Proposal to Transfer Some Members of the Genera Haemobartonella And
Total Page:16
File Type:pdf, Size:1020Kb
International Journal of Systematic and Evolutionary Microbiology (2001), 51, 891–899 Printed in Great Britain Proposal to transfer some members of the genera Haemobartonella and Eperythrozoon to the genus Mycoplasma with descriptions of ‘Candidatus Mycoplasma haemofelis’, ‘Candidatus Mycoplasma haemomuris’, ‘Candidatus Mycoplasma haemosuis’ and ‘Candidatus Mycoplasma wenyonii’ 1 Department of Harold Neimark,1 Karl-Erik Johansson,2 Yasuko Rikihisa3 Microbiology and 4 Immunology, Box 44, and Joseph G. Tully † College of Medicine, State University of New York, 450 Clarkson Avenue, Author for correspondence: Harold Neimark. Tel: j1 718 270 1242. Fax: j1 718 270 2656. Brooklyn, NY 11203, USA e-mail: neimah25!hscbklyn.edu 2 Department of Bacteriology, National Veterinary Institute, Cell-wall-less uncultivated parasitic bacteria that attach to the surface of host SE-751 89 Uppsala, erythrocytes currently are classified in the order Rickettsiales, family Sweden Anaplasmataceae, in the genera Haemobartonella and Eperythrozoon. Recently 3 Department of Veterinary 16S rRNA gene sequences have been determined for four of these species: Biosciences, College of Haemobartonella felis and Haemobartonella muris and Eperythrozoon suis and Veterinary Medicine, The Eperythrozoon wenyonii. Phylogenetic analysis of these sequence data shows Ohio State University, that these haemotrophic bacteria are closely related to species in the genus Columbus, OH 43210, USA Mycoplasma (class Mollicutes). These haemotrophic bacteria form a new 4 Mycoplasma Section, phylogenetic cluster within the so-called pneumoniae group of Mycoplasma Frederick Cancer Research and share properties with one another as well as with other members of the and Development Center, pneumoniae group. These studies clearly indicate that the classification of National Institute of Allergy and Infectious these taxa should be changed to reflect their phylogenetic affiliation and the Diseases, Frederick, following is proposed: (i) that Haemobartonella felis and Haemobartonella MD 21707, USA muris should be transferred to the genus Mycoplasma as ‘Candidatus Mycoplasma haemofelis’ and ‘Candidatus Mycoplasma haemomuris’ and (ii) that Eperythrozoon suis and Eperythrozoon wenyonii should be transferred to the genus Mycoplasma as ‘Candidatus Mycoplasma haemosuis’ and ‘Candidatus Mycoplasma wenyonii’. The former Haemobartonella and Eperythrozoon species described here represent a new group of parasitic mycoplasmas that possess a pathogenic capacity previously unrecognized among the mollicutes. These haemotrophic mycoplasmas have been given the trivial name haemoplasmas. These results call into question the affiliation of the remaining officially named species of Haemobartonella and Eperythrozoon which should be considered species of uncertain affiliation pending the resolution of their phylogenetic status. Keywords: 16S rRNA, Rickettsiales, Anaplasmataceae, Mollicutes, phylogeny INTRODUCTION species that infect the erythrocytes of a wide range of animal hosts (Kreier & Ristic, 1984; Kreier et al., The genera Haemobartonella and Eperythrozoon are 1992; Tyzzer, 1942; Weinman, 1944). Microscopic composed of uncultivated haemotrophic bacterial examination of blood smears stained with ................................................................................................................................................. Romanowsky-type stains show the presence of small † Present address: 16400 Black Rock Road, Germantown, MD 20874–3212, coccoid, ring or rod-shaped structures on the surface USA. of red cells (the rod-shaped structures probably are The GenBank accession numbers for the 16S rDNA sequences are given in composed of coccoidal bodies since coccoidal bodies the species descriptions. mainly have been observed by electron microscopy). 01711 # 2001 IUMS 891 H. Neimark and others The organisms stain blue to purple with Wright– assigned either to the genus Haemobartonella or Giemsa stain; acridine orange staining may reveal the Eperythrozoon on the basis of reports (i) that presence of these organisms when they are not demon- Eperythrozoon commonly occur as ‘ring forms’, while strated by Romanowsky-type staining procedures ring forms are rare or absent in Haemobartonella, and (Bobade & Nash, 1987). These parasitic bacteria have (ii) that Eperythrozoon occur with about equal fre- not been cultivated in cell-free media and up to now quency on erythrocytes and free in the plasma while have been maintained only by serial passage in animal Haemobartonella rarely occur free in plasma. We hosts. consider these characters for differentiating species of Many infections caused by Haemobartonella and Haemobartonella and Eperythrozoon to be weak and Eperythrozoon are clinically inapparent but some agree with Kreier & Ristic (1984) that the differen- species are pathogenic and cause overt disease in tiation of species in these genera is arbitrary. The healthy, immunocompetent hosts. Clinical disease currently approved Haemobartonella and Eperythro- usually includes haemolytic anaemia which varies from zoon species and their recognized hosts are: Haemo- mild to severe. Haemobartonella felis produces a bartonella canis (dog), Haemobartonella felis (cat) and severe, often fatal infectious anaemia in cats; mice Haemobartonella muris (rodents), and Eperythrozoon inoculated with Haemobartonella muris frequently are coccoides (mouse), Eperythrozoon ovis (sheep), killed; Eperythrozoon suis produces acute infections in Eperythrozoon suis (pig) Eperythrozoon parvum (pig) pigs and anaemia with mortality in young pigs; and and Eperythrozoon wenyonii (cattle). Eperythrozoon ovis can produce severe anaemia, poor More than 35 years ago Tanaka et al. (1965) showed weight gain and morbidity in lambs. by electron microscopy that Haemobartonella muris These bacteria can persist for years in latently infected and Eperythrozoon coccoides lack cell walls and noted animals without causing clinical disease. Apparently that these wall-less bacteria were indistinguishable they are cleared from the circulation by sequestration from mycoplasmas. Subsequently, all the other Haemo- in the spleen (Maede, 1979). Subjecting latently bartonella and Eperythrozoon species described in infected animals to splenectomy, stress or other pre- Bergey’s manual also were shown by electron mi- disposing factors often result in the appearance of croscopy to lack cell walls. large numbers of infected erythrocytes in the cir- Recently, four of these wall-less haemotrophic bac- culation; overt disease may also result. Haematologic teria, Haemobartonella felis, Haemobartonella muris, detection of inapparent infections may require daily Eperythrozoon suis and Eperythrozoon wenyonii, have sampling for prolonged periods, but molecular di- been subjected to phylogenetic investigation and their agnostic procedures should improve detection (Oberst 16S rRNA gene sequences have been published et al., 1990, 1993; Gawaltny & Oberst, 1994; Berent (Rikihisa et al., 1997; Neimark & Kocan, 1997). The et al., 1998; Messick et al., 1999). Transmission fre- data from these studies demonstrate clearly that these quently involves vectors and Haemobartonella and bacteria are not phylogenetically related to rickettsiae Eperythrozoon species have been shown to be trans- but instead that their closest relatives are species in the mitted by various blood-feeding arthropods, including genus Mycoplasma. These data also affirm that each of ticks, lice, fleas, flies and mosquitoes. these haemotrophic bacteria is a valid species. Thus, Haemobartonella and Eperythrozoon species are cur- these representatives of the genera Haemobartonella rently classified as rickettsiae (order Rickettsiales) and Eperythrozoon are in fact members of a single which they appeared to resemble because of their small genus, Mycoplasma (Rikihisa et al., 1997; Neimark & size and staining properties, their uncultivated status, Kocan, 1997; Johansson et al., 1999). The results do their transmission by arthropod vectors and their not support the current classification of these species haemotrophic character (Kreier & Ristic, 1984). The as rickettsiae and indicate that the classification of latter property seemed to relate these bacteria to these taxa should be changed. We propose the fol- Anaplasma (family Anaplasmataceae) which grow as lowing: (i) that the described Haemobartonella species inclusion bodies within erythrocytes; however Haemo- should be removed from the order Rickettsiales, family bartonella and Eperythrozoon differ from Anaplasma in Anaplasmataceae, and transferred to the genus Myco- that they are wall-less, attach to the surface of red cells plasma as ‘Candidatus Mycoplasma haemofelis’ and do not invade erythrocytes. and ‘Candidatus Mycoplasma haemomuris’ and (ii) that the described Eperythrozoon species should Until now, individual species were named after the be removed from the order Rickettsiales, family host in which they were identified. Host range was Anaplasmataceae, and transferred to the genus Myco- considered a useful character for speciation because plasma as ‘Candidatus Mycoplasma haemosuis’ and most organisms were thought to inhabit a single host. ‘Candidatus Mycoplasma wenyonii’. However, the impression of a single or primary host for each species may need to be modified since the host range of most species has not been fully examined. METHODS Eperythrozoon ovis for example, originally named after its sheep host, also infects goats. Phylogenetic analysis. The 16S rRNA gene sequences were