Establishment of Babesia vulpes n. sp. (Apicomplexa: Babesiidae), a piroplasmid species pathogenic for domestic dogs

Background Canine babesiosis is a severe disease caused by several Babesia spp. A number of names have been proposed for the canine-infecting piroplasmid pathogen initially named Theileria annae Zahler, Rinder, Schein & Gothe, 2000. It was shown to be a member of the Babesia (sensu lato) group infecting carnivores and is also closely related to the Babesia microti group. Subsequently, the same parasite species was reclassified as a member of the genus Babesia and the name Babesia vulpes Baneth, Florin-Christensen, Cardoso & Schnittger, 2015 was proposed for it. However, both names do not meet the requirements of the International Code of Zoological Nomenclature (no accompanying descriptions, no deposition of type-specimens) and cannot be recognized as available names from the nomenclatural point of view. The purpose of this study was to further characterize this parasite in order to confirm its validity, to provide its description and to introduce zoological nomenclature for it with the name Babesia vulpes n. sp. Results Morphological description of the parasite in canine erythrocytes demonstrated that it takes the shape of small (1.33 × 0.98 µm), round to oval forms reminiscent of the pyriform and ring shapes of other small canine Babesia spp., such as Babesia gibsoni Patton, 1910 and Babesia conradae Kjemtrup, Wainwright, Miller, Penzhorn & Carreno, 2006. However, these parasite forms were overall smaller than those measured for the latter two species and no tetrad (Maltese cross) form was reported. Furthermore, phylogenetic analysis using the cytochrome c oxidase subunit 1 (COX1) amino acid sequences substantiates the species identity of this parasite as previously demonstrated based on phylogenetic analysis of the 18S rRNA and β-tubulin genes. The holotype of the parasite species was designated and deposited in an accessible public collection. Conclusions This study ratifies the name Babesia vulpes n. sp. proposed for the parasite previously referred to as Theileria annae Zahler, Rinder, Schein & Gothe, 2000, Babesia annae (Zahler, Rinder, Schein & Gothe, 2000) or Babesia vulpes Baneth, Florin-Christensen, Cardoso & Schnittger, 2015, or mentioned as “Babesia microti-like piroplasm”, “Babesia Spanish dog isolate” and Babesia cf. microti. Electronic supplementary material The online version of this article (10.1186/s13071-019-3385-z) contains supplementary material, which is available to authorized users.


Background
Babesia Starcovici, 1893 and Theileria Bettencourt, França & Borges, 1907 are tick-borne protozoan genera classified in the phylum Apicomplexa, class Piroplasmea and order Piroplasmida, which infect domestic and wild animals, and humans, and may cause severe disease. Piroplasmids referred to as Theileria (sensu stricto) have been originally defined by the presence of a pre-erythrocytic life stage in leukocyte host cells and trans-stadial transmission in ticks. In contrast, schizont propagation is absent in species of Babesia (sensu stricto) and they display the characteristic of tick transovarial transmission [1,2]. An additional group of piroplasmids are referred to

Open Access
Parasites & Vectors *Correspondence: gad.baneth@mail.huji.ac.il 1 Koret School of Veterinary Medicine, Hebrew University, P.O. Box 12,76100 Rehovot, Israel Full list of author information is available at the end of the article as Babesia (sensu lato) since they cannot be assigned to either of the above groups [3].
A study published by our group has demonstrated that within the B. microti-like piroplasmids (Clade I), a new species named Babesia vulpes Baneth, Florin-Christensen, Cardoso & Schnittger, 2015 is positioned in a monophyletic group of Babesia parasites that exclusively infect carnivores and is closely related to the monophyletic B. microti group. Furthermore, we demonstrated that within the Babesia-infecting carnivore group, this parasite can be unequivocally delineated as a distinct species [5]. In the latter study, the name B. vulpes had been proposed as a new species designation [5]; nevertheless, as pointed out in a Letter to the Editor of this journal by Harris [19], according to Article 16.4 of the International Code of Zoological Nomenclature (ICZN), the naming was not statutory. This is because in order to name a species, the naming publication must contain the fixation of a holotype deposited in a specified collection and a description of the species, preferably containing morphological details. As Harris [19] mentioned, these details, as well as the designation of the proposed name as "sp. nov. " were also missing in the naming of T. annae by Zahler et al. [6]; therefore, both these names are currently considered nomina nuda (plural for nomen nudum, Latin for "naked name", a name not statutorily in force, but which can be made available in subsequent naming procedures) [19]. According to the glossary of ICZN, a nomen nudum is not an available name (in the meaning used in the zoological nomenclature) and therefore the same name may be made available later for the same or a different concept.
The purpose of this study is therefore to further characterize and provide the missing requirements (description, designation of a name-bearing type) in order to establish B. vulpes n. sp. as a valid species name.

Methods
Blood smears fixed with methanol and stained with Hemacolor ® (Merck, Darmstadt, Germany) were obtained from the Inno Veterinary Laboratory in Braga, Portugal, and evaluated for parasite morphology by light microscopy. Piroplasm parasites from these smears prepared in 2009 from two Portuguese dogs infected with this parasite then termed B. microti-like were previously examined, described and molecularly characterized [17]. The smears were examined by oil immersion microscopy (Zeiss, Jena, Germany) at 1000× magnification. Sizes of parasites were measured using a micrometer. Measurements are in micrometers and are given as the range followed by the mean and standard deviation in parentheses. A stained blood smear from one of these dogs containing the holotype was deposited in the National Natural History Collection of the Hebrew University of Jerusalem, Israel, and the remaining slides containing the paratypes were deposited in the Parasite Collection of the University of Oporto, Portugal.
PCR to amplify the cox1 gene was carried out using blood samples from three Israeli red foxes (V. vulpes) collected for a hemoparasite survey. The samples had been shown to be infected with the new species by PCR of the 18S rRNA gene followed by sequencing (GenBank: KJ871347, KJ871348, KJ871349), and for which a nearly complete longer gene sequence (GenBank: KJ871351) derived from one fox had been used in the phylogenetic analysis of Baneth et al. [5]. To this end, a region of the cox1 gene was amplified using primers cox1F133 and cox1R11130 essentially as previously described [20]. Conventional PCR was performed in a total volume of 25 μl using the PCR-ready High Specificity mix (Syntezza Bioscience, Jerusalem, Israel) with 400 nM of each primers and sterile DNase/RNase-free water (Sigma, St. Louis, MO, USA). Amplification was carried out using a programmable, conventional thermocycler (Biometra, Göttingen, Germany). PCR products were electrophoresed on 1.5% agarose gels stained with ethidium bromide and evaluated under UV light for the size of amplified fragments by comparison to a 100 bp DNA molecular weight marker. Direct sequencing of PCR allowed determining the cox1 nucleotide sequences (GenBank: KX169167, KX169168 and KX169169) and corresponding COX1 amino acid sequences (GenBank: APX55184, APX55185 and APX55186) for subsequent inclusion into phylogenetic analyses.

Differential diagnosis
Intraerythrocytic parasites presented as round to ovalshaped, and an eccentric, basophilic-staining; round nucleus was conspicuous in some parasites (Fig. 1a, b). Of 18 parasites measured, 16 presented as single parasites, whereas the remaining two were located in the same erythrocyte. Parasites occupied only a small portion of the erythrocyte and were reminiscent of the pyriform and ring shapes described for other small-form Babesia species that infect dogs [37,38]; however, no tetrad (Maltese cross) shapes were seen. The morphological shape of B. vulpes n. sp. described here from dog erythrocytes is similar to the ring and pyriform shapes described for other small, canine-infecting Babesia spp. [38][39][40] [38]. However, in contrast to B. conradae, no tetrad (Maltese cross) forms were observed in B. vulpes n. sp. Babesia gibsoni Patton, 1910, another small-form Babesia of dogs, which is also not reported to produce tetrads, is described as being considerably larger than B. vulpes n. sp. with the ring shape measuring 2.71 × 1.61 µm and the pyriform shape measuring 2.1 × 0.94 µm [40], or according to a different report, 1.9 × 1.2 µm, without a distinction between the shapes [39]. The above comparisons indicate that B. vulpes n. sp. is a distinct form consistent with the small-form piroplasms of canines. However, B. vulpes n. sp. tends to be smaller than B. conradae and B. gibsoni and has not been reported to form tetrads, thus further distinguishing it from B. conradae.

Molecular phylogeny
Phylogenetic analysis of amino acid COX1 sequences for Theileria spp. and Babesia spp. sp. and the B. microti group. The results based on the COX1 amino acid sequences coincide and support the previously presented results on the species identity of B. vulpes n. sp. by phylogenetic analysis of 18S RNA and β-tubulin gene sequences [5]. In addition, a neighbor-joining tree based on 25 cox1 nucleotide sequences with a final dataset of 879 positions of B. vulpes n. sp. and other piroplasmid species was inferred and corroborated results obtained by COX1 amino acid sequences. Specifically, the same topology and an identical bootstrap support were determined for the corresponding relevant clades of the trees inferred by amino acid and nucleotide sequences [B. microti group/B. vulpes n. sp. clade (bs = 100) and (B. vulpes n. sp. clade (bs = 100)] (Additional file 1: Figure S1).

Discussion
This study establishes B. vulpes n. sp. as a new taxon fulfilling the requirements of the ICZN guidelines. A morphological description with measurements of the parasite forms in canine erythrocytes and the deposition of the holotype and paratypes in suitable collections have been Fig. 2 Neighbor-joining tree of COX1 amino acid sequences of Babesia vulpes n. sp. and other piroplasmid species. Sequences analyzed in the context of this study are designated by bold accession numbers of taxon labels. Clade designations are presented as defined previously [3,50]. The percentage of replicate trees as determined by 1000 replicates of a bootstrap test are shown next to the branches. A Plasmodium falciparum COX1 sequence has been included as the outgroup. The scale-bar represents the evolutionary distance in the units of the number of amino acid substitutions per site. Gray dots designate Babesia species that infect domestic dogs [51] made in compliance with the ICZN guidelines [36]. The generic placement of B. vulpes n. sp. is derived from the molecular phylogenetic analysis of the 18S RNA and β-tubulin genes, and COX1 protein sequences, whereas the species name has been chosen because the red fox (V. vulpes) is considered the main natural host of this piroplasmid (see also [5]). As mentioned above, according to the ICZN regulations, "T. annae" [6] is considered a non-available name (nomen nudum), which has never been valid in terms of the Code, and thus the principle of priority does not apply in this case. Accordingly, as previously stated [19], the species name "annae" does not need to be carried into the proposed species designation. The renaming of "T. annae" as B. vulpes n. sp. should now replace the use of all synonyms for this species, such as "B. microti-like piroplasm", Babesia cf. microti, "B. annae" and "Babesia Spanish dog isolate", thus ending the confusion when referring to this parasite species. Furthermore, in accordance with recent findings on the molecular phylogeny of this and other piroplasmid species, the proposed name clearly distinguishes this parasite from species of the genus Theileria Bettencourt, França & Borges, 1907.
The COX1 has been increasingly applied in molecular phylogenetic studies of piroplasmids [20,46,47]. The phylogenetic analysis using COX1 demonstrated that B. vulpes n. sp. does not segregate into Theileria (s.s.) (Clade V) nor into Babesia (s.s.) (Clade VI), but into a group of Babesia (s.l.) species that is placed into Clade I (B. microti-like parasites or Archaeopiroplasmida; see [11]). Within Clade I, B. vulpes n. sp. is strongly supported as a distinct species of a subclade of Babesia (s.l.) species that has so far been found to exclusively infect carnivores of the families Mustelidae and Canidae. The subclade including B. vulpes n. sp. can be clearly distinguished from the subclades of the B. microti group and B. rodhaini together forming Clade I of B. microti-like piroplasmids ( Fig. 2; [3]). As previously outlined in detail, the 18S rRNA and β-tubulin gene phylogenetic analyzes are in accordance with this result [5]. Babesia vulpes n. sp. is the first species defined within its own subclade group and it is expected that additional species in this group will be described in the future (see also [5]).
Overall, the congruent phylogenetic analysis of the 18S and β-tubulin genes and the COXI protein-sequence encoded by the mitochondrial genome, and the fact that B. vulpes n. sp. has not been shown to infect rodents and humans, distinguishes it as a species from the zoonotic B. microti located in the B. microti group (Fig. 2). Furthermore, B. microti from mice belonging to the zoonotic B. microti group was not found to be infectious to dogs, pigs, chicken and goats in an experimental transmission study, while it was infectious to rats [48].
The mode of transmission and tick vectors of B. vulpes n. sp. have not been determined yet. Although the DNA of this parasite has been detected in several tick species (reviewed in [5]), including I. hexagonus, which was proposed as a vector [49], and D. reticulatus [35], no study to date has provided sufficient proof for the vectorial capacity of any particular tick species, and further research is needed to elucidate this issue.

Conclusions
The fixation of holotype and the morphological description and differentiation of the new species provided here establish the species name B. vulpes n. sp. by fulfilling the ICZN requirements for description of a new species. The name B. vulpes n. sp. should replace all the synonyms that have been used for this parasite including "Theileria annae", "Babesia annae", "B. microti-like piroplasm", Babesia cf. microti and "Babesia Spanish dog isolate".

Additional file
Additional file 1: Figure S1. A neighbor-joining tree of 25 cox1 nucleotide sequences of B. vulpes n. sp. and other piroplasmid species. Clade designations are presented as defined previously [3,50]. After alignment of nucleotide sequences, all positions containing gaps and missing data were eliminated, resulting in a final dataset of 879 positions. The T92 + G model with the shape parameter (G = 0.42) was selected based on Akaike information criterion (AIC) and the neighbor-joining tree inferred [23,24]. The percentages of replicate trees as determined by 1000 replicates of a bootstrap test are shown next to the branches. A Plasmodium falciparum cox1 sequence has been included as the outgroup. The scale-bar represents the evolutionary distance in the units of the number of nucleotide substitutions per site. Gray dots designate Babesia spp. that infect domestic dogs [51].