necrosis factor-alpha and interferon-gamma expression in the spleen and lungs of mice infected with the human Babesia isolate WA1. Parasitol. Res. 86, 121–128. Herwaldt, B.L., Kjemtrup, A.M., Conrad, P.A., Barnes, R.C., Wilson, M., McCarthy, M.G., Sayers, M.H., Eberhard, M.L., 1997. Transfusiontransmitted babesiosis in Washington State: first reported case caused by a WA1-type parasite. J. Infect. Dis. 175, 1259–1262. Herwaldt, B.L., de Bruyn, G., Pieniazek, N.J., Homer, M., Lofy, K.H., Slemenda, S.B., Fritsche, T.R., Persing, D.H., Limaye, A.P., 2004. Babesia divergens-like infection, Washington State. Emerg. Infect. Dis. 10, 622–629. Hwang, U.W., Kim, W., 1998. General properties and phylogenetic utilities of nuclear ribosomal DNA and mitochondrial DNA commonly used in molecular systematics. Korean J. Parasitol. 37, 215–228. Kjemtrup, A.M., Conrad, P.A., 2000. Human babesiosis: an emerging tickborne disease. Int. J. Parasitol. 30, 1323–1337. Kjemtrup, A.M., Conrad, P.A., 2001. Emerging perspectives on human babesiosis. In: Scheld, W.M., Craig, A., Hughes, J.M. (Eds.), Emerging Infections. ASM Press, Washington, DC, pp. 175–195. 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., 2000a. There are at least three genetically distinct small piroplasms from dogs. Int. J. Parasitol. 30, 1501–1505. Kjemtrup, A.M., Thomford, J., Robinson, T., Conrad, P.A., 2000b. Phylogenetic relationships of human and wildlife piroplasm isolates in the western United States inferred from the 18S nuclear small subunit RNA gene. Parasitology 120 (5), 487–493. Kjemtrup, A.M., Robinson, T., Conrad, P.A., 2001. Description and epidemiology of Theileria youngi n.sp. from a northern Californian dusky-footed woodrat (Neotoma fuscipes) population. J. Parasitol. 87, 373–378. Kjemtrup, A.M., Lee, B., Fritz, C.L., Evans, C., Chervenak, M., Conrad, P.A., 2002. Investigation of transfusion transmission of a WA1-type babesial parasite to a premature infant in California. Transfusion 42, 1482–1487. Kjemtrup, A.M., Conrad, P.A., 2006. A review of the small canine piroplasm from California: Babesia conradae in the literature. Vet Parasitol 138, 112– 117. Kjemtrup, A.M., Carreno, R., Wainwright, K., Miller, M., 2006. Babesia conradae, sp. nov., a small canine Babesia identified in California. Vet. Parasitol. 138, 103–111. Krause, P.J., Telford III.., S.R., Ryan, R., Conrad, P.A., Wilson, M., Thomford, J.W., Spielman, A., 1994. Diagnosis of babesiosis: evaluation of a serologic test for the detection of Babesia microti antibody. J. Infect. Dis. 169, 923–926. Levine, N.D., 1971. Taxonomy of piroplasms. Trans. Am. Microsc. Soc. 90, 2– 33. Levine, N.D., 1985. Veterinary Protozology, first ed. University Press, Ames, Iowa. Levine, N.D., 1988., first ed. The Protozoan Phylum Apicomplexa, vol. 2. CRC Press, Boca Raton, FL. Mackenstedt, U., Luton, K., Baverstock, P.R., Johnson, A.M., 1994. Phylogenetic relationships of Babesia divergens as determined from comparison of small subunit ribosomal RNA gene sequences. Mol. Biochem. Parasitol. 68, 161–165. Mehlhorn, H., Shein, E., 1984. The piroplasms: life cycle and sexual stages. Adv. Parasitol. 23, 37–103. Page, R.D.M., 1996. TREEVIEW: an application to display phylogenetic trees on personal computers. Comput. App. Biosci. 12, 357–358. Penzhorn, B.L., Kjemtrup, A.M., Lopez-Rebollar, L.M., Conrad, P.A., 2001. Babesia leo n. sp. from lions in the Kruger National Park, South Africa, and its relation to other small piroplasms. J. Parasitol. 87, 681–685. Persing, D.H., Conrad, P.A., 1995. Babesiosis: new insights from phylogenetic analysis. Infect. Agents Dis. 4, 182–195. Persing, D.H., Herwaldt, B.L., Glaser, C., Lane, R.S., Thomford, J.W., Mathiesen, D., Krause, P.J., Phillip, D.F., Conrad, P.A., 1995. Infection with a Babesia-like organism in northern California. N. Engl. J. Med. 332, 298–303. Quick, R.E., Herwaldt, B.L., Thomford, J.W., Garnett, M.E., Eberhard, M.L., Wilson, M., Spach, D.H., Dickerson, J.W., Telford III.., S.R., Steingart, K.R., Pollock, R., Persing, D.H., Kobayashi, J.M., Juranek, D.D., Conrad, P.A., 1993. Babesiosis in Washington State: a new species of Babesia? Ann. Intern. Med. 119, 284–290. Rudzinska, M.A., 1981. Morphologic aspects of host-cell-parasite relationships in babesiosis. In: Ristic, M., Kreier, J.P. (Eds.), Babesiosis. Academic Press, New York, NY, pp. 87–142. Scholtens, R.G., Braff, E.H., Healey, G.A., Gleason, N., 1968. A case of babesiosis in man in the United States. Am. J. Trop. Med. Hyg. 17, 810–813. Swofford, D.L., Olsen, G.J., Waddel, P.J., Hillis, D.M., 1996. Phylogenetic inference. In: Hillis, D.M., Moritz, C., Mable, B.K. (Eds.), Molecular Systematics. Sinauer Associates, Inc., Sunderland, MA, pp. 407–514. Telford III.., S.R., Gorenflot, A., Brasseur, P., Spielman, A., 1993. Babesial infections in humans and wildlife. In: Kreier, J.P. (Ed.), Parasitic Protozoa. Academic Press, San Diego, CA, pp. 1–47. Tenter, A.M., Barta, J.R., Beveridge, I., Duszynski, D.W., Mehlhorn, H., Morrison, D.A., Thompson, R.C.A., Conrad, P.A., 2002. The conceptual basis for a new classification of the coccidia. Int. J. Parasitol., 595–616. Thomford, J.W., Conrad, P.A., Telford III.., S.R., Mathiesen, D., Bowman, B.H., Spielman, A., Eberhard, M.L., Herwaldt, B.L., Quick, R.E., Persing, D.H., 1994. Cultivation and phylogenetic characterization of a newly recognized human pathogenic protozoan. J. Infect. Dis. 169, 1050–1056. Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. CLUSTAL W: improving the sensitivity of progressive muliple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673–4680. P.A. Conrad et al. / International Journal for Parasitology 36 (2006) 779–789 Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., Higgins, D.G., 1997. The CLUSTAL-X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25, 4876–4882. Van de Peer, Y., Caers, A., De Rijk, P., De Wachter, R., 1998. Database on the structure of small ribosomal subunit RNA. Nucleic Acids Res. 26, 179–182. 789 Van Peenen, P.F.D., Duncan, J.F., 1968. Piroplasms (Protozoa: Sarcodina) of wild mammals of California. Bull. Wildl. Dis. Assoc. 4, 3–8. Wood, S.F., 1952. Mammal blood parasite records from southwestern United States and Mexico. J. Parasitol. 38, 85–86. Wozniak, E.J., Lowenstine, L.J., Hemmer, R., Robinson, T., Conrad, P.A., 1996. Comparative pathogenesis of human WA1 and Babesia microti isolates in a Syrian hamster model. Lab Anim Sci. 46, 507–515.